Investigating the holistic processing of characters during Chinese reading

Investigating the holistic processing of characters during Chinese reading | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Investigating the holistic processing of characters during Chinese reading Lin Li, Yaning Ji, Xinhui Liu, Xinyu Zhao, Kevin B. Paterson This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5778975/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background. Chinese characters are complex visual objects created from a configuration of line strokes. Little is known about the visual input required to support their recognition. However, there is longstanding debate concerning whether characters are recognized holistically, without requiring access to detailed featural information, or recognized analytically by decomposing characters into their constituent parts. Method. The present research investigated this issue using measures of eye movements during sentence reading and employing a visual blurring technique. Characters in sentences were presented as normal or with an exterior or interior region visually degraded. As the exterior region was informative about overall character shape, this manipulation allowed us to investigate effects of preserving or degrading information that might support holistic processing. Additionally, to assess effects on the lexical processing of words, each sentence included one of a pair of interchangeable two-character target words that had either a high or low frequency of written usage. Results. Our findings showed that degrading either exterior or interior character information disrupted normal eye movement behavior, suggesting that both sources of information are required to support normal reading. Effects for the two-character target words additionally showed that this disruption had additive effects on visual and lexical processing. Evidence for a processing advantage for interior degradation (which preserved holistic processing) over exterior degradation was observed in skipping rates for target words. This was consistent with readers being more likely to skip (i.e., direct their gaze beyond a word without fixating it) when holistic information about overall character shape was available. Conclusion. These results contribute to an important debate concerning the role of holistic processes in reading, by revealing that both overall character shape and detailed featural information may be required to support normal reading. However, effects in word-skipping revealed a potential role for holistic processing during parafoveal processing, during which readers attempt to recognize the next word along in a sentence. The findings showed that this proceeded most efficiently when overall character shape information was available, suggesting a role for holistic processing in parafoveal character recognition. eye movements Chinese reading visual blurring holistic processing Figures Figure 1 Figure 2 Introduction Chinese words are composed of box-like logograms known as characters, each of which occupies the same square area of space and are created from a sequence of distinct line strokes [ 1 , 2 ]. These characters vary in complexity, ranging from simple characters containing only a few strokes, such as 一 ( yī 1 meaning "one"), or 人 ( rén , meaning "person"), to highly intricate characters containing many strokes, such as 赢 ( yíng , meaning "win") and 骉 ( biāo , meaning "galloping horses"). More complex characters usually are constructed from smaller units of strokes called radicals, many of which are characters themselves and contribute phonological or semantic information when part of more complex characters. From a language processing perspective, this complexity raises the question of what information is essential for a character to be recognized efficiently. Factors such as stroke patterns, structural composition, and the spatial relationships between components seem likely to play key roles in enabling the rapid and accurate identification of characters. However, there is ongoing debate concerning whether readers process Chinese characters analytically, by breaking them down into more basic components such as radicals and strokes; or, alternatively, whether readers process characters holistically, by recognizing their overall shape without focusing on individual parts [ 3 ]. A long-standing perspective in alphabetic language research is that word recognition operates compositionally - visual features are combined into letters, and words are recognized through their constituent letters (for an overview, see [ 4 ]). This view has shaped the development of influential computational models of word recognition, most notably the interactive-activation model [ 5 ]. According to such models, visual input is processed hierarchically: basic features are first detected, then integrated into abstract letter representations, which subsequently activate lexical-level word representations. A similar hierarchical framework has been proposed by Taft and colleagues to explain Chinese word identification [ 6 , 7 ], drawing inspiration from the interactive-activation model (and for an overview of computational models of Chinese word recognition, see [ 8 ]). According to this approach, Chinese word recognition also follows a compositional process, beginning with the detection of individual strokes, which are combined into sub-character components (such as radicals), then integrated into whole characters, and finally assembled into words. Proponents of this analytic approach point to evidence that character recognition is sensitive to the position and meaning of radicals, suggesting that this information is extracted during character recognition [ 6 , 7 , 9 , 10 , 11 , 12 , 13 ]. By comparison, advocates of holistic processing argue that characters function as the primary unit for orthographic processing, as each corresponds to a single syllable and conveys semantic information comparable to a phoneme in alphabetic scripts like English [ 14 ]. According to this view, characters are recognized as a whole, with their component radicals processed only when the task demands their decomposition [ 15 ]. Another possibility is that both analytic and holistic strategies are used during character recognition. For instance, familiar characters may be recognized rapidly through holistic processing, whereas less familiar characters might require a more analytic approach [ 15 , 16 ]. Moreover, the use of one strategy rather than another may depend on the reader’s level of expertise. Expert readers, who have extensive exposure to Chinese characters, may be more likely to process characters holistically, while novice or beginning readers may rely on an analytic strategy due to their limited familiarity with characters. Note, however, that other research suggests expertise in Chinese character recognition may involve reduced, rather than enhanced, holistic processing due to experienced readers developing a more detailed understanding of character construction [ 17 ]. With the present study, we investigated the use of holistic versus analytic processing during Chinese reading using a visual technique that involved selectively blurring either an interior or exterior character region. This approach was designed to control the availability of character information during reading, allowing us to investigate whether skilled Chinese readers can process characters holistically based on their overall shape. Our approach builds on studies investigating holistic processing in other visual domains, including recognition of faces, chessboards, fingerprints, and X-ray images [ 18 , 19 , 20 , 21 , 22 , 23 , 24 ]. Many such studies use visual blurring or similar experimental techniques to investigate the effects of limiting the availability of detailed visual information on the recognition of visual objects [ 25 , 26 ]. This includes studies that have manipulated spatial frequency content of visual input, based on the evidence that the visual system processes a range of information across different levels of detail, from high spatial frequencies that encode fine detail to lower spatial frequencies that convey more global information about the shape and location of an object [ 27 , 28 ]. Research using these technique to limit the availability of visual information show that participants can identify objects based solely on cues to their overall shape or configuration, without relying on detailed, feature-based cues [ 23 , 29 , 30 ]. Similar approaches have been taken to the study of visual word recognition, based on the assumption that words are a type of visual object. Classic research on the word superiority effect suggests that words can be identified even when insufficient visual information is available to unambiguously identify individual letters [ 31 ]. Studies using blurring techniques additionally suggest that words can be accurately recognized using only coarse-scale cues to their overall shape [ 32 ], though this may be strongest for familiar, high-frequency words. Moreover, this ability to read and recognize words accurately even when only holistic information is available has been show to extend to full sentence reading [ 33 , 34 , 35 , 36 ]. One possibility is that that coarse-scale cues to the overall shape of words may facilitate early stages of word recognition, while more detailed visual processing may be necessary for recognizing individual letters in words. This follows the idea that object perception proceeds over time from an analysis based on global to more detailed visual information [ 37 ], although note that recent findings suggest a different time course for word recognition [ 38 ]. Although research on the need for detailed visual information to recognize characters in logographic scripts like Chinese remains limited, Zhang and Reilly [ 39 ] conducted an internet-based study using a selective blurring technique to show that character recognition remains robust even with incomplete visual information. Their experiment used Gaussian blurring to obscure either the top/bottom or left/right halves of characters while measuring recognition accuracy. Participants achieved overall high character identification rates (88% accuracy), suggesting that full visual detail may not be essential for character recognition. These findings align with the view that skilled Chinese readers may be able to employ holistic visual processing strategies to facilitate character identification when presented with only partial character information. Zhang and Reilly [ 39 ] further suggested that future research may use measures of eye movements to gain insights into the effectiveness of partial information in supporting character recognition during reading. The present research builds upon this prior work by examining the effects of selectively blurring either the exterior or interior components of Chinese characters on eye movements during sentence reading. The rationale for this approach is that the exterior region of Chinese characters is likely to provide information about its overall shape, while the interior region provides more detailed visual information. In addition, we used eye movement measures to investigate the influence of this manipulation on reading behavior. Most investigations of holistic processes in word recognition have examined recognition accuracy [ 39 ] or employed tasks such as the lexical decision task or the Reicher-Wheeler two-alternative forced-choice task [ 31 , 40 ]. These provide insights into the cognitive processes that underpin the recognition of individually presented words. However, they require participants to make explicit decisions about the characteristics of stimuli, which may introduce task demands not present in typical reading situations and so may promote the use of specific processing strategies. In contrast, research using eye movement measures allows participants to engage in normal reading behavior, while the dynamic properties of eye movements have been shown to yield valuable information about the cognitive decision-making processes involved in recognizing words during reading [ 41 , 42 ]. We therefore employed this approach in the present research to investigate whether readers can use holistic information about overall character shape to facilitate processes of word recognition in reading. Previous eye-tracking studies employing blurring techniques in alphabetic scripts demonstrate that skilled readers can process sentences efficiently even when only coarse-scale cues to word identities are available. [ 33 , 34 , 35 , 36 ]. In these experiments, researchers manipulate text quality by blurring either specific target words within a sentence or the entire sentence. Participants read these texts while their eye movements are recorded. Analyses focused on reading speed, comprehension accuracy (assessed via post-sentence questions), and specific eye-tracking metrics (e.g., fixation durations, regressions) to examine how visual degradation impacts reading. Strikingly, participants seem able to read and comprehend sentences remarkably well even when these included degraded words, while showing minimal disruption to normal eye movement behavior. The findings suggest that readers can achieve fluent reading without relying on fine-grained visual details of individual letters. However, the generalizability of these effects to logographic systems like Chinese presently is unclear. We therefore used the same approach to investigate whether Chinese readers can use holistic information about overall character shape to support normal reading in the absence of fine-grained visual cues to character identity. This approach enabled us to investigate effects on both sentence-level eye movement behavior and eye movements for specific target words in sentences. Sentence-level metrics such as total reading time for sentences, and the number and average duration of eye fixations, length of forward-directed eye movements (forward saccade length), and frequency of regressions (backward eye movements) during sentence reading, provide an overall assessment of reading difficulty. We hypothesized that if readers encountered difficulty due to text degradation, they would exhibit slower sentence reading, characterized by an increase in the number and duration of fixations, shorter forward eye movements, and a higher frequency of regressions to reread parts of the sentence, compared to when text was shown normally. Crucially, this approach also allowed us to compare effects of degrading information about the visual detail in characters compared to configurational information about the overall shape of characters. If readers can process text effectively using only holistic information about overall word shape, they might read sentences more quickly, with fewer and shorter fixations, longer forward eye movements, and fewer regressions, when the exterior region of characters is preserved. To isolate effects of our manipulation on processes of word identification during reading, we included pairs of interchangeable two-character target words in each sentence frame. These words differed in terms of their lexical frequency, with one having a higher frequency of written usage compared to the other. Both words were selected to be equally plausible within the sentence context and the high and low frequency words were carefully matched in terms of their visual complexity (e.g., number of character strokes) and their predictability from the preceding text. Substantial research has established that effects of lexical frequency on eye movement behavior provide a robust marker of word recognition difficulty [ 41 , 42 ]. Similarly to effects in alphabetic scripts, studies show that high-frequency words are more likely to be skipped (i.e., not fixated during the initial, first-pass reading of a sentence) or receive shorter fixation durations compared to lower frequency words [ 43 , 44 , 45 ]. These findings also align with evidence from word recognition studies showing that high-frequency words are recognized more quickly than low-frequency words in both alphabetic languages and Chinese [ 45 , 46 , 47 ]. Research manipulating text quality - such as through blurring or contrast adjustments – has additionally shown that low-frequency words are disproportionately affected by text degradation, resulting in larger word frequency effects for degraded compared to normal text [ 48 , 49 ]. In the present study, by manipulating lexical frequency, we aimed to determine whether selectively blurring exterior versus interior regions of characters would differentially influence the word frequency effect. By analyzing eye movement measures sensitive to the initial (first-pass) processing of words or their later word processing, we sought to determine whether these differential effects occurred during early or late stages of word recognition. Word-skipping measures would provide the earliest indication of an effect, as this measure is sensitive to an early stage of lexical processing that takes place while a word is in the parafovea (i.e., lower acuity retinal vision outside of central vision; for a review of parafoveal processing in reading) [ 50 ]. When processing words parafoveally, readers acquire only relatively coarse-scale visual information. However, if this information is sufficient to support processes of word identification, the reader may “skip” the word by making a forward saccade that moves their gaze past this word, terminating with a fixation on a following word. Prior research shows that words that are of higher rather than lower frequency are more likely to be skipped. Accordingly, with the present research, we investigated whether selectively blurring exterior versus interior information in characters might support effects of word frequency on word-skipping rates. Effects of word frequency would also be expected to be observed in fixation times on target words, with shorter fixations for higher compared to the lower frequency words [ 43 , 44 , 45 ]. If word identification is disrupted by selective blurring, we would expect target words in blurred text to be skipped less frequently and to receive longer fixation durations compared to the same words in normal, unblurred text. Additionally, the increased difficulty caused by text degradation should amplify the word frequency effect, manifesting as larger differences in word-skipping rates and fixation durations between high- and low-frequency words in blurred text compared to unblurred text. Crucially, this approach enabled us to assess effects of selective character blurring on word identification processes during reading. In particular, it allowed us to determine whether readers can identify words more efficiently when holistic information about overall word shape is available. If holistic information is more effective for supporting word identification in reading, we might expect higher word-skipping rates, shorter fixation durations, and a smaller lexical frequency effect in these measures when exterior rather than interior character information is preserved. Method The research was approved by the Research Ethics Committee in the Faculty of Psychology at Tianjin Normal University and conducted in accordance with the principles of the Declaration of Helsinki. All participants gave written informed consent. Participants. Participants were 48 undergraduate students aged 18–23 years (M = 21.5 years; 30 female). All were native Chinese readers, screened for normal visual acuity (greater than 20/20 in Snellen values) using a Tumbling E chart [ 51 ]. Stimuli and Design . Stimuli consisted of 120 pairs of Chinese sentence frames, each containing a two-character target word with either high or low lexical frequency, based on the SUBTLEX-CH database [ 52 ]. Sentences were 20 to 25 characters long (M = 22.3 characters), including the target word, which was always positioned near the center of each sentence. These stimuli were based on the stimulus used by Wang et al. [ 53 ] to examine adult age differences in effects of text degradation on the word frequency effect in eye movements during Chinese reading. The high- and low-frequency target words were matched on key characteristics. First, the frequency of the individual characters (both first and second) within the target words was equivalent across the two frequency groups. Second, the visual complexity of the words was matched, measured by the number of strokes in the individual characters and the word as a whole. Third, cloze predictability tests conducted with 20 participants (who did not take part in the experiment) confirmed that high- and low-frequency words were equally unpredictable as sentence completions. This involved presenting sentence fragments truncated immediately before the target word and collecting written continuations, which showed that both word types were produced with similarly low frequency. Finally, the naturalness of the sentences was assessed separately by ten participants, who also did not participate in the main experiment. Using a 7-point scale ranging from highly natural to highly unnatural, participants rated all sentences as highly natural overall (M = 5.92, SD = 0.32), with no significant difference in naturalness ratings between sentences containing high- and low-frequency target words, t (119) = 1.52, p = 0.13. This careful matching of lexical, visual, and contextual properties ensured that the target words and sentences were comparable across all critical variables. The sentence stimuli were presented under three display conditions: normal (unmodified) or with blurring applied to either an interior or exterior region of each character. To implement these conditions, a 28×28 pixel grid was superimposed on each character. This grid enabled the definition of two distinct regions: an interior region at the center of the character, consisting of a square containing 400 cells, and an exterior region encompassing the character's surrounding envelope, containing 384 cells. For the blurring conditions, either the interior or exterior area was blurred using a Gaussian blur applied to each 3-pixel section using Adobe Photoshop. Figure 1 provides an illustration of this procedure, showing how it was applied to a sample character to create the interior and exterior blur conditions. Three versions of each sentence stimulus were created, and these were distributed across six lists. Each list included one version of every sentence, containing either a high- or low-frequency target word displayed in one of three conditions: normal, interior blur, or exterior blur (example displays are shown in Fig. 2 ). Sentences were allocated to the lists using a Latin square to ensure an equal number of stimuli from each condition in each list, with every sentence appearing in all conditions across the six lists. Participants were pseudo-randomly assigned to one of the six lists, with an equal number of participants assigned to each. Each participant viewed all 120 stimuli in a randomized order, with sentences presented either normally or with interior or exterior blur. The experiment began with four practice sentences to familiarize participants with the task. The study used a within-participants design, with two factors: lexical frequency (high, low) and display type (normal, interior blur, exterior blur). This ensured each participant was exposed to all display conditions and target word frequencies across the experimental stimuli. Apparatus and Procedure . An EyeLink 1000 Plus eye-tracker (SR Research, Canada) was used to record right-eye gaze location during binocular viewing, with a sampling rate of 1,000 Hz. This system provides high spatial resolution (< 0.01° RMS) and precise temporal resolution. Sentences were presented in Song font as black text on a gray background. At a viewing distance of 60 cm, each character subtended approximately 1° of visual angle horizontally, corresponding to a standard character size for normal reading [ 54 ]. Participants completed the experiment individually and were instructed to read the sentences normally and for comprehension. At the start of the experiment, a three-point horizontal calibration procedure was performed along the same line as the sentence presentations. This ensured spatial accuracy of 0.30° or better for all participants, corresponding to precision within half a character. Calibration accuracy was checked before each trial, and the eye-tracker was recalibrated as needed to maintain this high level of precision. At the beginning of each trial, a fixation square, equal in size to one character, appeared on the left side of the screen. Once participants fixated on the square, the sentence was displayed, with its first character replacing the square. Participants pressed a response key when they finished reading each sentence. On 25% of the trials, a yes/no comprehension question followed the sentence. Participants responded to these questions by pressing one of two designated keys, and their responses were recorded. The entire experiment took approximately 40 minutes to complete for each participant. Data analysis . Data were analyzed using linear mixed-effects models (LMEs) [ 55 ] for continuous variables and generalized linear mixed-effects models for binomial variables, implemented using the lme4 package (Version 1.1–21) [ 56 ] in R [ 57 ]. For all analyses, models incorporated a maximum random effects structure where possible [ 58 ]. Both sentence-level and word-level analyses were conducted. In sentence-level analyses, display type was treated as a fixed factor. For target word analyses, fixed factors included lexical frequency, display type, and their interaction, with participants and stimuli included as crossed random effects. If models failed to converge, the random effects structure was simplified by sequentially reducing the structure for stimuli, first removing random effect correlations, then random slopes. Effects were analyzed on log-transformed data, with results reported alongside untransformed means. Contrasts for main effects and interactions were defined using sliding contrasts ( contr.sdif function) from the MASS package (Version 7.3–47) [ 59 ]. Malsburg and Angele [ 60 ] argue that the use of multiple correlated dependent measures in eye movement research may increase Type 1 effects. Accordingly, to control for multiple comparisons, we applied Tukey's HSD correction. Fixation time effects on log-transformed data are reported, and significance was determined using the standard threshold of t/z values > 1.96. Results Accuracy answering comprehension questions was high across participants (> 90%), with no significant differences observed between sentences displayed in a normal format and those with interior or exterior blur (Normal: M = 95.7%, SD = 6.6%; Interior Blur: M = 96.1%, SD = 7.3%; Exterior Blur: M = 94.4%, SD = 7.7%; F (2, 94) = 0.76, p = .47, partial η² = .016). These results indicate that sentences remained comprehensible even when individual characters were partially obscured. Prior to conducting statistical analyses, the eye movement data were pre-procesed following standard procedures. Fixations shorter than 80 ms and longer than 1200 ms were excluded, accounting for 5% of all fixations. Trials with track loss were also removed (5 trials, less than 1% of the data), along with trials involving sentences that received fewer than six fixations (64 trials, less than 1% of the data). Note that an eye-movement recording comprising fewer than six fixations may indicate track-loss, where the eye-tracker failed to record a participants’ eye movements. Alternatively, given the perceptual span for Chinese (approximately 5 characters, [ 60 ]), this might indicate that a participant was not reading normally and for comprehension. Statistical analyses were conducted on both sentence-level eye movement data and eye movement data for the target words embedded in each sentence. Sentence-level analyses employed standard measures of reading behavior. Sentence reading time (SRT) was recorded as the duration from the onset of the sentence display to when the participant pressed a response key, indicating they had finished reading. The number of fixations (NF) referred to the total count of fixations made during sentence reading. Average fixation duration (AFD) represented the mean duration, in milliseconds, of all fixations throughout the reading of a sentence. Average forward saccade length (AFSL) measured the mean distance, in character spaces, of progressive eye movements. Finally, the number of regressions (NR) indicated the average number of backward eye movements participants made while reading a sentence. Together, these measures provided a detailed overview of the temporal and spatial dynamics of participants' reading behavior. For the word-level analyses, we report measures informative about the first-pass and later processing of words. First-pass reading describes the initial processing of a word before a fixation to the right or a regression from the word of interest. We report several key first-pass reading measures. Word-skipping (SKIP) refers to the probability that a word is not fixated during first-pass reading. First-fixation duration (FFD) is the length of the initial fixation on a word during first-pass reading. Single-fixation duration (SFD) is the duration of the first fixation on a word that receives only one fixation during first-pass reading. Gaze duration (GD) is the total time spent on all first-pass fixations on a word. In addition, we examined measures sensitive to later stages of word processing. Total reading time (TRT) reflects the sum of all fixations on a word, capturing both its initial and later processing. Finally, regressions-in probability (RI) is the likelihood of a regression back to an earlier word, providing insight into reanalysis or integration processes. In eye movement research, it is common practice to report multiple partially correlated dependent measures to provide a rich understanding of how variables of interest affect reading behavior. However, this approach carries the risk of inflating Type I errors. These concerns are examined in detail by von der Malsburg and Angele [ 61 ], who outline several approaches to minimize this risk. These include focusing hypothesis-driven analyses on measures most likely to show the expected effects, applying corrections for multiple comparisons across dependent variables (though such corrections may not be appropriate for partially correlated measures), or adopting the heuristic that an effect should be considered reliable only when observed in at least two dependent measures. We note that all effects of interest in the present research meet this reliability criterion, appearing in at least two dependent measures. Moreover, several key effects were observed across an even broader range of measures, further strengthening confidence in their robustness. Table 2 presents the mean sentence-level eye movement measures, and Table 3 provides a summary of the statistical analyses. Significant main effects of display type were observed across all measures. Compared to normal displays, reading times were longer for both interior and exterior blur conditions. These conditions were also associated with an increased number of fixations, longer average fixation durations, shorter forward saccades, and a greater number of regressions. Together, the findings indicate that interior and exterior blur displays both significantly disrupted readers' eye movement patterns. Average fixation durations were notably longer for sentences displayed with interior blur compared to those with exterior blur. However, no other significant differences emerged between the two blur conditions, suggesting that interior and exterior blur had broadly comparable effects on most eye movement measures. Target word analyses . Table 4 presents target-word level eye movement measures, and Table 5 provides a summary of the statistical analyses. Standard lexical frequency effects were obtained, with lower skipping rates, longer fixation times and more regressions back to low- compared to high-frequency target words. This fixation time effect was evident in both first-pass and later fixation time measures, indicating that word frequency influenced the initial stages of word identification and had a sustained impact on the processing of target words. These results align with previous studies investigating word frequency effects in Chinese reading [ 49 , 53 ]. Effects of display type were also observed, revealing a differential impact of interior and exterior blur the processing of words. Word-skipping rates were lower for words presented with exterior blur compared to normal displays, while interior blur produced an intermediate word-skipping rate that did not differ significantly from either the exterior blur or normal displays. During the very first fixations on words, as reflected in first-fixation duration (FFD) and single-fixation duration (SFD), fixation times were longer for interior compared to exterior blur. No significant differences were observed in fixation times during later stages of word processing between interior and exterior blur conditions, as measured by gaze duration (GD) and total reading time (TRT). However, both blur conditions produced longer fixation times in these measures compared to words show normally. This overall pattern of findings therefore point to an early, short-lived advantage for exterior blur over interior blur, but also emphasize that neither blur condition provides adequate visual information to support the normal processing of words. This underscores the disruptive effects of both interior and exterior blur on the processing of words during reading. These results might reflect differences in either the visual or lexical processing of words (or both). Accordingly, we also examined the interaction between display condition and lexical frequency to understand whether selective blurring affected lexical processing. An interaction between display condition and lexical frequency was observed in target word-skipping rates for normal versus exterior blur displays, but no significant interactions were found between interior blur and normal displays or between exterior and interior blur displays. Further analysis focused on the magnitude of word frequency effects on word-skipping rates across these display conditions. For normal displays, a significant word frequency effect was found, with a 7% effect (β = 0.39, SE = 0.11, z = 3.51, p .05). For interior blur displays, the word frequency effect was intermediate, at 3% (β = 0.22, SE = 0.11, z = 1.94, p = 0.05). These findings suggest that blurring reduced the availability of parafoveal information about word identity necessary for efficient word-skipping. Compared to normal displays, blurred displays disrupted readers' ability to skip words effectively. Among the blur conditions, exterior blur caused the greatest disruption to word-skipping compared with displays in which words were shown normally, suggesting that holistic character information plays a critical role in supporting parafoveal processes of word identification. This highlights the particular importance of exterior character information in supporting normal reading processes. No interactions between display conditions and word frequency were observed in measures sensitive to first-pass fixational processing of target words (i.e., FFD, SFD, and GD). However, interactions between these variables were evident in later measures. For total reading time (TRT), an interaction was found between exterior blur and normal displays, but no interactions were observed between interior blur and normal displays or between exterior and interior blur displays. Similarly, for regressions-in probability (RI), an interaction was observed for exterior blur versus normal displays, but not for interior blur versus normal displays or exterior versus interior blur displays. Further analysis revealed no significant word frequency effects in TRT or RI under normal display conditions (TRT: word frequency effect = 3 ms, β = 0.04, SE = 0.03, t = 1.33, p > .05; RI: word frequency effect = 0.8%, β = 0.06, SE = 0.15, z = 0.41, p > .05). By contrast, word frequency effects were present under both blur conditions, with effects being numerically larger in the exterior blur condition (TRT: word frequency effect = 51 ms, β = 0.11, SE = 0.03, t = 4.36, p < .001; RI: word frequency effect = 6.0%, β = 0.47, SE = 0.14, z = 3.29, p < .01) compared to the interior blur condition (TRT: word frequency effect = 46 ms, β = 0.11, SE = 0.03, t = 3.99, p < .001; RI: word frequency effect = 3.0%, β = 0.27, SE = 0.15, z = 1.83, p = 0.07). The lack of a word frequency effect in later processing measures (i.e., total reading time) for normal displays contrasts was surprising. Some prior investigations of word frequency effects in Chinese show robust effects in both early and late processing measures [ 44 ]. However, Wang et al. [ 53 ] also showed a non-significant effect of word frequency in total reading times for words, using the same stimulus set as the present study and young adult participants drawn from a similar population. This difference in the time course of word frequency effects may reflect variation in the stimuli and manipulation of word frequency across different experiments. For instance, while the present experiment carefully controlled for other characteristics that might influence word identification, including character frequency and stroke complexity, other studies either do not control these factors or have varied them systematically [ 43 , 44 , 62 ]. Moreover, many studies of word frequency effect in Chinese reading have focused on measures of early word processing and do not report effects in later measures such as total reading time [ 43 , 62 ]. Accordingly, while effects of word frequency appear to be observed robustly in both early and later processing measures in alphabetic reading [ 63 ], it is unclear if the same patterns of effects are observed consistently in character-based scripts like Chinese, especially when other factors affecting word identification are controlled. Indeed, the lack of word frequency effects in later processing measures for normal displays in the present research suggests that lexical frequency primarily influences the early stages of word identification, which are typically completed during first-pass reading for text displayed normally. By contrast, a larger word frequency effect for blurred displays in these later measures suggests that the process of word identification had a longer time-course when word information was partially occluded. Moreover, the stronger effects for interior blur compared to exterior blur suggest that information from the interior regions of words plays a more critical role in supporting word identification during reading. Discussion The present research employed a selective visual blurring technique alongside eye movement measurements to explore the role of holistic processing in Chinese reading. Participants were presented with sentences containing a specific target word, which varied in written frequency—either high or low. These sentences were displayed in three conditions; presented either as normal or with either an interior or an exterior region of their characters blurred. Crucially, exterior blurring degraded visual information about the overall shape of the characters while interior blurring preserved this holistic information. Perhaps the most striking finding was that participants could read and comprehend sentences relatively well even when only partial character information was available. Comprehension accuracy, assessed by presenting comprehension questions following 25% of sentences, was high (> 90% correct responses) with no difference across display conditions. This suggested that sentences were easily understood even when interior or exterior character information was degraded. Degrading this information nevertheless disrupted normal reading processes, with participants taking longer to read the blurred sentences, by making more and longer fixations, shorter forward-directed eye movements and more regressions for these sentences compared to the unblurred sentences. Crucially, there was little indication of a differential effect of interior versus exterior blurring in sentence-level eye movement measures. While the sentences appeared to be read a little more slowly following interior blurring, this effect was reliable only in average fixations durations, suggesting that exterior and interior blurring had comparable effects on sentence-level eye movement behavior. Overall, the pattern of sentence-level effects show that sentences can be read and understood well when either only interior or exterior character information is available, but that reading may be most efficient when both sources were present. An analysis of specific target words in each sentence enabled us to take a closer look at the effect of selective blurring on the processing of individual words during reading. These more focused analyses showed that blurring either interior or exterior character information slowed the processing of words, with readers making longer fixations on target words in blurred than unblurred sentences. This replicated our sentence-level findings showing that the blurred text was read more slowly. The word-level eye movement measures we used allowed us to separate effects occurring during the initial, first-pass processing of words from those affecting later processing. Crucially, fixation time measures sensitive to first-pass processing revealed a differentiation in effects of exterior versus interior blurring, with longer first-pass fixations on words following exterior blurring. This suggests that degrading information about the overall character shape caused short-lived disruption to an early stage of the word’s processing. Readers may therefore have used holistic processing about overall character shape during this early processing stage. Note that this finding is in line with other evidence that holistic processing facilitates an early stage of visual processing [ 37 ]. The absence of a similar effect in fixation time measures sensitive to the later processing of words additionally suggests that this influence of holistic processing is short-lived. The short-lived nature of this effect may also explain why it is observed only in measures sensitive to early word-level processing and not sentence-level measures. By comparison, the relative cost for either exterior or interior blurring compared to unblurred text was sustained across measures of both early and late fixational processing, providing further evidence that reading was most efficient when both sources of character information were available. The word-level analyses also allowed us to examine effects of visual blurring on the word frequency effect in reading. This describes the processing advantage for words that have a higher frequency of written usage and so are more familiar to readers [ 43 , 62 ]. In the present experiment, we observed standard word frequency effects in word-skipping rates and fixation times on the target words. Skipping rates were lower and fixation times were longer for words of low than high frequency, with readers also making more regressions back to the lower frequency words. Effects of word frequency were observed in fixation time measures sensitive to both early and late stages of a word’s processing, revealing that word frequency influenced an early stage of processing and has a sustained influence on the processing of words. The pattern of word frequency effects we obtained were in line with those observed in previous research [ 43 , 44 , 45 ], indicating that the present experiment produced typical effects of word frequency in eye movement behavior. There was no indication that blurring influenced the word frequency effect in fixation times for words, although we did observe such an influence in word-skipping. The lack of a modulating effect in fixation times is consistent with blurring primarily influencing the visual processing of words without impacting on subsequent processes of lexical access. This finding is consistent with findings showing that effects of visual degradation and word frequency can be additive [ 48 ], suggesting they involve distinct stages of processing. One possibility is that the partial information provided by selectively blurring interior or exterior character regions disrupted visual processing without affecting word recognition. This pattern of effects also is consistent with readers being able to use both interior and exterior character information to support word recognition. By comparison the influence of visual blurring on the word frequency effect in word-skipping rates was potentially informative about the influence of holistic influences on the parafoveal processing of words. Parafoveal processing describes the pre-processing of upcoming linguistic information to the right of where the reader currently is fixating, which influences the likelihood of a word being skipped during reading [ 50 ]. The main effect of word frequency we observed in this measure was due to readers being able to parafoveally process higher frequency words more efficiently and so skip these words more often, replicating prior research findings [ 44 ]. We also observed an interactive effect of character blurring and word frequency, such that the effect of word frequency on skipping rates varied across display conditions. Specifically, a 7% word-frequency effect for unblurred text diminished to 3% following interior character blurring and was eliminated (producing only a 1% effect) following exterior character blurring. The indication is that selectively degrading interior and exterior parts of characters made the parafoveal processing of word identities more difficulty, especially when this disrupted overall character shape information. Taken together, the present findings suggest both overall character shape and more detailed featural information is required to support normal reading processes. This finding is broadly in line with other evidence that holistic cues as well as detailed visual information may contribute to effective word recognition [ 39 ], in addition to the recognition of other visual objects such as faces [ 23 , 29 , 30 ]. The present findings are important in showing that such effects are observed in measures that do not require participants to perform a secondary task (e.g., lexical decision) which may introduce task demands not present in normal reading situations, leading them to employ specific processing strategies. Our findings additionally suggest that holistic information may facilitate the early processing of words. We have reported two sources of evidence for this. First, we observed a cost for exterior versus interior blurring in fixations times sensitive to early word processing, suggesting this early processing was facilitated when exterior character information was available. Second, we found that the word frequency effect in word-skipping rates was eliminated by exterior blurring (and reduced by interior blurring), suggesting that character shape information is required to support the parafoveal processing of words. As the earliest stage of word recognition occurs during its parafoveal processing, when readers acquire visual information about a word prior to it being fixated, it seems that holistic character information may facilitate this early stage of word identification. Limitations While the present research contributes to our understanding of holistic processing in character recognition, we also acknowledge several limitations. First, our approach to dividing characters into interior and exterior regions was based on approximately equal spatial areas. This method, while systematic, failed to accommodate individual differences in character structure. As a result, the interior blurring manipulation may have impinged on exterior regions for some characters, particularly those with ambiguous boundaries. Second, the inherent variation in character construction meant our defined regions contained unequal amounts of visual information. Interior regions typically encompassed more stroke details, while exterior regions consisted largely of surrounding white space. This imbalance likely affected how perceptual information was disrupted across different characters. Third, as noted by a reviewer, the complexity and variability of Chinese characters make it inherently difficult to objectively define a character's overall shape. This fundamental challenge limits our ability to systematically manipulate perceptual features. Finally, while our blurring technique successfully restricted access to fine visual detail, it may have also introduced confounding visual effects. The manipulation potentially created artificial boundaries and disrupted stroke continuity, which could have interfered with holistic processing mechanisms rather than simply limiting local feature analysis. We note that many of these limitations also apply to other studies investigating how blurring specific character regions affects recognition. For instance, Zhang and Reilly [ 39 ] demonstrated that character recognition remains relatively accurate even when blurring obscures either the upper/lower or left/right halves of characters. These findings suggest that while our blurring approach has limitations, it aligns with broader methodological challenges in this research area. Nevertheless, given these limitations, future research could explore alternative approaches to better understand holistic processing in character recognition. One promising direction might involve systematically manipulating the spatial frequency content of characters, inspired by similar approaches in studies with alphabetic scripts [ 32 , 33 , 34 , 35 , 36 , 64 ]. This could provide further insights while avoiding some of the artifacts introduced by blurring techniques. The rationale for this method stems from well-established findings that the visual system processes information across multiple spatial frequency channels, from high frequencies encoding fine visual details to low frequencies conveying global shape information [ 27 , 28 ]. As discussed in our Introduction, substantial evidence demonstrates that object recognition often can be achieved using only coarse-scale visual information. This suggests that fine visual details, including those required to identify individual character components, may not be necessary for successful word recognition. By carefully controlling the spatial frequency spectrum of character stimuli, researchers could more precisely investigate the relative contributions of global versus local information in character recognition, including how these contributions might differ across character types or proficiency levels [ 64 , 65 ]. Clearly, further work is required to replicate and extend the present findings. Within eye movement research, the role of exterior character information during parafoveal processing also might be investigated further using the boundary paradigm [ 66 ]. In this paradigm, an invisible boundary is placed directly in front of a word in a sentence. Prior to the reader’s gaze crossing this boundary, the word is replaced by a preview stimulus. Then, as soon as the reader’s gaze crosses the boundary, this preview is quickly restored to the original word, so that the experimenter can assess the influence of different previews on subsequent processing. This technique could be used to further investigate whether character shape information aids the parafoveal processing of word identities. This could be investigated, for example, by comparing effects of blurring interior versus exterior regions of character previews, similarly to the approach used in the current experiment. Based on the present findings, we might predict a larger preview benefit for interior versus exterior character blurring as the former preserves contour information about a character’s overall shape. Such findings might more directly reveal an influence of holistic cues on parafoveal processing. Further research may also take account of the internal structure of Chinese characters to investigate what information contributes to word recognition. As already noted, Chinese characters are constructed from substructures of strokes organized into radicals, which can also function as words if presented alone, and which convey phonological and semantic information when part of more complex characters. Theoretical models of Chinese word recognition [ 5 ], inspired by analytic models of word recognition in alphabetic scripts [ 6 , 7 ], assume that character recognition follows a compositional process, beginning with the detection of individual strokes, which combine into sub-character components, such as radicals, before being integrated into whole characters and finally assembled into words. The present research suggests that both global character shape and detailed featural information may contribute to efficient word processing, yet the literature reveals considerable complexity in character recognition. On one hand, substantial evidence demonstrates that characters are often processed compositionally through initial identification of their constituent radicals [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 ]. On the other hand, findings indicate that radical recognition may not occur automatically during natural reading [ 15 ], pointing to flexibility in compositional processing that varies with task demands. This overall pattern of results may support a dual-pathway model of character recognition. Beyond the well-established compositional route, there may be a direct pathway for holistic character identification that operates independently of radical decomposition. The development of this holistic route seems likely to be closely tied to reading expertise and experience. Specifically, high-frequency characters, which are more familiar to readers, may show stronger evidence of holistic recognition. Furthermore, comparisons across reader groups may reveal that skilled native readers depend less on radical information than both less-proficient readers and those acquiring Chinese as a second language. If correct, this may indicate that reading experience shapes the relative engagement of these different processing routes. Indeed, one possibility is that the balance between holistic and featural processing during Chinese reading dynamically adjusts according to multiple factors including character properties, reader proficiency, and situational demands. In sum, the present research provides novel insight into the visual basis of word recognition in Chinese, by showing that both overall character shape and detailed featural information are used during reading. Moreover, our findings suggest that holistic information about overall character shape might play a privileged role in the early processing of words, especially during parafoveal processing. While further work is needed to substantiate and extend these findings, it seems likely that such work may will contribute to our understanding of the relative use of holistic and featural processing in Chinese reading. Abbreviations SKIP skipping rate FFD first fixation duration SFD single fixation duration GD gaze duration TRT total reading time RI regressions in probability Declarations Ethics approval and consent to participate The research was approved by the Research Ethics Committee in the Faculty of Psychology at Tianjin Normal University and conducted in accordance with the principles of the Declaration of Helsinki. All participants gave written informed consent. Consent for publication All participants gave informed written consent that their data could be used for publication. Availability of data and materials Data sets and analytic code for analyses in R for the present experiment is available via the University of Leicester Figshare repository at https://figshare.com/s/d09c2b3cefc44a65f21c Competing interests The authors have no competing interests. Funding The research was supported by Tianjin Social Science Planning Grant [TJXL24-002] from Tianjin local government. Authors' contributions Lin Li played a lead role in conceptualization, data curation, formal analysis, funding acquisition, methodology, project administration, resources, and supervision and an equal role in investigation, writing–original draft, and writing–review and editing. Yaning Ji played a lead role in investigation and a supporting role in data curation. Xinhui Liu played an equal role in data curation and investigation. Xinyu Zhao played an equal role in conceptualization. Kevin B. Paterson played an equal role in conceptualization, writing–original draft, and writing–review and editing. Acknowledgement Not applicable. References Hoosain R. (1991). Psycholinguistic implications for Linguistic relativity: A case study of Chinese. Psychology Press . Hoosain R. Psychological reality of the word in Chinese. Adv Psychol. 1992;90:111–30. Hsiao JH, Cottrell GW. Not all visual expertise is holistic, but it may be leftist. Psychol Sci. 2009b;20(4):455–63. Grainger J. Orthographic processing: A ‘mid-level’ vision of reading. Q J Experimental Psychol. 2018;71(2):335–59. 10.1080/17470218.2017.1314515 . McClelland JL, Rumelhart DE. An interactive activation model of context effects in letter perception: Part 1. An account of basic findings. J Exp Psychol Hum Percept Perform. 1981;7(3):191–217. 10.1037/0096-1523.7.3.191 . Taft M, Zhu X. Submorphemic processing in reading Chinese. J Experimental Psychol Learn Memory Cognition. 1997;23(3):761–75. Taft M, Zhu X, Peng D. Positional specificity of radicals in Chinese character recognition. J Mem Lang. 1999;40:498–519. Reichle ED, Yu L. Models of Chinese Reading: Review and Analysis. Cogn Sci. 2018;42:1154–65. https://doi.org/10.1111/cogs.12564 . Chen Y, Yeh S. Binding radicals in Chinese character recognition: Evidence from repetition blindness. J Mem Lang. 2015;78:47–63. Fang SP, Horng RY, Tzeng OJL. Consistency effects in Chinese character and pseudo-character naming tasks. In: Kao HSR, Hoosain R, editors. Linguistics, psychology, and the Chinese language. Hong Kong: Hong Kong University; 1986. pp. 11–21. Feldman LB, Siok WW. Semantic radicals contribute to the visual identification of Chinese characters. J Mem Lang. 1999;40(4):559–76. Leck KJ, Weekes BS, Chen MJ. Visual and phonological pathways to the lexicon: Evidence from Chinese readers. Mem Cognit. 1995;23(4):468–76. Yeh SL, Li JL. Role of structure and radical in judgments of visual similarity of Chinese characters. J Exp Psychol Hum Percept Perform. 2002;28:933–47. Chen SC, Liu IM. Functional orthographic units in Chinese character recognition. Acta Physiol Sinica. 2000;32:235–68. Tao L, Healy AF. The unitization effect in reading Chinese and English text. Sci Stud Read. 2002;6:167–97. Healy AF. Letter detection: A window to unitization and other cognitive processes in reading text. Psychon Bull Rev. 1994;1(3):333–44. Tso RV, Au TK, Hsiao JH. (2022). Non-monotonic developmental trend of holistic processing in visual expertise: the case of Chinese character recognition. Cogn Res Principles Implications, 7 (1). Richler JJ, Cheung OS, Gauthier I. Holistic processing predicts face recognition. Psychol Sci. 2011;22(4):464–71. https://doi.org/10.1177/0956797611401753 . Vogelsang MD, Palmeri TJ, Busey TA. Holistic processing of fingerprints by expert forensic examiners. Cogn research: principles implications. 2017;2(1):15. https://doi.org/10.1186/s41235-017-0051-x . Sheridan H, Reingold EM. The Holistic Processing Account of Visual Expertise in Medical Image Perception: A Review. Front Psychol. 2017;8:1620. https://doi.org/10.3389/fpsyg.2017.01620 . Reingold EM, Sheridan H. Eye movements and visual expertise in chess and medicine. In: Liversedge SP, Gilchrist ID, Everling S, editors. The Oxford handbook of eye movements. Oxford University Press; 2011. pp. 523–50. Richler JJ, Palmeri TJ, Gauthier I. Meanings, Mechanisms, and Measures of Holistic Processing. Front Psychol. 2012;3:553. Balas B, Gable J, Pearson H. (2018). The Effects of Blur and Inversion on the Recognition of Ambient Face Images. Perception , 030100661881258. 10.1177/0301006618812581 Richler JJ, Gauthier I. A meta-analysis and review of holistic face processing. Psychol Bull. 2014;140(5):1281–302. Ruiz-Soler M, Beltran FS. Face perception: An integrative review of the role of spatial frequencies. Psychol Res. 2006;70:273–92. https://doi.org/10.1007/s00426-005-0215-z . Oliva A, Schyns P. Coarse blobs or fine edges? Evidence that information diagnosticity changes the perception of complex visual stimuli. Cogn Psychol. 1997;34:72–107. Blakemore C, Campbell FW. Adaptation to spatial stimuli. J Physiol. 1969;200(1):11–3. Pantle A, Sekuler R. Size-detecting mechanisms in human vision. Sci (New York N Y). 1968;162(3858):1146–8. https://doi.org/10.1126/science.162.3858.1146-a . Collishaw SM, Hole GJ. Featural and configurational processes in the recognition of faces of different familiarity. Perception. 2000;29(8):893–909. https://doi.org/10.1068/p2949 . Sandford A, Sarker T, Bernier T. Effects of geometric distortions, Gaussian blur, and contrast negation on recognition of familiar faces. Visual Cognition. 2017;26(3):207–22. https://doi.org/10.1080/13506285.2017.1407853 . Reicher GM. Perceptual recognition as a function of meaningfulness of stimulus material. J Exp Psychol. 1968;81(2):275–80. Patching GR, Jordan TR. Assessing the role of different spatial frequencies in word perception by good and poor readers. Mem Cognit. 2005a;33:961–71. Jordan TR, McGowan VA, Paterson KB. Reading with a filtered fovea: The influence of visual quality at the point of fixation during reading. Psychon Bull Rev. 2012;19(6):1078–84. Jordan TR, McGowan VA, Kurtev S, Paterson KB. A further look at postview effects in reading: An eye-movements study of influences from the left of fixation. J Experimental Psychol Learn Memory Cognition. 2016;42(2):296–307. Paterson KB, McGowan VA, Jordan TR. Filtered text reveals adult age differences in reading: Evidence from eye movements. Psychol Aging. 2013a;28(2):352–64. Paterson KB, McGowan VA, Jordan TR. Effects of adult aging on reading filtered text: evidence from eye movements. PeerJ. 2013b;1:e63. Schyns PG, Oliva A. (1994). From blobs to boundary edges: Evidence for time- and spatial-scale-dependent scene recognition. Psychol. Sci.5 (4), 195–200. https://doi.org/10.1111/j.1467-9280.1994.tb00500.x (1994). Bertrand Pilon C, Arguin M. The processing of spatial frequencies through time in visual word recognition. Sci Rep. 2024;14:6628. https://doi.org/10.1038/s41598-024-57219-3 . Zhang Q, Reilly R. What regions of Chinese characters are crucial for recognition? A web-based study. J Chin Writ Syst. 2020;4(4):297–311. Wheeler DD. Processes in word recognition. Cogn Psychol. 1970;1(1):59–85. Rayner K. Eye movements in reading and information processing: 20 years of research. Psychol Bull. 1998;124(3):372–422. Rayner K. The 35th Sir Frederick Bartlett Lecture: Eye movements and attention in reading, scene perception, and visual search. Q J Experimental Psychol. 2009;62(8):1457–506. Li X, Bicknell K, Liu P, Wei W, Rayner K. Reading is fundamentally similar across disparate writing systems: A systematic characterization of how words and characters influence eye movements in Chinese reading. J Experimental Psychol Gen. 2014;143(2):895–913. Yan G, Tian H, Bai X, Rayner K. The effect of word and character frequency on the eye movements of Chinese readers. Br J Psychol. 2006;97(2):259–68. Ma G, Zhuang X. Distributional analyses of word frequency effects in Chinese sentence reading and lexical decision tasks. J Res Reading. 2018;41:S183–96. https://doi.org/10.1111/1467-9817.12259 . Morrison CM, Ellis AW. Roles of word frequency and age of acquisition in word naming and lexical decision. J Experimental Psychol Learn Memory Cognition. 1995;21(1):116–33. Taft M. Recognition of affixed words and the word frequency effect. Mem Cognit. 1979;7(4):263–72. Becker C, Killion A, Thomas,H. Interaction of visual and cognitive effects in word recognition. J Exp Psychol Hum Percept Perform. 1977;3(3):389–401. Wang J, Li L, Li S, Xie F, Chang M, Paterson KB, et al. Adult Age Differences in Eye Movements During Reading: The Evidence From Chinese. Journals Gerontol Ser B: Psychol Sci Social Sci. 2018;73(4):584–93. Schotter ER, Angele B, Rayner K. Parafoveal Processing in Reading. Atten Percept Psychophysics. 2012;74:5–35. http://dx.doi.org/10.3758/s13414-011-0219-2 . Taylor HR. Applying new design principles to the construction of an illiterate e chart. Optom Vis Sci. 1978;55(5):348–51. Cai Q, Brysbaert M. (2010). SUBTLEX-CH: Chinese word and character frequencies based on film subtitles. PLoS ONE, 5(6), e10729. Wang J, Li L, Li S, Xie F, Liversedge SP, Paterson KB. Effects of aging and text-stimulus quality on the word-frequency effect during Chinese reading. Psychol Aging. 2018;33(4):693–712. Xu M, Jordan TR. Assessing effects of viewing distance on normal Chinese reading: Some methodological and theoretical implications. Behav Res Methods. 2009;41:971–6. Baayen RH, Davidson DJ, Bates DM. Mixed-effects modeling with crossed random effects for subjects and items. J Mem Lang. 2008;59(4):390–412. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models Using lme4. J Stat Softw. 2015;67(1):1–48. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2021. Barr DJ, Levy R, Scheepers C, Tily HJ. Random effects structure for confirmatory hypothesis testing: Keep it maximal. J Mem Lang. 2013;68(3):255–78. Venables WN, Ripley BD. (2002). Random and Mixed Effects: Springer New York. Modern applied statistics with S , 271–300. Inhoff AW, Liu W. The perceptual span and oculomotor activity during the reading of Chinese sentences. J Exp Psychol Hum Percept Perform. 1998;24(1):20–34. https://doi.org/10.1037/0096-1523.24.1.20 . Von Malsburg D, T., Angele B. False positives and other statistical errors in standard analyses of eye movements in reading. J Mem Lang. 2017;94:119–33. https://doi.org/10.1016/j.jml.2016.10.003 . Zang C, Zhang M, Bai X, Yan G, Paterson KB, Liversedge SP. Effects of word frequency and visual complexity on eye movements of young and older Chinese readers. Q J Experimental Psychol. 2015;69(7):1409–25. https://doi.org/10.1080/17470218.2015.1083594 . Kuperman V, Schroeder S, Gnetov D. Word length and frequency effects on text reading are highly similar in 12 alphabetic languages. J Mem Lang. 2024;135:104497. Jordan TR, Dixon J, McGowan VA, Kurtev S, Paterson KB. Fast and slow readers and the effectiveness of the spatial frequency content of text: Evidence from reading times and eye movements. J Exp Psychol Hum Percept Perform. 2016;42:1066–71. Jordan TR, Thomas SM, Patching GR, ScottBrown KC. Assessing the importance of letter pairs in initial, exterior, and interior positions in reading. J Experimental Psychology: Learn Memory Cognition. 2003;29:883–93. Rayner K. Eye Guidance in Reading: Fixation Locations within Words. Perception. 1979;8(1):21–30. Footnotes in Pinyin Tables Table 1 Target word characteristics. Frequency (counts/million) Visual Complexity (No. of Character Strokes) Lexical Predictability High Frequency Low Frequency t p High Frequency Low Frequency t p High Frequency Low Frequency t p First Character 752.5 (1083.3) 544.7 (929.6) 1.63 .10 8.0 (3.1) 8.2 (3.1) .43 .67 Second Character 719.0 (1371.5) 900.1 (1502.9) 1.07 .28 8.1 (3.2) 8.3 (3.1) .54 .59 Word 144.8 (223.7) 12.75 (19.6) 6.5 .001 16.2 (4.7) 16.5 (4.3) .71 .48 .09 (.08) .10 (.06) 1.72 .09 Note. The Standard Error of the mean is shown in parentheses. Table 2 Means for sentence-level measures Display type Measure Normal Interior Blur Exterior Blur Sentence reading time (ms) 3462 (40) 3901 (46) 3842 (46) Average fixation duration (ms) 252 (1) 264 (1) 256 (1) Number of fixations 12.9 (.13) 13.9 (.14) 14.1 (.14) Forward saccade length (chars) 1.75 (.02) 1.65 (.01) 1.62 (.01) Number of regressions 3.52 (.06) 3.73 (.06) 3.83 (.07) Note . Standard errors are reported in parentheses. Table 3 Statistical effects for sentence-level measures Comparison Statistic Sentence Reading Time Number of Fixations Average Fixation Duration Forward Saccade Length Number of Regressions Intercept β 8.11 13.64 5.54 1.67 3.40 SE 0.06 0.63 0.01 0.07 0.24 t/z 146.96* 21.49* 412.96 24.90 13.96 p 0.000 0.000 0.000 0.000 0.000 Display: Exterior Blur-Normal β 0.10 1.14 0.01 0.13 0.35 SE 0.01 0.13 0.00 0.01 0.07 t/z 9.92* 8.78* 4.02* 9.81* 4.85* p 0.000 0.000 0.000 0.000 0.000 Display: Interior Blur-Exterior Blur β 0.02 0.12 0.03 0.03 0.12 SE 0.01 0.13 0.00 0.01 0.07 t/z 1.69 0.91 8.80* 2.16 1.66 p 0.211 0.632 0.000 0.079 0.221 Display: Interior Blur-Normal β 0.12 1.02 0.05 0.10 0.23 SE 0.01 0.13 0.00 0.01 0.07 t/z 11.61* 7.86* 12.81* 7.65* 3.19* p 0.000 0.000 0.000 0.000 0.004 Note . Asterisks indicates statistically significant fixed-factor effects ( t/z > 1.96) Table 4 Means for target-word level measures High Frequency Low Frequency Measure normal interior blur exterior blur normal interior blur exterior blur Word Skipping (%) 31 (2) 27 (1) 24 (1) 24 (1) 24 (1) 23 (1) First-fixation duration (ms) 249 (3) 261 (3) 250 (3) 253 (3) 267 (3) 259 (4) Single-fixation duration (ms) 247 (4) 259 (4) 248 (3) 250 (3) 266 (4) 256 (4) Gaze duration (ms) 277 (5) 290 (6) 289 (6) 281 (5) 307 (6) 303 (6) Total reading time (ms) 376 (9) 380 (10) 383 (10) 379 (8) 426 (11) 434 (11) Regressions In (%) 17 (1) 16 (1) 15 (1) 17 (1) 19 (1) 21 (2) Note . Standard errors are reported in parentheses. Table 5 Statistical effects for target-word level measures Measure Statistic SKIP FFD SFD GD TRT RI Intercept β 1.28 5.48 5.48 5.56 5.80 1.78 SE 0.14 0.02 0.02 0.02 0.04 0.14 t/z 9.21* 325.42* 314.52* 223.45* 164.55* 12.75* p 0.000 0.000 0.000 0.000 0.000 0.000 Word Frequency β 0.21 0.02 0.02 0.04 0.09 0.27 SE 0.07 0.01 0.01 0.01 0.02 0.09 t/z 3.24* 2.22* 2.36* 3.15* 5.19* 3.15* p 0.001 0.032 0.023 0.003 0.000 0.002 Display: interior blur - normal β 0.11 0.05 0.05 0.06 0.04 0.04 SE 0.08 0.01 0.01 0.01 0.02 0.11 t/z 1.36 4.21* 4.30* 4.34* 2.42* 0.42 p 0.361 0.000 0.000 0.000 0.041 0.908 Display: exterior blur -normal Β 0.24 0.01 0.02 0.05 0.06 0.09 SE 0.08 0.01 0.01 0.01 0.02 0.10 t/z 2.92* 1.15 1.52 3.42* 3.15* 0.84 p 0.010 0.485 0.285 0.002 0.005 0.676 Display: interior blur - exterior blur Β 0.13 0.04 0.03 0.01 0.01 0.04 SE 0.08 0.01 0.01 0.01 0.02 0.10 t/z 1.56 3.11* 2.80* 0.96 0.72 0.42 p 0.264 0.005 0.014 0.601 0.753 0.906 Frequency high - low: Display exterior blur -normal Β 0.36 0.00 0.01 0.02 0.08 0.41 SE 0.16 0.02 0.02 0.03 0.04 0.21 t/z 2.27* 0.21 0.38 0.55 2.18* 1.98* p 0.023 0.836 0.702 0.583 0.029 0.047 Frequency high - low: Display interior blur – exterior blur Β 0.19 0.00 0.00 0.01 0.01 0.20 SE 0.16 0.02 0.02 0.03 0.04 0.21 t/z 1.20 0.16 0.05 0.30 0.23 0.99 p 0.231 0.871 0.961 0.766 0.819 0.321 Frequency high - low: Display interior blur - normal Β 0.17 0.00 0.01 0.02 0.07 0.21 SE 0.16 0.02 0.02 0.03 0.04 0.21 t/z 1.08 0.05 0.43 0.84 1.94 0.99 p 0.281 0.964 0.665 0.402 0.053 0.322 Note . Asterisks indicates statistically significant fixed-factor effects ( t/z > 1.96) Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5778975","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":446150653,"identity":"7b9239c4-5cf8-4521-9fed-6e23d8bc7508","order_by":0,"name":"Lin Li","email":"","orcid":"","institution":"Tianjin Normal University","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Li","suffix":""},{"id":446150654,"identity":"de1c3918-66b4-4d09-8125-e0109dc8771a","order_by":1,"name":"Yaning Ji","email":"","orcid":"","institution":"Tianjin Normal University","correspondingAuthor":false,"prefix":"","firstName":"Yaning","middleName":"","lastName":"Ji","suffix":""},{"id":446150655,"identity":"0e35d73f-1ab6-4fb2-aa55-ce369571cd49","order_by":2,"name":"Xinhui Liu","email":"","orcid":"","institution":"Tianjin Normal University","correspondingAuthor":false,"prefix":"","firstName":"Xinhui","middleName":"","lastName":"Liu","suffix":""},{"id":446150656,"identity":"2d07ccfc-00dc-4230-afd9-2c9006fbcac0","order_by":3,"name":"Xinyu Zhao","email":"","orcid":"","institution":"Tianjin Normal University","correspondingAuthor":false,"prefix":"","firstName":"Xinyu","middleName":"","lastName":"Zhao","suffix":""},{"id":446150657,"identity":"e17498e2-73b6-4b88-a087-493cff71b9ef","order_by":4,"name":"Kevin B. Paterson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6klEQVRIiWNgGAWjYBACCQkG5gcfChh4kAWZCWlhM5xhANbC2ECsFgZpHgMwm0gtkrN7DxjbGNjI8Le3P3/wcQ+DPH8Dj7EBPi3SMucSHucYpPFInDlj2DjjGYPhjAM8xgn4tMhJ5BgY5xgc5jGQyGFs5jnAwLiBgcf4ACEt0hYG/3kM5J8/bP5zgMGeoBZpkBYGgwNAWxgMmxkOMCSCtOB1mOScc2mGPQbJQL/kGM7sOSCRPOMwWzFe70vc7j384EeFnT1/+/EHH34csLHtb2/eLIFPCwNqvDNIEIoVTC2jYBSMglEwCjABACSMQZpsufOXAAAAAElFTkSuQmCC","orcid":"","institution":"University of Leicester","correspondingAuthor":true,"prefix":"","firstName":"Kevin","middleName":"B.","lastName":"Paterson","suffix":""}],"badges":[],"createdAt":"2025-01-07 07:38:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5778975/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5778975/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81205796,"identity":"e554e957-9783-4879-ba9a-23789075650d","added_by":"auto","created_at":"2025-04-23 12:10:59","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":986776,"visible":true,"origin":"","legend":"\u003cp\u003eExample of an interior blurred character\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5778975/v1/f3f18fae2821febb91d9bce6.jpeg"},{"id":81205797,"identity":"a0dac5c4-3b60-4387-97f9-88f2c0c09cb6","added_by":"auto","created_at":"2025-04-23 12:10:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":66193,"visible":true,"origin":"","legend":"\u003cp\u003eAn example stimulus in each display condition\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNote\u003c/em\u003e. High and low-frequency target words are shown underlined but were presented normally in the experiment.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5778975/v1/94aee9b6c3d89d0a491a7ff1.png"},{"id":90145442,"identity":"37c31730-9800-409a-8a3e-2c01931cc972","added_by":"auto","created_at":"2025-08-29 05:32:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2055439,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5778975/v1/3e5354cc-8aa6-440e-898e-bf3a12c2e980.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Investigating the holistic processing of characters during Chinese reading","fulltext":[{"header":"Introduction","content":"\u003cp\u003eChinese words are composed of box-like logograms known as characters, each of which occupies the same square area of space and are created from a sequence of distinct line strokes [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. These characters vary in complexity, ranging from simple characters containing only a few strokes, such as 一 (\u003cem\u003eyī\u003c/em\u003e\u003csup\u003e1\u003c/sup\u003e meaning \"one\"), or 人 (\u003cem\u003er\u0026eacute;n\u003c/em\u003e, meaning \"person\"), to highly intricate characters containing many strokes, such as 赢 (\u003cem\u003ey\u0026iacute;ng\u003c/em\u003e, meaning \"win\") and 骉 (\u003cem\u003ebiāo\u003c/em\u003e, meaning \"galloping horses\"). More complex characters usually are constructed from smaller units of strokes called radicals, many of which are characters themselves and contribute phonological or semantic information when part of more complex characters. From a language processing perspective, this complexity raises the question of what information is essential for a character to be recognized efficiently. Factors such as stroke patterns, structural composition, and the spatial relationships between components seem likely to play key roles in enabling the rapid and accurate identification of characters. However, there is ongoing debate concerning whether readers process Chinese characters analytically, by breaking them down into more basic components such as radicals and strokes; or, alternatively, whether readers process characters holistically, by recognizing their overall shape without focusing on individual parts [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eA long-standing perspective in alphabetic language research is that word recognition operates compositionally - visual features are combined into letters, and words are recognized through their constituent letters (for an overview, see [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]). This view has shaped the development of influential computational models of word recognition, most notably the interactive-activation model [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. According to such models, visual input is processed hierarchically: basic features are first detected, then integrated into abstract letter representations, which subsequently activate lexical-level word representations. A similar hierarchical framework has been proposed by Taft and colleagues to explain Chinese word identification [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], drawing inspiration from the interactive-activation model (and for an overview of computational models of Chinese word recognition, see [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]). According to this approach, Chinese word recognition also follows a compositional process, beginning with the detection of individual strokes, which are combined into sub-character components (such as radicals), then integrated into whole characters, and finally assembled into words.\u003c/p\u003e \u003cp\u003eProponents of this analytic approach point to evidence that character recognition is sensitive to the position and meaning of radicals, suggesting that this information is extracted during character recognition [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. By comparison, advocates of holistic processing argue that characters function as the primary unit for orthographic processing, as each corresponds to a single syllable and conveys semantic information comparable to a phoneme in alphabetic scripts like English [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. According to this view, characters are recognized as a whole, with their component radicals processed only when the task demands their decomposition [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Another possibility is that both analytic and holistic strategies are used during character recognition. For instance, familiar characters may be recognized rapidly through holistic processing, whereas less familiar characters might require a more analytic approach [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Moreover, the use of one strategy rather than another may depend on the reader\u0026rsquo;s level of expertise. Expert readers, who have extensive exposure to Chinese characters, may be more likely to process characters holistically, while novice or beginning readers may rely on an analytic strategy due to their limited familiarity with characters. Note, however, that other research suggests expertise in Chinese character recognition may involve reduced, rather than enhanced, holistic processing due to experienced readers developing a more detailed understanding of character construction [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWith the present study, we investigated the use of holistic versus analytic processing during Chinese reading using a visual technique that involved selectively blurring either an interior or exterior character region. This approach was designed to control the availability of character information during reading, allowing us to investigate whether skilled Chinese readers can process characters holistically based on their overall shape. Our approach builds on studies investigating holistic processing in other visual domains, including recognition of faces, chessboards, fingerprints, and X-ray images [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMany such studies use visual blurring or similar experimental techniques to investigate the effects of limiting the availability of detailed visual information on the recognition of visual objects [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. This includes studies that have manipulated spatial frequency content of visual input, based on the evidence that the visual system processes a range of information across different levels of detail, from high spatial frequencies that encode fine detail to lower spatial frequencies that convey more global information about the shape and location of an object [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Research using these technique to limit the availability of visual information show that participants can identify objects based solely on cues to their overall shape or configuration, without relying on detailed, feature-based cues [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSimilar approaches have been taken to the study of visual word recognition, based on the assumption that words are a type of visual object. Classic research on the word superiority effect suggests that words can be identified even when insufficient visual information is available to unambiguously identify individual letters [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Studies using blurring techniques additionally suggest that words can be accurately recognized using only coarse-scale cues to their overall shape [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e], though this may be strongest for familiar, high-frequency words. Moreover, this ability to read and recognize words accurately even when only holistic information is available has been show to extend to full sentence reading [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. One possibility is that that coarse-scale cues to the overall shape of words may facilitate early stages of word recognition, while more detailed visual processing may be necessary for recognizing individual letters in words. This follows the idea that object perception proceeds over time from an analysis based on global to more detailed visual information [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e], although note that recent findings suggest a different time course for word recognition [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAlthough research on the need for detailed visual information to recognize characters in logographic scripts like Chinese remains limited, Zhang and Reilly [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e] conducted an internet-based study using a selective blurring technique to show that character recognition remains robust even with incomplete visual information. Their experiment used Gaussian blurring to obscure either the top/bottom or left/right halves of characters while measuring recognition accuracy. Participants achieved overall high character identification rates (88% accuracy), suggesting that full visual detail may not be essential for character recognition. These findings align with the view that skilled Chinese readers may be able to employ holistic visual processing strategies to facilitate character identification when presented with only partial character information. Zhang and Reilly [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e] further suggested that future research may use measures of eye movements to gain insights into the effectiveness of partial information in supporting character recognition during reading.\u003c/p\u003e \u003cp\u003eThe present research builds upon this prior work by examining the effects of selectively blurring either the exterior or interior components of Chinese characters on eye movements during sentence reading. The rationale for this approach is that the exterior region of Chinese characters is likely to provide information about its overall shape, while the interior region provides more detailed visual information. In addition, we used eye movement measures to investigate the influence of this manipulation on reading behavior. Most investigations of holistic processes in word recognition have examined recognition accuracy [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e] or employed tasks such as the lexical decision task or the Reicher-Wheeler two-alternative forced-choice task [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. These provide insights into the cognitive processes that underpin the recognition of individually presented words. However, they require participants to make explicit decisions about the characteristics of stimuli, which may introduce task demands not present in typical reading situations and so may promote the use of specific processing strategies. In contrast, research using eye movement measures allows participants to engage in normal reading behavior, while the dynamic properties of eye movements have been shown to yield valuable information about the cognitive decision-making processes involved in recognizing words during reading [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. We therefore employed this approach in the present research to investigate whether readers can use holistic information about overall character shape to facilitate processes of word recognition in reading.\u003c/p\u003e \u003cp\u003ePrevious eye-tracking studies employing blurring techniques in alphabetic scripts demonstrate that skilled readers can process sentences efficiently even when only coarse-scale cues to word identities are available. [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. In these experiments, researchers manipulate text quality by blurring either specific target words within a sentence or the entire sentence. Participants read these texts while their eye movements are recorded. Analyses focused on reading speed, comprehension accuracy (assessed via post-sentence questions), and specific eye-tracking metrics (e.g., fixation durations, regressions) to examine how visual degradation impacts reading. Strikingly, participants seem able to read and comprehend sentences remarkably well even when these included degraded words, while showing minimal disruption to normal eye movement behavior. The findings suggest that readers can achieve fluent reading without relying on fine-grained visual details of individual letters. However, the generalizability of these effects to logographic systems like Chinese presently is unclear. We therefore used the same approach to investigate whether Chinese readers can use holistic information about overall character shape to support normal reading in the absence of fine-grained visual cues to character identity.\u003c/p\u003e \u003cp\u003eThis approach enabled us to investigate effects on both sentence-level eye movement behavior and eye movements for specific target words in sentences. Sentence-level metrics such as total reading time for sentences, and the number and average duration of eye fixations, length of forward-directed eye movements (forward saccade length), and frequency of regressions (backward eye movements) during sentence reading, provide an overall assessment of reading difficulty. We hypothesized that if readers encountered difficulty due to text degradation, they would exhibit slower sentence reading, characterized by an increase in the number and duration of fixations, shorter forward eye movements, and a higher frequency of regressions to reread parts of the sentence, compared to when text was shown normally. Crucially, this approach also allowed us to compare effects of degrading information about the visual detail in characters compared to configurational information about the overall shape of characters. If readers can process text effectively using only holistic information about overall word shape, they might read sentences more quickly, with fewer and shorter fixations, longer forward eye movements, and fewer regressions, when the exterior region of characters is preserved.\u003c/p\u003e \u003cp\u003eTo isolate effects of our manipulation on processes of word identification during reading, we included pairs of interchangeable two-character target words in each sentence frame. These words differed in terms of their lexical frequency, with one having a higher frequency of written usage compared to the other. Both words were selected to be equally plausible within the sentence context and the high and low frequency words were carefully matched in terms of their visual complexity (e.g., number of character strokes) and their predictability from the preceding text. Substantial research has established that effects of lexical frequency on eye movement behavior provide a robust marker of word recognition difficulty [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Similarly to effects in alphabetic scripts, studies show that high-frequency words are more likely to be skipped (i.e., not fixated during the initial, first-pass reading of a sentence) or receive shorter fixation durations compared to lower frequency words [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. These findings also align with evidence from word recognition studies showing that high-frequency words are recognized more quickly than low-frequency words in both alphabetic languages and Chinese [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eResearch manipulating text quality - such as through blurring or contrast adjustments \u0026ndash; has additionally shown that low-frequency words are disproportionately affected by text degradation, resulting in larger word frequency effects for degraded compared to normal text [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. In the present study, by manipulating lexical frequency, we aimed to determine whether selectively blurring exterior versus interior regions of characters would differentially influence the word frequency effect. By analyzing eye movement measures sensitive to the initial (first-pass) processing of words or their later word processing, we sought to determine whether these differential effects occurred during early or late stages of word recognition. Word-skipping measures would provide the earliest indication of an effect, as this measure is sensitive to an early stage of lexical processing that takes place while a word is in the parafovea (i.e., lower acuity retinal vision outside of central vision; for a review of parafoveal processing in reading) [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. When processing words parafoveally, readers acquire only relatively coarse-scale visual information. However, if this information is sufficient to support processes of word identification, the reader may \u0026ldquo;skip\u0026rdquo; the word by making a forward saccade that moves their gaze past this word, terminating with a fixation on a following word. Prior research shows that words that are of higher rather than lower frequency are more likely to be skipped. Accordingly, with the present research, we investigated whether selectively blurring exterior versus interior information in characters might support effects of word frequency on word-skipping rates.\u003c/p\u003e \u003cp\u003eEffects of word frequency would also be expected to be observed in fixation times on target words, with shorter fixations for higher compared to the lower frequency words [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. If word identification is disrupted by selective blurring, we would expect target words in blurred text to be skipped less frequently and to receive longer fixation durations compared to the same words in normal, unblurred text. Additionally, the increased difficulty caused by text degradation should amplify the word frequency effect, manifesting as larger differences in word-skipping rates and fixation durations between high- and low-frequency words in blurred text compared to unblurred text. Crucially, this approach enabled us to assess effects of selective character blurring on word identification processes during reading. In particular, it allowed us to determine whether readers can identify words more efficiently when holistic information about overall word shape is available. If holistic information is more effective for supporting word identification in reading, we might expect higher word-skipping rates, shorter fixation durations, and a smaller lexical frequency effect in these measures when exterior rather than interior character information is preserved.\u003c/p\u003e"},{"header":"Method","content":"\u003cp\u003eThe research was approved by the Research Ethics Committee in the Faculty of Psychology at Tianjin Normal University and conducted in accordance with the principles of the Declaration of Helsinki. All participants gave written informed consent.\u003c/p\u003e\n\u003cp\u003eParticipants. Participants were 48 undergraduate students aged 18\u0026ndash;23 years (M\u0026thinsp;=\u0026thinsp;21.5 years; 30 female). All were native Chinese readers, screened for normal visual acuity (greater than 20/20 in Snellen values) using a Tumbling E chart [\u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e].\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStimuli and Design\u003c/em\u003e. Stimuli consisted of 120 pairs of Chinese sentence frames, each containing a two-character target word with either high or low lexical frequency, based on the SUBTLEX-CH database [\u003cspan class=\"CitationRef\"\u003e52\u003c/span\u003e]. Sentences were 20 to 25 characters long (M\u0026thinsp;=\u0026thinsp;22.3 characters), including the target word, which was always positioned near the center of each sentence. These stimuli were based on the stimulus used by Wang et al. [\u003cspan class=\"CitationRef\"\u003e53\u003c/span\u003e] to examine adult age differences in effects of text degradation on the word frequency effect in eye movements during Chinese reading.\u003c/p\u003e\n\u003cp\u003eThe high- and low-frequency target words were matched on key characteristics. First, the frequency of the individual characters (both first and second) within the target words was equivalent across the two frequency groups. Second, the visual complexity of the words was matched, measured by the number of strokes in the individual characters and the word as a whole. Third, cloze predictability tests conducted with 20 participants (who did not take part in the experiment) confirmed that high- and low-frequency words were equally unpredictable as sentence completions. This involved presenting sentence fragments truncated immediately before the target word and collecting written continuations, which showed that both word types were produced with similarly low frequency. Finally, the naturalness of the sentences was assessed separately by ten participants, who also did not participate in the main experiment. Using a 7-point scale ranging from highly natural to highly unnatural, participants rated all sentences as highly natural overall (M\u0026thinsp;=\u0026thinsp;5.92, SD\u0026thinsp;=\u0026thinsp;0.32), with no significant difference in naturalness ratings between sentences containing high- and low-frequency target words, \u003cem\u003et\u003c/em\u003e(119)\u0026thinsp;=\u0026thinsp;1.52, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.13. This careful matching of lexical, visual, and contextual properties ensured that the target words and sentences were comparable across all critical variables.\u003c/p\u003e\n\u003cp\u003eThe sentence stimuli were presented under three display conditions: normal (unmodified) or with blurring applied to either an interior or exterior region of each character. To implement these conditions, a 28\u0026times;28 pixel grid was superimposed on each character. This grid enabled the definition of two distinct regions: an interior region at the center of the character, consisting of a square containing 400 cells, and an exterior region encompassing the character\u0026apos;s surrounding envelope, containing 384 cells. For the blurring conditions, either the interior or exterior area was blurred using a Gaussian blur applied to each 3-pixel section using Adobe Photoshop. Figure \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e provides an illustration of this procedure, showing how it was applied to a sample character to create the interior and exterior blur conditions.\u003c/p\u003e\n\u003cp\u003eThree versions of each sentence stimulus were created, and these were distributed across six lists. Each list included one version of every sentence, containing either a high- or low-frequency target word displayed in one of three conditions: normal, interior blur, or exterior blur (example displays are shown in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Sentences were allocated to the lists using a Latin square to ensure an equal number of stimuli from each condition in each list, with every sentence appearing in all conditions across the six lists. Participants were pseudo-randomly assigned to one of the six lists, with an equal number of participants assigned to each. Each participant viewed all 120 stimuli in a randomized order, with sentences presented either normally or with interior or exterior blur. The experiment began with four practice sentences to familiarize participants with the task.\u003c/p\u003e\n\u003cp\u003eThe study used a within-participants design, with two factors: lexical frequency (high, low) and display type (normal, interior blur, exterior blur). This ensured each participant was exposed to all display conditions and target word frequencies across the experimental stimuli.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eApparatus and Procedure\u003c/em\u003e. An EyeLink 1000 Plus eye-tracker (SR Research, Canada) was used to record right-eye gaze location during binocular viewing, with a sampling rate of 1,000 Hz. This system provides high spatial resolution (\u0026lt;\u0026thinsp;0.01\u0026deg; RMS) and precise temporal resolution. Sentences were presented in Song font as black text on a gray background. At a viewing distance of 60 cm, each character subtended approximately 1\u0026deg; of visual angle horizontally, corresponding to a standard character size for normal reading [\u003cspan class=\"CitationRef\"\u003e54\u003c/span\u003e].\u003c/p\u003e\n\u003cp\u003eParticipants completed the experiment individually and were instructed to read the sentences normally and for comprehension. At the start of the experiment, a three-point horizontal calibration procedure was performed along the same line as the sentence presentations. This ensured spatial accuracy of 0.30\u0026deg; or better for all participants, corresponding to precision within half a character. Calibration accuracy was checked before each trial, and the eye-tracker was recalibrated as needed to maintain this high level of precision. At the beginning of each trial, a fixation square, equal in size to one character, appeared on the left side of the screen. Once participants fixated on the square, the sentence was displayed, with its first character replacing the square. Participants pressed a response key when they finished reading each sentence. On 25% of the trials, a yes/no comprehension question followed the sentence. Participants responded to these questions by pressing one of two designated keys, and their responses were recorded. The entire experiment took approximately 40 minutes to complete for each participant.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData analysis\u003c/em\u003e. Data were analyzed using linear mixed-effects models (LMEs) [\u003cspan class=\"CitationRef\"\u003e55\u003c/span\u003e] for continuous variables and generalized linear mixed-effects models for binomial variables, implemented using the \u003cem\u003elme4\u003c/em\u003e package (Version 1.1\u0026ndash;21) [\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e] in R [\u003cspan class=\"CitationRef\"\u003e57\u003c/span\u003e]. For all analyses, models incorporated a maximum random effects structure where possible [\u003cspan class=\"CitationRef\"\u003e58\u003c/span\u003e]. Both sentence-level and word-level analyses were conducted. In sentence-level analyses, display type was treated as a fixed factor. For target word analyses, fixed factors included lexical frequency, display type, and their interaction, with participants and stimuli included as crossed random effects. If models failed to converge, the random effects structure was simplified by sequentially reducing the structure for stimuli, first removing random effect correlations, then random slopes. Effects were analyzed on log-transformed data, with results reported alongside untransformed means. Contrasts for main effects and interactions were defined using sliding contrasts (\u003cem\u003econtr.sdif\u003c/em\u003e function) from the \u003cem\u003eMASS\u003c/em\u003e package (Version 7.3\u0026ndash;47) [\u003cspan class=\"CitationRef\"\u003e59\u003c/span\u003e]. Malsburg and Angele [\u003cspan class=\"CitationRef\"\u003e60\u003c/span\u003e] argue that the use of multiple correlated dependent measures in eye movement research may increase Type 1 effects. Accordingly, to control for multiple comparisons, we applied Tukey\u0026apos;s HSD correction. Fixation time effects on log-transformed data are reported, and significance was determined using the standard threshold of t/z values\u0026thinsp;\u0026gt;\u0026thinsp;1.96.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eAccuracy answering comprehension questions was high across participants (\u0026gt;\u0026thinsp;90%), with no significant differences observed between sentences displayed in a normal format and those with interior or exterior blur (Normal: M\u0026thinsp;=\u0026thinsp;95.7%, SD\u0026thinsp;=\u0026thinsp;6.6%; Interior Blur: M\u0026thinsp;=\u0026thinsp;96.1%, SD\u0026thinsp;=\u0026thinsp;7.3%; Exterior Blur: M\u0026thinsp;=\u0026thinsp;94.4%, SD\u0026thinsp;=\u0026thinsp;7.7%; \u003cem\u003eF\u003c/em\u003e(2, 94)\u0026thinsp;=\u0026thinsp;0.76, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.47, partial \u0026eta;\u0026sup2; = .016). These results indicate that sentences remained comprehensible even when individual characters were partially obscured.\u003c/p\u003e\n\u003cp\u003ePrior to conducting statistical analyses, the eye movement data were pre-procesed following standard procedures. Fixations shorter than 80 ms and longer than 1200 ms were excluded, accounting for 5% of all fixations. Trials with track loss were also removed (5 trials, less than 1% of the data), along with trials involving sentences that received fewer than six fixations (64 trials, less than 1% of the data). Note that an eye-movement recording comprising fewer than six fixations may indicate track-loss, where the eye-tracker failed to record a participants\u0026rsquo; eye movements. Alternatively, given the perceptual span for Chinese (approximately 5 characters, [\u003cspan class=\"CitationRef\"\u003e60\u003c/span\u003e]), this might indicate that a participant was not reading normally and for comprehension.\u003c/p\u003e\n\u003cp\u003eStatistical analyses were conducted on both sentence-level eye movement data and eye movement data for the target words embedded in each sentence. Sentence-level analyses employed standard measures of reading behavior. Sentence reading time (SRT) was recorded as the duration from the onset of the sentence display to when the participant pressed a response key, indicating they had finished reading. The number of fixations (NF) referred to the total count of fixations made during sentence reading. Average fixation duration (AFD) represented the mean duration, in milliseconds, of all fixations throughout the reading of a sentence. Average forward saccade length (AFSL) measured the mean distance, in character spaces, of progressive eye movements. Finally, the number of regressions (NR) indicated the average number of backward eye movements participants made while reading a sentence. Together, these measures provided a detailed overview of the temporal and spatial dynamics of participants\u0026apos; reading behavior.\u003c/p\u003e\n\u003cp\u003eFor the word-level analyses, we report measures informative about the first-pass and later processing of words. First-pass reading describes the initial processing of a word before a fixation to the right or a regression from the word of interest. We report several key first-pass reading measures. Word-skipping (SKIP) refers to the probability that a word is not fixated during first-pass reading. First-fixation duration (FFD) is the length of the initial fixation on a word during first-pass reading. Single-fixation duration (SFD) is the duration of the first fixation on a word that receives only one fixation during first-pass reading. Gaze duration (GD) is the total time spent on all first-pass fixations on a word. In addition, we examined measures sensitive to later stages of word processing. Total reading time (TRT) reflects the sum of all fixations on a word, capturing both its initial and later processing. Finally, regressions-in probability (RI) is the likelihood of a regression back to an earlier word, providing insight into reanalysis or integration processes.\u003c/p\u003e\n\u003cp\u003eIn eye movement research, it is common practice to report multiple partially correlated dependent measures to provide a rich understanding of how variables of interest affect reading behavior. However, this approach carries the risk of inflating Type I errors. These concerns are examined in detail by von der Malsburg and Angele [\u003cspan class=\"CitationRef\"\u003e61\u003c/span\u003e], who outline several approaches to minimize this risk. These include focusing hypothesis-driven analyses on measures most likely to show the expected effects, applying corrections for multiple comparisons across dependent variables (though such corrections may not be appropriate for partially correlated measures), or adopting the heuristic that an effect should be considered reliable only when observed in at least two dependent measures. We note that all effects of interest in the present research meet this reliability criterion, appearing in at least two dependent measures. Moreover, several key effects were observed across an even broader range of measures, further strengthening confidence in their robustness.\u003c/p\u003e\n\u003cp\u003eTable \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e presents the mean sentence-level eye movement measures, and Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e provides a summary of the statistical analyses. Significant main effects of display type were observed across all measures. Compared to normal displays, reading times were longer for both interior and exterior blur conditions. These conditions were also associated with an increased number of fixations, longer average fixation durations, shorter forward saccades, and a greater number of regressions. Together, the findings indicate that interior and exterior blur displays both significantly disrupted readers\u0026apos; eye movement patterns.\u003c/p\u003e\n\u003cp\u003eAverage fixation durations were notably longer for sentences displayed with interior blur compared to those with exterior blur. However, no other significant differences emerged between the two blur conditions, suggesting that interior and exterior blur had broadly comparable effects on most eye movement measures.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTarget word analyses\u003c/em\u003e. Table \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e presents target-word level eye movement measures, and Table \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e provides a summary of the statistical analyses.\u003c/p\u003e\n\u003cp\u003eStandard lexical frequency effects were obtained, with lower skipping rates, longer fixation times and more regressions back to low- compared to high-frequency target words. This fixation time effect was evident in both first-pass and later fixation time measures, indicating that word frequency influenced the initial stages of word identification and had a sustained impact on the processing of target words. These results align with previous studies investigating word frequency effects in Chinese reading [\u003cspan class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e53\u003c/span\u003e].\u003c/p\u003e\n\u003cp\u003eEffects of display type were also observed, revealing a differential impact of interior and exterior blur the processing of words. Word-skipping rates were lower for words presented with exterior blur compared to normal displays, while interior blur produced an intermediate word-skipping rate that did not differ significantly from either the exterior blur or normal displays. During the very first fixations on words, as reflected in first-fixation duration (FFD) and single-fixation duration (SFD), fixation times were longer for interior compared to exterior blur.\u003c/p\u003e\n\u003cp\u003eNo significant differences were observed in fixation times during later stages of word processing between interior and exterior blur conditions, as measured by gaze duration (GD) and total reading time (TRT). However, both blur conditions produced longer fixation times in these measures compared to words show normally. This overall pattern of findings therefore point to an early, short-lived advantage for exterior blur over interior blur, but also emphasize that neither blur condition provides adequate visual information to support the normal processing of words. This underscores the disruptive effects of both interior and exterior blur on the processing of words during reading.\u003c/p\u003e\n\u003cp\u003eThese results might reflect differences in either the visual or lexical processing of words (or both). Accordingly, we also examined the interaction between display condition and lexical frequency to understand whether selective blurring affected lexical processing. An interaction between display condition and lexical frequency was observed in target word-skipping rates for normal versus exterior blur displays, but no significant interactions were found between interior blur and normal displays or between exterior and interior blur displays. Further analysis focused on the magnitude of word frequency effects on word-skipping rates across these display conditions. For normal displays, a significant word frequency effect was found, with a 7% effect (\u0026beta;\u0026thinsp;=\u0026thinsp;0.39, SE\u0026thinsp;=\u0026thinsp;0.11, \u003cem\u003ez\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.51, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). In contrast, the effect was minimal and non-significant for exterior blur displays, showing only a 1% effect (\u0026beta;\u0026thinsp;=\u0026thinsp;0.03, SE\u0026thinsp;=\u0026thinsp;0.12, \u003cem\u003ez\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.22, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;.05). For interior blur displays, the word frequency effect was intermediate, at 3% (\u0026beta;\u0026thinsp;=\u0026thinsp;0.22, SE\u0026thinsp;=\u0026thinsp;0.11, \u003cem\u003ez\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.94, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.05). These findings suggest that blurring reduced the availability of parafoveal information about word identity necessary for efficient word-skipping. Compared to normal displays, blurred displays disrupted readers\u0026apos; ability to skip words effectively. Among the blur conditions, exterior blur caused the greatest disruption to word-skipping compared with displays in which words were shown normally, suggesting that holistic character information plays a critical role in supporting parafoveal processes of word identification. This highlights the particular importance of exterior character information in supporting normal reading processes.\u003c/p\u003e\n\u003cp\u003eNo interactions between display conditions and word frequency were observed in measures sensitive to first-pass fixational processing of target words (i.e., FFD, SFD, and GD). However, interactions between these variables were evident in later measures. For total reading time (TRT), an interaction was found between exterior blur and normal displays, but no interactions were observed between interior blur and normal displays or between exterior and interior blur displays. Similarly, for regressions-in probability (RI), an interaction was observed for exterior blur versus normal displays, but not for interior blur versus normal displays or exterior versus interior blur displays. Further analysis revealed no significant word frequency effects in TRT or RI under normal display conditions (TRT: word frequency effect\u0026thinsp;=\u0026thinsp;3 ms, \u0026beta;\u0026thinsp;=\u0026thinsp;0.04, SE\u0026thinsp;=\u0026thinsp;0.03, \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.33, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;.05; RI: word frequency effect\u0026thinsp;=\u0026thinsp;0.8%, \u0026beta;\u0026thinsp;=\u0026thinsp;0.06, SE\u0026thinsp;=\u0026thinsp;0.15, \u003cem\u003ez\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.41, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;.05). By contrast, word frequency effects were present under both blur conditions, with effects being numerically larger in the exterior blur condition (TRT: word frequency effect\u0026thinsp;=\u0026thinsp;51 ms, \u0026beta;\u0026thinsp;=\u0026thinsp;0.11, SE\u0026thinsp;=\u0026thinsp;0.03, \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.36, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001; RI: word frequency effect\u0026thinsp;=\u0026thinsp;6.0%, \u0026beta;\u0026thinsp;=\u0026thinsp;0.47, SE\u0026thinsp;=\u0026thinsp;0.14, \u003cem\u003ez\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.29, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.01) compared to the interior blur condition (TRT: word frequency effect\u0026thinsp;=\u0026thinsp;46 ms, \u0026beta;\u0026thinsp;=\u0026thinsp;0.11, SE\u0026thinsp;=\u0026thinsp;0.03, \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.99, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001; RI: word frequency effect\u0026thinsp;=\u0026thinsp;3.0%, \u0026beta;\u0026thinsp;=\u0026thinsp;0.27, SE\u0026thinsp;=\u0026thinsp;0.15, \u003cem\u003ez\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.83, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.07).\u003c/p\u003e\n\u003cp\u003eThe lack of a word frequency effect in later processing measures (i.e., total reading time) for normal displays contrasts was surprising. Some prior investigations of word frequency effects in Chinese show robust effects in both early and late processing measures [\u003cspan class=\"CitationRef\"\u003e44\u003c/span\u003e]. However, Wang et al. [\u003cspan class=\"CitationRef\"\u003e53\u003c/span\u003e] also showed a non-significant effect of word frequency in total reading times for words, using the same stimulus set as the present study and young adult participants drawn from a similar population. This difference in the time course of word frequency effects may reflect variation in the stimuli and manipulation of word frequency across different experiments. For instance, while the present experiment carefully controlled for other characteristics that might influence word identification, including character frequency and stroke complexity, other studies either do not control these factors or have varied them systematically [\u003cspan class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e]. Moreover, many studies of word frequency effect in Chinese reading have focused on measures of early word processing and do not report effects in later measures such as total reading time [\u003cspan class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e]. Accordingly, while effects of word frequency appear to be observed robustly in both early and later processing measures in alphabetic reading [\u003cspan class=\"CitationRef\"\u003e63\u003c/span\u003e], it is unclear if the same patterns of effects are observed consistently in character-based scripts like Chinese, especially when other factors affecting word identification are controlled.\u003c/p\u003e\n\u003cp\u003eIndeed, the lack of word frequency effects in later processing measures for normal displays in the present research suggests that lexical frequency primarily influences the early stages of word identification, which are typically completed during first-pass reading for text displayed normally. By contrast, a larger word frequency effect for blurred displays in these later measures suggests that the process of word identification had a longer time-course when word information was partially occluded. Moreover, the stronger effects for interior blur compared to exterior blur suggest that information from the interior regions of words plays a more critical role in supporting word identification during reading.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present research employed a selective visual blurring technique alongside eye movement measurements to explore the role of holistic processing in Chinese reading. Participants were presented with sentences containing a specific target word, which varied in written frequency\u0026mdash;either high or low. These sentences were displayed in three conditions; presented either as normal or with either an interior or an exterior region of their characters blurred. Crucially, exterior blurring degraded visual information about the overall shape of the characters while interior blurring preserved this holistic information.\u003c/p\u003e \u003cp\u003e Perhaps the most striking finding was that participants could read and comprehend sentences relatively well even when only partial character information was available. Comprehension accuracy, assessed by presenting comprehension questions following 25% of sentences, was high (\u0026gt;\u0026thinsp;90% correct responses) with no difference across display conditions. This suggested that sentences were easily understood even when interior or exterior character information was degraded. Degrading this information nevertheless disrupted normal reading processes, with participants taking longer to read the blurred sentences, by making more and longer fixations, shorter forward-directed eye movements and more regressions for these sentences compared to the unblurred sentences. Crucially, there was little indication of a differential effect of interior versus exterior blurring in sentence-level eye movement measures. While the sentences appeared to be read a little more slowly following interior blurring, this effect was reliable only in average fixations durations, suggesting that exterior and interior blurring had comparable effects on sentence-level eye movement behavior. Overall, the pattern of sentence-level effects show that sentences can be read and understood well when either only interior or exterior character information is available, but that reading may be most efficient when both sources were present.\u003c/p\u003e \u003cp\u003eAn analysis of specific target words in each sentence enabled us to take a closer look at the effect of selective blurring on the processing of individual words during reading. These more focused analyses showed that blurring either interior or exterior character information slowed the processing of words, with readers making longer fixations on target words in blurred than unblurred sentences. This replicated our sentence-level findings showing that the blurred text was read more slowly. The word-level eye movement measures we used allowed us to separate effects occurring during the initial, first-pass processing of words from those affecting later processing. Crucially, fixation time measures sensitive to first-pass processing revealed a differentiation in effects of exterior versus interior blurring, with longer first-pass fixations on words following exterior blurring. This suggests that degrading information about the overall character shape caused short-lived disruption to an early stage of the word\u0026rsquo;s processing. Readers may therefore have used holistic processing about overall character shape during this early processing stage. Note that this finding is in line with other evidence that holistic processing facilitates an early stage of visual processing [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. The absence of a similar effect in fixation time measures sensitive to the later processing of words additionally suggests that this influence of holistic processing is short-lived. The short-lived nature of this effect may also explain why it is observed only in measures sensitive to early word-level processing and not sentence-level measures. By comparison, the relative cost for either exterior or interior blurring compared to unblurred text was sustained across measures of both early and late fixational processing, providing further evidence that reading was most efficient when both sources of character information were available.\u003c/p\u003e \u003cp\u003eThe word-level analyses also allowed us to examine effects of visual blurring on the word frequency effect in reading. This describes the processing advantage for words that have a higher frequency of written usage and so are more familiar to readers [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. In the present experiment, we observed standard word frequency effects in word-skipping rates and fixation times on the target words. Skipping rates were lower and fixation times were longer for words of low than high frequency, with readers also making more regressions back to the lower frequency words. Effects of word frequency were observed in fixation time measures sensitive to both early and late stages of a word\u0026rsquo;s processing, revealing that word frequency influenced an early stage of processing and has a sustained influence on the processing of words. The pattern of word frequency effects we obtained were in line with those observed in previous research [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e], indicating that the present experiment produced typical effects of word frequency in eye movement behavior.\u003c/p\u003e \u003cp\u003eThere was no indication that blurring influenced the word frequency effect in fixation times for words, although we did observe such an influence in word-skipping. The lack of a modulating effect in fixation times is consistent with blurring primarily influencing the visual processing of words without impacting on subsequent processes of lexical access. This finding is consistent with findings showing that effects of visual degradation and word frequency can be additive [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e], suggesting they involve distinct stages of processing. One possibility is that the partial information provided by selectively blurring interior or exterior character regions disrupted visual processing without affecting word recognition. This pattern of effects also is consistent with readers being able to use both interior and exterior character information to support word recognition.\u003c/p\u003e \u003cp\u003eBy comparison the influence of visual blurring on the word frequency effect in word-skipping rates was potentially informative about the influence of holistic influences on the parafoveal processing of words. Parafoveal processing describes the pre-processing of upcoming linguistic information to the right of where the reader currently is fixating, which influences the likelihood of a word being skipped during reading [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. The main effect of word frequency we observed in this measure was due to readers being able to parafoveally process higher frequency words more efficiently and so skip these words more often, replicating prior research findings [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. We also observed an interactive effect of character blurring and word frequency, such that the effect of word frequency on skipping rates varied across display conditions. Specifically, a 7% word-frequency effect for unblurred text diminished to 3% following interior character blurring and was eliminated (producing only a 1% effect) following exterior character blurring. The indication is that selectively degrading interior and exterior parts of characters made the parafoveal processing of word identities more difficulty, especially when this disrupted overall character shape information.\u003c/p\u003e \u003cp\u003eTaken together, the present findings suggest both overall character shape and more detailed featural information is required to support normal reading processes. This finding is broadly in line with other evidence that holistic cues as well as detailed visual information may contribute to effective word recognition [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], in addition to the recognition of other visual objects such as faces [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. The present findings are important in showing that such effects are observed in measures that do not require participants to perform a secondary task (e.g., lexical decision) which may introduce task demands not present in normal reading situations, leading them to employ specific processing strategies.\u003c/p\u003e \u003cp\u003eOur findings additionally suggest that holistic information may facilitate the early processing of words. We have reported two sources of evidence for this. First, we observed a cost for exterior versus interior blurring in fixations times sensitive to early word processing, suggesting this early processing was facilitated when exterior character information was available. Second, we found that the word frequency effect in word-skipping rates was eliminated by exterior blurring (and reduced by interior blurring), suggesting that character shape information is required to support the parafoveal processing of words. As the earliest stage of word recognition occurs during its parafoveal processing, when readers acquire visual information about a word prior to it being fixated, it seems that holistic character information may facilitate this early stage of word identification.\u003c/p\u003e"},{"header":"Limitations","content":"\u003cp\u003eWhile the present research contributes to our understanding of holistic processing in character recognition, we also acknowledge several limitations. First, our approach to dividing characters into interior and exterior regions was based on approximately equal spatial areas. This method, while systematic, failed to accommodate individual differences in character structure. As a result, the interior blurring manipulation may have impinged on exterior regions for some characters, particularly those with ambiguous boundaries. Second, the inherent variation in character construction meant our defined regions contained unequal amounts of visual information. Interior regions typically encompassed more stroke details, while exterior regions consisted largely of surrounding white space. This imbalance likely affected how perceptual information was disrupted across different characters. Third, as noted by a reviewer, the complexity and variability of Chinese characters make it inherently difficult to objectively define a character's overall shape. This fundamental challenge limits our ability to systematically manipulate perceptual features. Finally, while our blurring technique successfully restricted access to fine visual detail, it may have also introduced confounding visual effects. The manipulation potentially created artificial boundaries and disrupted stroke continuity, which could have interfered with holistic processing mechanisms rather than simply limiting local feature analysis.\u003c/p\u003e \u003cp\u003eWe note that many of these limitations also apply to other studies investigating how blurring specific character regions affects recognition. For instance, Zhang and Reilly [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e] demonstrated that character recognition remains relatively accurate even when blurring obscures either the upper/lower or left/right halves of characters. These findings suggest that while our blurring approach has limitations, it aligns with broader methodological challenges in this research area. Nevertheless, given these limitations, future research could explore alternative approaches to better understand holistic processing in character recognition. One promising direction might involve systematically manipulating the spatial frequency content of characters, inspired by similar approaches in studies with alphabetic scripts [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e]. This could provide further insights while avoiding some of the artifacts introduced by blurring techniques. The rationale for this method stems from well-established findings that the visual system processes information across multiple spatial frequency channels, from high frequencies encoding fine visual details to low frequencies conveying global shape information [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. As discussed in our Introduction, substantial evidence demonstrates that object recognition often can be achieved using only coarse-scale visual information. This suggests that fine visual details, including those required to identify individual character components, may not be necessary for successful word recognition. By carefully controlling the spatial frequency spectrum of character stimuli, researchers could more precisely investigate the relative contributions of global versus local information in character recognition, including how these contributions might differ across character types or proficiency levels [\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eClearly, further work is required to replicate and extend the present findings. Within eye movement research, the role of exterior character information during parafoveal processing also might be investigated further using the boundary paradigm [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e]. In this paradigm, an invisible boundary is placed directly in front of a word in a sentence. Prior to the reader\u0026rsquo;s gaze crossing this boundary, the word is replaced by a preview stimulus. Then, as soon as the reader\u0026rsquo;s gaze crosses the boundary, this preview is quickly restored to the original word, so that the experimenter can assess the influence of different previews on subsequent processing. This technique could be used to further investigate whether character shape information aids the parafoveal processing of word identities. This could be investigated, for example, by comparing effects of blurring interior versus exterior regions of character previews, similarly to the approach used in the current experiment. Based on the present findings, we might predict a larger preview benefit for interior versus exterior character blurring as the former preserves contour information about a character\u0026rsquo;s overall shape. Such findings might more directly reveal an influence of holistic cues on parafoveal processing.\u003c/p\u003e \u003cp\u003eFurther research may also take account of the internal structure of Chinese characters to investigate what information contributes to word recognition. As already noted, Chinese characters are constructed from substructures of strokes organized into radicals, which can also function as words if presented alone, and which convey phonological and semantic information when part of more complex characters. Theoretical models of Chinese word recognition [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], inspired by analytic models of word recognition in alphabetic scripts [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], assume that character recognition follows a compositional process, beginning with the detection of individual strokes, which combine into sub-character components, such as radicals, before being integrated into whole characters and finally assembled into words.\u003c/p\u003e \u003cp\u003eThe present research suggests that both global character shape and detailed featural information may contribute to efficient word processing, yet the literature reveals considerable complexity in character recognition. On one hand, substantial evidence demonstrates that characters are often processed compositionally through initial identification of their constituent radicals [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. On the other hand, findings indicate that radical recognition may not occur automatically during natural reading [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], pointing to flexibility in compositional processing that varies with task demands.\u003c/p\u003e \u003cp\u003eThis overall pattern of results may support a dual-pathway model of character recognition. Beyond the well-established compositional route, there may be a direct pathway for holistic character identification that operates independently of radical decomposition. The development of this holistic route seems likely to be closely tied to reading expertise and experience. Specifically, high-frequency characters, which are more familiar to readers, may show stronger evidence of holistic recognition. Furthermore, comparisons across reader groups may reveal that skilled native readers depend less on radical information than both less-proficient readers and those acquiring Chinese as a second language. If correct, this may indicate that reading experience shapes the relative engagement of these different processing routes. Indeed, one possibility is that the balance between holistic and featural processing during Chinese reading dynamically adjusts according to multiple factors including character properties, reader proficiency, and situational demands.\u003c/p\u003e \u003cp\u003eIn sum, the present research provides novel insight into the visual basis of word recognition in Chinese, by showing that both overall character shape and detailed featural information are used during reading. Moreover, our findings suggest that holistic information about overall character shape might play a privileged role in the early processing of words, especially during parafoveal processing. While further work is needed to substantiate and extend these findings, it seems likely that such work may will contribute to our understanding of the relative use of holistic and featural processing in Chinese reading.\u003c/p\u003e "},{"header":"Abbreviations","content":"\u003cp\u003eSKIP skipping rate\u003c/p\u003e\n\u003cp\u003eFFD first fixation duration\u003c/p\u003e\n\u003cp\u003eSFD single fixation duration\u003c/p\u003e\n\u003cp\u003eGD gaze duration\u003c/p\u003e\n\u003cp\u003eTRT total reading time\u003c/p\u003e\n\u003cp\u003eRI regressions in probability\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research was approved by the Research Ethics Committee in the Faculty of Psychology at Tianjin Normal University and conducted in accordance with the principles of the Declaration of Helsinki. All participants gave written informed consent.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll participants gave informed written consent that their data could be used for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData sets and analytic code for analyses in R for the present experiment is available via the University of Leicester Figshare repository at\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ehttps://figshare.com/s/d09c2b3cefc44a65f21c\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research was supported by Tianjin Social Science Planning Grant [TJXL24-002] from Tianjin local government.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLin Li played a lead role in conceptualization, data curation, formal analysis, funding acquisition, methodology, project administration, resources, and supervision and an equal role in investigation, writing–original draft, and writing–review and editing. Yaning Ji played a lead role in investigation and a supporting role in data curation. Xinhui Liu played an equal role in data curation and investigation. Xinyu Zhao played an equal role in conceptualization. Kevin B. Paterson played an equal role in conceptualization, writing–original draft, and writing–review and editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHoosain R. (1991). Psycholinguistic implications for Linguistic relativity: A case study of Chinese. \u003cem\u003ePsychology Press\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoosain R. Psychological reality of the word in Chinese. Adv Psychol. 1992;90:111\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsiao JH, Cottrell GW. Not all visual expertise is holistic, but it may be leftist. Psychol Sci. 2009b;20(4):455\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrainger J. Orthographic processing: A \u0026lsquo;mid-level\u0026rsquo; vision of reading. Q J Experimental Psychol. 2018;71(2):335\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/17470218.2017.1314515\u003c/span\u003e\u003cspan address=\"10.1080/17470218.2017.1314515\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcClelland JL, Rumelhart DE. An interactive activation model of context effects in letter perception: Part 1. An account of basic findings. J Exp Psychol Hum Percept Perform. 1981;7(3):191\u0026ndash;217. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1037/0096-1523.7.3.191\u003c/span\u003e\u003cspan address=\"10.1037/0096-1523.7.3.191\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaft M, Zhu X. Submorphemic processing in reading Chinese. J Experimental Psychol Learn Memory Cognition. 1997;23(3):761\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaft M, Zhu X, Peng D. Positional specificity of radicals in Chinese character recognition. J Mem Lang. 1999;40:498\u0026ndash;519.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReichle ED, Yu L. Models of Chinese Reading: Review and Analysis. Cogn Sci. 2018;42:1154\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/cogs.12564\u003c/span\u003e\u003cspan address=\"10.1111/cogs.12564\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Y, Yeh S. Binding radicals in Chinese character recognition: Evidence from repetition blindness. J Mem Lang. 2015;78:47\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFang SP, Horng RY, Tzeng OJL. Consistency effects in Chinese character and pseudo-character naming tasks. In: Kao HSR, Hoosain R, editors. Linguistics, psychology, and the Chinese language. Hong Kong: Hong Kong University; 1986. pp. 11\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeldman LB, Siok WW. Semantic radicals contribute to the visual identification of Chinese characters. J Mem Lang. 1999;40(4):559\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeck KJ, Weekes BS, Chen MJ. Visual and phonological pathways to the lexicon: Evidence from Chinese readers. Mem Cognit. 1995;23(4):468\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeh SL, Li JL. Role of structure and radical in judgments of visual similarity of Chinese characters. J Exp Psychol Hum Percept Perform. 2002;28:933\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen SC, Liu IM. Functional orthographic units in Chinese character recognition. Acta Physiol Sinica. 2000;32:235\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTao L, Healy AF. The unitization effect in reading Chinese and English text. Sci Stud Read. 2002;6:167\u0026ndash;97.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHealy AF. Letter detection: A window to unitization and other cognitive processes in reading text. Psychon Bull Rev. 1994;1(3):333\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTso RV, Au TK, Hsiao JH. (2022). Non-monotonic developmental trend of holistic processing in visual expertise: the case of Chinese character recognition. Cogn Res Principles Implications, \u003cem\u003e7\u003c/em\u003e(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichler JJ, Cheung OS, Gauthier I. Holistic processing predicts face recognition. Psychol Sci. 2011;22(4):464\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0956797611401753\u003c/span\u003e\u003cspan address=\"10.1177/0956797611401753\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVogelsang MD, Palmeri TJ, Busey TA. Holistic processing of fingerprints by expert forensic examiners. Cogn research: principles implications. 2017;2(1):15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s41235-017-0051-x\u003c/span\u003e\u003cspan address=\"10.1186/s41235-017-0051-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSheridan H, Reingold EM. The Holistic Processing Account of Visual Expertise in Medical Image Perception: A Review. Front Psychol. 2017;8:1620. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2017.01620\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2017.01620\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReingold EM, Sheridan H. Eye movements and visual expertise in chess and medicine. In: Liversedge SP, Gilchrist ID, Everling S, editors. The Oxford handbook of eye movements. Oxford University Press; 2011. pp. 523\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichler JJ, Palmeri TJ, Gauthier I. Meanings, Mechanisms, and Measures of Holistic Processing. Front Psychol. 2012;3:553.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBalas B, Gable J, Pearson H. (2018). The Effects of Blur and Inversion on the Recognition of Ambient Face Images. \u003cem\u003ePerception\u003c/em\u003e, 030100661881258. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/0301006618812581\u003c/span\u003e\u003cspan address=\"10.1177/0301006618812581\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichler JJ, Gauthier I. A meta-analysis and review of holistic face processing. Psychol Bull. 2014;140(5):1281\u0026ndash;302.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuiz-Soler M, Beltran FS. Face perception: An integrative review of the role of spatial frequencies. Psychol Res. 2006;70:273\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00426-005-0215-z\u003c/span\u003e\u003cspan address=\"10.1007/s00426-005-0215-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliva A, Schyns P. Coarse blobs or fine edges? Evidence that information diagnosticity changes the perception of complex visual stimuli. Cogn Psychol. 1997;34:72\u0026ndash;107.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlakemore C, Campbell FW. Adaptation to spatial stimuli. J Physiol. 1969;200(1):11\u0026ndash;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePantle A, Sekuler R. Size-detecting mechanisms in human vision. Sci (New York N Y). 1968;162(3858):1146\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.162.3858.1146-a\u003c/span\u003e\u003cspan address=\"10.1126/science.162.3858.1146-a\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollishaw SM, Hole GJ. Featural and configurational processes in the recognition of faces of different familiarity. Perception. 2000;29(8):893\u0026ndash;909. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1068/p2949\u003c/span\u003e\u003cspan address=\"10.1068/p2949\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandford A, Sarker T, Bernier T. Effects of geometric distortions, Gaussian blur, and contrast negation on recognition of familiar faces. Visual Cognition. 2017;26(3):207\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13506285.2017.1407853\u003c/span\u003e\u003cspan address=\"10.1080/13506285.2017.1407853\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReicher GM. Perceptual recognition as a function of meaningfulness of stimulus material. J Exp Psychol. 1968;81(2):275\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatching GR, Jordan TR. Assessing the role of different spatial frequencies in word perception by good and poor readers. Mem Cognit. 2005a;33:961\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJordan TR, McGowan VA, Paterson KB. Reading with a filtered fovea: The influence of visual quality at the point of fixation during reading. Psychon Bull Rev. 2012;19(6):1078\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJordan TR, McGowan VA, Kurtev S, Paterson KB. A further look at postview effects in reading: An eye-movements study of influences from the left of fixation. J Experimental Psychol Learn Memory Cognition. 2016;42(2):296\u0026ndash;307.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaterson KB, McGowan VA, Jordan TR. Filtered text reveals adult age differences in reading: Evidence from eye movements. Psychol Aging. 2013a;28(2):352\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaterson KB, McGowan VA, Jordan TR. Effects of adult aging on reading filtered text: evidence from eye movements. PeerJ. 2013b;1:e63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchyns PG, Oliva A. (1994). From blobs to boundary edges: Evidence for time- and spatial-scale-dependent scene recognition. \u003cem\u003ePsychol. Sci.5\u003c/em\u003e(4), 195\u0026ndash;200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1467-9280.1994.tb00500.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1467-9280.1994.tb00500.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBertrand Pilon C, Arguin M. The processing of spatial frequencies through time in visual word recognition. Sci Rep. 2024;14:6628. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-024-57219-3\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-57219-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Q, Reilly R. What regions of Chinese characters are crucial for recognition? A web-based study. J Chin Writ Syst. 2020;4(4):297\u0026ndash;311.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWheeler DD. Processes in word recognition. Cogn Psychol. 1970;1(1):59\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRayner K. Eye movements in reading and information processing: 20 years of research. Psychol Bull. 1998;124(3):372\u0026ndash;422.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRayner K. The 35th Sir Frederick Bartlett Lecture: Eye movements and attention in reading, scene perception, and visual search. Q J Experimental Psychol. 2009;62(8):1457\u0026ndash;506.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Bicknell K, Liu P, Wei W, Rayner K. Reading is fundamentally similar across disparate writing systems: A systematic characterization of how words and characters influence eye movements in Chinese reading. J Experimental Psychol Gen. 2014;143(2):895\u0026ndash;913.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan G, Tian H, Bai X, Rayner K. The effect of word and character frequency on the eye movements of Chinese readers. Br J Psychol. 2006;97(2):259\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa G, Zhuang X. Distributional analyses of word frequency effects in Chinese sentence reading and lexical decision tasks. J Res Reading. 2018;41:S183\u0026ndash;96. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1467-9817.12259\u003c/span\u003e\u003cspan address=\"10.1111/1467-9817.12259\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorrison CM, Ellis AW. Roles of word frequency and age of acquisition in word naming and lexical decision. J Experimental Psychol Learn Memory Cognition. 1995;21(1):116\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaft M. Recognition of affixed words and the word frequency effect. Mem Cognit. 1979;7(4):263\u0026ndash;72.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBecker C, Killion A, Thomas,H. Interaction of visual and cognitive effects in word recognition. J Exp Psychol Hum Percept Perform. 1977;3(3):389\u0026ndash;401.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang J, Li L, Li S, Xie F, Chang M, Paterson KB, et al. Adult Age Differences in Eye Movements During Reading: The Evidence From Chinese. Journals Gerontol Ser B: Psychol Sci Social Sci. 2018;73(4):584\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchotter ER, Angele B, Rayner K. Parafoveal Processing in Reading. Atten Percept Psychophysics. 2012;74:5\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.3758/s13414-011-0219-2\u003c/span\u003e\u003cspan address=\"10.3758/s13414-011-0219-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor HR. Applying new design principles to the construction of an illiterate e chart. Optom Vis Sci. 1978;55(5):348\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai Q, Brysbaert M. (2010). SUBTLEX-CH: Chinese word and character frequencies based on film subtitles. PLoS ONE, 5(6), e10729.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang J, Li L, Li S, Xie F, Liversedge SP, Paterson KB. Effects of aging and text-stimulus quality on the word-frequency effect during Chinese reading. Psychol Aging. 2018;33(4):693\u0026ndash;712.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu M, Jordan TR. Assessing effects of viewing distance on normal Chinese reading: Some methodological and theoretical implications. Behav Res Methods. 2009;41:971\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaayen RH, Davidson DJ, Bates DM. Mixed-effects modeling with crossed random effects for subjects and items. J Mem Lang. 2008;59(4):390\u0026ndash;412.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBates D, M\u0026auml;chler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models Using lme4. J Stat Softw. 2015;67(1):1\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarr DJ, Levy R, Scheepers C, Tily HJ. Random effects structure for confirmatory hypothesis testing: Keep it maximal. J Mem Lang. 2013;68(3):255\u0026ndash;78.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVenables WN, Ripley BD. (2002). Random and Mixed Effects: Springer New York. \u003cem\u003eModern applied statistics with S\u003c/em\u003e, 271\u0026ndash;300.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eInhoff AW, Liu W. The perceptual span and oculomotor activity during the reading of Chinese sentences. J Exp Psychol Hum Percept Perform. 1998;24(1):20\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/0096-1523.24.1.20\u003c/span\u003e\u003cspan address=\"10.1037/0096-1523.24.1.20\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVon Malsburg D, T., Angele B. False positives and other statistical errors in standard analyses of eye movements in reading. J Mem Lang. 2017;94:119\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jml.2016.10.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jml.2016.10.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZang C, Zhang M, Bai X, Yan G, Paterson KB, Liversedge SP. Effects of word frequency and visual complexity on eye movements of young and older Chinese readers. Q J Experimental Psychol. 2015;69(7):1409\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17470218.2015.1083594\u003c/span\u003e\u003cspan address=\"10.1080/17470218.2015.1083594\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuperman V, Schroeder S, Gnetov D. Word length and frequency effects on text reading are highly similar in 12 alphabetic languages. J Mem Lang. 2024;135:104497.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJordan TR, Dixon J, McGowan VA, Kurtev S, Paterson KB. Fast and slow readers and the effectiveness of the spatial frequency content of text: Evidence from reading times and eye movements. J Exp Psychol Hum Percept Perform. 2016;42:1066\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJordan TR, Thomas SM, Patching GR, ScottBrown KC. Assessing the importance of letter pairs in initial, exterior, and interior positions in reading. J Experimental Psychology: Learn Memory Cognition. 2003;29:883\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRayner K. Eye Guidance in Reading: Fixation Locations within Words. Perception. 1979;8(1):21\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e in Pinyin\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":" \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTarget word characteristics.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"13\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eFrequency\u003c/p\u003e \u003cp\u003e(counts/million)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c9\" namest=\"c6\"\u003e \u003cp\u003eVisual Complexity\u003c/p\u003e \u003cp\u003e(No. of Character Strokes)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c13\" namest=\"c10\"\u003e \u003cp\u003eLexical Predictability\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003cp\u003eFrequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLow Frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003et\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eHigh Frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eLow Frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cem\u003et\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eHigh Frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eLow Frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003e\u003cem\u003et\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFirst Character\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e752.5\u003c/p\u003e \u003cp\u003e(1083.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e544.7\u003c/p\u003e \u003cp\u003e(929.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e8.0 (3.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e8.2 (3.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" morerows=\"1\" nameend=\"c13\" namest=\"c10\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSecond Character\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e719.0\u003c/p\u003e \u003cp\u003e(1371.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e900.1\u003c/p\u003e \u003cp\u003e(1502.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e8.1 (3.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e8.3 (3.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e.59\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWord\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e144.8\u003c/p\u003e \u003cp\u003e(223.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.75\u003c/p\u003e \u003cp\u003e(19.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e16.2 (4.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e16.5 (4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e.09 (.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e.10 (.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e1.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"13\"\u003e\u003cem\u003eNote.\u003c/em\u003e The Standard Error of the mean is shown in parentheses.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMeans for sentence-level measures\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eDisplay type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeasure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNormal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInterior Blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExterior Blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSentence reading time (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3462 (40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3901 (46)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3842 (46)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAverage fixation duration (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e252 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e264 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e256 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of fixations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.9 (.13)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13.9 (.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.1 (.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward saccade length (chars)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.75 (.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.65 (.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003e1.62 (.01)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of regressions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.52 (.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.73 (.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.83 (.07)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c5\" namest=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNote\u003c/em\u003e. Standard errors are reported in parentheses.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStatistical effects for sentence-level measures\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComparison\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStatistic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSentence Reading Time\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNumber of Fixations\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAverage Fixation\u003c/p\u003e \u003cp\u003eDuration\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eForward Saccade\u003c/p\u003e \u003cp\u003eLength\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNumber of\u003c/p\u003e \u003cp\u003eRegressions\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3.40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.24\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e146.96*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.49*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e412.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e24.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e13.96\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDisplay: Exterior Blur-Normal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.35\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9.92*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.78*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.02*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9.81*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e4.85*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDisplay: Interior Blur-Exterior Blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e8.80*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.66\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.211\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.632\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.079\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.221\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDisplay: Interior Blur-Normal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e11.61*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.86*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e12.81*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e7.65*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3.19*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eNote\u003c/em\u003e. Asterisks indicates statistically significant fixed-factor effects (\u003cem\u003et/z\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;1.96)\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMeans for target-word level measures\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eHigh Frequency\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eLow Frequency\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeasure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003enormal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003einterior blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eexterior blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003enormal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003einterior blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eexterior blur\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWord Skipping (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31 (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e24 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e23 (1)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFirst-fixation duration (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e249 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e261 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e253 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e267 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e259 (4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSingle-fixation duration (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e247 (4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e259 (4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e248 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e250 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e266 (4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e256 (4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGaze duration (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e277 (5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e290 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e289 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e281 (5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e307 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e303 (6)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal reading time (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e376 (9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e380 (10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e383 (10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e379 (8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e426 (11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e434 (11)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRegressions In (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e17 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e19 (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e21 (2)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eNote\u003c/em\u003e. Standard errors are reported in parentheses.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStatistical effects for target-word level measures\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeasure\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStatistic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSKIP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFFD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSFD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eGD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eTRT\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eRI\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e5.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.78\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9.21*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e325.42*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e314.52*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e223.45*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e164.55*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e12.75*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eWord Frequency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.24*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.22*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.36*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3.15*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e5.19*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e3.15*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDisplay: interior blur - normal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.21*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.30*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4.34*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.42*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.361\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.908\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDisplay: exterior blur -normal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eΒ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.92*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3.42*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3.15*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.485\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.676\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eDisplay: interior blur - exterior blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eΒ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.11*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.80*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.264\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.601\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.906\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eFrequency high - low:\u003c/p\u003e \u003cp\u003eDisplay exterior blur -normal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eΒ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.41\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.27*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.18*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.98*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.836\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.583\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eFrequency high - low:\u003c/p\u003e \u003cp\u003eDisplay interior blur \u0026ndash; exterior blur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eΒ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.231\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.871\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.961\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.766\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.819\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.321\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eFrequency high - low:\u003c/p\u003e \u003cp\u003eDisplay interior blur - normal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eΒ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003et/z\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.281\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.964\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.665\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.402\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.322\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003cem\u003eNote\u003c/em\u003e. Asterisks indicates statistically significant fixed-factor effects (\u003cem\u003et/z\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;1.96)\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"eye movements, Chinese reading, visual blurring, holistic processing","lastPublishedDoi":"10.21203/rs.3.rs-5778975/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5778975/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground. \u003c/strong\u003eChinese characters are complex visual objects created from a configuration of line strokes. Little is known about the visual input required to support their recognition. However, there is longstanding debate concerning whether characters are recognized holistically, without requiring access to detailed featural information, or recognized analytically by decomposing characters into their constituent parts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethod. \u003c/strong\u003eThe present research investigated this issue using measures of eye movements during sentence reading and employing a visual blurring technique. Characters in sentences were presented as normal or with an exterior or interior region visually degraded. As the exterior region was informative about overall character shape, this manipulation allowed us to investigate effects of preserving or degrading information that might support holistic processing. Additionally, to assess effects on the lexical processing of words, each sentence included one of a pair of interchangeable two-character target words that had either a high or low frequency of written usage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults. \u003c/strong\u003eOur findings showed that degrading either exterior or interior character information disrupted normal eye movement behavior, suggesting that both sources of information are required to support normal reading. Effects for the two-character target words additionally showed that this disruption had additive effects on visual and lexical processing. Evidence for a processing advantage for interior degradation (which preserved holistic processing) over exterior degradation was observed in skipping rates for target words. This was consistent with readers being more likely to skip (i.e., direct their gaze beyond a word without fixating it) when holistic information about overall character shape was available.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion. \u003c/strong\u003eThese results contribute to an important debate concerning the role of holistic processes in reading, by revealing that both overall character shape and detailed featural information may be required to support normal reading. However, effects in word-skipping revealed a potential role for holistic processing during parafoveal processing, during which readers attempt to recognize the next word along in a sentence. The findings showed that this proceeded most efficiently when overall character shape information was available, suggesting a role for holistic processing in parafoveal character recognition.\u003c/p\u003e","manuscriptTitle":"Investigating the holistic processing of characters during Chinese reading","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-23 12:10:55","doi":"10.21203/rs.3.rs-5778975/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9d27bada-4b36-4388-bbcf-a3f50277ce69","owner":[],"postedDate":"April 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-29T05:24:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-23 12:10:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5778975","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5778975","identity":"rs-5778975","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}