Cognitive-load residual is found in Chinese EFL readers’ word comprehension

Cognitive-load residual is found in Chinese EFL readers’ word comprehension | 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 Cognitive-load residual is found in Chinese EFL readers’ word comprehension Ling Wang, Juan Yang, Wei Zhang, Peilin Yang, Yihong Duan, Bo Sun, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6407039/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 Although brain functions surrounding the prefrontal cortex (PFC) has been widely acknowledged as a signature of the instant cognitive load (CL) induced by task demands, inconsistent results of the existing studies imply that the hemodynamic changes in the PFC may not serve as a stable direct indicator of second-language (L2) readers’ instant CL. This study employed 61 Chinese university students to comprehend the identical English words in three different groups to reveal the nature of CL. By observing both the PFC-centred brain functions and the comprehension performance of the participants in the three groups, the following conclusions were reached: 1) The PFC can serve only as an individualised signature of the instant CL that afforded by L2 readers; 2) The unsteadiness of the instant CL is due to residual load carried over from previous tasks; and 3) The residual CL cannot be linearly modelled. Cognitive load (CL) English as foreign language (EFL) second language (L2) brain function functional near infrared spectroscopy (fNIRS) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Cognitive load (CL), a term which is also referred to by many other studies as mental effort or mental workload, has a long research history. Sweller and Paas [1-4] introduced the concept within the context of education and learning, characterising it as a 3-dimensional construct comprising intrinsic load, extraneous load, and germane load. CL theory has since guided teachers and educational practitioners to develop pedagogical designs in terms of how to facilitate learners’ learning via modulating learners’ cognitive load (e.g., [5-9]). In recent years, over 10,000 papers with high relevance to this topic were published, which highlights the continuing importance of application CL theory in education, as well as the importance of effectively monitoring and measuring cognitive load for learners. The development of CL measurement techniques has accelerated in recent years, especially through methods that integrate participants’ physiological data. Physiological data reflects the activity of sympathetic nervous system caused by CL, such as heart rate variability (HRV; e.g., [10]), skin conductance or resistance (e.g. [11]), and eye-tracking related indices: pupil dilation, eye gaze (e.g., [12]) and blink rate (e.g., [13]), as well as facial temperature (e.g., [14]). Unobtrusive brain-function-monitoring devices such as functional near infrared spectroscopy (fNIRS; e.g., [15]) and electroencephalography (EEG; e.g., [16]) have also been used to monitor alterations in brain functions and network synchronisations due to the changed CL. Although cognitive load has been identified as a vital factor in second language (L2) learning, modelling L2 learners’ cognitive load remains a challenge due to the varied cognitive strategies that may be recruited by learners for L2 recognition and comprehension. This study explores the possibility of revealing and modelling the instability of CL experienced by Chinese readers during English word comprehension, based on the CL residual assumption proposed by [17]. A review of the related work Cognitive load models There are several relative models in addition to the Sweller and Paas’s CL model [1-4], such as the working-memory-based model [18] and the mental workload model [19]. As a multidimensional construct, mental workload is generally defined in terms of the cognitive resources that are demanded and competed for in a task due to human’s limited online processing ability [19]. Mental effort is characterised as the cognitive capacity that is allocated to accommodate the demands imposed by a given task [20], and sometimes mental effort is taken as a composition of mental workload that a task requires, e.g., NASA-TLX [21-23]. In addition to the terms “mental effort” and “mental workload” that are generally treated as synonyms for CL, terms like perceived difficulty [24] and mental stress [25] are likewise used synonymously . For decades, the CL theory model has been the subject of continuous study and discussion, particularly concerning its applications in educational contexts, where it provides guidance and evaluation for instructional designs and pedagogical activities. Recent theoretical studies have expanded to a wider range, such as studying the relationship between critical thinking and cognitive load [26], and the relationship between philosophy and cognitive load [27]. CL theory has also been adopted to explore how information and communication technologies reshape people’s cognitive structures [28]. However, the rapid development of CL measuring methods poses challenges for CL theory. For example, Kruger and Doherty [13] conceptualised cognitive load as a dynamic process which should be assessed with a combination of several measurements. They recommended modelling learners’ extraneous, intrinsic, germane and instant load with EEG and eye-movement data, while assessing learners’ overall and average load with psychometric instrument and eye-tracking. However, Bernhardt et al. [29] showed that ocular measures may be more sensitive to increased visual load than EEG workload indices, leading to inconsistencies in CL evaluation. The aforementioned contradictions can be attributed to the high instability of the CL in authentic tasks. Therefore, how to model the unsteadiness of cognitive load is a problem which remains unsolved. This unsteadiness results in significant changes in brain function involvement when brain activity is used as an indicator of the CL imposed by task demands. For example, Wang et al. [17] found that learners’ brain activation showed significant differences in the same task, which was inferred as a result of the “CL residuals” inherited from the priorly conducted mathematical tasks. A similar finding was found in Chan et al. [30], in which they found task operators received photoneuromodulation reduced the brain activations related to WM, implying that the previously conducted operating (photoneuromodulation) may interfere participants’ following tasks’ implementation without behavioural differences. Measuring CL via monitoring learners’ brain functions with fNIRS In recent years, fNIRS has been explored as a potential method of measuring learners’ cognitive load by analysing learners’ brain functions during tasks. Compared with EEG, fNIRS has higher spatial and temporal resolution and is more robust against motion artefacts. Till now, applications of fNIRS in CL modelling can be classified into three different patterns. The dominant research pattern is to observe participants’ brain function changes based on an n-back task to provide a general quantification of the cognitive load with fNIRS data (e.g., [30-35]). Participants’ fNIRS data is collected based on the n-back task and the brain function changes are discussed within the working-memory-based model. For example, in Zhuang et al. ’s study [32], significant differences in brain activation were found among participants who were asked to implement recall tasks with different difficulty level, indicating that higher task difficulty is associated with increased prefrontal cortex (PFC) activation. (e.g., [32-35]). The second research pattern of using fNIRS to model participants’ CL is comparing participants’ hemodynamic (such as blood oxygen) changes in the PFC during the non-n-back tasks to their subjective feelings obtained from scales or questionnaires by using the latter as a calibration tool. It has been suggested that the hemodynamic changes in brain regions associated with working memory (WM) can be directly linked with operators’ performance. Take Alyan et al.’s study [32] for instance, fNIRS was used to monitor participants’ oxygenated hemoglobin (HbO) changes. The results suggested that participants working at a nonergonomic workstation exhibited a significant association between prefrontal cortex deactivation and stressful conditions. That is, the participants who performed the montreal imaging stress task at a nonergonomic workstation showed a lower HbO concentration in the prefrontal cortex (PFC), longer response time, and higher NASA- TLX scores compared to those working at an ergonomic workstation. However, the relationship between hemodynamic changes in PFC and operating performance is not always significant. In an extreme situation, HbO concentration in the PFC even showed no clear correlation with task performance [36]. Therefore, using subjective reports as a calibration tool can make the CL model based on hemodynamic changes in brain regions more reliable. For example, in [37], participants’ HbO changes in the PFC were verified with their self-reported mental effort (NASA- TLX) in reading and writing tasks, showing that the HbO changes of PFC only represented CL in the reading task, but not in the writing task. The third research pattern of measuring CL via fNIRS is to compare participants’ hemodynamic changes in relative brain regions across parallel non-n-back tasks, with one of the hemodynamic patterns as a benchmark [17, 38-39]. For example, Wang and Aryadoust [39] examined the activation patters of the PFC in English-as-second-language readers by manipulating task difficulty through the nature of L2 words (positive vs. negative). By using learners’ brain function data from a benchmark task, the workload can also be assessed by observing brain function changes caused by previously conducted tasks [17]. It is noted that most of the studies we mentioned above have focused on participants’ hemodynamic changes in the PFC to represent instant CL, based on the neurovasuclar coupling principle, which states that active brain regions require increased blood flow to meet enhanced energy demands. The PFC is the brain region associated with executive functions required by all kinds of tasks with online resource requirements. Nevertheless, other interested functional brain regions, such as the middle temporal gyrus [39] and the right ventrolateral cortex [32] that are also important for relative studies, have also been identified as CL components to provide a more comprehensive view of the CL construction in specific tasks. In sum, although the application of fNIRS in CL modelling is booming in ergonomic and medical research areas, its use in modelling learners’ CL in educational area remains limited . This study employs the third research pattern to capture and measure the instant CL experienced by Chinese EFL readers when they comprehend English words, considering interference from previously conducted tasks. Mental workload imposed on English-as-foreign-language (EFL) readers Cognitive load is presumed to be a key factor that impacts learners’ L2 reading comprehension performance. It is assumed that verbal redundancy, where learners are concurrently provided with content, may lead to cognitive overload in L2 reading [40] and L2 listening [41]. However, the construction of CL imposed on EFL reading is controversial, as the process of recognising and comprehending L2 words is not as simple as it is for native speakers. Models regarding L2 reading, such as BIA+ model [42], Revised Hierarchical Model (RHM model; [43]), and Distributed Conceptual Feature Model (DCFM; [44-45]), all suggest that readers’ first language (L1) plays an important role in their L2’s comprehension. The BIA+ model suggests that L1 and L2 lexicons are stored together, and are activated simultaneously when L2 stimuli come in. The RHM model assumes that L2 word meaning retrieval requires mediation, which is the equivalent L1 translation [43]. Although the DCFM [44] suggests that the conceptual memory of L1 and L2 words is shared by different languages, the link strengths among different languages and conceptual memory nodes are unstable and can be reshaped according to learners’ application and learning of L1 and L2. In sum, CL imposed on EFL readers is not only determined by English word comprehension itself, but may also be incurred by the impacts from L1/L2 semantic networks, which are highly dynamic and individualised. The left medial fusiform gyrus serves as a critical neural substrate for orthographic-to-phonological conversion during reading, specifically mediating the grapheme-phoneme mapping essential for decoding English words [46-47]. Neuroimaging evidence further delineates the functional specialisation of the left posterior fusiform cortex, where Dietz et al. [48] observed task-dependent activation patterns modulated by phonological processing complexity. Concurrently, the superior temporal gyrus (STG; Brodmann Area 22), anatomically situated within Wernicke’s area, demonstrates comparable functional significance in English reading. This region orchestrates dual linguistic operations: acoustic-phonetic analysis of speech stimuli [49-50] and articulatory planning for phonological output [51- 52]. Notably, both the left fusiform gyrus and STG exhibit contributions to reading fluency, with the former prioritising visual-linguistic transformation and the latter supporting auditory-oral integration of phonological representations. In addition to the left fusiform gyrus and STG, many researchers have proposed that the middle temporal gyrus (MTG) also plays a vital role in native English speakers’ reading [53-58]. Left MTG was found to be involved in the processing of lexical- [55-56] and sentence comprehension [57-58]. Chinese readers, whose L1 is non-alphabetic, show a unique pattern in Chinese characters’ recognition. Unlike alphabetical languages, Chinese characters do not adhere to letter-sound conversion rules [46]. Chinese characters, with their strokes packed into square shapes, correspond to morphemes and typically have pronunciations suggested by their visual configurations (for reviews, see [59] and [60]). Therefore, a distinctive neurocognitive pattern was revealed among Chinese readers in their English lexical processing, that is, semantic access predominantly occurs through holistic visual processing rather than sequential grapheme-to-phoneme conversion as LSTG was not involved in Chinese (L1) readers’ English words processing [46, 61]; instead, right fusiform gyrus was employed. Such characteristic arises from readers’ implicit tendency to establish direct form-meaning mappings at the orthographic level, a cognitive strategy that diminishes the functional involvement of the brain regions (STG/MTG) that are important for English native speakers while amplifying the right fusiform gyrus’s role as a critical neural substrate for cross-linguistic visual word recognition. In sum, to model Chinese EFL readers’ cognitive load, it is necessary to consider not only the PFC and brain regions that are vital important for English words’ lexical comprehension but also other brain regions that have been revealed to show significant influences in Chinese readers’ English reading. This study The related works illustrate the challenges and complexities of capturing and measuring EFL readers’ instant CL through observing readers’ brain functions; therefore, the research questions of this study are as follows: RQ1. Can the brain function surrounding the PFC be taken as a signature of the CL experienced by Chinese readers in an authentic English word comprehension task? RQ2. How can the unsteadiness of the brain function mechanism regarding the same word comprehension task be explained by the influence of previously conducted non-language tasks? RQ3. Can the instability of the CL be modelled by alteration of the relative brain functions? Method Participants In this study, 61 Mandarin speaking university students (37 females and 24 males) were invited to participate in the experiment, with an average age of 20.03 years ( sd = 0.97 year), ranging from 18 to 23 years. All participants were right-handed and enrolled in formal English classes. Their English proficiency was assessed using their CET-4 scores, where CET stands for the National College English Test, which is a standardised English examination administered by the Ministry of Education of China and is highly correlated with learners’ near-future English proficiency during their university studies . Participants were randomly clustered into three groups: Group 1 (baseline task group, n=21) , Group 2 (GEFT group, n=20), and Group 3 (calculation group, n=20) . The CET-4 scores of participants in the three separate groups showed no significant difference. The study was formally approved by the Ethics Review Committee of Sichuan Normal University and all participants gave written informed consent. Apparatus Functional near-infrared spectroscopy (fNIRS) is one of the most commonly used neuroimaging methods to detect the concentration changes of oxyhaemoglobin (HbO) and deoxyhaemoglobin (HbR). In this study, the fNIRS signals were collected by NirSmart II-3000A, which is a portable device developed by Huichuang of China. The sampling rate of the device is 11 Hz and the wavelengths are 730 nm and 850 nm. An fNIRS cap with 49 channels was placed on each participant’s scalp to collect the data of the participant’s brain functions when he/she conducted the task. Both the device and the cap were the same ones used in Wang and Yang’s study [17, 60]. The initial processing of fNIRS signals was conducted through NirSpark, a software of NirSmart. This preprocessing pipeline employed a three-stage approach: (1) Motion artifacts’ influence in the optical density trajectories was mitigated through cubic spline interpolation-based differential computation; (2) physiological interference (e.g., heart beating, respiratory movements and blood pressure fluctuations) was suppressed by a band-pass filter with a passing frequency range of 0.01–0.2 Hz; and (3) according to the modified Beer–Lambert law, target indices (relative changes in HbO and HbR) were calculated from the optical density signals. In the modified Beer–Lambert equations, the value of differential path factor is set to be 6 for wavelengths of 730 nm and 850 nm. Design of the quasi-experiment and the tasks Experimental design The quasi-experiment was designed as shown in Figure 1.Participants in Group 1 directly conducted the word comprehension task (baseline task group), participants in Group 2 conducted the word comprehension task 5 minutes after they implemented the Group Embedded Figures Test (GEFT) task, and participants in Group 3 conducted the word comprehension task 5 minutes after completing a mathematical calculation task. The words required to be comprehended by participants were displayed on a monitor. Participants were required to complete the task sequentially while wearing an fNIRS cap when he/she carried out the task. Each word was at most displayed for 5 seconds. Participants would press a button to switch to the next word if they finished comprehending a word within 5 s; otherwise, the screen would automatically switch to the next word. Group Embedded Figures Test (GEFT) This study adapted the CL residual assumption proposed by [17]; therefore, the GEFT from Wang et al.’s study was also used here as a non-language task. Field independence, subsumed under spatial ability (for a review, see [62]), and GEFT is a commonly used tool to reveal learners’ visuospatial abilities in two dimensions [63]. As mentioned earlier, visuospatial recognition is a vitally important ability for native Chinese speakers in character recognition. Therefore, for Chinese L2 learners, the GEFT represents one of the most commonly and frequently confronted tasks in educational settings. The evaluation adhered to the protocol outlined in the GEFT manual [64]. Participants in Group 2 completed the GEFT. Mathematical calculation task The mathematical calculation task was also derived from Wang et al.’s study [17], which is a test focusing on learners’ mathematical thinking on calculation. In order to disassociate this process from spatial recognition, the task only employs addition and multiplication as its basic operations. Therefore, participants in Group 3 were presented with 3 multiplication problems: 1234×5678; 3333×6666; 9999×7777. They were required to solve each problem within 2 minutes. The calculation task represents another type of the most confronted tasks for Chinese L2 learners in school. Word comprehension task This study used a word-picture mapping task (as shown in Figure 2) to represent the word comprehension process of participants. In this task, a common animal name in L2 was displayed at the top of the screen, with two candidate animal images presented at the bottom. Participants were required to respond using a keyboard. The reason of using word-picture mapping task instead of word-word translation task is that the former directly links the lexical layer of L2 to the concept layer that is shared both by L1 and L2 [45] to maximally avoid applying equivalent L1 translations as mediation. FNIRS data collection and processing Figures 3 (a)-(c) illustrate the location of the 22 infrared light sources, 16 detectors, as well as the 49 fNIRS channels on the fNIRS cap, and the arrangement of the sources and detectors is in line with [60]. The 49 fNIRS channels were divided into 22 brain regions basically according to Brodmann’s division, covering Broca’s area, Wernicke’s area, the central executive control area, the motor related area, and other brain areas that are important for language processing. The division of the 22 brain regions with further details provided in the Appendix. The brain regions of interest (ROIs) of this study are defined as follows: the PFC (channels F, G, H, I), the left MTG and STG (B1), the left fusiform gyrus (E1), the right MTG and STG (B2), and the right fusiform gyrus (E2), as shown in Figures 3 (d)-(f). Given that many previous studies have demonstrated that a rise in HbO concentration is more sensitive than a drop in HbR, as an indicator of brain activation, this study used relative HbO concentration changes to determine whether an fNIRS channel was significantly activated. The definitions of baseline HbO data, task-related HbO data, as well as the computation of functional connectivity between each pair of fNIRS channels followed the same research paradigm described in [17] and [60]. Results Behavioural performance comparisons among three groups Neurological comparisons among the three groups in word comprehension task Brain activation comparison The significant activation of the brain regions (after false discovery rate, RDF correction) regarding three different groups was displayed in Figure 4. A one-way ANOVA about participants’ word comprehension performance and response time in word comprehension was conducted among the three groups, as shown in Table 1. Both word comprehension performance and response time of participants showed no significant differences among the three groups. Table 1. One-way ANOVA of the learners’ word comprehension performance among the three groups Base-line (group 1) GEFT (group 2) Calculation (group 3) M SD M SD M SD Sig. Ƞ2 Word comprehension performance (score) 16.67 1.88 17.8 1.58 17.7 1.42 0.96 0.00 Response time (seconds) 43.47 36.35 43.73 65.52 44.92 44.53 0.78 0.01 Note. Ƞ 2 =.01 (small effect); Ƞ 2 =.06 (medium effect); Ƞ 2 =.14 (large effect); An ANOVA of the HbO concentration with False Discovery Rate (FDR) correction in different interested brain functions among the three groups was conducted. As shown in Table 2, there were no significant differences in learners’ HbO concentration regarding 5 different ROIs among the three groups. Table 2. An ANOVA of the learners’ HbO concentration regarding 5 brain ROIs in word comprehension task among the three groups Base-line (group 1) GEFT (group 2) Calculation (group 3) M SD M SD M SD Sig. Ƞ2 PFC (F, G,H, I) 0.1247 0.05 0.0971 0.0806 0.2104 0.0739 0.37 0.03 Left MTG and STG (B1) -0.0009 0.0003 0.0023 0.0007 0.0058 0.0005 0.62 0.04 Left fusiform gyrus (E1) 0.0093 0.0012 0.0147 0.0011 0.0148 0.0005 0.81 0.01 Right MTG and STG (B2) 0.0021 0.0002 -0.0016 0.0005 -0.0041 0.0005 0.62 0.02 Right fusiform gyrus (E2) 0.0178 0.0010 0.0064 0.0010 0.0092 0.0006 0.43 0.03 Note. Ƞ 2 =.01 (small effect); Ƞ 2 =.06 (medium effect); Ƞ 2 =.14 (large effect); After conducting ANOVAs within 3 groups among five ROIs, it was found that the participants’ HbO concentration in the PFC (with FDR correction) was significantly higher than that in B1, B2, E1 and E2 in Group 1 and 3 (paired samples t-test; p 0.05). Networks’ synchronisation comparison Figure 5 illustrates the learners’ brain functions synchronisation within and between the PFC (F1-I2), B1, B2, E1, E2 and other brain regions in Group 1 (baseline task group), Group 2 (word comprehension after GEFT), and Group 3 (word comprehension after GEFT). In order to clarify the altered synchronisations among these ROIs among the three different groups, an one-way ANOVA of the Pearson correlations among the PFC, B1, B2, E1 and E2 were performed. It was found that the synchronisation among all brain regions within the PFC was significantly enhanced in Group 2 ( p =0.00, F (2,81)=5.76) compared with that in Group 1 and 3 (posthoc LSD). Additionally, the synchronisations between the PFC and other brain regions in Group 2 were significantly higher than that in Group 1 (two samples t-test; p =0.03, one tailed). Significantly enhanced synchronisations among PFC, B1, B2, E1 and E2 in Group 2 ( p =0.02, F (2,27)=4.29) were found compared with those in Group 1 and 3 ( post hoc LSD). Modelling differential brain function mechanisms regarding the same L2 comprehension task By using prefrontal cortex (F, G,H, I), B1, B2, E1, E2, as well as the interactive effects among PFC and other brain regions, i.e., PFC*B1, PFC*B2, PFC*E1, and PFC*E2, as the variables, and participants’ word comprehension performance (S) and response time (T) as the regression targets for three groups, the details regarding regression models are listed in Table 3, Table 4 and Figure 6. Table 3. Multiple regression models targeted at word comprehension performance (S) with 5 independent variables and 4 interactive variables. Variables Base-line task group (Group 1) GEFT group (Group 2) Calculation group (Group 3) Standard coefficient p Multiple regression model 1 Standard coefficient p Multiple regression model 2 Standard coefficient p Multiple regression model 3 PFC -0.70 0.01 Adjusted R 2 =0.50 p = 0.04 b Durbin-Watson=2.10 95% CI=[14.86, 21.75] -0.50 0.50 Adjusted R 2 =0.13 p =0.34 b Durbin-Watson=1.91 95% CI=[16.12, 20.97] -1.18 0.05 Adjusted R 2 =0.06 p =0.42 b Durbin-Watson=2.27 95% CI=[15.91,19.42] B1 0.71 0.12 1.35 0.12 -0.43 0.36 B2 0.02 0.95 -1.32 0.24 0.18 0.64 E1 -0.19 0.95 -0.85 0.12 -0.34 0.64 E2 0.33 0.57 0.86 0.22 0.15 0.79 PFC*B1 -0.54 0.24 0.03 0.98 0.31 0.57 PFC*B2 -0.86 0.06 0.12 0.90 -0.72 0.17 PFC*E1 0.19 0.71 1.18 0.36 2.52 0.10 PFC*E2 0.99 0.07 -1.14 0.08 -1.29 0.26 As illustrated in Table 3, Table 4 and Figure 6, only the linear multi-regression model targeted at (S) for the baseline task (Group 1) was significant (Durbin-Watson=2.10; p =0.04 b ). In this significant linear model, activity in the PFC ( p =0.01) was significantly and negatively correlated with the learners’ score (S) with adjusted R 2 of 0.50. This result indicates that a higher HbO concentration in the PFC is associated with lower performance, suggesting that a lower cognitive load may enhance learners’ comprehension. Table 4. Multiple regression models targeted at word comprehension performance (T) with 5 independent variables and 4 interactive variables. Variables Base-line task group (Group 1) GEFT group (Group 2) Calculation group (Group 3) Standard coefficient p Multiple regression model 1 Standard coefficient p Multiple regression model 2 Standard coefficient p Multiple regression model 3 PFC -0.24 0.53 Adjusted R 2 =-0.22 p =0.77 b Durbin-Watson=1.91 95% CI=[1555.21,2043.62] -0.56 0.53 Adjusted R 2 =-0.29 p =0.83 b Durbin-Watson=2.31 95% CI=[1458.84,2111.82] 0.59 0.27 Adjusted R 2 =0.17 p =0.29 b Durbin-Watson=1.63 95% CI=[1410.56,2272.38] B1 1.07 0.13 -0.58 0.56 0.18 0.67 B2 -0.21 0.63 0.40 0.76 -0.37 0.31 E1 -0.19 0.72 -0.14 0.82 -0.35 0.61 E2 0.08 0.88 1.01 0.34 0.22 0.69 PFC*B1 -0.08 0.91 0.92 0.54 -0.01 0.98 PFC*B2 -0.39 0.55 -0.18 0.88 -0.13 0.79 PFC*E1 -0.53 0.96 -0.84 0.59 0.18 0.89 PFC*E2 0.15 0.19 0.13 0.86 -0.75 0.48 However, this significant influence of PFC activation on English word comprehension was not observed in Group 2 and 3, likely due to the effects of previously conducted tasks (GEFT and calculation). Although the regression model for Group 3 yielded an almost identical linear relationship to that of Group 1 in a two-dimensional space, where principal component analysis (PCA) transformed 9 variables as the horizontal coordinate and the word comprehension score as the vertical coordinate, the regression model for Group 3 was statistically insignificant ( p =0.42, R 2 =0.06). Discussion RQ1. Brain functions surrounding the PFC may serve as an individualised signature of the CL afforded by L2 readers Firstly, we address whether the brain functions surrounding the PFC can be taken as a signature of the cognitive load induced in an authentic English word comprehension task. This study’s results regarding Group 1 support this assumption. The multiple regression model for the baseline task group displayed a significant negative correlation between readers’ PFC function and comprehension performance. Meantime, other brain regions of interest, i.e., the STG, MTG, and fusiform gyrus, showed varied impacts on participants’ L2 word comprehension. By controlling task’s difficulty (with participants in all three groups comprehending the same set of English words), learners’ individualised L2 proficiency can be reflected by their PFC-centred HbO concentrations. Numerous studies have reported that outcomes linked to PFC function, which is indicative of working memory (WM), are highly correlated with overall English language competences, such as reading, listening, speaking and the use of English (e.g., [65]). For example, Malone [66] concluded in his study that WM outcomes were correlated with vocabulary performance derived from reading to form recognition. However, the lexical network of L2 readers is highly dynamic and individualised [67], leading Chinese EFL readers to employ varied strategies stemming from their L1 when comprehending English words. Such strategies may include “directly mapping L2 print to its meaning,” “mapping L2 print to its meaning via the word’s phonological representation,” and “mapping L2 print to its meaning via its equivalent L1 translation”, etc. All those strategies can be summarised in terms of paths between L2 and L1 in a lexical network, each characterised by different linking strengths, path lengths, and transfer powers (for a review, see [68]). Therefore, the diverse strategies employed by different readers may involve distinct brain functions and give rise to different CL constructions. The varied impacts of the STG, MTG, and fusiform gyrus on participants’ comprehension scores observed in this study further support this hypothesis. Nevertheless, irrespective of the variability in the cognitive processes employed by L2 readers, retrieving word meaning via the PFC appears to be a compulsive final step for all readers. That is, while the PFC’s function may not be a highly correlated factor in EFL readers’ long-term achievement, it is crucial for readers’ instant comprehending activities. Given that working memory (WM) acts as a container with limited capacity for the cognitive processes required in online L2 processing [69-70], it is reasonable to use PFC activation to quantify the instant CL experienced by readers during a reading task under WM constraints [69]. The multiple regression model regarding the baseline task group further supports this view by illustrating a significant and negative correlation between readers’ PFC function and comprehension performance. Our findings in Group 1 are also in line with many existing neurological studies. For instance, Wang and Aryadoust [39] reported that EFL participants generally exhibited higher extraneous cognitive load than their L1 counterparts, i.e., for L2 readers, lower PFC activation is associated with higher language competence. However, the instant CL mediated by task difficulty remains variable. For example, Soares et al. [71] revealed a positive relationship between the using of PFC and the difficulty of a recall task, suggesting that the more elements (n) required to be recalled (i.e., higher CL), the greater the activation of the PFC; while in Wang and Aryadoust ’s study [39], there was no significant difference in participants’ PFC activation between positive words (task with high CL) and negative words (task with low CL). This variability in instant CL will be discussed in the next section. RQ2. The unsteadiness of the instant CL can be explained as CL residual inherited from previous tasks The former section has confirmed the signature role of PFC-centred brain functions that plays in representing EFL readers’ instant CL, and in this section we discuss the unsteadiness of the instant CL based on the CL residual assumption proposed by [17]. Although the overall PFC activation among the three groups showed no significant difference, initially suggesting that there was no CL residual inherited from previous tasks, a closer examination of the PFC-centred networks among the three groups revealed that readers’ PFC centred network in Group 2 was significantly altered compared with those in Groups 1 and 3. This finding suggests that previously conducted tasks changed readers’ brain function mechanism regarding L2 word comprehension when all groups performed an exactly same task. Additional powerful evidence comes from the comparison of the regression models among the three groups. The relationship between PFC-centred functions and learners’ performance, observed in Group 1, disappeared in the other two groups. Significantly enhanced synchronisations among the PFC, B1, B2, E1 and E2 in Group 2 were found compared with those in Groups 1 and 3, as readers’ PFC activation in Group 2 overly decreased with an insignificant level. Although this finding is consistent with [17], the results cannot lead to the conclusion that readers’ cognitive load decreased, as the regression model regarding Group 2 lost its significance (R 2 =0.13) and the activation of PFC among the three groups also showed no significant differences. Thus, we prefer to describe this phenomenon as a negative biological reaction reflected by readers in Group 2 to the word comprehension, potentially due to the CL residual influence from the previously conducted GEFT task. This phenomenon is similar to the “go on strike” effect found by [36], where the brain ceases effective processing of information when overwhelmed by cognitive load induced by the task. A disrupted relationship between PFC-centred functions and readers’ performance was also observed in Group 3, in which the R 2 value of the regression model dropped to 0.06, although the alteration of brain functions was less pronounced than that in Group 2. In the regression model regarding Group 2, although the PFC remained a negative influence on learners’ comprehension scores, the effect from the PFC was no longer significant, suggesting that comprehension performance in Group 2 was no longer dominated by the cognitive load represented by the PFC. Therefore, the results of this study verify the existence of a cognitive-load residual from previously conducted tasks, however, in the form of disturbing the function of PFC. In Groups 2 and 3, the PFC does not reliably represent learners’ L2 vocabulary proficiency, suggesting that readers with lower proficiency also exert lower mental effort (i.e., inactivation of the PFC) in L2 word comprehension. RQ3. The CL residual cannot be linearly modelled by alterations of relative brain functions Finally, we would like to discuss whether the unsteadiness of the instant CL can be modelled through the brain function changes surrounding the PFC. The answer is negative. The results found in this study suggest that the brain functions surrounding the PFC in Groups 2 and 3 underwent nonlinear transformations due to previously conducted tasks, rendering the PFC an unreliable indicator of CL signature in an authentic reading context. We cannot find an explanatory model to describe the working mechanism of the CL residuals, and at the same time accommodate PFC alterations within the same framework. This study employed a primary task as a baseline benchmark to model participants’ instant CL with an fNIRS device to collect physiological data. Task-performance-based methods, such as calculating the number of elements that comprise a task, provide an intuitive yet simplistic way to map task complexity to the CL produced [2], and typically adopt response time or accuracy of learners’ performance on tasks as indices of the CL generated by the relative tasks (e.g., see [72]). The task-based CL measuring methods are based on a naïve prerequisite, i.e., the greater the mental effort devoted to a task, the lower the performance achieved. However, as pointed by [73], changes of the workload cannot always be simply and linearly mapped to participants’ performance. Similarly, our study found that the altered brain functions of readers in Groups 2 and 3 could not be linearly mapped to their unchanged comprehension performance. There is no available model that both can accommodate the PFC-centred functions in L2 comprehension (Group 1) and the altered brain functions due to the previously conducted tasks (Groups 2 and 3) within the same explanatory framework. This problem may be solved in the future by employing nonlinear models with explanatory components. Conclusion This study explored the possibility of monitoring and modelling Chinese EFL readers’ instant CL in their L2 words comprehension tasks via observing participants’ PFC-centred brain function changes. The results of this study conclude that the brain functions surrounding the PFC represent an individualised and highly unstable signature of the instant CL experienced by L2 readers. This study also found that the unsteadiness of the instant CL appears to be due to CL residual inherited from previous tasks, as revealed by significantly altered brain functions surrounding the PFC; finally, the CL residual cannot be linearly modelled, but future research may explore the possibility of employing nonlinear models to better explain this phenomenon. Declarations Ethics approval and consent to participate We obtained all participants’ consents, and this study was approved by the Research Ethics Review Committee of Sichuan Normal University. Availability of data and materials Stimuli, datasets and analysis code are available through a public OSF repository: https://osf.io/b7en5/ Competing interests Authors declare that there is no competing interests. Consent for publication Not applicable Funding This research is supported by the Major Science and Technology Special Program of Jiangsu Province [BG2024025]; Research project of Ministry of Education of China[23YJC880062]; Teaching reform and research project of Sichuan Normal University [JWC20240107; JWC20240116]. Authors contributions Wang, L., Yang, J. and Wang, D. F. conducted and administrated the experiment, and wrote the manuscript; Zhang, W., Yang, P. L., and Duan, Y. H. processed the fNIRS data and prepared the related figures and tables; Sun, B. and Chen, D. designed the experiment; Liang, Z. J. and Zhang, Y. L. processed the behavioural data and prepared the related figures and tables. Acknowledgement Not applicable. References Sweller, J., Cognitive load during problem solving: Effects on learning. Cognitive Science, 1988. 12 : p. 257-285. Sweller, J., Cognitive load theory and educational technology. Etr&D-Educational Technology Research and Development, 2020. 68 (1): p. 1-16. Sweller, J., et al., Cognitive load and selective attention as factors in the structuring of technical material. Journal of Experimental Psychology, 1990. General (119): p. 176-192. Sweller, J. and F. Paas, Should self-regulated learning be integrated with cognitive load theory? A commentary. Learning and Instruction, 2017. 51 : p. 85-89. Kennedy, M.J., et al., Improving teacher candidates' knowledge of phonological awareness: A multimedia approach. Computers & Education, 2013. 64 : p. 42-51. Khalil, M.K. and I.A. Elkhider, Applying learning theories and instructional design models for effective instruction. Advances in Physiology Education, 2016. 40 (2): p. 147-156. Ngu, B.H. and H.P. Phan, Learning linear equations: capitalizing on cognitive load theory and learning by analogy. International Journal of Mathematical Education in Science and Technology, 2022. 53 (10): p. 2686-2702. Nyachwaya, J.M. and M. Gillaspie, Features of representations in general chemistry textbooks: a peek through the lens of the cognitive load theory. Chemistry Education Research and Practice, 2016. 17 (1): p. 58-71. Offerdahl, E.G., J.B. Arneson, and N. Byrne, Lighten the Load: Scaffolding Visual Literacy in Biochemistry and Molecular Biology. Cbe-Life Sciences Education, 2017. 16 (1). Duffy, P.L., et al., Form versus Function: A Comparison of Lewis Structure Drawing Tools and the Extraneous Cognitive Load They Induce. Journal of Chemical Education, 2019. 96 (2): p. 238-247. Parong, J. and R.E. Mayer, Cognitive and affective processes for learning science in immersive virtual reality. Journal of Computer Assisted Learning, 2021. 37 (1): p. 226-241. Kim, J. and I.-H. Jo, Feasibility and use of psychophysiological responses based on cognitive load theory. Australasian Journal of Educational Technology, 2019. 35 (3): p. 150-165. Kruger, J.-L. and S. Doherty, Measuring cognitive load in the presence of educational video: Towards a multimodal methodology. Australasian Journal of Educational Technology, 2016. 32 (6): p. 19-31. Marinescu, A.C., et al., Physiological Parameter Response to Variation of Mental Workload. Human Factors: The Journal of the Human Factors and Ergonomics Society, 2017. 60 (1): p. 31-56. Midha, S., et al., Measuring Mental Workload Variations in Office Work Tasks using fNIRS. International Journal of Human - Computer Studies, 2021. 147 : p. 102580. Al-Samarraie, H., et al., How reading in single- and multiple-column types influence our cognitive load: an EEG study. Electronic Library, 2019. 37 (4): p. 593-606. Wang, L., et al., Influence of high-level mathematical thinking on L2 phonological processing of Chinese EFL learners: Evidence from an fNIRS study. Thinking Skills and Creativity, 2023. 47 : p. 101242. Baddeley, A., The episodic buffer: a new component of working memory? Trends in cognitive sciences, 2000. 4 (11): p. 417-423. Wickens, C.D., Multiple Resources and Mental Workload. Hum Factors, 2008. 50 (3): p. 449-455. Paas, F., Training strategies for attaining transfer of problem-solving skill in statistics: A cognitive-load approach. Journal of Educational Psychology,, 1992. 84 : p. 429–434. Cao, A., et al., NASA TLX: software for assessing subjective mental workload. Behav Res Methods, 2009. 41 (1): p. 113–117. Hart, S.G., NASA-task load index (NASA-TLX); 20 years later , in Human factors and ergonomics society annual meeting . 2006, Sage Los Angeles. p. 904–908. Hart, S.G. and L.E. Staveland, Development of NASA-TLX (task load index): Results of empirical and theoretical research. Advances in psychology, 1988. 52 : p. 139–183. Amadieu, F., C. Mariné, and C. Laimay, The attention-guiding effect and cognitive load in the comprehension of animations. Computers in Human Behavior, 2011. 27 (1): p. 36-40. Gaillard, A., Comparing the concepts of mental load and stress. Ergonomics, 1993. 36 (9): p. 991–1005. Ellerton, P., On critical thinking and content knowledge: A critique of the assumptions of cognitive load theory. Thinking Skills and Creativity, 2022. 43 . Derry, J., A Problem for Cognitive Load Theory-the Distinctively Human Life-form. Journal of Philosophy of Education, 2020. 54 (1): p. 5-22. Ferraz Lorenzo, M. and C. Machado Trujillo, Cognitive Processes, ICT, and Education: A Critical Analysis. Computers in the Schools, 2018. 35 (3): p. 186-203. Bernhardt, K.A., et al., The effects of dynamic workload and experience on commercially available EEG cognitive state metrics in a high-fidelity air traffic control environment. Applied Ergonomics, 2019. 77 : p. 83-91. Chan, A.S., et al., Photoneuromodulation makes a difficult cognitive task less arduous. Scientific Reports, 2021. 11 (1). Zhuang, C., et al., Scale invariance in fNIRS as a measurement of cognitive load. Cortex, 2022. 154 : p. 62-76. Alyan, E., N.M. Saad, and N. Kamel, Effects of Workstation Type on Mental Stress: fNIRS Study. Human Factors, 2021. 63 (7): p. 1230-1255. Csipo, T., et al., Increased cognitive workload evokes greater neurovascular coupling responses in healthy young adults. Plos One, 2021. 16 (5). Fishburn, F.A., et al., Sensitivity of fNIRS to cognitive state and load. Frontiers in Human Neuroscience, 2014. 8 : p. 76. Mandrick, K., et al., Neural and psychophysiological correlates of human performance under stress and high mental workload. Biological Psychology, 2016. 121 : p. 62-73. Causse, M., et al., Mental workload and neural efficiency quantified in the prefrontal cortex using fNIRS. Scientific Reports, 2017. 7 (1): p. 5222. Midha, S., et al., Measuring Mental Workload Variations in Office Work Tasks using fNIRS. International Journal of Human-Computer Studies, 2021. 147 . Karmakar, S., et al., Real time detection of extraneous cognitive load using fNIRS: A deep learning approach. Biomedical Signal Processing and Control, 2023. 80 : p. 104227. Wang, X. and V. Aryadoust, An eye-tracking and neuroimaging study of negative wording effects on cognitive load in a metacognitive awareness tool. System, 2025. 128 : p. 103552. Diao, Y. and J. Sweller, Redundancy in foreign language reading comprehension instruction: Concurrent written and spoken presentations. Learning and Instruction, 2007. 17 : p. 78-88. Moussa-Inaty, J., P. Ayres, and J. Sweller, Improving Listening Skills in English as a Foreign Language by Reading Rather than Listening: A Cognitive Load Perspective. Applied Cognitive Psychology, 2012. 26 (3): p. 391-402. Dijkstra, T. and W.J. Van Heuven, The architecture of the bilingual word recognition system: From identification to decision. Bilingualism: Language and cognition, 2002. 5 (3): p. 175-197. Kroll, J.F., et al., The revised hierarchical model: A critical review and assessment. Bilingualism: Language and Cognition, 2010. 13 : p. 373–381. De Groot, A.M.B., Bilingual lexical representation: A closer look at conceptual representations , in Orthography, phonology, morphology, and meaning , R. Frost and L. Katz, Editors. 2012, Elsevier. p. 389–412. Van Hell, J.G. and A.M.B. De Groot, Conceptual representation in bilingual memory: Effects of concreteness and cognate status in word association. Bilingualism: Language and Cognition, 1998. 1 : p. 193–211. Tan, L.H., et al., Neural systems of second language reading are shaped by native language. Human brain mapping, 2003. 18 (3): p. 158-166. Donald, J.B., A.P. Charles, and S. Walter, Cross-cultural effect on the brain revisited: Universal structures plus writing system variation. Hum Brain Mapp, 2005. 25 (1): p. 92-104. Dietz, N.A.E., et al., Phonological decoding involves left posterior fusiform gyrus. Hum. Brain Mapp., 2005. 26 : p. 81–93. Mesgarani, N., et al., Phonetic Feature Encoding in Human Superior Temporal Gyrus. Science, 2014. 343 (6174): p. 1006–1010. Yi, H.G., M.K. Leonard, and E.F. Chang, The encoding of speech sounds in the superior temporal gyrus. Neuron, 2019. 102 (6): p. 1096–1110. Ramos Nuñez, A.I., et al., The role of left vs. right superior temporal gyrus in speech perception: An fMRI-guided TMS study. Brain and Language, 2020. 209 : p. 104838. Maguinness, C., C. Roswandowitz, and K. von Kriegstein, Understanding the mechanisms of familiar voice-identity recognition in the human brain. Neuropsychologia, 2018. 116 : p. 179–193. Zhang, J., J. Chen, and G.S. Ding, Universality and language specificity of brain reading networks: A developmental perspective. Developmental Science, 2023. in press . Henseler, I., et al., Modulating brain mechanisms resolving Lexico-semantic interference during word production: a transcranial direct current stimulation study. Cognitive Neurosci, 2014. 26 : p. 1403–1417. Lau, E., C. Phillips, and D. Poeppel, A cortical network for semantics: (de)constructing the N400. Nat Rev Neurosci 2008. 9 : p. 920–933. Turken, A.U. and N.F. Dronkers, The neural architecture of the language comprehension network: converging evidence from lesion and connectivity analyses. Front. Syst. Neurosci., 2011. 5 : p. 1. Di Liberto, G.M., E.C. Lalor, and R.E. Millman, Causal cortical dynamics of a predictive enhancement of speech intelligibility. NEUROIMAGE, 2018. 166 : p. 247-258. Tuennerhoff, J. and U. Noppeney, When sentences live up to your expectations. NeuroImage, 2016. 124 : p. 641–653. Gottardo, A., et al., Models of English and Chinese word reading for adolescent Chinese–English bilinguals. Reading and Writing, 2017. 30 : p. 1377–1406. Yang, J., et al., Mobile application-based phonetic training facilitates Chinese-English learners’ learning of L2. Learning and Instruction, 2024. 93 : p. Article number 101967. Cao, F., et al., Similarities and differences in brain activation and functional connectivity in first and second language reading: Evidence from Chinese learners of English. NEUROPSYCHOLOGIA, 2014. 63 : p. 275-284. Rittschof, K.A., Field dependence–independence as visuospatial and executive functioning in working memory: implications for instructional systems design and research. Education Tech Research Dev, 2010. 58 : p. 99–114. Boccia, M., et al., Does field independence predict visuo-spatial abilities underpinning human navigation? Behavioural evidence. Experimental Brain Research, 2016. 234 (10): p. 2799-2807. Witkin, H.A., et al., A manual for the Embedded Figures Tests , M. Park, Editor. 1971, Mind Garden Inc.: CA. Kormos, J. and A. Safar, Phonological Short-Term Memory, Working Memory and Foreign Language Performance in Intensive Language Learning. Bilingualism Language & Cognition, 2008. 11 (2): p. 261-271. Malone, J., Incidental vocabulary learning in SLA: Effects of frequency, aural enhancement, and working memory. Studies in Second Language Acquisition, 2018. 40 (3): p. 651-675. Pan, X. and D. Jared, Cross-language activation of culture-specific features in Chinese–English bilinguals. J Cult Cogn Sci, 2021. 5 : p. 325–344. Malt, B.C., et al., Bidirectional lexical interaction in late immersed Mandarin-English bilinguals. Journal of Memory and Language. Journal of Memory and Language, 2015. 82 : p. 86–104. Ellis, N.C., At the interface: Dynamic interactions of explicit and implicit language knowledge. Studies in second language acquisition, 2005. 27 (2): p. 305-352. Sweller, J., Instructional Design Consequences of an Analogy between Evolution by Natural Selection and Human Cognitive Architecture. Instructional Science, 2004. 32 : p. 9-31. Soares, S.M.P., et al., Brain correlates of attentional load processing reflect degree of bilingual engagement: Evidence from EEG. NeuroImage, 2024. 298 : p. 120786. Blissett, S., et al., Optimizing self-regulation of performance: is mental effort a cue? Advances in Health Sciences Education, 2018. 23 (5): p. 891-898. Leppink, J. and P. Pérez-Fuster, Mental Effort, Workload, Time on Task, and Certainty: Beyond Linear Models. Educ Psychol Rev, 2019. 31 : p. 421–438. Additional Declarations No competing interests reported. Supplementary Files Appendix.docx 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-6407039","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":451752487,"identity":"452a7324-7660-4448-a3c4-40359d054649","order_by":0,"name":"Ling Wang","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Ling","middleName":"","lastName":"Wang","suffix":""},{"id":451752488,"identity":"e5fa0cbe-57eb-45b2-a7d4-ecd21c57c3d4","order_by":1,"name":"Juan Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYBAC9h4wxQbEzAcYEiCCBni1MCK0sCWQpAUEeOAqCWk5YybB2MaX2D+75/OHBzV3EhvYm7dJMNTcwa2lty0NqIUtccads9skEo49S2zgOVYmwXDsGW4t/czHwFoabuRuY0hsOJzYIJEDtLfhMB4tjG1gLfNv5Dz+ANYi/wa/FsHeZogtG27kMEhAbOHBr0Wa51iyBcM5NuONN9LMgH45bNzGk1ZsAWTg1MLHk2N4g6HsmOy8G8mPP/6oOSzbz354440PNbi1AAGL9F+2YwguKFZhcYoLMH9g+FODX8koGAWjYBSMbAAAWzFWu3C6kiwAAAAASUVORK5CYII=","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":true,"prefix":"","firstName":"Juan","middleName":"","lastName":"Yang","suffix":""},{"id":451752489,"identity":"db53c7f4-2227-4def-b4ca-5e92ca82368d","order_by":2,"name":"Wei Zhang","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Zhang","suffix":""},{"id":451752490,"identity":"9b7aced8-93ad-4179-860e-cf15cd18b2ff","order_by":3,"name":"Peilin Yang","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Peilin","middleName":"","lastName":"Yang","suffix":""},{"id":451752491,"identity":"8f8805a4-1571-491f-91c7-791741b528ae","order_by":4,"name":"Yihong Duan","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Yihong","middleName":"","lastName":"Duan","suffix":""},{"id":451752492,"identity":"1d5ed73b-5bcb-403e-8b06-4298cdb5c04d","order_by":5,"name":"Bo Sun","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Sun","suffix":""},{"id":451752493,"identity":"53093866-dad1-42ba-99b6-eacf7906d114","order_by":6,"name":"Zhijie Liang","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Zhijie","middleName":"","lastName":"Liang","suffix":""},{"id":451752494,"identity":"5f7e1d40-663a-4d1b-b9fc-348cf78dc81e","order_by":7,"name":"Yuexiang Gao","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Yuexiang","middleName":"","lastName":"Gao","suffix":""},{"id":451752495,"identity":"a60e5342-ba69-4abe-b8fc-aa178db8477e","order_by":8,"name":"Yangli Zhang","email":"","orcid":"","institution":"Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Yangli","middleName":"","lastName":"Zhang","suffix":""},{"id":451752496,"identity":"ab96567a-9016-4822-a82c-a650af8ee66a","order_by":9,"name":"Die Chen","email":"","orcid":"","institution":"Experimental School affiliated to Sichuan Normal University","correspondingAuthor":false,"prefix":"","firstName":"Die","middleName":"","lastName":"Chen","suffix":""},{"id":451752497,"identity":"1663a182-5b29-44a0-9463-ccdee346e7a2","order_by":10,"name":"Daifa Wang","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Daifa","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-04-09 01:53:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6407039/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6407039/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82143323,"identity":"39503584-8262-44dc-9b2e-bce949e68973","added_by":"auto","created_at":"2025-05-07 06:41:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":152962,"visible":true,"origin":"","legend":"\u003cp\u003eThe process of the quasi-experiment\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/7dd60891856d10a542a113e3.png"},{"id":82143325,"identity":"7e2f1460-fc80-4079-8b56-dbac91a4e7b8","added_by":"auto","created_at":"2025-05-07 06:41:08","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":305606,"visible":true,"origin":"","legend":"\u003cp\u003eAn illustration of word comprehension test\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/201243065ec70b2827d127ca.jpeg"},{"id":82143326,"identity":"7461f765-0f99-45ec-a21a-5d27b777cb2e","added_by":"auto","created_at":"2025-05-07 06:41:08","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":607712,"visible":true,"origin":"","legend":"\u003cp\u003efNIRS measurement images. (a) , (b) and (c) show the locations of the light sources, detectors, as well as 49 fNIRS channels from the front, left and right hemisphere angles, respectively [60]; (d), (e) and (f) illustrate 5 defined ROIs in the front, left- and right temporal cortices, respectively.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/33c735979136d4cafd0a4e9c.jpeg"},{"id":82143330,"identity":"b4cf2be7-3a0e-423f-841a-db3b809ee4c3","added_by":"auto","created_at":"2025-05-07 06:41:08","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1132769,"visible":true,"origin":"","legend":"\u003cp\u003eA comparison of the significantly activated brain regions among three groups\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/7df441cc9f19d9811b998a69.png"},{"id":82146753,"identity":"04a15ad0-63b2-401b-ab9f-eb22d544149b","added_by":"auto","created_at":"2025-05-07 06:57:08","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1177028,"visible":true,"origin":"","legend":"\u003cp\u003eActivation correlations among different brain regions in 3 groups\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/d590077b11ae0b1dd1a2fdef.png"},{"id":82146754,"identity":"1a5e4cee-b284-4223-9e9a-0829dae4a713","added_by":"auto","created_at":"2025-05-07 06:57:08","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":328725,"visible":true,"origin":"","legend":"\u003cp\u003eLinear illustration of the multiple regression models\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/35abe431009edae4badec6b7.jpeg"},{"id":89893050,"identity":"de1dd683-cbac-435b-9a0b-6c3f4e5d629b","added_by":"auto","created_at":"2025-08-26 07:54:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5266897,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/ffa6002a-70c9-40a2-87af-e1ba90818272.pdf"},{"id":82143327,"identity":"bd32e6a8-f571-4b7c-abe5-fab232a911aa","added_by":"auto","created_at":"2025-05-07 06:41:08","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":927989,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-6407039/v1/f46e3441066563e34440a329.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Cognitive-load residual is found in Chinese EFL readers’ word comprehension","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCognitive load (CL), a term which is also referred to by many other studies as mental effort or mental workload, has a long research history. Sweller and Paas [1-4] introduced the concept within \u0026nbsp;the context of education and learning, characterising it as a 3-dimensional construct \u0026nbsp;comprising intrinsic load, extraneous load, and germane load. CL theory has since guided \u0026nbsp;teachers and educational practitioners to develop pedagogical designs in terms of how to facilitate learners\u0026rsquo; learning via \u0026nbsp;modulating learners\u0026rsquo; cognitive load (e.g., [5-9]). In recent years, over 10,000 papers with high relevance to this topic were published, which highlights the continuing importance of application CL theory in education, as well as the importance of effectively monitoring and measuring cognitive load for learners.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe development of CL measurement techniques has accelerated \u0026nbsp;in recent years, especially \u0026nbsp;through methods that integrate \u0026nbsp;participants\u0026rsquo; physiological data. Physiological data reflects the activity of sympathetic nervous system caused by CL, such as heart rate variability (HRV; e.g., [10]), skin conductance or resistance (e.g. [11]), and eye-tracking related indices: pupil dilation, eye gaze (e.g., [12]) and blink rate (e.g., [13]), as well as facial temperature (e.g., [14]). Unobtrusive brain-function-monitoring devices such as functional near infrared spectroscopy (fNIRS; e.g., [15]) and electroencephalography (EEG; e.g., [16]) have also been used to monitor alterations in brain functions and network synchronisations due to the changed CL. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAlthough cognitive load \u0026nbsp;has been identified as a vital factor in second language (L2) learning, modelling L2 learners\u0026rsquo; cognitive load remains a challenge due to the varied cognitive strategies that may be recruited by learners for L2 recognition and comprehension. This study explores the possibility of revealing and modelling the instability of CL experienced by Chinese readers during English word comprehension, based on the CL residual assumption proposed by [17]. \u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eA review of the related work\u003c/h2\u003e\n\u003ch2\u003eCognitive load models\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eThere are several relative models in addition to the Sweller and Paas\u0026rsquo;s CL model [1-4], such as the working-memory-based model [18] and the mental workload model [19]. As a multidimensional construct, mental workload is generally defined in terms of the cognitive resources that are demanded and competed for in a task due to human\u0026rsquo;s limited online processing ability [19]. Mental effort is characterised as the cognitive capacity that is allocated to accommodate the demands imposed by a given task [20], and sometimes mental effort is taken as a composition of mental workload that a task requires, e.g., NASA-TLX [21-23]. In addition to the terms \u0026ldquo;mental effort\u0026rdquo; and \u0026ldquo;mental workload\u0026rdquo; that are generally treated as synonyms for CL, terms like perceived difficulty [24] and mental stress [25] are likewise used synonymously .\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor decades, the CL theory model has been the subject of continuous study and discussion, particularly concerning its applications in educational contexts, where \u0026nbsp;it provides guidance and evaluation for instructional designs and pedagogical activities. Recent theoretical studies have expanded to a wider range, such as studying the relationship between critical thinking and cognitive load [26], and the relationship between philosophy and cognitive load [27]. CL theory has also been adopted to explore how information and communication technologies reshape \u0026nbsp;people\u0026rsquo;s cognitive structures [28].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHowever, the rapid development of CL measuring methods poses challenges for CL theory. For example, Kruger and Doherty [13] conceptualised cognitive load as a dynamic process which should be assessed with a combination of several measurements. They recommended modelling learners\u0026rsquo; extraneous, intrinsic, germane and instant load with EEG and eye-movement data, while assessing \u0026nbsp;learners\u0026rsquo; overall and average load with psychometric instrument and eye-tracking. However, Bernhardt et al. [29] showed that ocular measures may be more sensitive to increased visual load than EEG workload indices, leading to inconsistencies in CL evaluation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe aforementioned contradictions can be attributed to the high instability of the CL in authentic tasks. Therefore, how to model the unsteadiness of cognitive load is a problem which remains unsolved. This unsteadiness results in significant changes in brain function involvement when \u0026nbsp;brain activity is used as an indicator of the CL \u0026nbsp;imposed by task demands. For example, Wang et al. [17] found that learners\u0026rsquo; brain activation showed significant differences in the same task, which was inferred as a result of the \u0026ldquo;CL residuals\u0026rdquo; inherited from the priorly conducted mathematical tasks. A similar finding was found in Chan et al. [30], in which they found task operators received photoneuromodulation reduced the brain activations related to WM, implying that the previously conducted operating (photoneuromodulation) may interfere participants\u0026rsquo; following tasks\u0026rsquo; implementation without behavioural differences.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eMeasuring CL via monitoring learners\u0026rsquo; brain functions with fNIRS\u003c/h2\u003e\n\u003cp\u003eIn recent years, fNIRS has been explored as a potential method of measuring learners\u0026rsquo; cognitive load by analysing learners\u0026rsquo; brain functions during tasks. Compared with EEG, fNIRS has higher spatial and temporal resolution and is more robust against motion artefacts. Till now, applications of fNIRS in CL modelling can be classified into three different patterns. The dominant research pattern is to observe participants\u0026rsquo; brain function changes based on an n-back task to provide a general quantification of the cognitive load with fNIRS data (e.g., [30-35]). \u0026nbsp;Participants\u0026rsquo; fNIRS data is collected based on the n-back task and the brain function changes are discussed within the working-memory-based model. For example, in \u0026nbsp;Zhuang et al. \u0026rsquo;s study [32], significant differences in brain activation were found among participants who were asked to implement recall tasks with different difficulty level, indicating that higher task difficulty is associated with increased prefrontal cortex (PFC) activation. (e.g., [32-35]).\u003c/p\u003e\n\u003cp\u003eThe second research pattern of using fNIRS to model participants\u0026rsquo; CL is comparing participants\u0026rsquo; hemodynamic (such as blood oxygen) changes in the PFC during the non-n-back tasks to their subjective feelings obtained from scales or questionnaires by using the latter as a calibration tool. It has been suggested that the hemodynamic changes in brain regions associated with working memory (WM) can be directly linked with operators\u0026rsquo; performance. Take Alyan et al.\u0026rsquo;s study [32] for instance, fNIRS was used to monitor participants\u0026rsquo; oxygenated hemoglobin (HbO) changes. The results suggested that participants working at a nonergonomic workstation exhibited a significant association between prefrontal cortex deactivation and \u0026nbsp;stressful conditions. That is, the participants who performed the montreal imaging stress task at a nonergonomic workstation showed a lower HbO concentration in the prefrontal cortex (PFC), longer response time, and higher NASA- TLX scores compared to those working at an ergonomic workstation. However, the relationship between hemodynamic changes in PFC and operating performance is not always significant. In an extreme situation, HbO concentration in the PFC even showed no clear correlation with task performance [36]. \u0026nbsp;Therefore, using subjective reports as a calibration tool can make the CL model based on hemodynamic changes in brain regions more reliable. For example, \u0026nbsp;in [37], participants\u0026rsquo; HbO changes in the PFC were verified with their self-reported mental effort (NASA- TLX) in reading and writing tasks, showing that the HbO changes of PFC only represented CL in the reading task, but not in the writing task.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe third research pattern of measuring CL via fNIRS is to compare participants\u0026rsquo; hemodynamic changes in relative brain regions across \u0026nbsp;parallel non-n-back tasks, with one of the hemodynamic patterns as a benchmark \u0026nbsp;[17, 38-39]. For example, Wang and Aryadoust [39] examined the activation patters of the PFC in English-as-second-language readers by manipulating task difficulty through the nature of L2 words (positive vs. negative). By using learners\u0026rsquo; brain function data from a benchmark task, the workload can also be assessed by observing brain function changes caused by previously conducted tasks [17].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt is noted that most of the studies we mentioned above have focused on participants\u0026rsquo; hemodynamic changes in the PFC to represent \u0026nbsp;instant CL, based on the neurovasuclar coupling principle, which states that active brain regions require increased blood flow to meet enhanced energy demands. The PFC is the brain region associated with executive functions required by all kinds of tasks with online resource requirements. Nevertheless, other interested functional brain regions, such as the middle temporal gyrus [39] and the right ventrolateral cortex [32] that are also important for relative studies, have also been identified as CL components to provide a more comprehensive view of the CL construction in specific tasks.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn sum, although the application of fNIRS in CL modelling is booming in ergonomic and medical research areas, its use in modelling learners\u0026rsquo; CL in educational area remains limited . This study employs the third research pattern to capture and measure the instant CL experienced by Chinese EFL readers when they comprehend English words, considering interference from previously conducted tasks.\u003c/p\u003e\n\u003ch2\u003eMental workload imposed on English-as-foreign-language (EFL) readers\u003c/h2\u003e\n\u003cp\u003eCognitive load is presumed to be a key factor that impacts learners\u0026rsquo; L2 reading comprehension performance. It is assumed that verbal redundancy, where learners are concurrently provided \u0026nbsp;with content, may lead to \u0026nbsp;cognitive overload in L2 reading [40] and L2 listening [41]. However, the construction of CL imposed on EFL reading is controversial, as the process of recognising and comprehending L2 words is not as simple as it is for native speakers.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eModels regarding L2 reading, such as BIA+ model [42], Revised Hierarchical Model (RHM model; [43]), and Distributed Conceptual Feature Model (DCFM; [44-45]), all suggest that readers\u0026rsquo; first language (L1) plays an important role in their L2\u0026rsquo;s comprehension. The BIA+ model \u0026nbsp;suggests that L1 and L2 lexicons are stored together, and \u0026nbsp;are activated simultaneously when L2 stimuli come in. The RHM model assumes that \u0026nbsp;L2 word meaning retrieval requires mediation, which is the equivalent L1 translation [43]. Although the DCFM [44] suggests that the conceptual memory of L1 and L2 words is shared by different languages, the link strengths among different languages and conceptual memory nodes are unstable and can be reshaped according to learners\u0026rsquo; application and learning of L1 and L2. In sum, CL imposed on EFL readers is not only determined \u0026nbsp;by English word comprehension itself, but may also be incurred by the impacts from L1/L2 semantic networks, which are highly dynamic and individualised. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe left medial fusiform gyrus serves as a critical neural substrate for orthographic-to-phonological conversion during reading, specifically mediating the grapheme-phoneme mapping essential for decoding English words [46-47]. Neuroimaging evidence further delineates the functional specialisation of the left posterior fusiform cortex, where Dietz et al. [48] observed task-dependent activation patterns modulated by phonological processing complexity. Concurrently, the superior temporal gyrus (STG; Brodmann Area 22), anatomically situated within Wernicke\u0026rsquo;s area, demonstrates comparable functional significance in English reading. This region orchestrates dual linguistic operations: acoustic-phonetic analysis of speech stimuli [49-50] and articulatory planning for phonological output [51- 52]. Notably, both the left fusiform gyrus and STG exhibit contributions to reading fluency, with the former prioritising visual-linguistic transformation and the latter supporting auditory-oral integration of phonological representations. In addition to the left fusiform gyrus and STG, many researchers \u0026nbsp; have proposed that the middle temporal gyrus (MTG) also plays a vital role in native English speakers\u0026rsquo; reading [53-58]. Left MTG was found to be involved in the processing of lexical- [55-56] and sentence comprehension [57-58].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eChinese readers, whose L1 is non-alphabetic, show a unique pattern in Chinese characters\u0026rsquo; recognition. Unlike alphabetical languages, Chinese characters do not adhere to letter-sound conversion rules [46]. Chinese characters, with their strokes packed into square shapes, correspond to morphemes and typically have pronunciations suggested by their visual configurations (for reviews, see [59] and [60]). Therefore, a distinctive neurocognitive pattern was revealed among Chinese readers in their English lexical processing, that is, semantic access predominantly occurs through holistic visual processing rather than sequential grapheme-to-phoneme conversion as LSTG was not involved in Chinese (L1) readers\u0026rsquo; English words processing [46, 61]; instead, right fusiform gyrus was employed. Such characteristic arises from readers\u0026rsquo; implicit tendency to establish direct form-meaning mappings at the orthographic level, a cognitive strategy that diminishes the functional involvement of the brain regions (STG/MTG) that are important for English native speakers while amplifying the right fusiform gyrus\u0026rsquo;s role as a critical neural substrate for cross-linguistic visual word recognition.\u003c/p\u003e\n\u003cp\u003eIn sum, to model Chinese EFL readers\u0026rsquo; cognitive load, it is necessary to consider not only the \u0026nbsp;PFC and \u0026nbsp;brain regions that are vital important for English words\u0026rsquo; lexical comprehension but also other brain regions that have been revealed to show significant influences in Chinese readers\u0026rsquo; English reading.\u003c/p\u003e\n\u003ch2\u003eThis study\u003c/h2\u003e\n\u003cp\u003eThe related works illustrate the challenges \u0026nbsp;and complexities \u0026nbsp;of capturing and measuring EFL readers\u0026rsquo; instant CL through observing readers\u0026rsquo; brain functions; therefore, the research questions of this study are as follows:\u003c/p\u003e\n\u003cp\u003eRQ1. Can the brain function surrounding the PFC be taken as a signature of the CL experienced by Chinese readers in an authentic English word comprehension task?\u003c/p\u003e\n\u003cp\u003eRQ2. How can the unsteadiness of the brain function mechanism regarding the same word comprehension task be explained by the influence of previously conducted non-language tasks?\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRQ3. Can the instability of the CL be modelled by alteration of the relative brain functions?\u003c/p\u003e"},{"header":"Method","content":"\u003ch2\u003eParticipants\u003c/h2\u003e\n\u003cp\u003eIn this study, 61 Mandarin speaking university students (37 females and 24 males) were invited to participate in the experiment, with an average age of 20.03 years (\u003cem\u003esd\u003c/em\u003e = 0.97 year), ranging from 18 to 23 years. All participants were right-handed and enrolled in formal English classes. Their English proficiency was assessed using their \u0026nbsp;CET-4 scores, where CET stands for the National College English Test, which is a standardised English examination administered by the Ministry of Education of China and is highly correlated with learners\u0026rsquo; near-future English proficiency during their university studies . Participants \u0026nbsp;were randomly clustered into three groups: Group 1 (baseline task group, n=21) , Group 2 (GEFT group, n=20), \u0026nbsp;and Group 3 (calculation group, n=20) . The CET-4 scores of participants in the three separate groups showed no significant difference. The study was formally approved by the Ethics Review Committee of Sichuan Normal University and all participants gave written informed consent.\u003c/p\u003e\n\u003ch2\u003eApparatus\u003c/h2\u003e\n\u003cp\u003eFunctional near-infrared spectroscopy (fNIRS) is one of the most commonly used neuroimaging methods to detect the concentration changes of oxyhaemoglobin (HbO) and deoxyhaemoglobin (HbR).\u0026nbsp;In this study,\u0026nbsp;the\u0026nbsp;fNIRS signals were collected by NirSmart II-3000A,\u0026nbsp;which is a portable device developed by\u0026nbsp;Huichuang\u0026nbsp;of\u0026nbsp;China. The\u0026nbsp;sampling rate of\u0026nbsp;the device is\u0026nbsp;11 Hz and\u0026nbsp;the\u0026nbsp;wavelengths\u0026nbsp;are\u0026nbsp;730\u0026nbsp;nm and 850 nm.\u0026nbsp;An\u0026nbsp;fNIRS cap\u0026nbsp;with 49 channels\u0026nbsp;was placed on\u0026nbsp;each participant\u0026rsquo;s\u0026nbsp;scalp to collect the data of the participant\u0026rsquo;s brain functions when he/she conducted the task.\u0026nbsp;Both the device and the cap were the same ones used in Wang and Yang\u0026rsquo;s study [17, 60].\u003c/p\u003e\n\u003cp\u003eThe initial processing of fNIRS signals was conducted through NirSpark, a \u0026nbsp;software of NirSmart. This preprocessing pipeline employed a three-stage approach: (1) Motion artifacts\u0026rsquo; influence in the optical density trajectories was mitigated through cubic spline interpolation-based differential computation; (2) physiological interference (e.g., heart beating, respiratory movements and blood pressure fluctuations) was suppressed by a band-pass filter with a passing frequency range of 0.01\u0026ndash;0.2 Hz; and (3) according to the modified Beer\u0026ndash;Lambert law, target indices (relative changes in HbO and HbR) were calculated from the optical density signals. In the modified Beer\u0026ndash;Lambert equations, the value of differential path factor is set to be 6 for wavelengths of 730 nm and 850 nm.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eDesign of the quasi-experiment and the tasks\u003c/h2\u003e\n\u003ch3\u003eExperimental design\u003c/h3\u003e\n\u003cp\u003eThe quasi-experiment was designed as shown in Figure 1.Participants in Group 1 directly conducted the word comprehension task (baseline task group), participants in Group 2 conducted the word comprehension task 5 minutes after they implemented the Group Embedded Figures Test (GEFT) task, and participants in Group 3 conducted the word comprehension task 5 minutes after \u0026nbsp;completing a mathematical calculation task. The words required to be comprehended by participants were displayed on a monitor. Participants were required to complete \u0026nbsp;the task sequentially \u0026nbsp;while wearing an fNIRS cap when he/she carried out the task. Each word was at most displayed for 5 seconds. \u0026nbsp; Participants would press a button to switch to the next word if they finished comprehending a word within 5 s; otherwise, the screen would automatically switch to the next word.\u003c/p\u003e\n\u003ch3\u003eGroup Embedded Figures Test (GEFT)\u003c/h3\u003e\n\u003cp\u003eThis study \u0026nbsp;adapted the CL residual assumption proposed by [17]; therefore, the GEFT from Wang et al.\u0026rsquo;s study was also used here as a non-language task. Field independence, subsumed under spatial ability (for a review, see [62]), and GEFT is a commonly used tool to reveal learners\u0026rsquo; visuospatial abilities in two dimensions [63]. As mentioned earlier, \u0026nbsp;visuospatial recognition is a vitally important ability for native Chinese speakers in character recognition. Therefore, for Chinese L2 learners, the GEFT represents one of the most commonly and frequently confronted tasks in educational settings. The evaluation adhered to the protocol outlined in the GEFT manual [64]. Participants in Group 2 completed the GEFT.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eMathematical calculation task\u003c/h3\u003e\n\u003cp\u003eThe mathematical calculation task was also derived from Wang et al.\u0026rsquo;s study [17], which is a test focusing on learners\u0026rsquo; mathematical thinking on calculation. In order to disassociate this process from spatial recognition, the task only employs addition and multiplication as its basic operations. Therefore, participants in Group 3 were presented with 3 multiplication problems: 1234\u0026times;5678; 3333\u0026times;6666; 9999\u0026times;7777. They were required to solve each problem within 2 minutes. The calculation task represents another type of the most confronted tasks for Chinese L2 learners in school.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eWord comprehension task\u003c/h3\u003e\n\u003cp\u003eThis study used a word-picture mapping task (as shown in Figure 2) to represent the word comprehension process of participants. In this task, a \u0026nbsp;common animal name in L2 was displayed at the top of the screen, with two candidate animal images presented \u0026nbsp;at the bottom. Participants were required to respond using a keyboard. The reason of using word-picture mapping task instead of word-word translation task is that the former directly links the lexical layer of L2 to the concept layer that is shared both by L1 and L2 [45] to maximally avoid applying equivalent L1 translations as mediation.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eFNIRS data collection and processing\u003c/h3\u003e\n\u003cp\u003eFigures 3 (a)-(c) illustrate the location of the 22 infrared light sources, 16 detectors, as well as the 49 fNIRS channels on the fNIRS cap, and the arrangement of the sources and detectors is in line with [60]. The 49 fNIRS channels were divided into 22 brain regions basically according to Brodmann\u0026rsquo;s division, covering Broca\u0026rsquo;s area, Wernicke\u0026rsquo;s area, the central executive control area, the motor related area, and other brain areas that are important for language processing. The division of the 22 brain regions with further details provided in the Appendix. The brain regions of interest (ROIs) of this study are defined as follows: the PFC (channels F, G, H, I), the left MTG and STG (B1), the left fusiform gyrus (E1), the right MTG and STG (B2), and the right fusiform gyrus (E2), as shown in Figures 3 (d)-(f).\u003c/p\u003e\n\u003cp\u003eGiven that many previous studies have demonstrated that a rise in HbO concentration is more sensitive than a drop in HbR, as an indicator of brain activation, this study used relative HbO concentration changes to determine whether an fNIRS channel was significantly activated. The definitions of baseline HbO data, task-related HbO data, as well as the computation of functional connectivity between each pair of fNIRS channels followed the same research paradigm described in [17] and [60].\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eBehavioural performance comparisons among three groups\u003c/h2\u003e\n\u003ch2\u003eNeurological comparisons among the three groups\u0026nbsp;in word comprehension task\u003c/h2\u003e\n\u003ch3\u003eBrain activation comparison\u003c/h3\u003e\n\u003cp\u003eThe significant activation of the brain regions (after false discovery rate, RDF correction) regarding three different groups was displayed in Figure 4. A one-way ANOVA about participants\u0026rsquo; word comprehension performance and \u0026nbsp;response time in word comprehension was conducted among the three groups, as shown in Table 1. Both word comprehension performance and response time of participants showed no significant differences among the three groups.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 1. One-way ANOVA of the learners\u0026rsquo; word comprehension performance among the three groups\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eBase-line\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(group 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eGEFT\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(group 2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eCalculation\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(group 3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;M\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSig.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eȠ2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eWord comprehension performance (score)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e16.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;17.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e1.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eResponse time (seconds)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e43.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e36.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e43.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e65.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; 44.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e44.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eNote.\u003c/em\u003e \u003cem\u003eȠ\u003csup\u003e2\u003c/sup\u003e\u003c/em\u003e=.01 (small effect); \u003cem\u003eȠ\u003csup\u003e2\u003c/sup\u003e\u003c/em\u003e=.06 (medium effect); \u003cem\u003eȠ\u003csup\u003e2\u003c/sup\u003e\u003c/em\u003e=.14 (large effect);\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAn ANOVA of the HbO concentration with False Discovery Rate (FDR) correction in different interested brain functions among the three groups was conducted. As shown in Table 2, there were \u0026nbsp;no significant differences in learners\u0026rsquo; HbO concentration regarding 5 different ROIs among the three groups. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 2. An ANOVA of the learners\u0026rsquo; HbO concentration regarding 5 brain ROIs in word comprehension task among the three groups\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eBase-line\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(group 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eGEFT\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(group 2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eCalculation\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(group 3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;M\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSig.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eȠ2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003ePFC (F, G,H, I)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.1247\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0971\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0806\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.2104\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.0739\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eLeft MTG and STG (B1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.0009\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0023\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0007\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0058\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.0005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eLeft fusiform gyrus (E1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0093\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0147\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0148\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.0005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eRight MTG and STG (B2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0021\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.0016\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.0041\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.0005\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eRight fusiform gyrus (E2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0178\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0010\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0064\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0010\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.0092\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.0006\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eNote.\u003c/em\u003e \u003cem\u003eȠ\u003csup\u003e2\u003c/sup\u003e\u003c/em\u003e=.01 (small effect); \u003cem\u003eȠ\u003csup\u003e2\u003c/sup\u003e\u003c/em\u003e=.06 (medium effect); \u003cem\u003eȠ\u003csup\u003e2\u003c/sup\u003e\u003c/em\u003e=.14 (large effect);\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAfter conducting ANOVAs within 3 groups among five ROIs, it was found that the participants\u0026rsquo; HbO concentration in the PFC (with FDR correction) was significantly higher than that in B1, B2, E1 and E2 in Group 1 and \u0026nbsp;3 (paired samples t-test; \u003cem\u003ep\u003c/em\u003e\u0026lt;0.01, two tailed); however, such significant difference was not observed in Group 2 (paired samples t-test; \u003cem\u003ep\u003c/em\u003e\u0026gt;0.05). \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eNetworks\u0026rsquo; synchronisation comparison\u003c/h3\u003e\n\u003cp\u003eFigure 5 illustrates the learners\u0026rsquo; brain functions synchronisation within and between the PFC (F1-I2), B1, B2, E1, E2 and other brain regions in Group 1 (baseline task group), Group 2 (word comprehension after GEFT), and Group 3 (word comprehension after GEFT). In order to clarify the altered synchronisations among these ROIs among the three different groups, an one-way ANOVA \u0026nbsp;of the Pearson correlations among the PFC, B1, B2, E1 and E2 \u0026nbsp; were performed. It was found that the synchronisation among all brain regions within the PFC was significantly enhanced in Group 2 (\u003cem\u003ep\u003c/em\u003e=0.00, \u003cem\u003eF\u003c/em\u003e(2,81)=5.76) compared with that in Group 1 and \u0026nbsp;3 (posthoc LSD). Additionally, the synchronisations between the PFC and other brain regions in Group 2 \u0026nbsp;were significantly higher than that in Group 1 (two samples t-test; \u003cem\u003ep\u003c/em\u003e=0.03, one tailed). Significantly enhanced synchronisations among PFC, B1, B2, E1 and E2 in Group 2 (\u003cem\u003ep\u003c/em\u003e=0.02, \u003cem\u003eF\u003c/em\u003e(2,27)=4.29) were found compared with those in Group 1 and \u0026nbsp;3 ( post hoc LSD).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eModelling \u0026nbsp; differential brain function mechanisms regarding the same L2 comprehension task\u003c/h3\u003e\n\u003cp\u003eBy using prefrontal cortex (F, G,H, I), B1, B2, E1, E2, as well as the interactive effects among PFC and other brain regions, i.e., PFC*B1, PFC*B2, PFC*E1, and PFC*E2, as the variables, and participants\u0026rsquo; word comprehension performance (S) and response time (T) as the regression targets for three groups, the details regarding regression models are listed in Table 3, Table 4 and Figure 6.\u003c/p\u003e\n\u003cp\u003eTable 3. Multiple regression models targeted at word comprehension performance (S) with 5 independent variables and 4 interactive variables.\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eBase-line task group\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(Group 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eGEFT group\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(Group 2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eCalculation group\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(Group 3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eStandard coefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMultiple regression model 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eStandard coefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMultiple regression model 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eStandard coefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMultiple regression model 3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePFC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"9\" valign=\"top\"\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e=0.50\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e=\u003cstrong\u003e0.04\u003c/strong\u003e\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eDurbin-Watson=2.10\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e95% CI=[14.86, 21.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"9\" valign=\"top\"\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e=0.13\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e=0.34\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e\u0026nbsp;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003eDurbin-Watson=1.91\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e95% CI=[16.12, 20.97]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-1.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"9\" valign=\"top\"\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e=0.06\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e=0.42\u003csup\u003eb\u003cbr\u003e\u0026nbsp;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003eDurbin-Watson=2.27\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e95% CI=[15.91,19.42]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eB1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eB2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-1.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eE1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eE2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePFC*B1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePFC*B2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePFC*E1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePFC*E2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-1.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-1.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAs illustrated in Table 3, Table 4 and Figure 6, only the linear multi-regression model targeted at (S) for the baseline task (Group 1) was significant (Durbin-Watson=2.10; \u003cem\u003ep\u003c/em\u003e=0.04\u003csup\u003eb\u003c/sup\u003e). In this significant linear model, activity in the PFC (\u003cem\u003ep\u003c/em\u003e=0.01) was significantly and negatively correlated with the learners\u0026rsquo; score (S) with adjusted R\u003csup\u003e2\u003c/sup\u003e of 0.50. This result indicates that a higher \u0026nbsp; HbO concentration in the PFC is associated with lower performance, suggesting that \u0026nbsp;a lower cognitive load \u0026nbsp;may enhance \u0026nbsp;learners\u0026rsquo; comprehension.\u003c/p\u003e\n\u003cp\u003eTable 4. Multiple regression models targeted at word comprehension performance (T) with 5 independent variables and 4 interactive variables.\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 30px;\"\u003e\n \u003cp\u003eBase-line task group\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(Group 1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 30px;\"\u003e\n \u003cp\u003eGEFT group\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(Group 2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 30px;\"\u003e\n \u003cp\u003eCalculation group\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(Group 3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003eStandard coefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eMultiple regression model 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003eStandard coefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eMultiple regression model 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003eStandard coefficient\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eMultiple regression model 3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003ePFC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"9\" valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e=-0.22\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e=0.77\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eDurbin-Watson=1.91\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e95% CI=[1555.21,2043.62]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"9\" valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e=-0.29\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e=0.83\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e\u0026nbsp;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003eDurbin-Watson=2.31\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e95% CI=[1458.84,2111.82]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"9\" valign=\"top\" style=\"width: 16px;\"\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e=0.17\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e=0.29\u003csup\u003eb\u003cbr\u003e\u0026nbsp;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003eDurbin-Watson=1.63\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e95% CI=[1410.56,2272.38]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003eB1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e1.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003eB2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003eE1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003eE2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e1.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003ePFC*B1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003ePFC*B2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003ePFC*E1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8px;\"\u003e\n \u003cp\u003ePFC*E2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 9px;\"\u003e\n \u003cp\u003e-0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 5px;\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;However, this significant influence of PFC activation on English word comprehension was not observed in Group 2 and 3, likely due to the effects of previously conducted tasks (GEFT and calculation). Although the regression model for Group 3 yielded an almost identical linear relationship to that of Group 1 in a two-dimensional space, where principal component analysis (PCA) transformed 9 variables as the horizontal coordinate and the word comprehension score as the vertical coordinate, the regression model for Group 3 was statistically insignificant (\u003cem\u003ep\u003c/em\u003e=0.42, R\u003csup\u003e2\u003c/sup\u003e=0.06).\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003ch2\u003eRQ1. Brain functions surrounding the PFC \u0026nbsp;may serve as an individualised signature of the CL afforded by L2 readers\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eFirstly, we address \u0026nbsp;whether the brain functions surrounding the PFC can be taken as a signature of the cognitive load induced \u0026nbsp;in an authentic English word comprehension task. This study\u0026rsquo;s results regarding Group 1 support \u0026nbsp;this assumption. The multiple regression model for the baseline \u0026nbsp;task group displayed \u0026nbsp;a significant \u0026nbsp; negative correlation between readers\u0026rsquo; PFC function and \u0026nbsp;comprehension performance. Meantime, other \u0026nbsp;brain regions of interest, i.e., the STG, MTG, and fusiform gyrus, showed varied impacts on participants\u0026rsquo; L2 word comprehension. By controlling task\u0026rsquo;s difficulty (with participants in all three groups comprehending the same set of English words), learners\u0026rsquo; individualised L2 proficiency can be reflected by their PFC-centred HbO concentrations.\u003c/p\u003e\n\u003cp\u003eNumerous studies have reported that \u0026nbsp;outcomes \u0026nbsp;linked to PFC function, which is indicative of \u0026nbsp;working memory (WM), \u0026nbsp;are highly correlated with \u0026nbsp;overall English language competences, such as reading, listening, speaking and the use of English (e.g., [65]). For example, Malone [66] concluded in his study that WM outcomes were correlated with vocabulary performance \u0026nbsp;derived from reading to form recognition. However, the lexical network of L2 readers is highly dynamic and individualised [67], leading Chinese EFL readers to employ varied strategies stemming from their L1 \u0026nbsp;when comprehending English words. \u0026nbsp;Such strategies may include \u0026ldquo;directly mapping L2 print to its meaning,\u0026rdquo; \u0026ldquo;mapping L2 print to its meaning via the word\u0026rsquo;s phonological representation,\u0026rdquo; and \u0026ldquo;mapping L2 print to its meaning via its equivalent L1 translation\u0026rdquo;, etc. All those strategies can be summarised in terms of paths between L2 and L1 in a lexical network, each characterised by \u0026nbsp;different linking strengths, path lengths, and transfer powers (for a review, see [68]). \u0026nbsp; Therefore, \u0026nbsp;the diverse strategies \u0026nbsp;employed by different readers may involve distinct \u0026nbsp;brain functions and give rise to different CL constructions. The varied impacts of the STG, MTG, and fusiform gyrus on participants\u0026rsquo; comprehension scores observed \u0026nbsp;in this study further support this hypothesis.\u003c/p\u003e\n\u003cp\u003eNevertheless, \u0026nbsp;irrespective of the variability in the cognitive processes employed by L2 readers, retrieving word meaning \u0026nbsp;via the PFC appears to be a compulsive final step \u0026nbsp;for all readers. That is, while the PFC\u0026rsquo;s function may not be a highly correlated factor in EFL readers\u0026rsquo; long-term achievement, \u0026nbsp;it is crucial for readers\u0026rsquo; instant comprehending activities. Given that working memory (WM) acts as a container with limited capacity \u0026nbsp;for the cognitive processes required in online L2 processing [69-70], it is reasonable to use \u0026nbsp;PFC activation to quantify the instant CL \u0026nbsp;experienced by readers \u0026nbsp;during a reading task under \u0026nbsp;WM constraints [69]. The multiple regression model regarding the baseline task group further supports this view by illustrating a significant and negative correlation between readers\u0026rsquo; PFC function and comprehension performance.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur findings in Group 1 are also in line with many existing neurological studies. For instance, Wang and Aryadoust [39] reported that EFL participants generally exhibited higher extraneous cognitive load than their L1 counterparts, i.e., for L2 readers, lower PFC activation is associated with \u0026nbsp;higher language competence. However, the instant CL mediated by task difficulty remains variable. For example, \u0026nbsp;Soares et al. [71] revealed a positive relationship between the using of PFC and the difficulty of a recall task, suggesting that the more elements (n) required to be recalled (i.e., higher CL), the greater the activation of the PFC; while in Wang and Aryadoust \u0026rsquo;s study [39], there was no significant difference in participants\u0026rsquo; PFC activation between positive words (task with high CL) and negative words (task with low CL). This \u0026nbsp;variability in instant CL will be discussed in the next section.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eRQ2. The unsteadiness of the instant CL can be explained as\u0026nbsp;CL residual inherited from previous tasks\u003c/h2\u003e\n\u003cp\u003eThe former section has confirmed the signature role of PFC-centred brain functions that plays in representing EFL readers\u0026rsquo; instant CL, and in this section we discuss the unsteadiness of the instant CL based on the CL residual assumption proposed by [17]. Although the overall \u0026nbsp;PFC activation among the three groups showed no significant difference, \u0026nbsp;initially suggesting that there was no CL residual inherited from previous tasks, \u0026nbsp;a closer examination of the PFC-centred networks among the three groups revealed that readers\u0026rsquo; PFC centred network in Group 2 was significantly altered compared with those in Groups 1 and 3. This finding suggests that previously conducted tasks changed readers\u0026rsquo; brain function mechanism regarding L2 word comprehension when \u0026nbsp;all groups performed an exactly same task.\u003c/p\u003e\n\u003cp\u003eAdditional powerful evidence comes from the comparison of the regression models among the three groups. The relationship between PFC-centred functions and learners\u0026rsquo; performance, observed in Group 1, disappeared in the other two groups. Significantly enhanced synchronisations among the PFC, B1, B2, E1 and E2 in Group 2 were found compared with those in Groups 1 and 3, as readers\u0026rsquo; PFC activation in Group 2 overly decreased with an insignificant level. Although this finding is consistent with [17], the results cannot lead to the conclusion that readers\u0026rsquo; cognitive load decreased, as the regression model regarding Group 2 lost its significance (R\u003csup\u003e2\u003c/sup\u003e=0.13) and the activation of PFC among the three groups also showed no significant differences. Thus, we prefer to describe this phenomenon as a negative biological reaction reflected by readers in Group 2 to the word comprehension, potentially due to the CL residual influence from the previously conducted GEFT task. This phenomenon is similar to the \u0026ldquo;go on strike\u0026rdquo; effect found by [36], where the brain ceases effective processing of information when overwhelmed by cognitive load induced by the task. A disrupted relationship between PFC-centred functions and readers\u0026rsquo; performance was also observed in Group 3, in which the R\u003csup\u003e2\u003c/sup\u003e value of the regression model dropped to 0.06, although the alteration of brain functions was less pronounced than that in Group 2.\u003c/p\u003e\n\u003cp\u003eIn the regression model regarding Group 2, although the PFC remained a negative influence on learners\u0026rsquo; comprehension scores, the effect from the PFC was no longer significant, suggesting that comprehension performance in Group 2 was no longer dominated by the cognitive load represented by the PFC. Therefore, the results of this study verify the existence of a cognitive-load residual from previously conducted tasks, however, in the form of disturbing the function of PFC. In Groups 2 and \u0026nbsp;3, the PFC does not reliably represent learners\u0026rsquo; L2 vocabulary proficiency, suggesting that readers with lower proficiency also exert lower mental effort (i.e., inactivation of the PFC) in L2 word comprehension. \u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eRQ3. The CL residual cannot be linearly modelled by \u0026nbsp; alterations of \u0026nbsp;relative brain functions\u003c/h2\u003e\n\u003cp\u003eFinally, we would like to discuss whether the unsteadiness of the instant CL can be modelled through the brain function changes surrounding the PFC. The answer is negative. The results found in this study suggest that the brain functions surrounding the PFC in Groups 2 and 3 underwent nonlinear transformations due to previously conducted tasks, rendering the PFC \u0026nbsp;an unreliable indicator of CL signature in an authentic reading context. We cannot find an explanatory model to describe the working mechanism of the CL residuals, and \u0026nbsp;at \u0026nbsp;the same time accommodate PFC alterations within \u0026nbsp; the same framework.\u003c/p\u003e\n\u003cp\u003eThis study employed a primary task as a baseline benchmark to model participants\u0026rsquo; instant CL with an fNIRS device \u0026nbsp;to collect physiological data. Task-performance-based methods, such as calculating the number of elements that comprise a task, \u0026nbsp;provide an intuitive \u0026nbsp;yet simplistic way to map task complexity to the CL produced [2], and typically adopt response time or accuracy of learners\u0026rsquo; performance on tasks as indices of the CL generated by the relative tasks (e.g., see [72]).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe task-based CL measuring methods are based on a na\u0026iuml;ve prerequisite, i.e., the greater the mental effort devoted to a task, the lower the performance achieved. However, as pointed by [73], changes of the workload cannot always be simply and linearly mapped to participants\u0026rsquo; performance. \u0026nbsp;Similarly, our study found that the altered brain functions of readers in Groups 2 and 3 could not be linearly mapped to their unchanged comprehension performance. There is no available model that both can accommodate the PFC-centred functions in L2 comprehension (Group 1) and the altered brain functions due to the previously conducted tasks (Groups 2 and 3) within \u0026nbsp;the same explanatory framework. This problem may be solved in the future by employing nonlinear models with explanatory components.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003e This study explored the possibility of monitoring and modelling Chinese EFL readers\u0026rsquo; instant CL in their L2 words comprehension tasks via observing participants\u0026rsquo; PFC-centred brain function changes. The results of this study conclude that the brain functions surrounding the PFC represent an individualised and highly unstable signature of the instant CL experienced by L2 readers. This study also found that the unsteadiness of the instant CL appears to be due to CL residual inherited from previous tasks, as revealed by significantly altered brain functions surrounding the PFC; finally, the CL residual cannot be linearly modelled, but future research may explore the possibility of employing nonlinear models to better explain this phenomenon.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e\n\u003cp\u003eWe obtained all participants\u0026rsquo; consents, and this study was approved by the Research Ethics Review Committee of Sichuan Normal University.\u003c/p\u003e\n\u003ch2\u003eAvailability\u0026nbsp;of\u0026nbsp;data\u0026nbsp;and\u0026nbsp;materials\u003c/h2\u003e\n\u003cp\u003eStimuli, datasets and analysis code are available through a public OSF repository: https://osf.io/b7en5/\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003e\u0026nbsp; Authors declare that there is no\u0026nbsp;competing interests.\u003c/p\u003e\n\u003ch2\u003eConsent for publication\u003c/h2\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis research is supported by the Major Science and Technology Special Program of Jiangsu Province [BG2024025]; Research project of Ministry of Education of China[23YJC880062]; Teaching reform and research project of Sichuan Normal University [JWC20240107; JWC20240116]. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAuthors contributions\u003c/h2\u003e\n\u003cp\u003eWang, L., Yang, J. and Wang, D. F. conducted and administrated the experiment, and wrote the manuscript; \u0026nbsp;Zhang, W., Yang, P. L., and Duan, Y. H. processed the fNIRS data and prepared the related figures and tables; Sun, B. and Chen, D. designed the experiment; Liang, Z. J. and Zhang, Y. L. processed the behavioural data and prepared the related figures and tables.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSweller, J., \u003cem\u003eCognitive load during problem solving: Effects on learning.\u003c/em\u003e Cognitive Science, 1988. \u003cstrong\u003e12\u003c/strong\u003e: p. 257-285.\u003c/li\u003e\n\u003cli\u003eSweller, J., \u003cem\u003eCognitive load theory and educational technology.\u003c/em\u003e Etr\u0026amp;D-Educational Technology Research and Development, 2020. \u003cstrong\u003e68\u003c/strong\u003e(1): p. 1-16.\u003c/li\u003e\n\u003cli\u003eSweller, J., et al., \u003cem\u003eCognitive load and selective attention as factors in the structuring of technical material.\u003c/em\u003e Journal of Experimental Psychology, 1990. \u003cstrong\u003eGeneral\u003c/strong\u003e(119): p. 176-192.\u003c/li\u003e\n\u003cli\u003eSweller, J. and F. Paas, \u003cem\u003eShould self-regulated learning be integrated with cognitive load theory? A commentary.\u003c/em\u003e Learning and Instruction, 2017. \u003cstrong\u003e51\u003c/strong\u003e: p. 85-89.\u003c/li\u003e\n\u003cli\u003eKennedy, M.J., et al., \u003cem\u003eImproving teacher candidates\u0026apos; knowledge of phonological awareness: A multimedia approach.\u003c/em\u003e Computers \u0026amp; Education, 2013. \u003cstrong\u003e64\u003c/strong\u003e: p. 42-51.\u003c/li\u003e\n\u003cli\u003eKhalil, M.K. and I.A. Elkhider, \u003cem\u003eApplying learning theories and instructional design models for effective instruction.\u003c/em\u003e Advances in Physiology Education, 2016. \u003cstrong\u003e40\u003c/strong\u003e(2): p. 147-156.\u003c/li\u003e\n\u003cli\u003eNgu, B.H. and H.P. Phan, \u003cem\u003eLearning linear equations: capitalizing on cognitive load theory and learning by analogy.\u003c/em\u003e International Journal of Mathematical Education in Science and Technology, 2022. \u003cstrong\u003e53\u003c/strong\u003e(10): p. 2686-2702.\u003c/li\u003e\n\u003cli\u003eNyachwaya, J.M. and M. Gillaspie, \u003cem\u003eFeatures of representations in general chemistry textbooks: a peek through the lens of the cognitive load theory.\u003c/em\u003e Chemistry Education Research and Practice, 2016. \u003cstrong\u003e17\u003c/strong\u003e(1): p. 58-71.\u003c/li\u003e\n\u003cli\u003eOfferdahl, E.G., J.B. Arneson, and N. Byrne, \u003cem\u003eLighten the Load: Scaffolding Visual Literacy in Biochemistry and Molecular Biology.\u003c/em\u003e Cbe-Life Sciences Education, 2017. \u003cstrong\u003e16\u003c/strong\u003e(1).\u003c/li\u003e\n\u003cli\u003eDuffy, P.L., et al., \u003cem\u003eForm versus Function: A Comparison of Lewis Structure Drawing Tools and the Extraneous Cognitive Load They Induce.\u003c/em\u003e Journal of Chemical Education, 2019. \u003cstrong\u003e96\u003c/strong\u003e(2): p. 238-247.\u003c/li\u003e\n\u003cli\u003eParong, J. and R.E. Mayer, \u003cem\u003eCognitive and affective processes for learning science in immersive virtual reality.\u003c/em\u003e Journal of Computer Assisted Learning, 2021. \u003cstrong\u003e37\u003c/strong\u003e(1): p. 226-241.\u003c/li\u003e\n\u003cli\u003eKim, J. and I.-H. Jo, \u003cem\u003eFeasibility and use of psychophysiological responses based on cognitive load theory.\u003c/em\u003e Australasian Journal of Educational Technology, 2019. \u003cstrong\u003e35\u003c/strong\u003e(3): p. 150-165.\u003c/li\u003e\n\u003cli\u003eKruger, J.-L. and S. Doherty, \u003cem\u003eMeasuring cognitive load in the presence of educational video: Towards a multimodal methodology.\u003c/em\u003e Australasian Journal of Educational Technology, 2016. \u003cstrong\u003e32\u003c/strong\u003e(6): p. 19-31.\u003c/li\u003e\n\u003cli\u003eMarinescu, A.C., et al., \u003cem\u003ePhysiological Parameter Response to Variation of Mental Workload.\u003c/em\u003e Human Factors: The Journal of the Human Factors and Ergonomics Society, 2017. \u003cstrong\u003e60\u003c/strong\u003e(1): p. 31-56.\u003c/li\u003e\n\u003cli\u003eMidha, S., et al., \u003cem\u003eMeasuring Mental Workload Variations in Office Work Tasks using fNIRS.\u003c/em\u003e International Journal of Human - Computer Studies, 2021. \u003cstrong\u003e147\u003c/strong\u003e: p. 102580.\u003c/li\u003e\n\u003cli\u003eAl-Samarraie, H., et al., \u003cem\u003eHow reading in single- and multiple-column types influence our cognitive load: an EEG study.\u003c/em\u003e Electronic Library, 2019. \u003cstrong\u003e37\u003c/strong\u003e(4): p. 593-606.\u003c/li\u003e\n\u003cli\u003eWang, L., et al., \u003cem\u003eInfluence of high-level mathematical thinking on L2 phonological processing of Chinese EFL learners: Evidence from an fNIRS study.\u003c/em\u003e Thinking Skills and Creativity, 2023. \u003cstrong\u003e47\u003c/strong\u003e: p. 101242.\u003c/li\u003e\n\u003cli\u003eBaddeley, A., \u003cem\u003eThe episodic buffer: a new component of working memory?\u003c/em\u003e Trends in cognitive sciences, 2000. \u003cstrong\u003e4\u003c/strong\u003e(11): p. 417-423.\u003c/li\u003e\n\u003cli\u003eWickens, C.D., \u003cem\u003eMultiple Resources and Mental Workload.\u003c/em\u003e Hum Factors, 2008. \u003cstrong\u003e50\u003c/strong\u003e(3): p. 449-455.\u003c/li\u003e\n\u003cli\u003ePaas, F., \u003cem\u003eTraining strategies for attaining transfer of problem-solving skill in statistics: A cognitive-load approach.\u003c/em\u003e Journal of Educational Psychology,, 1992. \u003cstrong\u003e84\u003c/strong\u003e: p. 429\u0026ndash;434.\u003c/li\u003e\n\u003cli\u003eCao, A., et al., \u003cem\u003eNASA TLX: software for assessing subjective mental workload.\u003c/em\u003e Behav Res Methods, 2009. \u003cstrong\u003e41\u003c/strong\u003e(1): p. 113\u0026ndash;117.\u003c/li\u003e\n\u003cli\u003eHart, S.G., \u003cem\u003eNASA-task load index (NASA-TLX); 20 years later\u003c/em\u003e, in \u003cem\u003eHuman factors and ergonomics society annual meeting\u003c/em\u003e. 2006, Sage Los Angeles. p. 904\u0026ndash;908.\u003c/li\u003e\n\u003cli\u003eHart, S.G. and L.E. Staveland, \u003cem\u003eDevelopment of NASA-TLX (task load index): Results of empirical and theoretical research.\u003c/em\u003e Advances in psychology, 1988. \u003cstrong\u003e52\u003c/strong\u003e: p. 139\u0026ndash;183.\u003c/li\u003e\n\u003cli\u003eAmadieu, F., C. Marin\u0026eacute;, and C. Laimay, \u003cem\u003eThe attention-guiding effect and cognitive load in the comprehension of animations.\u003c/em\u003e Computers in Human Behavior, 2011. \u003cstrong\u003e27\u003c/strong\u003e(1): p. 36-40.\u003c/li\u003e\n\u003cli\u003eGaillard, A., \u003cem\u003eComparing the concepts of mental load and stress.\u003c/em\u003e Ergonomics, 1993. \u003cstrong\u003e36\u003c/strong\u003e(9): p. 991\u0026ndash;1005.\u003c/li\u003e\n\u003cli\u003eEllerton, P., \u003cem\u003eOn critical thinking and content knowledge: A critique of the assumptions of cognitive load theory.\u003c/em\u003e Thinking Skills and Creativity, 2022. \u003cstrong\u003e43\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eDerry, J., \u003cem\u003eA Problem for Cognitive Load Theory-the Distinctively Human Life-form.\u003c/em\u003e Journal of Philosophy of Education, 2020. \u003cstrong\u003e54\u003c/strong\u003e(1): p. 5-22.\u003c/li\u003e\n\u003cli\u003eFerraz Lorenzo, M. and C. Machado Trujillo, \u003cem\u003eCognitive Processes, ICT, and Education: A Critical Analysis.\u003c/em\u003e Computers in the Schools, 2018. \u003cstrong\u003e35\u003c/strong\u003e(3): p. 186-203.\u003c/li\u003e\n\u003cli\u003eBernhardt, K.A., et al., \u003cem\u003eThe effects of dynamic workload and experience on commercially available EEG cognitive state metrics in a high-fidelity air traffic control environment.\u003c/em\u003e Applied Ergonomics, 2019. \u003cstrong\u003e77\u003c/strong\u003e: p. 83-91.\u003c/li\u003e\n\u003cli\u003eChan, A.S., et al., \u003cem\u003ePhotoneuromodulation makes a difficult cognitive task less arduous.\u003c/em\u003e Scientific Reports, 2021. \u003cstrong\u003e11\u003c/strong\u003e(1).\u003c/li\u003e\n\u003cli\u003eZhuang, C., et al., \u003cem\u003eScale invariance in fNIRS as a measurement of cognitive load.\u003c/em\u003e Cortex, 2022. \u003cstrong\u003e154\u003c/strong\u003e: p. 62-76.\u003c/li\u003e\n\u003cli\u003eAlyan, E., N.M. Saad, and N. Kamel, \u003cem\u003eEffects of Workstation Type on Mental Stress: fNIRS Study.\u003c/em\u003e Human Factors, 2021. \u003cstrong\u003e63\u003c/strong\u003e(7): p. 1230-1255.\u003c/li\u003e\n\u003cli\u003eCsipo, T., et al., \u003cem\u003eIncreased cognitive workload evokes greater neurovascular coupling responses in healthy young adults.\u003c/em\u003e Plos One, 2021. \u003cstrong\u003e16\u003c/strong\u003e(5).\u003c/li\u003e\n\u003cli\u003eFishburn, F.A., et al., \u003cem\u003eSensitivity of fNIRS to cognitive state and load.\u003c/em\u003e Frontiers in Human Neuroscience, 2014. \u003cstrong\u003e8\u003c/strong\u003e: p. 76.\u003c/li\u003e\n\u003cli\u003eMandrick, K., et al., \u003cem\u003eNeural and psychophysiological correlates of human performance under stress and high mental workload.\u003c/em\u003e Biological Psychology, 2016. \u003cstrong\u003e121\u003c/strong\u003e: p. 62-73.\u003c/li\u003e\n\u003cli\u003eCausse, M., et al., \u003cem\u003eMental workload and neural efficiency quantified in the prefrontal cortex using fNIRS.\u003c/em\u003e Scientific Reports, 2017. \u003cstrong\u003e7\u003c/strong\u003e(1): p. 5222.\u003c/li\u003e\n\u003cli\u003eMidha, S., et al., \u003cem\u003eMeasuring Mental Workload Variations in Office Work Tasks using fNIRS.\u003c/em\u003e International Journal of Human-Computer Studies, 2021. \u003cstrong\u003e147\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eKarmakar, S., et al., \u003cem\u003eReal time detection of extraneous cognitive load using fNIRS: A deep learning approach.\u003c/em\u003e Biomedical Signal Processing and Control, 2023. \u003cstrong\u003e80\u003c/strong\u003e: p. 104227.\u003c/li\u003e\n\u003cli\u003eWang, X. and V. Aryadoust, \u003cem\u003eAn eye-tracking and neuroimaging study of negative wording effects on cognitive load in a metacognitive awareness tool.\u003c/em\u003e System, 2025. \u003cstrong\u003e128\u003c/strong\u003e: p. 103552.\u003c/li\u003e\n\u003cli\u003eDiao, Y. and J. Sweller, \u003cem\u003eRedundancy in foreign language reading comprehension instruction: Concurrent written and spoken presentations.\u003c/em\u003e Learning and Instruction, 2007. \u003cstrong\u003e17\u003c/strong\u003e: p. 78-88.\u003c/li\u003e\n\u003cli\u003eMoussa-Inaty, J., P. Ayres, and J. Sweller, \u003cem\u003eImproving Listening Skills in English as a Foreign Language by Reading Rather than Listening: A Cognitive Load Perspective.\u003c/em\u003e Applied Cognitive Psychology, 2012. \u003cstrong\u003e26\u003c/strong\u003e(3): p. 391-402.\u003c/li\u003e\n\u003cli\u003eDijkstra, T. and W.J. Van Heuven, \u003cem\u003eThe architecture of the bilingual word recognition system: From identification to decision.\u003c/em\u003e Bilingualism: Language and cognition, 2002. \u003cstrong\u003e5\u003c/strong\u003e(3): p. 175-197.\u003c/li\u003e\n\u003cli\u003eKroll, J.F., et al., \u003cem\u003eThe revised hierarchical model: A critical review and assessment.\u003c/em\u003e Bilingualism: Language and Cognition, 2010. \u003cstrong\u003e13\u003c/strong\u003e: p. 373\u0026ndash;381.\u003c/li\u003e\n\u003cli\u003eDe Groot, A.M.B., \u003cem\u003eBilingual lexical representation: A closer look at conceptual representations\u003c/em\u003e, in \u003cem\u003eOrthography, phonology, morphology, and meaning\u003c/em\u003e, R. Frost and L. Katz, Editors. 2012, Elsevier. p. 389\u0026ndash;412.\u003c/li\u003e\n\u003cli\u003eVan Hell, J.G. and A.M.B. De Groot, \u003cem\u003eConceptual representation in bilingual memory: Effects of concreteness and cognate status in word association.\u003c/em\u003e Bilingualism: Language and Cognition, 1998. \u003cstrong\u003e1\u003c/strong\u003e: p. 193\u0026ndash;211.\u003c/li\u003e\n\u003cli\u003eTan, L.H., et al., \u003cem\u003eNeural systems of second language reading are shaped by native language.\u003c/em\u003e Human brain mapping, 2003. \u003cstrong\u003e18\u003c/strong\u003e(3): p. 158-166.\u003c/li\u003e\n\u003cli\u003eDonald, J.B., A.P. Charles, and S. Walter, \u003cem\u003eCross-cultural effect on the brain revisited: Universal structures plus writing system variation.\u003c/em\u003e Hum Brain Mapp, 2005. \u003cstrong\u003e25\u003c/strong\u003e(1): p. 92-104.\u003c/li\u003e\n\u003cli\u003eDietz, N.A.E., et al., \u003cem\u003ePhonological decoding involves left posterior fusiform gyrus.\u003c/em\u003e Hum. Brain Mapp., 2005. \u003cstrong\u003e26\u003c/strong\u003e: p. 81\u0026ndash;93.\u003c/li\u003e\n\u003cli\u003eMesgarani, N., et al., \u003cem\u003ePhonetic Feature Encoding in Human Superior Temporal Gyrus.\u003c/em\u003e Science, 2014. \u003cstrong\u003e343\u003c/strong\u003e(6174): p. 1006\u0026ndash;1010.\u003c/li\u003e\n\u003cli\u003eYi, H.G., M.K. Leonard, and E.F. Chang, \u003cem\u003eThe encoding of speech sounds in the superior temporal gyrus.\u003c/em\u003e Neuron, 2019. \u003cstrong\u003e102\u003c/strong\u003e(6): p. 1096\u0026ndash;1110.\u003c/li\u003e\n\u003cli\u003eRamos Nu\u0026ntilde;ez, A.I., et al., \u003cem\u003eThe role of left vs. right superior temporal gyrus in speech perception: An fMRI-guided TMS study.\u003c/em\u003e Brain and Language, 2020. \u003cstrong\u003e209\u003c/strong\u003e: p. 104838.\u003c/li\u003e\n\u003cli\u003eMaguinness, C., C. Roswandowitz, and K. von Kriegstein, \u003cem\u003eUnderstanding the mechanisms of familiar voice-identity recognition in the human brain.\u003c/em\u003e Neuropsychologia, 2018. \u003cstrong\u003e116\u003c/strong\u003e: p. 179\u0026ndash;193.\u003c/li\u003e\n\u003cli\u003eZhang, J., J. Chen, and G.S. Ding, \u003cem\u003eUniversality and language specificity of brain reading networks: A developmental perspective.\u003c/em\u003e Developmental Science, 2023. \u003cstrong\u003ein press\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eHenseler, I., et al., \u003cem\u003eModulating brain mechanisms resolving Lexico-semantic interference during word production: a transcranial direct current stimulation study.\u003c/em\u003e Cognitive Neurosci, 2014. \u003cstrong\u003e26\u003c/strong\u003e: p. 1403\u0026ndash;1417.\u003c/li\u003e\n\u003cli\u003eLau, E., C. Phillips, and D. Poeppel, \u003cem\u003eA cortical network for semantics: (de)constructing the N400.\u003c/em\u003e Nat Rev Neurosci 2008. \u003cstrong\u003e9\u003c/strong\u003e: p. 920\u0026ndash;933.\u003c/li\u003e\n\u003cli\u003eTurken, A.U. and N.F. Dronkers, \u003cem\u003eThe neural architecture of the language comprehension network: converging evidence from lesion and connectivity analyses.\u003c/em\u003e Front. Syst. Neurosci., 2011. \u003cstrong\u003e5\u003c/strong\u003e: p. 1.\u003c/li\u003e\n\u003cli\u003eDi Liberto, G.M., E.C. Lalor, and R.E. Millman, \u003cem\u003eCausal cortical dynamics of a predictive enhancement of speech intelligibility.\u003c/em\u003e NEUROIMAGE, 2018. \u003cstrong\u003e166\u003c/strong\u003e: p. 247-258.\u003c/li\u003e\n\u003cli\u003eTuennerhoff, J. and U. Noppeney, \u003cem\u003eWhen sentences live up to your expectations.\u003c/em\u003e NeuroImage, 2016. \u003cstrong\u003e124\u003c/strong\u003e: p. 641\u0026ndash;653.\u003c/li\u003e\n\u003cli\u003eGottardo, A., et al., \u003cem\u003eModels of English and Chinese word reading for adolescent Chinese\u0026ndash;English bilinguals.\u003c/em\u003e Reading and Writing, 2017. \u003cstrong\u003e30\u003c/strong\u003e: p. 1377\u0026ndash;1406.\u003c/li\u003e\n\u003cli\u003eYang, J., et al., \u003cem\u003eMobile application-based phonetic training facilitates Chinese-English learners\u0026rsquo; learning of L2.\u003c/em\u003e Learning and Instruction, 2024. \u003cstrong\u003e93\u003c/strong\u003e: p. Article number 101967.\u003c/li\u003e\n\u003cli\u003eCao, F., et al., \u003cem\u003eSimilarities and differences in brain activation and functional connectivity in first and second language reading: Evidence from Chinese learners of English.\u003c/em\u003e NEUROPSYCHOLOGIA, 2014. \u003cstrong\u003e63\u003c/strong\u003e: p. 275-284.\u003c/li\u003e\n\u003cli\u003eRittschof, K.A., \u003cem\u003eField dependence\u0026ndash;independence as visuospatial and executive functioning in working memory: implications for instructional systems design and research.\u003c/em\u003e Education Tech Research Dev, 2010. \u003cstrong\u003e58\u003c/strong\u003e: p. 99\u0026ndash;114.\u003c/li\u003e\n\u003cli\u003eBoccia, M., et al., \u003cem\u003eDoes field independence predict visuo-spatial abilities underpinning human navigation? Behavioural evidence.\u003c/em\u003e Experimental Brain Research, 2016. \u003cstrong\u003e234\u003c/strong\u003e(10): p. 2799-2807.\u003c/li\u003e\n\u003cli\u003eWitkin, H.A., et al., \u003cem\u003eA manual for the Embedded Figures Tests\u003c/em\u003e, M. Park, Editor. 1971, Mind Garden Inc.: CA.\u003c/li\u003e\n\u003cli\u003eKormos, J. and A. Safar, \u003cem\u003ePhonological Short-Term Memory, Working Memory and Foreign Language Performance in Intensive Language Learning.\u003c/em\u003e Bilingualism Language \u0026amp; Cognition, 2008. \u003cstrong\u003e11\u003c/strong\u003e(2): p. 261-271.\u003c/li\u003e\n\u003cli\u003eMalone, J., \u003cem\u003eIncidental vocabulary learning in SLA: Effects of frequency, aural enhancement, and working memory.\u003c/em\u003e Studies in Second Language Acquisition, 2018. \u003cstrong\u003e40\u003c/strong\u003e(3): p. 651-675.\u003c/li\u003e\n\u003cli\u003ePan, X. and D. Jared, \u003cem\u003eCross-language activation of culture-specific features in Chinese\u0026ndash;English bilinguals.\u003c/em\u003e J Cult Cogn Sci, 2021. \u003cstrong\u003e5\u003c/strong\u003e: p. 325\u0026ndash;344.\u003c/li\u003e\n\u003cli\u003eMalt, B.C., et al., \u003cem\u003eBidirectional lexical interaction in late immersed Mandarin-English bilinguals. Journal of Memory and Language.\u003c/em\u003e Journal of Memory and Language, 2015. \u003cstrong\u003e82\u003c/strong\u003e: p. 86\u0026ndash;104.\u003c/li\u003e\n\u003cli\u003eEllis, N.C., \u003cem\u003eAt the interface: Dynamic interactions of explicit and implicit language knowledge.\u003c/em\u003e Studies in second language acquisition, 2005. \u003cstrong\u003e27\u003c/strong\u003e(2): p. 305-352.\u003c/li\u003e\n\u003cli\u003eSweller, J., \u003cem\u003eInstructional Design Consequences of an Analogy between Evolution by Natural Selection and Human Cognitive Architecture.\u003c/em\u003e Instructional Science, 2004. \u003cstrong\u003e32\u003c/strong\u003e: p. 9-31.\u003c/li\u003e\n\u003cli\u003eSoares, S.M.P., et al., \u003cem\u003eBrain correlates of attentional load processing reflect degree of bilingual engagement: Evidence from EEG.\u003c/em\u003e NeuroImage, 2024. \u003cstrong\u003e298\u003c/strong\u003e: p. 120786.\u003c/li\u003e\n\u003cli\u003eBlissett, S., et al., \u003cem\u003eOptimizing self-regulation of performance: is mental effort a cue?\u003c/em\u003e Advances in Health Sciences Education, 2018. \u003cstrong\u003e23\u003c/strong\u003e(5): p. 891-898.\u003c/li\u003e\n\u003cli\u003eLeppink, J. and P. P\u0026eacute;rez-Fuster, \u003cem\u003eMental Effort, Workload, Time on Task, and Certainty: Beyond Linear Models.\u003c/em\u003e Educ Psychol Rev, 2019. \u003cstrong\u003e31\u003c/strong\u003e: p. 421\u0026ndash;438.\u003c/li\u003e\n\u003c/ol\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":"Cognitive load (CL), English as foreign language (EFL), second language (L2), brain function, functional near infrared spectroscopy (fNIRS)","lastPublishedDoi":"10.21203/rs.3.rs-6407039/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6407039/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAlthough brain functions surrounding the prefrontal cortex (PFC) has been widely acknowledged as a signature of the instant cognitive load (CL) induced by task demands, inconsistent results of the existing studies imply that the hemodynamic changes in the PFC may not serve as a stable direct indicator of second-language (L2) readers\u0026rsquo; instant CL. This study employed 61 Chinese university students to comprehend the identical English words in three different groups to reveal the nature of CL. By observing both the PFC-centred brain functions and the comprehension performance of the participants in the three groups, the following conclusions were reached: 1) The PFC can serve only as an individualised signature of the instant CL that afforded by L2 readers; 2) The unsteadiness of the instant CL is due to residual load carried over from previous tasks; and 3) The residual CL cannot be linearly modelled.\u003c/p\u003e","manuscriptTitle":"Cognitive-load residual is found in Chinese EFL readers’ word comprehension","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-07 06:41:03","doi":"10.21203/rs.3.rs-6407039/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":"a97a1458-1959-4445-997c-a34cdf1590ad","owner":[],"postedDate":"May 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-26T07:54:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-07 06:41:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6407039","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6407039","identity":"rs-6407039","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}