Factors Contributing to Visual Discomfort in Standards-Compliant Urban Road Tunnels: Evidence from Field Measurements and Immersive Driving Experiments

Factors Contributing to Visual Discomfort in Standards-Compliant Urban Road Tunnels: Evidence from Field Measurements and Immersive Driving Experiments | 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 Article Factors Contributing to Visual Discomfort in Standards-Compliant Urban Road Tunnels: Evidence from Field Measurements and Immersive Driving Experiments ZHENGHAO JIN, YANDAN LIN, HAIPING SHEN This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9290694/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Visual discomfort is still reported in urban road tunnels even when conventional lighting criteria are satisfied. This exploratory study investigated factors associated with such discomfort from a human-factors perspective. Four representative urban tunnels were examined using field measurements and immersive laboratory replay. Field data included eye-level illuminance, roadway and sidewall illuminance, synchronized video, and glare-related metrics including threshold increment (TI). Eye-tracking indicators, including pupil diameter and area-of-interest hit rate, were used to characterize visual response and attentional allocation. In the laboratory, typical tunnel scenes were replayed in an immersive driving cabin to obtain ratings of visual comfort, discomfort glare, and visual fatigue. Although all four tunnels showed TI values below the commonly used 15% threshold, subjective comfort differed markedly across scenes. Compared with average illuminance or TI alone, discomfort was more sensitive to spatiotemporal lighting instability. Greater illuminance fluctuation and reflected light from highly reflective sidewalls and decorative lighting were associated with stronger pupil responses, higher AOI hit rates, and lower comfort. These findings suggest that temporal instability, reflective interference, and attentional capture may help explain visual discomfort in standards-compliant tunnels. Physical sciences/Engineering Biological sciences/Neuroscience Biological sciences/Psychology Social science/Psychology Urban road tunnel tunnel lighting visual discomfort eye tracking illuminance fluctuation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1 Introduction 1.1 Existing studies Urban underground roads, including cross-river tunnels, underpasses, and subsurface corridors, play an important role in alleviating surface congestion and improving traffic efficiency. However, their enclosed or semi-enclosed spatial characteristics also make the light field more prone to superposition and amplification, which may give rise to visual disturbances such as glare, flicker, zebra effects, and sidewall reflections. These disturbances not only affect visibility, but may also increase visual workload and attentional competition during dynamic driving, thereby reducing driving comfort and safety margins. Existing studies have mainly approached tunnel lighting from the perspectives of visual adaptation and glare control. Previous research has proposed strategies such as threshold-zone transition lighting and segmented lighting design, and has established predictive relationships between perceived glare and photometric variables from a human-factors perspective[ 1 – 4 ]. In parallel, temporal luminance variations caused by the combined effects of luminaire arrangement and vehicle speed have been regarded as an important source of time-varying illuminance fluctuations. However, a single modulation-frequency-based indicator is often insufficient to capture inter-individual differences in driver preference as well as differences in the overall spatial luminance structure of tunnels[ 5 , 6 ]. To quantify discomfort during tunnel driving more systematically, previous studies have introduced multidimensional evaluation approaches. From the perspective of human response mechanisms, pupil dynamics and eye-movement indicators can be used to characterize the coupled effects of changing light stimuli and cognitive load, and to reveal how salient bright regions, such as decorative light bands and highly reflective sidewalls, may trigger attentional capture and gaze shifts[ 4 , 7 , 8 ]. At the same time, existing standards and engineering assessment frameworks continue to emphasize baseline indicators such as luminance uniformity, illuminance level, luminance level, and threshold increment (TI) limits for glare control[ 9 – 12 ]. In practice, however, it is still common to observe substantial differences in driver or passenger experience even when engineering criteria are formally satisfied. This suggests that no single indicator is sufficient to explain subjective discomfort under compound visual disturbances. 1.2 Current standards and their limitations From a standards perspective, current tunnel lighting design requirements are primarily centered on segmented lighting strategies and baseline photometric quality constraints. For example, the Standard for Lighting Design of Urban Roads[ 13 ] requires daytime lighting in urban tunnels to be designed by dividing the tunnel into entrance, transition, interior, and exit zones, with lighting criteria determined according to vehicle speed and traffic volume. The standard also specifies that non-functional decorative lighting installed along the roadside should not interfere with functional lighting, and that frequently changing dynamic lighting should be avoided to prevent visual disturbance to drivers. Similarly, technical specifications for highway tunnel lighting[ 14 ], as well as their adoption in local standards, indicate that engineering evaluation and acceptance still focus mainly on baseline indicators such as luminance level, uniformity, and glare control across the entrance, transition, interior, and exit zones. With regard to glare, international guidelines such as CIE 88 commonly use threshold increment (TI) as a key indicator for controlling disability glare and propose upper-limit constraints for TI[ 15 ]. In the tunnels investigated in this study, TI remained within this type of recommended threshold range overall, while subjective comfort still differed substantially across scenes. This suggests that, under conditions where disability-glare thresholds are not exceeded, experiential differences may be driven instead by factors that are relatively underrepresented in current standards, such as temporal instability of illuminance or luminance, sidewall reflections, and attentional competition induced by decorative bright zones. At the national level, GB/T 50034 − 2024 has already incorporated flicker and stroboscopic effects into lighting-quality provisions, reflecting an increasing concern over risks associated with temporally modulated light[ 16 ]. However, this standard is intended for architectural lighting and is not sufficient to directly address the process-based mechanisms involved in dynamic tunnel-driving tasks. Against this background, the present study is positioned as a mechanism-oriented exploratory study and variable-screening effort. Rather than establishing universal thresholds or a predictive model, the study aims to collect multidimensional data under standards-compliant tunnel conditions, build a coupled database integrating physical lighting parameters and human-response indicators, and statistically identify the optical and human-factors variables that better explain differences in subjective discomfort. The findings are expected to provide hypotheses, candidate variables, and directional evidence for subsequent controlled experiments and model development. 2 Materials and methods This study adopted a three-stage design consisting of field measurement of optical stimuli, laboratory validation of human responses, and multidimensional correlation analysis. First, physical lighting parameters were synchronously collected in real tunnel-driving environments to characterize typical visually uncomfortable stimuli and to identify representative tunnel lighting scenes. Second, representative scenes were replayed in an immersive driving cabin to obtain eye-tracking indicators, including pupil diameter and gaze distribution, together with subjective evaluations. Finally, the physical lighting parameters, environmental features, and human-factor indicators were aligned by timestamp for statistical testing and factor screening. 2.1 Measured variables The variable system in this study comprised three categories: physical optical-stimulus indicators, glare-related indicators, and human-response indicators. Physical optical stimuli mainly included eye-level illuminance at the driver position, horizontal illuminance at the instrument panel, and luminance of key surfaces such as the ceiling, sidewalls, and road surface. Temporal fluctuation characteristics along the travel direction were further extracted from these measurements. Threshold increment (TI) was used to characterize disability glare, following the CIE-recommended model[ 17 ](Eq. (1)). In this equation, \(\:{E}_{eye}\) represent the measured eye-level illuminance; \(\:\theta\:\) represent the angle between the glare source and the line of sight; \(\:K\) represent an empirical constant; \(\:{L}_{av}\) represent the average background luminance.In the present study, the commonly used TI reference threshold of 15% was adopted as a safety-related benchmark rather than a direct predictor of subjective comfort. Eq. (1) The calculation of \(\:\:TI\) $$\:TI=\frac{65{L}_{v}}{{\left({L}_{av}\right)}^{0.8}}$$ $$\:{L}_{v}=K\frac{{E}_{eye}}{{\theta\:}^{n}}$$ $$\:n=\left\{\begin{array}{c}2.3-0.7\text{log}\theta\:,\theta\:<2^\circ\:\\\:2,\theta\:\ge\:2^\circ\:\end{array}\right.$$ To characterize discomfort glare, the De Boer nine-point rating framework was adopted[ 18 ]. Because complete luminance-field measurements were not available for all scenes, the study used the approximate De Boer Rating approach proposed by Bullough et al. based on environmental lighting parameters[ 19 ] (Eq. (2)). In this formulation, \(\:{E}_{l}\) represent direct illuminance from the glare source at driver eye level; \(\:{E}_{s}\) represent additional illuminance caused by wall-reflected light; \(\:{E}_{a}\) represent ambient illuminance measured at eye level; \(\:{L}_{s}\) represent luminaire luminance. This proxy indicator was used for correlation analysis and mechanism-oriented interpretation together with eye-tracking and subjective ratings, rather than as a replacement for direct subjective assessment. Eq. (2) The calculation of \(\:DeBoerRating\) $$\:DeBoerRating=\left\{\begin{array}{c}6.6-6.41\text{log}DG,\:when\:the\:visual\:angle\:of\:the\:glare\:source\:is\:below\:0.3^\circ\:\\\:6.6-6.4\text{log}DG+1.4\text{log}\left(\frac{50000}{{L}_{s}}\right),\:when\:the\:visual\:angle\:of\:the\:glare\:source\:is\:above\:0.3^\circ\:\end{array}\right.$$ $$\:DG=\text{log}\left({E}_{l}+{E}_{s}\right)+0.6\text{log}\left({E}_{l}/{E}_{s}\right)-0.5\text{log}{E}_{a}$$ Human-response indicators included the mean pupil diameter and variance of pupil diameter change, which reflect visual adaptation and load regulation, as well as gaze-distribution characteristics used to characterize attentional capture and attentional shift. 2.2 Field measurements Field measurements were conducted during real nighttime driving through multiple urban tunnels, with the aim of synchronously obtaining both optical stimulus input and time-series data during vehicle movement. The measured variables included eye-level illuminance at the driver’s eye height, horizontal illuminance at the instrument panel, and luminance of the ceiling, sidewalls, and road surface. A KONICA MINOLTA CL-500A spectrophotometer was used to measure illuminance and spectral distribution, while a KONICA MINOLTA CS-160 luminance meter was used to measure the luminance of key surfaces. In addition, an in-vehicle multi-angle video-recording system was employed for scene review and subsequent extraction of luminance-related features, while vehicle speed was recorded simultaneously. To reduce interference from external ambient light, field measurements were carried out between 23:00 and 02:00. The illuminance meter operated in continuous mode along predefined routes, and both illuminance and video data were automatically recorded with timestamps. During data acquisition, all in-vehicle lighting and unnecessary light-emitting devices were turned off to minimize stray light inside the cabin. The driver followed the predefined route at a stable driving condition, while the remaining team members were responsible for instrument monitoring and data logging to ensure consistency across tunnels. Four representative urban tunnels were selected: Beiheng Passage, Bund Tunnel, Fuxing East Road Tunnel, and Xinjian Road Tunnel. These tunnels covered different lengths, alignments, interior decorative features, and reflective conditions. Measurements were completed for all four tunnels following a unified sequence. Representative segments with notable lighting fluctuations and reflective interference were identified from continuous sequences at the entrance, middle, and exit zones, providing the basis for the subsequent selection of replay stimuli in the laboratory experiment (Fig. 1 ). 2.3 Laboratory experiment 2.3.1 Experimental setup Based on the field data, an immersive driving-cabin replay experiment was conducted to verify, under more controlled conditions, the proposed mechanism linking spatiotemporal illuminance variation and tunnel-wall reflection to attentional capture and comfort differences, while reducing the influence of random field factors. The experimental setup consisted of a driving cabin, steering wheel, pedals, and a large-screen projection system, as shown in Fig. 2 . Representative tunnel-driving videos acquired in the field were replayed in the cabin. During the experiment, a Tobii eye-tracking system was used to synchronously record pupil diameter, gaze trajectory, and fixation events. 2.3.2 Experimental procedure A total of eight participants were recruited; however, owing to equipment malfunction and participant-related reasons, valid test data were ultimately obtained from only six participants. Therefore, all subsequent analyses were conducted based on the valid data from these six participants, resulting in a repeated-measures dataset of 6 participants × 4 scenes = 24 observations. All participants were students at Fudan University and signed an informed consent form before the experiment. The experimental protocol involving human participants was approved by the Ethical Committee of Fudan University (Approval No. FE255051). All methods were performed in accordance with the relevant guidelines and regulations and in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to participation. The participants were aged 20–24 years, with a mean age of 21.8 years and a standard deviation of 1.2 years. All participants met the following criteria: (1) no visual disorders such as color blindness or color weakness; (2) no history of photosensitive epilepsy or related conditions; (3) no other eye-related diseases; and (4) corrected visual acuity of 5.0. The external stimuli consisted of representative driving video segments from the four tunnels. To reduce order and learning effects, the four scenes were coded as A-D and presented in a Latin-square-based randomized order. The tunnel codes and replay durations were as follows: A, Beiheng Passage, 5 min 0 s; B, Bund Tunnel, 5 min 0 s; C, Xinjian Road Tunnel, 2 min 35 s; and D, Fuxing East Road Tunnel, 4 min 03 s. Participants entered the driving cabin in pairs and were seated in the driver and front-passenger positions, respectively. They were instructed to watch the replay in a normal driving posture while keeping both hands on the steering wheel. After each video, they completed an online questionnaire covering subjective comfort, sense of safety, immersion, and trust, which was then aggregated into a composite score. In the pilot test, the questionnaire showed high internal consistency and acceptable structural validity (Cronbach’s α = 0.989, KMO = 0.857, Bartlett’s test, p < 0.0001), supporting its suitability for subsequent statistical analysis. To control the initial adaptation state, participants underwent 15 min of dark adaptation before the formal experiment. The four tunnel videos were then presented in randomized order, with the duration of each video not exceeding 5 min. A rest period of approximately 5 min was provided between consecutive videos to reduce short-term fatigue and emotional carryover. The detailed experimental procedure is illustrated in Fig. 3 . 2.3.3 Statistical analysis of experimental data Eye-tracking data were preprocessed using the default algorithms in Tobii Studio to remove blink-related segments and signal-loss intervals, followed by gaze calibration and event segmentation. The mean pupil diameter and variance of pupil diameter change were then calculated. In addition, the proportion and hit rate of fixations falling within predefined areas of interest (AOIs) were computed to characterize attentional capture. To examine the associations between subjective comfort and objective indicators, repeated-measures correlation (rmcorr) was used to estimate within-subject relationships between the composite comfort score and each variable. Linear mixed-effects models (LMMs), with participant specified as a random intercept, were further applied as a robustness check. Holm correction was used to adjust p-values for multiple testing. Given the repeated-measures design with six participants under four tunnel conditions, the present analyses were mainly powered to detect medium-to-large within-subject effects, whereas smaller effects may have remained undetected. 2.4 Multidimensional data alignment To support subsequent mechanism-oriented analysis while ensuring consistency and comparability between subjective and objective data, the illuminance sequences, eye-tracking data, and questionnaire responses were aligned to a unified temporal reference. Segments with obvious missing values or insufficient quality were excluded. All indicators, including mean values, fluctuation intensity, and attention-related measures, were then calculated according to unified rules. This procedure was intended solely to support the subsequent association analysis and mechanism interpretation, rather than to construct a general evaluation model. 3 Results 3.1 Field illuminance distribution and temporal fluctuation characteristics 3.1.1 Longitudinal illuminance profiles and statistical characteristics Field measurements were conducted in four representative urban tunnels: Beiheng Passage, Bund Tunnel, Fuxing East Road Tunnel, and Xinjian Road Tunnel. During steady vehicle passage through each tunnel, the driver’s eye-level illuminance and the horizontal illuminance at the instrument panel were synchronously recorded and aligned using a unified timestamp. For cross-tunnel comparison, the mean and variance of illuminance were calculated for each tunnel to characterize overall illuminance level and temporal stability, respectively. The longitudinal variation curves of eye-level and horizontal illuminance are presented in Fig. 4 and the corresponding summary statistics are given in Appendix A (Table A.1(a) and A.1(b)). In the subsequent analysis, fluctuation-related variables were treated as key indicators of the temporal instability of the light field and were jointly analyzed together with glare metrics and eye-tracking indicators. As shown in Fig. 4 , eye-level illuminance and horizontal illuminance showed similar patterns. Bund Tunnel exhibited the highest mean horizontal illuminance (170.02 lx) and the largest variance (4557.92), whereas Fuxing East Road Tunnel and Xinjian Road Tunnel showed lower mean levels and substantially smaller variances. Overall, the latter two tunnels displayed smaller changes in light intensity at tunnel entry and weaker fluctuations during passage, suggesting a relatively more stable light environment and, accordingly, a lower likelihood of inducing pronounced pupil adjustment. 3.2 Calculation and comparison of glare indicators 3.2.1 De Boer Rating To enable a comparable assessment of tunnel discomfort glare in the absence of full-field luminance measurements, this study used an approximate De Boer Rating derived from optical parameters as a proxy indicator. The background luminance required for the calculation was obtained from measured road-surface luminance. Under field constraints, glare-source luminance was inversely estimated from illuminance measurements and geometric assumptions, including the height difference between the observation point and the road surface, the distance between the light source and the observation point, and the equivalent emitting area. This approximation was used only for relative comparison and ranking, rather than for absolute engineering classification of glare level. The calculated values and the comparison results are shown in Fig. 5 . Among the 12 measurement locations across the four tunnels, Xinjian Road Tunnel showed the best overall performance, with a full-section mean De Boer Rating of 6.51 ± 0.52 and a maximum value of 7.09 at the exit section. In contrast, Bund Tunnel showed the lowest overall rating, with a full-section mean of 4.36 ± 0.30, remaining below the comfort threshold of 5.0 throughout the tunnel. A Kruskal-Wallis test indicated a significant overall difference among tunnels (H(3) = 9.46, p = 0.024), and post hoc Dunn testing with adjusted p-values showed that Xinjian Road Tunnel performed significantly better than Bund Tunnel (p = 0.013). 3.2.2 Threshold increment (TI) Threshold increment (TI) was further calculated to characterize the risk of disability glare. In the calculation, eye-level illuminance and background luminance were based on field measurements, while parameters such as the angle between the line of sight and the glare source were assigned using unified assumptions to ensure cross-tunnel comparability. The calculated TI values are summarized in Fig. 6 . None of the tunnels exceeded the commonly used threshold of 15%. This indicates that, within the range covered by the present sample, TI is better interpreted as a lower-bound constraint on disability-glare risk than as a sole explanatory variable for differences in subjective comfort. 3.3 Eye-tracking indicators: attentional allocation and visual load To examine the saliency-driven attentional capture potentially induced by sidewall reflections and decorative light bands, the tunnel scenes were divided into several predefined areas of interest (AOIs), including the left and right sidewalls, as illustrated in Fig. 7 . Fixation events were identified using the default I-VT algorithm in Tobii Glasses, and the AOI hit rate was calculated based on whether a fixation fell within a predefined AOI. AOI indices were first calculated at the participant level for each scene and then summarized as scene-level means and standard deviations. Participant-level data were used for the subsequent rmcorr and LMM analyses. The descriptive AOI and pupil diameter results are presented in supplementary Table A3. The mean AOI hit rates were 0.71% for Tunnel A, 0.79% for Tunnel B, 0.56% for Tunnel C, and 0.12% for Tunnel D. The higher AOI hit rates observed in Tunnels A and B indicate that drivers more frequently directed their gaze toward the predefined sidewall-related regions in these scenes. Combined with field observations showing decorative luminaires and sidewall reflection in Tunnels A and B, this result suggests that the coupled effect of decorative lighting and wall reflection may be one of the factors that divert attention away from the forward driving path. Pupil diameter reflects both changes in light stimulation and the regulation of cognitive load. After preprocessing the pupil data in Tobii Studio, the mean pupil diameter and the variance of pupil fluctuation were extracted as objective indicators of visual load. These variables were first calculated at the participant level by scene and then summarized as scene-level means and variances to describe how different scenes affected pupil dynamics. Participant-level data were again used for inferential analyses. As shown in Table 3, Tunnels A and B, which were associated with poorer light environments, generally exhibited larger mean pupil diameters and stronger fluctuations than Tunnels C and D. These results indicate that the visually less comfortable tunnel scenes were associated with greater pupil dilation and stronger temporal instability in pupil response, whereas the better-performing scenes showed smaller and more stable pupil responses. 3.4 Associations between comfort rating and objective indicators To identify the factors most strongly associated with visual comfort, the composite questionnaire score was linked with eye-tracking, gaze-behavior, and lighting-environment indicators. Specifically, the analysis included mean pupil diameter, pupil-diameter fluctuation, AOI hit rate, mean and variance of eye-level illuminance, mean and variance of horizontal illuminance, TI, and the De Boer Rating. The repeated-measures correlation (rmcorr) and linear mixed-effects model (LMM) results are summarized in Table 1 . Table 1 rmcorr and LMM results for the associations between experimental parameters and mean questionnaire scores Variable r value Holm-adjusted p value (r) LMM β Holm-adjusted p value (β) Mean pupil diameter -0.657 0.009 -11.63 0.001 Variance of pupil diameter -0.647 0.009 -11.44 0.001 AOI hit rate -0.659 0.009 -11.66 0.001 TI value -0.261 0.281 -4.61 0.252 De Boer rating 0.668 0.009 11.82 0.001 Mean illuminance at eye position -0.764 < 0.001 -13.52 < 0.001 Variance of illuminance at eye position -0.820 < 0.001 -14.50 < 0.001 Mean horizontal illuminance -0.777 < 0.001 -13.75 < 0.001 Variance of horizontal illuminance -0.783 < 0.001 -13.85 < 0.001 As shown in Table 1 , comfort rating was significantly associated with multiple objective indicators. Pupil-related variables and AOI hit rate were all negatively correlated with comfort, indicating that visually less comfortable tunnel scenes were accompanied by greater pupil response and more frequent gaze shifts toward non-primary regions. De Boer Rating was positively correlated with comfort, whereas TI was not significant in the present dataset. Because TI values in all four tunnels remained below the commonly used threshold of 15%, TI is better interpreted here as a lower-bound safety-related constraint than as a continuous explanatory variable for comfort differences. Among all indicators, illuminance-related variables showed the strongest associations with comfort. In particular, the variance of eye-level illuminance and the variance of horizontal illuminance had the largest negative correlations, suggesting that temporal instability of the light field may be more important for discomfort than mean illuminance level alone. The mean illuminance values were also negatively associated with comfort, but these relationships are more likely to reflect combined effects of glare, reflection, and spatial contrast under brighter conditions rather than a direct effect of illuminance itself. The fixed-effect directions in the LMM were consistent with the rmcorr results, and the significance conclusions for the main indicators remained unchanged after Holm correction, indicating that the observed associations were directionally stable under the present sample and model specification. Detailed LMM results are provided in Appendix A (Table A2 ). 4 Discussion Based on the field measurements, immersive replay, and coupled subjective-objective analysis, a preliminary pathway may explain why visual discomfort persists in standards-compliant tunnels. In the present dataset, comfort was not determined solely by average illuminance, but appeared to depend mainly on temporal instability and spatial saliency. Stronger illuminance fluctuation and more pronounced bright-dark alternation were associated with larger pupil responses, suggesting more frequent visual adaptation. At the same time, sidewall reflections and decorative light bands created salient bright regions that attracted lateral gaze. Together, these factors may increase attentional capture and discomfort even when disability glare thresholds are not exceeded. Under the present conditions, TI is better interpreted as a lower-bound indicator of glare-related safety risk than as a strong predictor of subjective comfort. By contrast, fluctuation-related optical indicators, reflection-related spatial characteristics, and attention-allocation measures were more informative for explaining discomfort. Tunnel-lighting evaluation may therefore be understood as a two-layer framework: one layer for safety and visibility, and another for visual comfort and attentional stability. This proposed mechanism should be regarded as an interpretive framework rather than a validated causal model. Several limitations should be noted. Real-tunnel measurements were influenced by traffic, reflective objects, and dynamic background lighting. Laboratory replay improved scene comparability, but display refresh rate and dynamic-range mapping may have altered some original stimulus characteristics. Temporal instability was represented mainly by statistical descriptors rather than frequency-domain features, and AOI analysis relied on static region definitions. In addition, the sample size was limited, so the findings are more suitable for mechanism exploration than for threshold definition or predictive modeling. Future work should combine controlled experiments, higher-fidelity simulations, larger samples, and broader tunnel conditions to test the causal validity of the proposed pathway. Particular attention should be given to luminaire arrangement and modulation characteristics, sidewall reflectance and geometry, and the saliency of decorative lighting. More robust attention-related indicators, together with mixed-effects or mediation-path analyses, may further clarify the relative contributions of different stimulus dimensions. 5 Conclusions This study investigated visual discomfort in urban road tunnels under standards-compliant conditions by combining field measurements with immersive driving-cabin replay. Physical characteristics of the light stimuli, eye-tracking responses, pupil dynamics, and subjective evaluations were synchronously analyzed within a unified time-aligned framework. The results showed that, across four representative tunnel samples, drivers’ comfort ratings still differed substantially even though threshold-based glare indicators such as TI remained within commonly accepted limits. This indicates that compliance with conventional engineering criteria does not necessarily guarantee a comfortable visual experience. Compared with TI alone, visual discomfort in the present dataset was more consistently associated with temporal instability of illuminance and luminance, as well as with attention shifts induced by sidewall reflection and decorative bright regions. In particular, greater illuminance fluctuation was linked to stronger pupil responses and lower comfort, while higher AOI hit rates suggested increased attentional capture by task-irrelevant bright regions. Taken together, the coupled subjective and objective evidence suggests that temporal light fluctuation, reflective interference from tunnel walls, and attention capture by decorative lighting may be important factors explaining why discomfort can still occur within nominally acceptable glare limits. Within the optical range covered in this study, TI is therefore better interpreted as a lower-bound safety constraint, whereas differences in comfort require explanatory variables related to spatiotemporal instability and visual saliency. Given the limited sample size and stimulus range, the present conclusions should be understood primarily as mechanism-oriented findings and hypothesis-generating evidence. Future studies with expanded tunnel samples, more complete luminance-field measurements, and controlled-variable experiments are needed to further verify the causal validity and interactions of the proposed mechanism. Declarations CRediT author contributions Zhenghao Jin: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Visualization, Writing – original draft. Yandan Lin: Conceptualization, Supervision, Methodology, Writing – review, Project administration. Haiping Shen: Methodology, Validation, Writing – review. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Author Contribution Zhenghao Jin: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Visualization, Writing – original draft.Yandan Lin: Conceptualization, Supervision, Methodology, Writing – review, Project administration.Haiping Shen: Methodology, Validation, Writing – review. Data Availability The data supporting the findings of this study are available from the corresponding author upon reasonable request, subject to ethical and privacy restrictions. References Chen, W., Chen, D. & Lin, Y. Discussion and practice of tunnel lighting. in Proceedings of the 11th Cross-Strait Symposium on Lighting Technology and Marketing, 279–285Ningbo, China, (2004). Schmidt-Clausen, H. J. et al. Mesopic vision and ambient light inside automobiles. China Light Lighting . 8 , 7–10 (2006). Lin, Y. et al. Model predicting discomfort glare caused by LED road lights. Opt. Express . 22 , 18056–18071 (2014). Ji, Z., Li, N. & Lin, Y. Decoding visual fatigue through pupil dynamics. In 2023 20th China International Forum on Solid State Lighting & 2023 9th International Forum on Wide Bandgap Semiconductors (SSLCHINA: IFWS), 424–428 DOI: 10.1109/SSLChinaIFWS60785.2023.10399698 . (2023). Zhao, X., Li, M., Hou, D., Lin, Y. & Xu, W. The Impact of Temporal Light Modulation on Psychological Responses Assessed by the Stroboscopic Visibility Measure. LEUKOS (2024). 10.1080/15502724.2024.2365686 Su, B., Hu, J., Zeng, J. & Wang, R. Traffic Safety Improvement via Optimizing Light Environment in Highway Tunnels. Int. J. Environ. Res. Public. Health . 19 , 8517 (2022). Yi, X., Zhao, R. & Lin, Y. The impact of nighttime car body lighting on pedestrians’ distraction: A virtual reality simulation based on bottom-up attention mechanism. Saf. Sci. 180 , 106633 (2024). He, S., Du, Z., Han, L., Wang, S. & Chen, Y. Impact of urban tunnel sidewall effect on drivers: Examining visual characteristics and driving behavior across different lanes. Tunn. Undergr. Space Technol. 143 , 105476 (2024). Cai, X., Tian, X. & Weng, J. Study on threshold luminance difference in highway tunnel lighting. Comput. Simul. 41 , 154–159 (2024). Han, Z., Wang, J. & Guan, Y. Lighting spacing of LED lamps in curved tunnels. J. Southwest. Jiaotong Univ. 57 , 434–439 (2022). Li, M. & Liu, M. Study on tunnel lighting flicker frequency based on drivers’ heart rate variation. China Illum Eng. J. 26 , 80–86 (2015). He, X. & Ning, J. Impact of LED display brightness on urban traffic safety: a case study of Chongqing. China Illum Eng. J. 36 , 16–22 (2025). Ministry of Housing and Urban-Rural Development of the People's Republic of China. Standard for lighting design of urban roads (CJJ 45-2015). (2015). Ministry of Transport of the People's Republic of China. Guidelines for lighting design of highway tunnels (JTG/T D70/2-01- (2014). (2014). Commission Internationale de l’Éclairage. Guide for the lighting of road tunnels and underpasses, 2nd ed. (CIE 88:2004). (2004). Ministry of Housing and Urban-. Rural Development of the People's Republic of China. Standard for lighting design of buildings (GB/T 50034 – 2024, 2024). CIE TC 4–15. Road lighting calculations, 2nd ed. (CIE 140:2019, 2019). 10.25039/TR.140.2019 de Boer, J. B. & Schreuder, D. A. Glare as a Criterion for Quality in Street Lighting. Trans. Illum Eng. Soc. 32 , 117–135 (1967). Bullough, J. D., Brons, J. A., Qi, R. & Rea, M. S. Predicting discomfort glare from outdoor lighting installations. Light Res. Technol. 40 , 225–242 (2008). Appendix Table A1 Mean and variance of illuminance in the four test tunnels: (a) eye-level illuminance; (b) horizontal illuminance. (a) Eye-level illuminance Beiheng Passage Bund Tunnel Fuxing East Road Tunnel Xinjian Road Tunnel Mean 2.99 6.40 2.92 2.27 Variance 1.61 4.02 0.28 0.45 (b) Horizontal illuminance Beiheng Passage Bund Tunnel Fuxing East Road Tunnel Xinjian Road Tunnel Mean 70.29 170.02 65.92 29.99 Variance 874 4557.92 218.56 129.56 Table A.2 Robustness check results of the LMM Indicator LMM β (score per + 1 SD) 95% CI t p p Holm Variance of illuminance at eye position -14.50 [-19.18, -9.82] -6.07 1.23E-05 1.12E-04 Variance of horizontal illuminance -13.85 [-18.93, -8.77] -5.34 5.39E-05 4.31E-04 Mean horizontal illuminance -13.75 [-18.89, -8.61] -5.24 6.60E-05 4.63E-04 Mean illuminance at eye position -13.52 [-18.79, -8.25] -5.03 1.04E-04 6.21E-04 De Boer rating 11.82 [5.74, 17.90] 3.81 0.00139 0.00696 AOI hit rate -11.66 [-17.80, -5.52] -3.72 0.00170 0.00696 Mean pupil diameter -11.63 [-17.78, -5.47] -3.70 0.00178 0.00696 Variance of pupil diameter -11.44 [-17.67, -5.21] -3.60 0.00222 0.00696 TI value -4.61 [-12.50, 3.28] -1.15 0.268 0.268 Table A.3. AOI hit rates and pupil diameter across the four tunnel scenes Tunnel ID Mean AOI hit rate ± standard deviation Mean pupil diameter (pixels) Variance of pupil diameter change A 0.71% ± 0.11% 115.65 28.57 B 0.79% ± 0.21% 111.37 27.75 C 0.56% ± 0.07% 100.47 23.57 D 0.12% ± 0.1% 99.33 16.84 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 11 May, 2026 Reviewers agreed at journal 01 May, 2026 Reviewers invited by journal 01 May, 2026 Editor assigned by journal 01 May, 2026 Editor invited by journal 10 Apr, 2026 Submission checks completed at journal 07 Apr, 2026 First submitted to journal 07 Apr, 2026 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-9290694","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":637595640,"identity":"d3adb4a6-a911-4b80-afd7-23b1ab7e7ee6","order_by":0,"name":"ZHENGHAO JIN","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"ZHENGHAO","middleName":"","lastName":"JIN","suffix":""},{"id":637595642,"identity":"29f7d0b2-95c8-4ca6-a6f4-1af4001d9cb7","order_by":1,"name":"YANDAN LIN","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYFCC5IYDDBVyDAwHQBw2orQkArWcMSZRCwNjGylaDI4nNh4unGeQ2Hf87AGGD2WHGfhnNxDQcuZhw+GZ2wwSZ57JS2Ccce4wg8SdAwS03EhsOMy77U/ihgM5Bsy8bYcZDCQSiNEyxyBxw/k3Bsx/idfSANRyA2gLIzFaJEF+4TlmYDzzxhuDgz3n0nkkbhDQwnc8+fBnnhoD2b7zOYYPfpRZy/HPIKBF4QASB8Tmwa8eCOQbCCoZBaNgFIyCEQ8Ai5hO1nlXnJMAAAAASUVORK5CYII=","orcid":"","institution":"Fudan University","correspondingAuthor":true,"prefix":"","firstName":"YANDAN","middleName":"","lastName":"LIN","suffix":""},{"id":637595644,"identity":"eeeb0b5e-5eae-49ba-88fc-96781732fdb7","order_by":2,"name":"HAIPING SHEN","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"HAIPING","middleName":"","lastName":"SHEN","suffix":""}],"badges":[],"createdAt":"2026-04-01 10:51:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9290694/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9290694/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108995081,"identity":"2fb83456-d067-4beb-b487-6792eab85848","added_by":"auto","created_at":"2026-05-11 14:01:16","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":888352,"visible":true,"origin":"","legend":"\u003cp\u003eField measurement route for the tunnel lighting environment\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/5fb3c8837ba8c6ed381bc288.png"},{"id":108995089,"identity":"9bc535a2-da07-4cd5-86a1-ee1c99c276b0","added_by":"auto","created_at":"2026-05-11 14:01:23","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":383713,"visible":true,"origin":"","legend":"\u003cp\u003eExperimental setup of the simulated driving cabin: (a) side view; (b) front view\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/7c40a1e8234ef2c9b799fb0b.jpeg"},{"id":108995036,"identity":"b9779824-2a74-4449-8c68-7d6ab54f6451","added_by":"auto","created_at":"2026-05-11 14:01:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":15978,"visible":true,"origin":"","legend":"\u003cp\u003eExperimental procedure\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/e13820e91ade9da171062d30.png"},{"id":108995107,"identity":"3f92ada9-0da3-4258-a36b-3097501bac0c","added_by":"auto","created_at":"2026-05-11 14:01:34","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":163439,"visible":true,"origin":"","legend":"\u003cp\u003eLongitudinal variations in eye-level and horizontal illuminance in the four tunnels: (a) Beiheng Passage; (b) Bund Tunnel; (c) Fuxing East Road Tunnel; (d) Xinjian Road Tunnel\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/2da7296795f177e7e2bf9edc.png"},{"id":108995079,"identity":"0012f1e4-499d-4399-b911-56249159d091","added_by":"auto","created_at":"2026-05-11 14:01:16","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":124904,"visible":true,"origin":"","legend":"\u003cp\u003eApproximate De Boer Rating in the four tunnels\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/9d0d1e91a0521520036e1ba9.png"},{"id":108995080,"identity":"c47be386-77de-4a9c-9aba-e412c275381b","added_by":"auto","created_at":"2026-05-11 14:01:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":36550,"visible":true,"origin":"","legend":"\u003cp\u003eTI values in the four tunnels\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/a5cb9ff101f8bfad3facd365.png"},{"id":108995084,"identity":"326d3177-075b-4030-928d-219fd619a3eb","added_by":"auto","created_at":"2026-05-11 14:01:19","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":465590,"visible":true,"origin":"","legend":"\u003cp\u003eDefinition of the areas of interest (AOIs)\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/fae9148c212c3bb71abf1695.png"},{"id":108995130,"identity":"6338fa0e-08c0-45ff-8f1e-b6c13344953d","added_by":"auto","created_at":"2026-05-11 14:02:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2414114,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9290694/v1/4ca5fc1e-571c-45d1-87d8-e3d3e29e417c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Factors Contributing to Visual Discomfort in Standards-Compliant Urban Road Tunnels: Evidence from Field Measurements and Immersive Driving Experiments","fulltext":[{"header":"1 Introduction","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e1.1 Existing studies\u003c/h2\u003e \u003cp\u003eUrban underground roads, including cross-river tunnels, underpasses, and subsurface corridors, play an important role in alleviating surface congestion and improving traffic efficiency. However, their enclosed or semi-enclosed spatial characteristics also make the light field more prone to superposition and amplification, which may give rise to visual disturbances such as glare, flicker, zebra effects, and sidewall reflections. These disturbances not only affect visibility, but may also increase visual workload and attentional competition during dynamic driving, thereby reducing driving comfort and safety margins.\u003c/p\u003e \u003cp\u003eExisting studies have mainly approached tunnel lighting from the perspectives of visual adaptation and glare control. Previous research has proposed strategies such as threshold-zone transition lighting and segmented lighting design, and has established predictive relationships between perceived glare and photometric variables from a human-factors perspective[\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. In parallel, temporal luminance variations caused by the combined effects of luminaire arrangement and vehicle speed have been regarded as an important source of time-varying illuminance fluctuations. However, a single modulation-frequency-based indicator is often insufficient to capture inter-individual differences in driver preference as well as differences in the overall spatial luminance structure of tunnels[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTo quantify discomfort during tunnel driving more systematically, previous studies have introduced multidimensional evaluation approaches. From the perspective of human response mechanisms, pupil dynamics and eye-movement indicators can be used to characterize the coupled effects of changing light stimuli and cognitive load, and to reveal how salient bright regions, such as decorative light bands and highly reflective sidewalls, may trigger attentional capture and gaze shifts[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. At the same time, existing standards and engineering assessment frameworks continue to emphasize baseline indicators such as luminance uniformity, illuminance level, luminance level, and threshold increment (TI) limits for glare control[\u003cspan additionalcitationids=\"CR10 CR11\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. In practice, however, it is still common to observe substantial differences in driver or passenger experience even when engineering criteria are formally satisfied. This suggests that no single indicator is sufficient to explain subjective discomfort under compound visual disturbances.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e1.2 Current standards and their limitations\u003c/h2\u003e \u003cp\u003eFrom a standards perspective, current tunnel lighting design requirements are primarily centered on segmented lighting strategies and baseline photometric quality constraints. For example, the Standard for Lighting Design of Urban Roads[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] requires daytime lighting in urban tunnels to be designed by dividing the tunnel into entrance, transition, interior, and exit zones, with lighting criteria determined according to vehicle speed and traffic volume. The standard also specifies that non-functional decorative lighting installed along the roadside should not interfere with functional lighting, and that frequently changing dynamic lighting should be avoided to prevent visual disturbance to drivers. Similarly, technical specifications for highway tunnel lighting[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], as well as their adoption in local standards, indicate that engineering evaluation and acceptance still focus mainly on baseline indicators such as luminance level, uniformity, and glare control across the entrance, transition, interior, and exit zones.\u003c/p\u003e \u003cp\u003eWith regard to glare, international guidelines such as CIE 88 commonly use threshold increment (TI) as a key indicator for controlling disability glare and propose upper-limit constraints for TI[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. In the tunnels investigated in this study, TI remained within this type of recommended threshold range overall, while subjective comfort still differed substantially across scenes. This suggests that, under conditions where disability-glare thresholds are not exceeded, experiential differences may be driven instead by factors that are relatively underrepresented in current standards, such as temporal instability of illuminance or luminance, sidewall reflections, and attentional competition induced by decorative bright zones. At the national level, GB/T 50034\u0026thinsp;\u0026minus;\u0026thinsp;2024 has already incorporated flicker and stroboscopic effects into lighting-quality provisions, reflecting an increasing concern over risks associated with temporally modulated light[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. However, this standard is intended for architectural lighting and is not sufficient to directly address the process-based mechanisms involved in dynamic tunnel-driving tasks.\u003c/p\u003e \u003cp\u003eAgainst this background, the present study is positioned as a mechanism-oriented exploratory study and variable-screening effort. Rather than establishing universal thresholds or a predictive model, the study aims to collect multidimensional data under standards-compliant tunnel conditions, build a coupled database integrating physical lighting parameters and human-response indicators, and statistically identify the optical and human-factors variables that better explain differences in subjective discomfort. The findings are expected to provide hypotheses, candidate variables, and directional evidence for subsequent controlled experiments and model development.\u003c/p\u003e \u003c/div\u003e"},{"header":"2 Materials and methods","content":"\u003cp\u003eThis study adopted a three-stage design consisting of field measurement of optical stimuli, laboratory validation of human responses, and multidimensional correlation analysis. First, physical lighting parameters were synchronously collected in real tunnel-driving environments to characterize typical visually uncomfortable stimuli and to identify representative tunnel lighting scenes. Second, representative scenes were replayed in an immersive driving cabin to obtain eye-tracking indicators, including pupil diameter and gaze distribution, together with subjective evaluations. Finally, the physical lighting parameters, environmental features, and human-factor indicators were aligned by timestamp for statistical testing and factor screening.\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Measured variables\u003c/h2\u003e \u003cp\u003eThe variable system in this study comprised three categories: physical optical-stimulus indicators, glare-related indicators, and human-response indicators. Physical optical stimuli mainly included eye-level illuminance at the driver position, horizontal illuminance at the instrument panel, and luminance of key surfaces such as the ceiling, sidewalls, and road surface. Temporal fluctuation characteristics along the travel direction were further extracted from these measurements.\u003c/p\u003e \u003cp\u003eThreshold increment (TI) was used to characterize disability glare, following the CIE-recommended model[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e](Eq.\u0026nbsp;(1)). In this equation, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{eye}\\)\u003c/span\u003e\u003c/span\u003e represent the measured eye-level illuminance;\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\theta\\:\\)\u003c/span\u003e\u003c/span\u003e represent the angle between the glare source and the line of sight;\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:K\\)\u003c/span\u003e\u003c/span\u003e represent an empirical constant; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{L}_{av}\\)\u003c/span\u003e\u003c/span\u003e represent the average background luminance.In the present study, the commonly used TI reference threshold of 15% was adopted as a safety-related benchmark rather than a direct predictor of subjective comfort.\u003c/p\u003e \u003cp\u003eEq.\u0026nbsp;(1) The calculation of\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\:TI\\)\u003c/span\u003e\u003c/span\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:TI=\\frac{65{L}_{v}}{{\\left({L}_{av}\\right)}^{0.8}}$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{L}_{v}=K\\frac{{E}_{eye}}{{\\theta\\:}^{n}}$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:n=\\left\\{\\begin{array}{c}2.3-0.7\\text{log}\\theta\\:,\\theta\\:\u0026lt;2^\\circ\\:\\\\\\:2,\\theta\\:\\ge\\:2^\\circ\\:\\end{array}\\right.$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eTo characterize discomfort glare, the De Boer nine-point rating framework was adopted[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Because complete luminance-field measurements were not available for all scenes, the study used the approximate De Boer Rating approach proposed by Bullough et al. based on environmental lighting parameters[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] (Eq.\u0026nbsp;(2)). In this formulation, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{l}\\)\u003c/span\u003e\u003c/span\u003e represent direct illuminance from the glare source at driver eye level; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{s}\\)\u003c/span\u003e\u003c/span\u003e represent additional illuminance caused by wall-reflected light; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{a}\\)\u003c/span\u003e\u003c/span\u003e represent ambient illuminance measured at eye level; \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{L}_{s}\\)\u003c/span\u003e\u003c/span\u003e represent luminaire luminance. This proxy indicator was used for correlation analysis and mechanism-oriented interpretation together with eye-tracking and subjective ratings, rather than as a replacement for direct subjective assessment.\u003c/p\u003e \u003cp\u003eEq.\u0026nbsp;(2) The calculation of \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:DeBoerRating\\)\u003c/span\u003e\u003c/span\u003e\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$$\\:DeBoerRating=\\left\\{\\begin{array}{c}6.6-6.41\\text{log}DG,\\:when\\:the\\:visual\\:angle\\:of\\:the\\:glare\\:source\\:is\\:below\\:0.3^\\circ\\:\\\\\\:6.6-6.4\\text{log}DG+1.4\\text{log}\\left(\\frac{50000}{{L}_{s}}\\right),\\:when\\:the\\:visual\\:angle\\:of\\:the\\:glare\\:source\\:is\\:above\\:0.3^\\circ\\:\\end{array}\\right.$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$$\\:DG=\\text{log}\\left({E}_{l}+{E}_{s}\\right)+0.6\\text{log}\\left({E}_{l}/{E}_{s}\\right)-0.5\\text{log}{E}_{a}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHuman-response indicators included the mean pupil diameter and variance of pupil diameter change, which reflect visual adaptation and load regulation, as well as gaze-distribution characteristics used to characterize attentional capture and attentional shift.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Field measurements\u003c/h2\u003e \u003cp\u003eField measurements were conducted during real nighttime driving through multiple urban tunnels, with the aim of synchronously obtaining both optical stimulus input and time-series data during vehicle movement. The measured variables included eye-level illuminance at the driver\u0026rsquo;s eye height, horizontal illuminance at the instrument panel, and luminance of the ceiling, sidewalls, and road surface. A KONICA MINOLTA CL-500A spectrophotometer was used to measure illuminance and spectral distribution, while a KONICA MINOLTA CS-160 luminance meter was used to measure the luminance of key surfaces. In addition, an in-vehicle multi-angle video-recording system was employed for scene review and subsequent extraction of luminance-related features, while vehicle speed was recorded simultaneously.\u003c/p\u003e \u003cp\u003eTo reduce interference from external ambient light, field measurements were carried out between 23:00 and 02:00. The illuminance meter operated in continuous mode along predefined routes, and both illuminance and video data were automatically recorded with timestamps. During data acquisition, all in-vehicle lighting and unnecessary light-emitting devices were turned off to minimize stray light inside the cabin. The driver followed the predefined route at a stable driving condition, while the remaining team members were responsible for instrument monitoring and data logging to ensure consistency across tunnels.\u003c/p\u003e \u003cp\u003eFour representative urban tunnels were selected: Beiheng Passage, Bund Tunnel, Fuxing East Road Tunnel, and Xinjian Road Tunnel. These tunnels covered different lengths, alignments, interior decorative features, and reflective conditions. Measurements were completed for all four tunnels following a unified sequence. Representative segments with notable lighting fluctuations and reflective interference were identified from continuous sequences at the entrance, middle, and exit zones, providing the basis for the subsequent selection of replay stimuli in the laboratory experiment (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Laboratory experiment\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Experimental setup\u003c/h2\u003e \u003cp\u003eBased on the field data, an immersive driving-cabin replay experiment was conducted to verify, under more controlled conditions, the proposed mechanism linking spatiotemporal illuminance variation and tunnel-wall reflection to attentional capture and comfort differences, while reducing the influence of random field factors. The experimental setup consisted of a driving cabin, steering wheel, pedals, and a large-screen projection system, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Representative tunnel-driving videos acquired in the field were replayed in the cabin. During the experiment, a Tobii eye-tracking system was used to synchronously record pupil diameter, gaze trajectory, and fixation events.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Experimental procedure\u003c/h2\u003e \u003cp\u003eA total of eight participants were recruited; however, owing to equipment malfunction and participant-related reasons, valid test data were ultimately obtained from only six participants. Therefore, all subsequent analyses were conducted based on the valid data from these six participants, resulting in a repeated-measures dataset of 6 participants \u0026times; 4 scenes\u0026thinsp;=\u0026thinsp;24 observations. All participants were students at Fudan University and signed an informed consent form before the experiment. The experimental protocol involving human participants was approved by the Ethical Committee of Fudan University (Approval No. FE255051). All methods were performed in accordance with the relevant guidelines and regulations and in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to participation.\u003c/p\u003e \u003cp\u003eThe participants were aged 20\u0026ndash;24 years, with a mean age of 21.8 years and a standard deviation of 1.2 years. All participants met the following criteria:\u003c/p\u003e \u003cp\u003e(1) no visual disorders such as color blindness or color weakness;\u003c/p\u003e \u003cp\u003e(2) no history of photosensitive epilepsy or related conditions;\u003c/p\u003e \u003cp\u003e(3) no other eye-related diseases; and\u003c/p\u003e \u003cp\u003e(4) corrected visual acuity of 5.0.\u003c/p\u003e \u003cp\u003eThe external stimuli consisted of representative driving video segments from the four tunnels. To reduce order and learning effects, the four scenes were coded as A-D and presented in a Latin-square-based randomized order. The tunnel codes and replay durations were as follows: A, Beiheng Passage, 5 min 0 s; B, Bund Tunnel, 5 min 0 s; C, Xinjian Road Tunnel, 2 min 35 s; and D, Fuxing East Road Tunnel, 4 min 03 s.\u003c/p\u003e \u003cp\u003eParticipants entered the driving cabin in pairs and were seated in the driver and front-passenger positions, respectively. They were instructed to watch the replay in a normal driving posture while keeping both hands on the steering wheel. After each video, they completed an online questionnaire covering subjective comfort, sense of safety, immersion, and trust, which was then aggregated into a composite score. In the pilot test, the questionnaire showed high internal consistency and acceptable structural validity (Cronbach\u0026rsquo;s α\u0026thinsp;=\u0026thinsp;0.989, KMO\u0026thinsp;=\u0026thinsp;0.857, Bartlett\u0026rsquo;s test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), supporting its suitability for subsequent statistical analysis.\u003c/p\u003e \u003cp\u003eTo control the initial adaptation state, participants underwent 15 min of dark adaptation before the formal experiment. The four tunnel videos were then presented in randomized order, with the duration of each video not exceeding 5 min. A rest period of approximately 5 min was provided between consecutive videos to reduce short-term fatigue and emotional carryover. The detailed experimental procedure is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 Statistical analysis of experimental data\u003c/h2\u003e \u003cp\u003eEye-tracking data were preprocessed using the default algorithms in Tobii Studio to remove blink-related segments and signal-loss intervals, followed by gaze calibration and event segmentation. The mean pupil diameter and variance of pupil diameter change were then calculated. In addition, the proportion and hit rate of fixations falling within predefined areas of interest (AOIs) were computed to characterize attentional capture.\u003c/p\u003e \u003cp\u003eTo examine the associations between subjective comfort and objective indicators, repeated-measures correlation (rmcorr) was used to estimate within-subject relationships between the composite comfort score and each variable. Linear mixed-effects models (LMMs), with participant specified as a random intercept, were further applied as a robustness check. Holm correction was used to adjust p-values for multiple testing.\u003c/p\u003e \u003cp\u003eGiven the repeated-measures design with six participants under four tunnel conditions, the present analyses were mainly powered to detect medium-to-large within-subject effects, whereas smaller effects may have remained undetected.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Multidimensional data alignment\u003c/h2\u003e \u003cp\u003eTo support subsequent mechanism-oriented analysis while ensuring consistency and comparability between subjective and objective data, the illuminance sequences, eye-tracking data, and questionnaire responses were aligned to a unified temporal reference. Segments with obvious missing values or insufficient quality were excluded. All indicators, including mean values, fluctuation intensity, and attention-related measures, were then calculated according to unified rules. This procedure was intended solely to support the subsequent association analysis and mechanism interpretation, rather than to construct a general evaluation model.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Field illuminance distribution and temporal fluctuation characteristics\u003c/h2\u003e\n \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\n \u003ch2\u003e3.1.1 Longitudinal illuminance profiles and statistical characteristics\u003c/h2\u003e\n \u003cp\u003eField measurements were conducted in four representative urban tunnels: Beiheng Passage, Bund Tunnel, Fuxing East Road Tunnel, and Xinjian Road Tunnel. During steady vehicle passage through each tunnel, the driver\u0026rsquo;s eye-level illuminance and the horizontal illuminance at the instrument panel were synchronously recorded and aligned using a unified timestamp. For cross-tunnel comparison, the mean and variance of illuminance were calculated for each tunnel to characterize overall illuminance level and temporal stability, respectively. The longitudinal variation curves of eye-level and horizontal illuminance are presented in Fig. \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and the corresponding summary statistics are given in Appendix A (Table A.1(a) and A.1(b)). In the subsequent analysis, fluctuation-related variables were treated as key indicators of the temporal instability of the light field and were jointly analyzed together with glare metrics and eye-tracking indicators.\u003c/p\u003e\n \u003cp\u003eAs shown in Fig. \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, eye-level illuminance and horizontal illuminance showed similar patterns. Bund Tunnel exhibited the highest mean horizontal illuminance (170.02 lx) and the largest variance (4557.92), whereas Fuxing East Road Tunnel and Xinjian Road Tunnel showed lower mean levels and substantially smaller variances. Overall, the latter two tunnels displayed smaller changes in light intensity at tunnel entry and weaker fluctuations during passage, suggesting a relatively more stable light environment and, accordingly, a lower likelihood of inducing pronounced pupil adjustment.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Calculation and comparison of glare indicators\u003c/h2\u003e\n \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e\n \u003ch2\u003e3.2.1 De Boer Rating\u003c/h2\u003e\n \u003cp\u003eTo enable a comparable assessment of tunnel discomfort glare in the absence of full-field luminance measurements, this study used an approximate De Boer Rating derived from optical parameters as a proxy indicator. The background luminance required for the calculation was obtained from measured road-surface luminance. Under field constraints, glare-source luminance was inversely estimated from illuminance measurements and geometric assumptions, including the height difference between the observation point and the road surface, the distance between the light source and the observation point, and the equivalent emitting area. This approximation was used only for relative comparison and ranking, rather than for absolute engineering classification of glare level. The calculated values and the comparison results are shown in Fig. \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e\n \u003cp\u003eAmong the 12 measurement locations across the four tunnels, Xinjian Road Tunnel showed the best overall performance, with a full-section mean De Boer Rating of 6.51\u0026thinsp;\u0026plusmn;\u0026thinsp;0.52 and a maximum value of 7.09 at the exit section. In contrast, Bund Tunnel showed the lowest overall rating, with a full-section mean of 4.36\u0026thinsp;\u0026plusmn;\u0026thinsp;0.30, remaining below the comfort threshold of 5.0 throughout the tunnel. A Kruskal-Wallis test indicated a significant overall difference among tunnels (H(3)\u0026thinsp;=\u0026thinsp;9.46, p\u0026thinsp;=\u0026thinsp;0.024), and post hoc Dunn testing with adjusted p-values showed that Xinjian Road Tunnel performed significantly better than Bund Tunnel (p\u0026thinsp;=\u0026thinsp;0.013).\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec17\" class=\"Section3\"\u003e\n \u003ch2\u003e3.2.2 Threshold increment (TI)\u003c/h2\u003e\n \u003cp\u003eThreshold increment (TI) was further calculated to characterize the risk of disability glare. In the calculation, eye-level illuminance and background luminance were based on field measurements, while parameters such as the angle between the line of sight and the glare source were assigned using unified assumptions to ensure cross-tunnel comparability. The calculated TI values are summarized in Fig. \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e\n \u003cp\u003eNone of the tunnels exceeded the commonly used threshold of 15%. This indicates that, within the range covered by the present sample, TI is better interpreted as a lower-bound constraint on disability-glare risk than as a sole explanatory variable for differences in subjective comfort.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Eye-tracking indicators: attentional allocation and visual load\u003c/h2\u003e\n \u003cp\u003eTo examine the saliency-driven attentional capture potentially induced by sidewall reflections and decorative light bands, the tunnel scenes were divided into several predefined areas of interest (AOIs), including the left and right sidewalls, as illustrated in Fig. \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e. Fixation events were identified using the default I-VT algorithm in Tobii Glasses, and the AOI hit rate was calculated based on whether a fixation fell within a predefined AOI. AOI indices were first calculated at the participant level for each scene and then summarized as scene-level means and standard deviations. Participant-level data were used for the subsequent rmcorr and LMM analyses. The descriptive AOI and pupil diameter results are presented in supplementary Table A3.\u003c/p\u003e\n \u003cp\u003eThe mean AOI hit rates were 0.71% for Tunnel A, 0.79% for Tunnel B, 0.56% for Tunnel C, and 0.12% for Tunnel D. The higher AOI hit rates observed in Tunnels A and B indicate that drivers more frequently directed their gaze toward the predefined sidewall-related regions in these scenes. Combined with field observations showing decorative luminaires and sidewall reflection in Tunnels A and B, this result suggests that the coupled effect of decorative lighting and wall reflection may be one of the factors that divert attention away from the forward driving path.\u003c/p\u003e\n \u003cp\u003ePupil diameter reflects both changes in light stimulation and the regulation of cognitive load. After preprocessing the pupil data in Tobii Studio, the mean pupil diameter and the variance of pupil fluctuation were extracted as objective indicators of visual load. These variables were first calculated at the participant level by scene and then summarized as scene-level means and variances to describe how different scenes affected pupil dynamics. Participant-level data were again used for inferential analyses.\u003c/p\u003e\n \u003cp\u003eAs shown in Table\u0026nbsp;3, Tunnels A and B, which were associated with poorer light environments, generally exhibited larger mean pupil diameters and stronger fluctuations than Tunnels C and D. These results indicate that the visually less comfortable tunnel scenes were associated with greater pupil dilation and stronger temporal instability in pupil response, whereas the better-performing scenes showed smaller and more stable pupil responses.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4 Associations between comfort rating and objective indicators\u003c/h2\u003e\n \u003cp\u003eTo identify the factors most strongly associated with visual comfort, the composite questionnaire score was linked with eye-tracking, gaze-behavior, and lighting-environment indicators. Specifically, the analysis included mean pupil diameter, pupil-diameter fluctuation, AOI hit rate, mean and variance of eye-level illuminance, mean and variance of horizontal illuminance, TI, and the De Boer Rating. The repeated-measures correlation (rmcorr) and linear mixed-effects model (LMM) results are summarized in Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003ermcorr and LMM results for the associations between experimental parameters and mean questionnaire scores\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003er value\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eHolm-adjusted p value (r)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eLMM \u0026beta;\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003eHolm-adjusted p value (\u0026beta;)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean pupil diameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.657\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-11.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance of pupil diameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.647\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-11.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eAOI hit rate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.659\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-11.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eTI value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.261\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e0.281\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-4.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e0.252\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eDe Boer rating\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e0.668\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e11.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean illuminance at eye position\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.764\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-13.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance of illuminance at eye position\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-14.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean horizontal illuminance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.777\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-13.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance of horizontal illuminance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-0.783\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-13.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, comfort rating was significantly associated with multiple objective indicators. Pupil-related variables and AOI hit rate were all negatively correlated with comfort, indicating that visually less comfortable tunnel scenes were accompanied by greater pupil response and more frequent gaze shifts toward non-primary regions. De Boer Rating was positively correlated with comfort, whereas TI was not significant in the present dataset. Because TI values in all four tunnels remained below the commonly used threshold of 15%, TI is better interpreted here as a lower-bound safety-related constraint than as a continuous explanatory variable for comfort differences.\u003c/p\u003e\n \u003cp\u003eAmong all indicators, illuminance-related variables showed the strongest associations with comfort. In particular, the variance of eye-level illuminance and the variance of horizontal illuminance had the largest negative correlations, suggesting that temporal instability of the light field may be more important for discomfort than mean illuminance level alone. The mean illuminance values were also negatively associated with comfort, but these relationships are more likely to reflect combined effects of glare, reflection, and spatial contrast under brighter conditions rather than a direct effect of illuminance itself.\u003c/p\u003e\n \u003cp\u003eThe fixed-effect directions in the LMM were consistent with the rmcorr results, and the significance conclusions for the main indicators remained unchanged after Holm correction, indicating that the observed associations were directionally stable under the present sample and model specification. Detailed LMM results are provided in Appendix A (Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003eA2\u003c/span\u003e).\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eBased on the field measurements, immersive replay, and coupled subjective-objective analysis, a preliminary pathway may explain why visual discomfort persists in standards-compliant tunnels. In the present dataset, comfort was not determined solely by average illuminance, but appeared to depend mainly on temporal instability and spatial saliency. Stronger illuminance fluctuation and more pronounced bright-dark alternation were associated with larger pupil responses, suggesting more frequent visual adaptation. At the same time, sidewall reflections and decorative light bands created salient bright regions that attracted lateral gaze. Together, these factors may increase attentional capture and discomfort even when disability glare thresholds are not exceeded.\u003c/p\u003e \u003cp\u003eUnder the present conditions, TI is better interpreted as a lower-bound indicator of glare-related safety risk than as a strong predictor of subjective comfort. By contrast, fluctuation-related optical indicators, reflection-related spatial characteristics, and attention-allocation measures were more informative for explaining discomfort. Tunnel-lighting evaluation may therefore be understood as a two-layer framework: one layer for safety and visibility, and another for visual comfort and attentional stability.\u003c/p\u003e \u003cp\u003eThis proposed mechanism should be regarded as an interpretive framework rather than a validated causal model. Several limitations should be noted. Real-tunnel measurements were influenced by traffic, reflective objects, and dynamic background lighting. Laboratory replay improved scene comparability, but display refresh rate and dynamic-range mapping may have altered some original stimulus characteristics. Temporal instability was represented mainly by statistical descriptors rather than frequency-domain features, and AOI analysis relied on static region definitions. In addition, the sample size was limited, so the findings are more suitable for mechanism exploration than for threshold definition or predictive modeling.\u003c/p\u003e \u003cp\u003eFuture work should combine controlled experiments, higher-fidelity simulations, larger samples, and broader tunnel conditions to test the causal validity of the proposed pathway. Particular attention should be given to luminaire arrangement and modulation characteristics, sidewall reflectance and geometry, and the saliency of decorative lighting. More robust attention-related indicators, together with mixed-effects or mediation-path analyses, may further clarify the relative contributions of different stimulus dimensions.\u003c/p\u003e"},{"header":"5 Conclusions","content":"\u003cp\u003eThis study investigated visual discomfort in urban road tunnels under standards-compliant conditions by combining field measurements with immersive driving-cabin replay. Physical characteristics of the light stimuli, eye-tracking responses, pupil dynamics, and subjective evaluations were synchronously analyzed within a unified time-aligned framework. The results showed that, across four representative tunnel samples, drivers\u0026rsquo; comfort ratings still differed substantially even though threshold-based glare indicators such as TI remained within commonly accepted limits. This indicates that compliance with conventional engineering criteria does not necessarily guarantee a comfortable visual experience.\u003c/p\u003e \u003cp\u003eCompared with TI alone, visual discomfort in the present dataset was more consistently associated with temporal instability of illuminance and luminance, as well as with attention shifts induced by sidewall reflection and decorative bright regions. In particular, greater illuminance fluctuation was linked to stronger pupil responses and lower comfort, while higher AOI hit rates suggested increased attentional capture by task-irrelevant bright regions. Taken together, the coupled subjective and objective evidence suggests that temporal light fluctuation, reflective interference from tunnel walls, and attention capture by decorative lighting may be important factors explaining why discomfort can still occur within nominally acceptable glare limits. Within the optical range covered in this study, TI is therefore better interpreted as a lower-bound safety constraint, whereas differences in comfort require explanatory variables related to spatiotemporal instability and visual saliency.\u003c/p\u003e \u003cp\u003eGiven the limited sample size and stimulus range, the present conclusions should be understood primarily as mechanism-oriented findings and hypothesis-generating evidence. Future studies with expanded tunnel samples, more complete luminance-field measurements, and controlled-variable experiments are needed to further verify the causal validity and interactions of the proposed mechanism.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCRediT author contributions\u003c/h2\u003e \u003cp\u003eZhenghao Jin: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Visualization, Writing \u0026ndash; original draft.\u003c/p\u003e \u003cp\u003eYandan Lin: Conceptualization, Supervision, Methodology, Writing \u0026ndash; review, Project administration.\u003c/p\u003e \u003cp\u003eHaiping Shen: Methodology, Validation, Writing \u0026ndash; review.\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eDeclaration of Competing Interest\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eZhenghao Jin: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Visualization, Writing \u0026ndash; original draft.Yandan Lin: Conceptualization, Supervision, Methodology, Writing \u0026ndash; review, Project administration.Haiping Shen: Methodology, Validation, Writing \u0026ndash; review.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data supporting the findings of this study are available from the corresponding author upon reasonable request, subject to ethical and privacy restrictions.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eChen, W., Chen, D. \u0026amp; Lin, Y. Discussion and practice of tunnel lighting. in Proceedings of the 11th Cross-Strait Symposium on Lighting Technology and Marketing, 279\u0026ndash;285Ningbo, China, (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt-Clausen, H. J. et al. Mesopic vision and ambient light inside automobiles. \u003cem\u003eChina Light Lighting\u003c/em\u003e. \u003cb\u003e8\u003c/b\u003e, 7\u0026ndash;10 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, Y. et al. Model predicting discomfort glare caused by LED road lights. \u003cem\u003eOpt. Express\u003c/em\u003e. \u003cb\u003e22\u003c/b\u003e, 18056\u0026ndash;18071 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJi, Z., Li, N. \u0026amp; Lin, Y. Decoding visual fatigue through pupil dynamics. In 2023 20th China International Forum on Solid State Lighting \u0026amp; 2023 9th International Forum on Wide Bandgap Semiconductors (SSLCHINA: IFWS), 424\u0026ndash;428 DOI: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/SSLChinaIFWS60785.2023.10399698\u003c/span\u003e\u003cspan address=\"10.1109/SSLChinaIFWS60785.2023.10399698\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao, X., Li, M., Hou, D., Lin, Y. \u0026amp; Xu, W. The Impact of Temporal Light Modulation on Psychological Responses Assessed by the Stroboscopic Visibility Measure. \u003cem\u003eLEUKOS\u003c/em\u003e (2024). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/15502724.2024.2365686\u003c/span\u003e\u003cspan address=\"10.1080/15502724.2024.2365686\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu, B., Hu, J., Zeng, J. \u0026amp; Wang, R. Traffic Safety Improvement via Optimizing Light Environment in Highway Tunnels. \u003cem\u003eInt. J. Environ. Res. Public. Health\u003c/em\u003e. \u003cb\u003e19\u003c/b\u003e, 8517 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYi, X., Zhao, R. \u0026amp; Lin, Y. The impact of nighttime car body lighting on pedestrians\u0026rsquo; distraction: A virtual reality simulation based on bottom-up attention mechanism. \u003cem\u003eSaf. Sci.\u003c/em\u003e \u003cb\u003e180\u003c/b\u003e, 106633 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe, S., Du, Z., Han, L., Wang, S. \u0026amp; Chen, Y. Impact of urban tunnel sidewall effect on drivers: Examining visual characteristics and driving behavior across different lanes. \u003cem\u003eTunn. Undergr. Space Technol.\u003c/em\u003e \u003cb\u003e143\u003c/b\u003e, 105476 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai, X., Tian, X. \u0026amp; Weng, J. Study on threshold luminance difference in highway tunnel lighting. \u003cem\u003eComput. Simul.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e, 154\u0026ndash;159 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, Z., Wang, J. \u0026amp; Guan, Y. Lighting spacing of LED lamps in curved tunnels. \u003cem\u003eJ. Southwest. Jiaotong Univ.\u003c/em\u003e \u003cb\u003e57\u003c/b\u003e, 434\u0026ndash;439 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, M. \u0026amp; Liu, M. Study on tunnel lighting flicker frequency based on drivers\u0026rsquo; heart rate variation. \u003cem\u003eChina Illum Eng. J.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 80\u0026ndash;86 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe, X. \u0026amp; Ning, J. Impact of LED display brightness on urban traffic safety: a case study of Chongqing. \u003cem\u003eChina Illum Eng. J.\u003c/em\u003e \u003cb\u003e36\u003c/b\u003e, 16\u0026ndash;22 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinistry of Housing and Urban-Rural Development of the People's Republic of China. Standard for lighting design of urban roads (CJJ 45-2015). (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinistry of Transport of the People's Republic of China. Guidelines for lighting design of highway tunnels (JTG/T D70/2-01- (2014). (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCommission Internationale de l\u0026rsquo;\u0026Eacute;clairage. Guide for the lighting of road tunnels and underpasses, 2nd ed. (CIE 88:2004). (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinistry of Housing and Urban-. \u003cem\u003eRural Development of the People's Republic of China. Standard for lighting design of buildings\u003c/em\u003e (GB/T 50034\u0026thinsp;\u0026ndash;\u0026thinsp;2024, 2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCIE TC 4\u0026ndash;15. Road lighting calculations, 2nd ed. (CIE 140:2019, 2019). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.25039/TR.140.2019\u003c/span\u003e\u003cspan address=\"10.25039/TR.140.2019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Boer, J. B. \u0026amp; Schreuder, D. A. Glare as a Criterion for Quality in Street Lighting. \u003cem\u003eTrans. Illum Eng. Soc.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, 117\u0026ndash;135 (1967).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBullough, J. D., Brons, J. A., Qi, R. \u0026amp; Rea, M. S. Predicting discomfort glare from outdoor lighting installations. \u003cem\u003eLight Res. Technol.\u003c/em\u003e \u003cb\u003e40\u003c/b\u003e, 225\u0026ndash;242 (2008).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Appendix","content":"\u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable A1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMean and variance of illuminance in the four test tunnels: (a) eye-level illuminance; (b) horizontal illuminance. (a)\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eEye-level illuminance\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eBeiheng Passage\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eBund Tunnel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eFuxing East Road Tunnel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003eXinjian Road Tunnel\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e2.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e6.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e2.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e2.27\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e1.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e4.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003ctfoot\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\"\u003e(b)\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tfoot\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003e\u003cbr\u003e\u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eHorizontal illuminance\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eBeiheng Passage\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eBund Tunnel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eFuxing East Road Tunnel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003eXinjian Road Tunnel\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e70.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e170.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e65.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e29.99\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e874\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e4557.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e218.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e129.56\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable A.2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eRobustness check results of the LMM\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eIndicator\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eLMM \u0026beta;\u003c/p\u003e\n \u003cp\u003e(score per +\u0026thinsp;1 SD)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e95% CI\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003et\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003ep\u003csub\u003eHolm\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance of illuminance at eye position\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-14.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-19.18, -9.82]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-6.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e1.23E-05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e1.12E-04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance of horizontal illuminance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-13.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-18.93, -8.77]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-5.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e5.39E-05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e4.31E-04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean horizontal illuminance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-13.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-18.89, -8.61]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-5.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e6.60E-05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e4.63E-04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean illuminance at eye position\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-13.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-18.79, -8.25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-5.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e1.04E-04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e6.21E-04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eDe Boer rating\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e11.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[5.74, 17.90]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e3.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e0.00139\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e0.00696\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eAOI hit rate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-11.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-17.80, -5.52]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-3.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e0.00170\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e0.00696\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMean pupil diameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-11.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-17.78, -5.47]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-3.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e0.00178\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e0.00696\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eVariance of pupil diameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-11.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-17.67, -5.21]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-3.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e0.00222\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e0.00696\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eTI value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\n \u003cp\u003e-4.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e[-12.50, 3.28]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e-1.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003e0.268\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c6\"\u003e\n \u003cp\u003e0.268\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cp align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003cstrong\u003eTable A.3.\u003c/strong\u003e AOI hit rates and pupil diameter across the four tunnel scenes\u003c/p\u003e\n \u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eTunnel ID\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eMean AOI hit rate\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eMean pupil diameter (pixels)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eVariance of pupil diameter change\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\"±\" colname=\"c2\"\u003e\n \u003cp\u003e0.71% \u0026plusmn; 0.11%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e115.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e28.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\"±\" colname=\"c2\"\u003e\n \u003cp\u003e0.79% \u0026plusmn; 0.21%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e111.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e27.75\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\"±\" colname=\"c2\"\u003e\n \u003cp\u003e0.56% \u0026plusmn; 0.07%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e100.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e23.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\"±\" colname=\"c2\"\u003e\n \u003cp\u003e0.12% \u0026plusmn; 0.1%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\n \u003cp\u003e99.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\n \u003cp\u003e16.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Urban road tunnel, tunnel lighting, visual discomfort, eye tracking, illuminance fluctuation","lastPublishedDoi":"10.21203/rs.3.rs-9290694/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9290694/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eVisual discomfort is still reported in urban road tunnels even when conventional lighting criteria are satisfied. This exploratory study investigated factors associated with such discomfort from a human-factors perspective. Four representative urban tunnels were examined using field measurements and immersive laboratory replay. Field data included eye-level illuminance, roadway and sidewall illuminance, synchronized video, and glare-related metrics including threshold increment (TI). Eye-tracking indicators, including pupil diameter and area-of-interest hit rate, were used to characterize visual response and attentional allocation. In the laboratory, typical tunnel scenes were replayed in an immersive driving cabin to obtain ratings of visual comfort, discomfort glare, and visual fatigue. Although all four tunnels showed TI values below the commonly used 15% threshold, subjective comfort differed markedly across scenes. Compared with average illuminance or TI alone, discomfort was more sensitive to spatiotemporal lighting instability. Greater illuminance fluctuation and reflected light from highly reflective sidewalls and decorative lighting were associated with stronger pupil responses, higher AOI hit rates, and lower comfort. These findings suggest that temporal instability, reflective interference, and attentional capture may help explain visual discomfort in standards-compliant tunnels.\u003c/p\u003e","manuscriptTitle":"Factors Contributing to Visual Discomfort in Standards-Compliant Urban Road Tunnels: Evidence from Field Measurements and Immersive Driving Experiments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-11 13:58:08","doi":"10.21203/rs.3.rs-9290694/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"106405993459549963121522995818537401183","date":"2026-05-11T08:52:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"155732997858147796699791597845012684165","date":"2026-05-01T15:44:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-01T15:41:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-01T15:38:33+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-10T10:20:38+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-07T09:37:54+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-04-07T08:01:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b2d5a0c9-8783-4e5f-b3ef-108c1bfd2868","owner":[],"postedDate":"May 11th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"106405993459549963121522995818537401183","date":"2026-05-11T08:52:30+00:00","index":71,"fulltext":""},{"type":"reviewerAgreed","content":"155732997858147796699791597845012684165","date":"2026-05-01T15:44:31+00:00","index":66,"fulltext":""},{"type":"reviewersInvited","content":"6","date":"2026-05-01T15:41:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-01T15:38:33+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67863519,"name":"Physical sciences/Engineering"},{"id":67863520,"name":"Biological sciences/Neuroscience"},{"id":67863521,"name":"Biological sciences/Psychology"},{"id":67863522,"name":"Social science/Psychology"}],"tags":[],"updatedAt":"2026-05-11T13:58:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-11 13:58:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9290694","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9290694","identity":"rs-9290694","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}