Integrating identity and emotion information in faces in ASD – capacity limits of information processing

Integrating identity and emotion information in faces in ASD – capacity limits of information processing | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Integrating identity and emotion information in faces in ASD – capacity limits of information processing Marianna Constantinou, Dena Al-Thani, Marwa Qaraqe, Ala Yankouskaya This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7612091/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The ability to combine identity and emotional information when viewing faces allows typically developing individuals to process facial information efficiently. The purpose of this study was to examine whether adults with autism spectrum disorder (ASD) share this ability, and test if it reflects a broader pattern of integrative information processing. Nineteen adults with ASD and nineteen typically developing adults completed a divided attention task where they had to indicate when a target was present. Each participant completed two separate experiments: i) the face experiment, which required detecting targets based on identity and emotional expression; and ii) the object experiment, which required detecting targets based on colour and shape. Analyses assessed whether, and how, participants responded more efficiently in the presence of both targets, compared to either single target. By employing mathematical modelling tools developed in the System Factorial Technology (SFT) framework, results showed that ASD individuals process faces at the rate of control individuals, however, failed to exhibit integrative properties. Contrary to controls, adults with ASD showed limited capacity to engage in integrative processing, even for object attributes, suggesting a domain-general integrative processing difficulty. Implications of these findings generalise to personalisation practices in education and can be used to inform the development of assistive technologies in ASD. Psychology ASD capacity processing identity emotional expression integration Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Autism spectrum disorder (ASD) is a neurodevelopmental condition associated with a broad range of expressions, some of which include persistent deficits in social communication and social interaction (American Psychiatric Association, 2013). Social exchanges depend heavily on the processing of facial cues, making face perception a key process in contributing to the social challenges in ASD. Even though evidence reveals that individuals with ASD show impairments in face perception (Nomi & Uddin, 2015; Schultz, 2005; Stantic et al., 2021) the specific nature of these impairments remains unclear. The ambiguity is related to the dynamic information expressed in faces, which simultaneously carry both information about one’s identity and one’s emotional expressions. Therefore, distinguishing between these aspects, and whether they independently or cooperatively affect individuals with ASD, is necessary for identifying their specific perceptual challenges. Clarifying whether difficulties stem from impaired processing of facial identity, emotional expression, or the integration of both can inform how such deficits contribute to broader challenges in social development and inform appropriate and targeted interventions. Existing evidence shows that ASD is linked to a reduced sensitivity in perceiving either facial identity (Blair, 2002; Jemel et al., 2006), or processing emotional expressions (Høyland et al., 2017; Uljarevic & Hamilton, 2013). However, it could also be the case that both dimensions are affected in ASD (Greimel et al., 2014; Ventura et al., 2023; Weigelt et al., 2012). Ventura et al. (2023) used surgical face masks to disrupt recognition of identity and emotion in typically developing adults and individuals with ASD. Regarding identity, only typically developing participants showed a context congruence effect – better recognition for faces when learned and tested with masks. Individuals with ASD had poorer recognition regardless of the whether the testing condition matched the learning condition. Regarding emotion, while masks impaired recognition in both groups, the specific emotions affected varied between groups (Ventura et al., 2023). Such findings indicate that both identity and emotion, when tested separately, are affected in ASD in the context of face processing. What remains to be examined is whether, when identity and emotion interact, typically developing individuals and individuals with ASD differ – and if so, how these differences are expressed. The theoretical framework justifying the above experimental work proposes the existence of two pathways in face processing: one responsible for identity-related processing and another for emotion-related processing (Bruce & Young, 1986). These pathways are structurally and functionally separate, with identity processing being associated with activity in the fusiform gyrus and inferior occipital gyri, while emotion processing being linked to activity in the limbic system (Haxby et al., 2000). This distinction is further supported by neuropsychological evidence with impairments in the former being linked to prosopagnosia (i.e., selective inability to recognise faces; Farah et al., 1995; Schiltz et al., 2006), while impairments in the latter to alexithymia (i.e., deficits in recognising one’s emotions; Taylor & Bagby, 2000; Van Der Velde et al., 2014). Notably, developmental prosopagnosia does not necessarily co-occur with alexithymia (Towler et al., 2025), a separation also reported in individuals with autism (Gehdu et al., 2024). Even though early models treated identity and emotion as separate systems, more recent behavioural, electrophysiological, and neural evidence recognises their interconnectivity, challenging the idea of strict independence, at least in neurotypical populations (Bach et al., 2014; D’Argembeau & Van Der Linden, 2011; Dobs et al., 2018; Alguacil et al., 2017; Fisher et al., 2016; Li et al., 2022; Yankouskaya et al., 2012). Despite this, experimentally capturing the interaction is challenging. Specifically, isolating identity-related information from emotion-related information is difficult, as emotional expressions involve both identity-independent and identity-dependent features (Fox & Barton, 2007). Moreover, identity recognition can also be influenced by emotional expressions as affective states can impact how identity is perceived (Chen et al., 2015; Redfern & Benton, 2017). Therefore, attempts to control one factor might interfere the processing of the other. Rather than supporting a modular system, the observed interconnectivity between identity and emotion processing suggests a functional interdependence. In typically developing individuals, these two factors are processed interactively. Critically, the integration of facial identity and facial expressions happens only when facial expressions are emotionally valanced (Vrancken et al., 2019; Yankouskaya et al., 2012), a mechanism which is supported by functional connections between the orbitofrontal cortex and the inferior occipital gyri (Yankouskaya et al., 2017). This suggests that executive mechanisms may allow for the integration between identity and emotion, enhancing perceptual processing. This would explain why performance is faster in response to faces which contain both identity and emotion target information, rather than each factor alone (Yankouskaya, et al., 2014). Such interactions are independent of cognitive ageing as they are found across adulthood (Yankouskaya, et al., 2014) and even older adults can sometimes perform better than younger adults, which could be explained by a related level of experience and familiarity with facial stimuli (Ganel & Goshen-Gottstein, 2004; Yankouskaya, et al., 2014). Therefore, the interactive nature of these factors is consistently embedded in typically developing individuals, possibly enhancing perceptual efficiency. Interestingly, the effect also generalises to object processing, suggesting that redundant target effects happen with responses being faster and more accurate when two target signals are presented simultaneously, rather than each target alone (e.g., shape and colour; see Miller, 1986). In terms of social context, integration has been found to also facilitate economic decision-making (Alguacil e al., 2017). However, to assess the beneficial nature of this interaction, it is first necessary to examine populations that may not engage this mechanism in the same way. This comparison can reveal whether the observed effects are related to the interaction itself, or to other differences in face processing strategies, such as processing features serially (one after the other, in sequence) or in parallel (independently but at the same time), rather than through an integrated process. This study aims to re-assess face processing abilities in adults with ASD. Specifically, the objectives of the current work were to address whether: i) individuals with ASD exhibit the ability to integrate information of facial identity and facial emotional expressions; and ii) whether this ability is exclusively linked to face processing, or rather reflects a general ability to process information in an integrative manner. To achieve these objectives, a divided attention task was used where participants were required to monitor the presence of two targets and respond quickly and accurately when they identified any target information (‘target present’). Alternatively, they were instructed to indicate the absence of a target (‘target absent’). Two experiments were utilised. In the ‘Face experiment’ face stimuli were used containing one stimulus displaying the target identity, one stimulus displaying the target emotional expression, one stimulus displaying both targets, and three stimuli containing no target information. In the ‘Object experiment’ object stimuli were utilised, manipulating the colour and shape to include one stimulus displaying the target colour, one stimulus displaying the target shape, one stimulus displaying both targets, and three stimuli with no target attributes. The modelling and comparison of response time distributions for single versus double targets, for each experiment, allowed for the estimation of capacity processing when the number of targets increased. Capacity processing, as developed in a human information processing framework – the System Factorial Technology (Townsend & Eidels, 2011; Townsend & Wenger, 2004; Wenger & Townsend, 2000), refers to how efficiently individuals handle increased information load. Here, reduced response times for double-target trials indicate more efficient information integration. By examining these effects in ASD, the current work investigates whether integrative processing is preserved in ASD, and whether it is unique to face processing. Methods Participants Nineteen adults with high functioning ASD were recruited via learning support committees at Bournemouth University, Aspens Charity Brighton, Liverpool Hope University, University of Liverpool, Liverpool John Moores University. Nineteen typically developing adults were recruited via email distribution lists across Bournemouth University, three universities in Liverpool, and the general public. Participants between groups were matched in age, sex, and level of education (see Supplementary Information, Table S1). All participants were over 18 years of age (age range 19-52 years), right-handed and were not on any medication at the time of the study. All participants in the ASD group had confirmed diagnosis of autism spectrum disorder. Exclusion criteria across groups included self-report of major neurological disorders, presence of systemic diseases, lack of normal or corrected-to-normal vision, and presence of colour blindness. All participants completed the Wechsler Abbreviated Scale of Intelligence (WAIS-III; Wechsler, 1999) and the Autism Spectrum Quotient (AQ; Baron-Cohen et al., 2001) to estimate intellectual abilities and autism symptom severity respectively. No group differences were found in intellectual abilities (see Supplementary Information, Fig. S1, A). Regarding the AQ scores, the control group scored lower (mean = 12.0, SD = 2.95, min = 7, max = 15.7), compared to the ASD group (mean = 33.0, SD = 5.0, min = 23, max = 42; see Supplementary Information, Fig. S1, B), with the scores of the latter being in line with previously reported literature (mean AQ score = 35.8, SD = 6.5; see Baron-Cohen et al., 2001). Ethical approval has been cleared by the Faculty of Research Ethics Committee at Liverpool Hope University, and Bournemouth University and the Access Review Group of Autism Together. Experiments and Stimuli Two experiments were used – the face experiment and the object experiment. In each experiment participants performed a divided attention task which required the simultaneous monitoring of two targets. Participants had to respond with either ‘target present’ when any target appeared on the screen, or ‘target absent’ when a trial displayed no target information. Participants were instructed to respond as fast and as accurately as possible. The experiments differed only in stimuli and stimulus onset parameters. In the face experiment, the targets were identity and emotional expression. Three sets of stimuli were randomly assigned to participants in each group. Each set included six images of faces: a face containing target identity only (I), a face containing target emotion only (E), a face containing both target identity and target emotion (IE), and three faces containing no target information (NT1-NT3). All face stimuli were sourced from The NimStim Face Stimuli Set (Tottenham et al., 2009). Due to restricted permission to publish some face images, identification of actors and posed emotional expressions can be found in Supplementary Information (Table S2). The procedure of stimuli selection is described in detail in Yankouskaya et al. (2012). To eliminate possible target judgments based on hairstyle, stimuli were cropped around the hairline. Background colour was adjusted to black. For an example of a stimulus set see Figure 1(A). In the object experiment, the targets were geometrical shape and colour. Three sets of stimuli were randomly assigned to participants. Each set consisted of six images of objects: an object containing target shape only (S), an object containing target colour only (C), an object containing target shape and target colour (SC), and three objects containing no target information (NT1-NT3). A list of the used stimuli sets is presented in Supplementary Information (Table S2). For an example of a stimulus set see Figure 1(B). Face stimuli were 8 x 11 cm and object stimuli were 6 x 6 cm when displayed on a 15.5-inch Dell monitor. All stimuli were displayed at a viewing distance of 0.8 m and the angular width subtended by the stimulus was approximately 10°. The presentation of the stimuli and data acquisition were controlled using Cogent 2000 and Cogent Graphics developed by the Cogent 2000 team at the FIL and ICN (http://www.vislab.ucl.ac.uk/cogent_2000.php). For each experiment, stimuli were presented in a random order. A single trial began with a fixation cross at the centre of the screen (100 ms). Immediately after, the fixation cross was replaced by one of six images which remained for 300 ms in the face experiment, and 200 ms in the object experiment. A blank screen followed by a stimulus offset until a response was made. A jittered (100-600 ms) inter-trial blank display was presented after each trial. If the response time exceeded 2000 ms, the feedback message ‘Missed response’ appeared, and the trail was repeated at the end of the experimental block. Each experiment consisted of 420 trials (70 trials per stimulus). Half of these trials required a ‘target present’ response. Participants were asked to have a short break (2 minutes) after completing 1/3 and 2/3 of each experiment. Procedure Each participant took part in two separate testing sessions. In session one, participants completed the Wechsler Abbreviated Scale of Intelligence and a short demographic questionnaire. Session two started with either the face experiment or object experiment (the order was randomly assigned to participants in the ASD group; the same order was kept for matching participants in the Control group). Participants began the experimental session with a written instruction for the first experiment. The instruction was repeated verbally by the investigator to ensure that the written instruction was understood correctly. A practice block of 36 trials followed (6 trials per stimulus) and feedback on accuracy and response speed was provided after each trial during the practice. If a participant made more that 50% of incorrect responses in practice blocks, the instruction was repeated, and a new data set was launched. Three participants (C1, C8, C16 Supplementary Information, Table S1) in the Control group and two participants in the ASD group (A8, A18, Supplementary Information, Table S1) required to repeat practice trials in the face experiment. All participants performed practice trials for the object experiment with an accuracy above 79%. Participants completed the face, and object experiments in a random order, with the Autism Spectrum Quotient being administered between the two experiments. Data Analysis The core analysis of the current study involved the assessment of capacity processing across the face experiment and object experiment in the ASD and control groups. Before that, to inform the reader of group-level accuracy and response time performance, and to demonstrate how the traditional approach supports capacity analysis, first non-parametric modelling of the group data is reported. Non-parametric modelling of the group data First, to gain an idea about overall accuracy performance, the proportion of correct responses for each group and each experiment was calculated. A nested ranks test (implemented via the nestedRanksTest package; Scofield, 2015) was used to perform a non-parametric comparison between groups. This method extends the Mann-Whitney-Wilcoxon test to a mixed-model framework, enabling two analyses: i) treating group as a fixed effect, and Subject as a random effect, and ii) treating task as a fixed effect, and Subject as a random effect. Test statistics are expressed as z-scores based on rank comparisons between groups or tasks. Significance was determined by comparing the observed z-score to a null distribution generated through 10,000 bootstrap iterations. (Scofield, 2015). Second, a non-parametric mixed analysis of variance (ANOVA) was carried out to examine the effects of group on response times for stimuli of the face and object experiments, using the aligned ranks transformation (ART) approach implemented in the ARTool R package (Kay & Wobbrock, 2016). This approach involves computing residuals via a regression model, adding the effect of interest, and ranking the resulting sum prior to performing a parametric factorial ANOVA (Mansouri et al., 2004). The model included two fixed effects – Group (ASD, Control) and Stimuli (images containing target information) – the interaction between Group and Stimuli, and a random effect for Subject. Prior to model testing, the appropriateness of the ART procedure was confirmed by verifying that an ANOVA on the ranked data yielded values close to F = 0.00 and p = 1.00 for all effects except the one corresponding to the effect of interest (Wobbrock et al., 2011). It should be noted that, unlike linear model tests which provide expected mean differences between factor level combinations, the ART approach interprets interactions as “differences of differences” (Levin & Marascuilo, 1971). In the present study this means assessing whether the difference between face stimuli in the ASD group significantly differs from that in the control group, and whether the difference between object stimuli in the ASD group significantly differs from that in the control group. Interaction effects were evaluated using the testInteractions function form the phia-package (https://cran.r-project.org/web/packages/phia/phia.pdf), applying the Hommel-adjusted step-up modification of the Bonferroni method to correct for multiple comparisons (Blakesley et al., 2009). Main effects were tested using Friedman’s test with Post Hoc analysis (Wilcoxon signed rank test) and Tukey HSD correction for multiple comparisons. Given the relatively small sample size, exact p-values were computed for the Wilcoxon signed rank based on the true distribution, providing conservative estimates that help control Type I error at the nominal significance level. For each comparison, the Hodges-Lehmann estimator was used to quantify the shift in location (i.e., estimated difference between medians of two distributions), along with 95% confidence intervals using a permutation approach. An effect size for the Wilcoxon signed rank test was calculated by dividing the absolute (positive) standardized test statistic Z by the square root of the number of pairs. To quantify evidence in favour of the null hypothesis (H0), Bayes factors (BFs) were calculated for all contrasts using the non-parametric approach proposed by (Yuan & Johnson, 2008), as implemented in JASP software (JASP Team, 2025, Version 0.95). Unlike standard Bayesian methods that rely on the sampling distribution of original data, this approach uses asymptotic approximations to the distributions of non-parametric test statistics. Computing capacity processing To test changes in efficiency of processing for dual target displays compared to single target displays, capacity coefficients were calculated for each participant in each experiment. The coefficient represents a ratio between RT performance for a stimulus containing two targets and a joined distribution for two single-target stimuli at a time interval (t) of 10 msec. The RT performance for each condition containing target information (I, E, IE in the face experiment, and S, C, SC in the object experiment) was estimated using the cumulative hazard function. The hazard function, representing a conditional probability that a response will occur in the next interval (i.e., t + D t ), captures the instantaneous and time-invariant likelihood of terminating an ongoing process at each unit of time. Computing the total amount of accumulated evidence for a response represents, therefore, the amount of work completed by time t and can be considered as a measure of efficiency (Townsend & Eidels, 2011). In this sense, the ratio between the cumulative hazard function for a stimulus containing two targets and the sum of cumulative hazard functions for single targets indicates whether, and how, adding a target affects the overall efficiency (capacity) of the system at each time interval: H 12 ( t ) C(t) = ------------------------ , H 1 ( t ) + H 2 ( t ) where C(t) - capacity coefficient at time t ; H 12 ( t ) – the cumulative hazard function for a stimulus containing two targets (denoted as 1 – target 1, 2 – target 2); H 1 ( t ) - the cumulative hazard function for a stimulus containing target 1 H 2 ( t ) - the cumulative hazard function for a stimulus containing target 2 A benchmark in an interpretation of capacity coefficient is the prediction derived from a standard parallel model with unlimited capacity (UCIP). In an experiment where participants are required to respond to either target, the standard parallel model predicts C(t) = 1 (i.e., the amount of work done on two units of information is equal to the sum of work done on every single unit). In the case when a parallel model has two items to process and the processing of both of them speeds up compared to when only one is processed, the system is said to exhibit super capacity (C(t) >1). If increasing a workload slows down processing time, the system will show limited capacity (C(t) < 1), indicating that processing of two-targets is less efficient compared to the baseline UCIP model (Neufeld et al., 2007). A special case of interest occurs if the system operates at fixed capacity (i.e., divides up a fixed resource across two targets); or serially processes two units of information (for example, working first on the identity target, and then on target emotion or vice versa). Although in both cases capacity will be hovering around C(t) = ½, the former assumes equal distribution parameters with H 12 ( t ) = H 1 ( t ) = H 2 ( t ). Finally, to estimate the population-level distributions of mean capacity ability, bootstrapping was conducted. Bootstrapping is a statistical method used for estimating the sampling distribution of a statistic, here the mean capacity ability, by resampling a dataset with replacement to create multiple simulated samples (Efron & Tibshirani, 1994). 2000 bootstrap samples were simulated by randomly sampling data points from the original dataset with replacement. For each of these resampled datasets the mean capacity for faces and objects was calculated per group. Results Accuracy performance Participants in the ASD and Control groups were more accurate in the object experiment (Mdn = 92.0, S.D. = 6.1 in ASD group; Mdn = 92.0, S.D. = 2.7 in Control group) compared to face experiment (Mdn = 84.5, S.D. = 7.4 in ASD group; Mdn = 86.0, S.D. = 7.9 in Control group; see details in Supplementary Information, Fig. S2). Accuracy performance was too high to allow for meaningful comparisons across the stimuli in each experiment. Therefore, analyses were conducted on i) differences between experiments in each group, ii) group differences in accuracy performance for each experiment, and iii) differences between stimuli for each experiment containing target information and stimuli for each experiment with no targets, in each group. The results of nested ranks tests indicate significantly higher accuracy in the object experiment compared to the face experiment in each group (Z = 0.499, n = 21, p = .0001; Z = 0.42, n = 21, p = .0001 respectively for ASD and Control). No significant differences were found between groups in accuracy for the face experiment (Z = 0.091, n = 21, p = .12) and for the object experiment (Z = 0.061, n = 21, p = .21). There were also no significant differences in accuracy performance between targets and non-targets in ASD (Z=-0.009, n = 21, p = .5) and Control (Z=-0.06, n = 21, p = .8) in both experiments (Z = 0.003, n = 21, p = .68; Z = 0.02, n = 21, p = .52 in the face and object experiment respectively). Response time performance Response time performance is displayed in Table 1 . Table 1 Median response time (*[95% CI]) for correct responses to **stimuli in face and object experiments in the ASD and Control groups. Face experiment groups I E IE NT1 NT2 NT3 ASD 725.7 [607, 786] 784.8 [641, 838] 670.0 [625, 801] 769.6 [678, 878] 742.2 [687, 811] 781.0 [679, 862] Control 696.3 [593, 866] 696.6 [580, 838] 609.9 [461,838] 647.3 [504, 747] 709.5 [623, 796] 655.9 [572, 801] Object experiment S C SC NTo1 NTo2 NTo3 ASD 721.1 [579, 888] 707.7 [600, 793] 619.7 [538,685] 634.3 [600, 788] 663.1 [605, 796] 655.9 [598, 768] Control 601.2 [565, 642] 576.4 [538, 646] 506.4 [471, 572] 611.5 [546, 677] 632.4 [552, 667] 621.5 [549, 655] * Confidence Interval (95% CI) was calculated using the bootstrap percentile method with 10,000 iterations. ** Stimuli in face experiment: I - target identity alone, E - target identity alone, IE – both target identity and target emotion, NT1-NT3 – three faces containing no target information. Stimuli in object experiment: S – target shape alone, C – target colour alone, SC – both target shape and target colour, NTo1-NTo3 – displays containing no target information. Nested ranks tests showed no significant differences in RT performance between face stimuli containing target (I, E, IE) and non-target (NT1, NT2, NT3) information in ASD (Z=-0.133, n = 21, p = .88) and Control (Z=-0.054, n = 21, p = .68) groups. Similar results were obtained in the object experiment where ASD and Control groups responded to stimuli containing targets (S, C, SC) as fast as to stimuli containing no targets (NTo1-NTo3) (Z=-0.043, n = 21, p = .65; Z=-0.191, n = 21, p = .95 for ASD and Control groups respectively). The main interest of the current study is responses to stimuli containing target information. Therefore, all consequent analyses will focus on stimuli required responses ‘target present’. To examine the effects of group on response times, analysis of variance of aligned rank transformed data was performed separately for the face and object experiments. The sums of aligned responses for main effects and F values of ANOVAs on aligned responses of no interest showed values closer to zero in both experiments indicating that data were suitable for the aligned ranks transformation. A linear mixed-effects model with Group (ASD, Control), Stimuli (I, E, IE in the face experiment and S, C, SC in the object experiment) and the interaction between Group and Stimuli as fixed effects and subjects as a random effect was fitted to aligned ranks data. The results of both analyses are summarised in Table 2 . As it can be seen in Table 2 , the results of the ANOVA in the face experiment showed a significant interaction between Group and Stimuli and a significant main effect of Stimuli, but a main effect of group did not reach significance. Post Hoc analysis indicated that this interaction stems from significant differences between the ASD and Control groups across all stimuli (see Fig. 2 ). Table 2 Analysis of variance of aligned ranks transformed RT data and Post Hoc comparisons in the face and object experiments Face experiment Object experiment Main effects *F df p *F df p Group 0.30 1 .58 4.71 1 .04 Stimuli 26.99 2 < .001 37.2 2 < .001 Interaction Group*Stimuli 18.31 2 < .001 0.38 2 .68 Post Hoc for Group*Stimuli χ 2 df p ASD-Control:I-E 9.07 1 .005 ASD-Control:I-IE 36.62 1 < .001 ASD-Control:E-IE 8.76 1 .005 *Type III Wald F-tests using the Kenward-Roger coefficient covariance matrix and Satterthwaite degrees of freedom ( df ) In the object experiment, the results of the ANOVA did not reveal a significant interaction between Group and Stimuli. Next, main effects in each experiment were explored using Friedman’s test as a non-parametric version of a one-way repeated measures ANOVA with Post Hoc analysis (Wilcoxon signed rank test and Tukey HSD correction for multiple comparisons). In the face experiment, the ASD group ( χ 2 (2) = 9.36, p = .009) was faster in responding to target identity alone compared to target emotional expression alone (p = .009). The differences between faces containing both targets and either single target were non-significant (p = .74; BF 01 = 3.19 for contrast between I and IE; p = .08, BF 01 = 1.36 for contrast between E and IE). In contrast, Control group ( χ 2 (2) = 28.74, p < .001) responded significantly faster to faces containing both targets compared to either single target (p < .001; BF 01 < 0.001 for contrast between I and IE; p < .001, BF 01 < 0.001 for contrast between E and IE), but the difference between the two single targets was non-significant (p = .87, BF 01 = 1.9) (Fig. 3 ). In the object experiment, ASD ( χ 2 (2) = 9.79, p = .007) and Control ( χ 2 (2) = 24.0, p < .001) groups responded faster to stimuli containing dual-targets (Fig. 3 ) compared to either single target display (contrast [SC-S] (ASD group: p = .001, BF 01 = 0.05; Control group: p < .001, BF 01 < 0.001), contrast [SC-C] (ASD group: p < .001, BF 01 = 0.044; Control group: p < .001, BF 01 = 0.008). Capacity analysis Individual capacity coefficients were computed in the face and object experiments and plotted as a function of response time divided into 10msec time bins (Fig. 4 – 5 ). Figure 4 shows that face processing capacity, in fifteen out of nineteen control individuals (see Supplementary Information Table S3), ranges from a value (C(t) = 1.5 to 4.5) at the fastest response times to somewhere just close to C(t) = 1 for the slowest RTs. In contrast, all ASD participants showed limited capacity with curves places well below C(t) = 1.0. Figure 5 shows that in the object experiment, the average capacity curve in ASD group was around C(t) = 1 which is in line with the prediction of parallel processing model. For the control group, faster responses were accompanied by C(t) values above 1. To inform the reader about stability of the estimated capacity function, we demonstrated examples of individual capacity plots for ASD and Control participants (Supplementary Information, Fig. S3-S6). Next, the overall capacity coefficients (averaged across all time bins) were calculated for each participant. Although the overall capacity value does not arbitrate between processing architectures for the ASD and Control groups, it provides our reader with an idea of whether increasing workload affects overall efficiency of the system (individual overall capacity coefficients are presented in Supplementary Information, Table S3). To test the effects of groups and experiments on overall capacity coefficients, a 2 (Group: ASD, Control) x 2 (Experiment: face, object) analysis of variance was conducted of aligned rank transformed overall capacity coefficients. The sums of aligned responses for main effects and F values of ANOVAs on aligned responses not of interest showed values close to zero in both experiments indicating that the data were suitable for the aligned ranks transformation. A linear mixed-effects model with Group, Experiment and the interaction between Group and Experiment as fixed effects and Subjects as a random effect was fitted to aligned ranks data. There was no significant interaction between Group and Experiment (F(1,54) = 0.76, p = .38) but two main effects were significant. A main effect of Group (F(1,54) = 75.9, p < .001) showed significantly lower overall capacity in ASD group compared to Control (W = 10.0, p < .001, BF 01 = 0.002). A main effect of Experiment ((F(1,54) = 4.37, p = .04) indicates that overall capacity for the object experiment was higher compared to faces (V = 35, p = .014), however, Bayes Factor provides moderate evidence for the hull hypothesis (BF 01 = 0.158). Further analyses to explore the main effects showed that in the Control group, the difference between overall processing capacity for faces (Mdn = 1.298) and object attributes (Mdn = 1.43) was non-significant (V = 74.5, p = .42, r = 0.21, BF 01 = 2.48). In contrast, the ASD group showed significantly lower overall processing capacity for faces (Mdn = 0.579) compared to object attributes (Mdn = 0.836) (V = 35, p = .014, BF 01 = 0.166). Compared to controls, individuals with ASD exhibit significantly lower processing capacity for both object attributes (W = 63.0; p < .001; BF 01 = 0.016) and face processing (W = 3.0; p < .001 BF 01 = 0.00078) (Fig. 6 ). Bootstrapped population-level distributions of the mean capacity are plotted for each group in Fig. 7 , by pairing face and object capacities. Results revealed a clear separation between the two groups. Specifically, the ASD capacity values were positioned predominantly below 1 for both faces and objects, whereas the mean general capacity of the control group was above 1. In addition, the ASD group distribution was more tightly clustered, compared to the control group, indicating less heterogeneity. The complete separation between ASD and CTRL clouds highlights their distinction in general capacity ability. The bootstrapped 95% CI were: ASD-faces [0.54–0.55], ASD-objects [0.89–0.90], CTRL-faces [1.41–1.41], ASD-objects [1.42–1.54]. Discussion The present study aimed to provide insights into two questions: i) whether individuals with ASD exhibit the ability to integrate information of facial identity and facial emotional expressions, and ii) whether this ability is linked to face processing only, or instead reflects a general ability to process information in an integrative manner. Using a divided attention task to assess facilitation effects when two types of information are presented, and the System Factorial Technology to estimate processing capacity of the system (Townsend & Eidels, 2011 ), results demonstrate differences between the ASD group and control group in the ability to integrate identity and emotional information in faces. Before discussing the findings, it is important to note that across groups, participants demonstrated high accuracy in detecting target information in both the face and object experiments. These high accuracy rates highlight that participants successfully identified the relevant stimuli, suggesting that group differences in response speed were not influenced by poor signal discrimination. Limited ability to integrate identity and emotion in faces in ASD The current findings indicated no overall group differences in overall speed when responding in the face experiment. However, a distinct pattern emerged when examining responses to faces containing target information. Control participants responded significantly faster to faces with both targets compared to those with a single target, suggesting a processing advantage for dual-target faces. In contrast, participants with ASD did not show an advantage for dual targets. This then raises the question: what underlies these contrasting effects in response to dual versus single face targets? Such between-group differences may be related to difficulties in emotion processing commonly reported in ASD. The current findings show that participants with ASD were slower to respond to faces containing only emotional targets, compared to those containing only identity targets. This is consistent with previous research suggesting emotion processing difficulties (Uljarevic & Hamilton, 2013 ; Yeung, 2022 ). However, the current work goes a step further by offering a new observation: if impaired emotion processing was the main contributing factor, then responses to emotional targets alone should be significantly slower compared to responses for faces containing both emotional and identity targets. Yet, the non-parametric analyses did not support this, suggesting that emotion processing difficulties may not fully explain the absence of a dual-target advantage in ASD individuals. Still, the results should be interpreted with caution as a Bayes factor does not provide strong evidence for the null hypothesis here. A second possible explanation emerges when considering the relationship between identity and emotion processing using estimates of processing efficiency. The capacity analysis in typically developing adults suggests that enhanced performance for dual targets reflects increased processing efficiency. Indeed, they showed higher capacity than predicted by a standard parallel model, which assumes independent processing of identity and emotions. Super capacity processing was evident when individual capacity coefficients were averaged across time bins (Table S3, Supplementary Information). The increase in capacity with higher task demands (i.e., dual versus single targets) suggests that target identity cues may facilitate the processing of emotional cues and vice versa, leading to an interaction between the two prior the “target present” decision is made. This finding aligns with previous studies reporting enhanced detection of one type of cue when accompanied by the other (Chen et al., 2015 ; Fitousi, 2017 ; M. H. Johnson et al., 2015 ; Yankouskaya et al., 2012 ). In contrast, the relationship between identity and emotion processing in ASD participants yielded limited processing capacity. Limited capacity indicates that processing of identity and emotions slows down as either channel is engaged. Such a slowdown may be caused by an inhibitory mechanism between identity and emotion (negative cross-talk; Wenger & Townsend, 2000 ), where one type of information may actively interfere with the processing of the other type, making both slower. Alternatively, parallel processing with fixed, evenly shared capacity might occur, where both identity and emotion are processed at the same time, but the resources are split equally between them, so neither gets processed as efficiently as when processed individually. Instead of being processed in parallel, identity and emotion may be processed serially, which would also contribute to slower overall processing. Both parallel, and serial architectures predict C≅ 0.5, and, potentially, may explain capacity curves in our ASD participants (Fig. 4 ). Clarifying the precise mechanism is essential for advancing how identity and emotion are processed. Although this remains open for future research, the current findings provide evidence of restricted resources in the ASD group in face processing. Is the limited ability to integrate information face-specific in ASD? The literature presents mixed findings, with some results supporting a face-specific processing mechanism that appears to be affected in ASD (Ewing et al., 2013 ; Khorrami et al., 2013 ; Weigelt et al., 2013 ), while other studies propose a more general processing deficit, not unique to faces (Bathelt et al., 2022 ; Kleinhans et al., 2008 ). The current findings on medians of response time support the former account by demonstrating that ASD participants were faster (i) in the object experiment compared to the face experiment; and (ii) for double-target displays compared to either single-target displays in the object experiment, but not in the face experiment. However, a closer examination of the processing architecture using a fine-grained scale of analysis revealed group differences in the object experiment. SFT is specifically designed for analysing individual participant data to uncover underlying cognitive architectures, and recent research continues to extend SFT’s application in studies with small numbers of participants (e.g., Fan et al., 2025 ). This demonstrates that even small sample sizes can yield robust individual-level insights with the appropriate analytical approach. Here, individuals with ASD showed capacity estimates consistent with a prediction by a standard parallel model, indicating no processing advantage. In contrast, typically developing individuals showed evidence for super capacity, suggesting more efficient integration of shape and colour when processing objects. These results suggest that the ASD group process colour and shape information simultaneously, but the presence of both attributes does not benefit processing rates. In contrast, control participants showed a facilitatory interaction, where the presence of target colour enhanced the processing of target shape, and the presence of target shape enhanced the processing of target colour. Could these results be attributed to a diminished ability of the ASD group to generate an interactive type of processing? Our data suggest that we should not exclude this possibility. The ASD group had significantly lower overall capacity processing compared to control participants. This limited processing capacity could restrict the interaction between simple object attributes, and prompt reliance on a parallel type of processing. In the context of more complex information factors, such as identity and emotions, this limitation may lead to even greater difficulty, resulting in a more restricted and less efficient processing style. Implications of current findings The central question across competing theories of information processing in ASD is whether characteristic patterns in task performance reflect weak integrative processing, strong local processing, or both (Happé et al., 1996; Happé & Frith, 2006 ; Nayar et al., 2017 ; Proff et al., 2022 ; Ventura et al., 2023 ). We suggest that the inconsistency in findings may arise from the loose operationalisation of experimental analyses. Our findings show that using the framework of System Factorial Technology offers a quantifiable way to assess integrative processing in the context of divided attention tasks. Specifically, the results indicate that individuals with ASD were able to detect and process identity and emotion cues when processing faces, with their accuracy close to controls. However, they did not benefit from integrating these cues – pointing towards an impairment in interactive processing. Furthermore, this reduced integrative ability extended to simple, non-social stimuli as ASD participants showed limited capacity to integrate features of colour and shape when processing objects. These results support the core prediction of the weak central coherence theory. Therefore, while local processing remains intact, the ability to integrate information at a global level is impaired. Implications of the current findings point to insights for understanding face processing difficulties in ASD. First, individuals with ASD showed a limited ability to integrate identity and emotion when processing faces. Could this deficit contribute to the broader difficulties in face processing observed in autism? While people vary in their ability to recognise identity and emotional cues depending on experience and familiarity (Yankouskaya, et al., 2014 ), all typically developing participants in the current work were more efficient when integrating these factors. This highlights the reliance on integrative processing and suggests an underlying mechanism to support this integration. Notably, supporting evidence comes from functional Magnetic Resonance Imaging (fMRI) findings showing that the interaction between identity and emotion during face processing is supported by synchronised activity between the right orbitofrontal cortex (OFC) and the inferior occipital gyrus (iOG), while also OFC activity is positively associated with processing capacity (Yankouskaya et al., 2017 ). The whole-brain functional hypoconnectivity in autism (Moseley et al., 2015 ) with the reduced synchronised activity between frontal and posterior regions (Just et al., 2012 ; Zhao et al., 2022 ) might explain difficulties in the integration of identity and emotion in individuals with autism. Recently, Kleinhans et al. ( 2025 ) showed that even during resting state fMRI, functional connectivity of the fusiform gyrus was altered in individuals with ASD, compared to neurotypically developing adults, and these differences predicted later memory performance for face identity. This suggests that intrinsic neural organisation related to identity processing is already altered in ASD. In this study, the lack of integrative processes in the ASD group – despite similar accuracy performance with the control group – points towards a disruption in integration between these facial dimensions. What remains to be investigated is which specific mechanism explains this difficulty. Second, the present work demonstrates how advances in human information processing developed in a System Factorial Technology (Johnson et al., 2010 ; Townsend & Eidels, 2011 ; Townsend et al., 2007 ) framework can be applied to quantify the efficiency with which ASD participants (and others) process multiple signals when presented simultaneously. The capacity coefficient measures how efficiently a system speeds up across different stages of processing in a given task. Our individual and group level analyses showed that the capacity coefficients yield additional information than mean RT about system performance. Although initial analyses suggested comparable performance, the ASD group showed reduced integration efficiency, highlighting the limitations of solely relying on standard analyses and the value of relevant approaches when examining underlying cognitive mechanisms. Importantly, while it is possible that the average pattern of responses observed at the group level in ASD will also be observed at the individual level, this cannot be assumed without testing at the individual level (widely acknowledged but overlooked issue in face processing research). Studies frequently report large variability between individuals in the degree of these deficits, with subtle impairments and mixed features (see for discussion Ozonoff et al., 2008 ). Therefore, group inferences can provide more meaningful information by assessing the dynamics of within-person processes as was demonstrated in our study (see also study by Johnson et al., 2010 ). The System Factorial Technology offers ‘ready to use’ tools for assessing individual characteristics of information processing which can be implemented in clinical settings for evaluating face processing deficits in ASD. Finally, the current findings also raise important implications for learning contexts and educational practices. The limited capacity to simultaneously process and integrate, both social and non-social information may hinder how individuals with ASD engage with instructional environments. Effective learning often depends on integrating multimodal inputs – such as facial expressions, gestures, tone of voice, visual materials – which guide attention and support comprehension. Integrative processing deficits suggest that individuals with ASD may struggle to fully benefit from such environments. Furthermore, because the difficulties extend beyond social cues to non-social features, like shape and colour, these integrative challenges may affect how learners with ASD process complex information across academic domains, where simultaneous integration of multiple information is often required. Therefore, one future direction is to investigate how autistic learners perform when information is presented sequentially. This could inform the development of assistive technologies, aiming to reduce integrative load by presenting elements in sequence (e.g., flashcards showing words and images separately, rather than combining the two elements). Such adjustments could be implemented via personalised learning, aligning with the cognitive processing profiles of autistic learners. To conclude, Fifić and Townsend ( 2010 ) suggested that an interactive dependency between features is the critical concept for modelling face perception. Our findings demonstrate that the ability to integrate personal and emotional information in faces is impaired in people with ASD and this, possibly, reflects a limited general ability of interactive processing. While supporting the main idea of the weak coherence hypothesis, our study, however, illustrates the need for further research into cognitive and neural mechanisms underlying the impairments in capacity processing. Such insights will underscore the importance of aligning educational approaches with cognitive processing profiles, to better support learning outcomes. References Alguacil S, Madrid E, Espín AM, Ruz M (2017) Facial identity and emotional expression as predictors during economic decisions. Cogn Affect Behav Neurosci 17:315–329. https://doi.org/10.3758/s13415-016-0481-9 American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596 Bach DR, Schmidt-Daffy M, Dolan RJ (2014) Facial expression influences face identity recognition during the attentional blink. Emotion 14(6):1007–1013. https://doi.org/10.1037/a0037945 Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E (2001) The Autism Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism. Malesand Females, Scientists and Mathematicians Bathelt J, Koolschijn PCM, Geurts HM (2022) Atypically slow processing of faces and non-faces in older autistic adults. Autism 26(7):1737–1751. https://doi.org/10.1177/13623613211065297 Blair R (2002) Fractionation of visual memory: Agency detection and its impairment in autism. Neuropsychologia 40(1):108–118. https://doi.org/10.1016/S0028- 3932(01)00069 – 0 Blakesley RE, Mazumdar S, Dew MA, Houck PR, Tang G, Reynolds CF, Butters MA (2009) Comparisons of methods for multiple hypothesis testing in neuropsychological research. Neuropsychology 23(2):255–264. https://doi.org/10.1037/a0012850 Bruce V, Young A (1986) Understanding face recognition. Br J Psychol 77(3):305–327. https://doi.org/10.1111/j.2044-8295.1986.tb02199.x Chen W, Liu CH, Li H, Tong K, Ren N, Fu X (2015) Facial expression at retrieval affects recognition of facial identity. Front Psychol. 6 https://doi.org/10.3389/fpsyg.2015.00780 D’Argembeau A, Van Der Linden M (2011) Influence of facial expression on memory for facial identity: Effects of visual features or emotional meaning? Emotion 11(1):199–202. https://doi.org/10.1037/a0022592 Dobs K, Schultz J, Bülthoff I, Gardner JL (2018) Task-dependent enhancement of facial expression and identity representations in human cortex. NeuroImage 172:689–702. https://doi.org/10.1016/j.neuroimage.2018.02.013 Efron B, Tibshirani RJ (1994) An introduction to the bootstrap, 1st edn. Chapman and Hall/CRC. https://doi.org/10.1201/9780429246593 Ekman P, O'Sullivan M (1988) The role of context in interpreting facial expression: comment on Russell and Fehr (1987). Journal of Experimental Psychology: General, 117 (1), 86–90 Ewing L, Pellicano E, Rhodes G (2013) Reevaluating the selectivity of face-processing difficulties in children and adolescents with autism. J Exp Child Psychol 115(2):342–355. https://doi.org/10.1016/j.jecp.2013.01.009 Fan G, Collins AJ, Thomas RD (2025) Visual Processing Within and Across Hemispheres: A Systems Factorial Technology Analysis. Comput Brain Behav. https://doi.org/10.1007/s42113-025-00249-5 Farah MJ, Levinson KL, Klein KL (1995) Face perception and within-category discrimination in prosopagnosia. Neuropsychologia 33(6):661–674. https://doi.org/10.1016/0028-3932(95)00002-K Fifić M, Townsend JT (2010) Information-processing alternatives to holistic perception: Identifying the mechanisms of secondary-level holism within a categorization paradigm. J Experimental Psychology: Learn Memory Cognition 36(5):1290–1313. https://doi.org/10.1037/a0020123 Fisher K, Towler J, Eimer M (2016) Facial identity and facial expression are initially integrated at visual perceptual stages of face processing. Neuropsychologia 80:115–125. https://doi.org/10.1016/j.neuropsychologia.2015.11.011 Fitousi D (2017) What’s in a face file? Feature binding with facial identity, emotion, and gaze direction. Psychol Res 81(4):777–794. https://doi.org/10.1007/s00426-016-0783-0 Fox CJ, Barton JJS (2007) What is adapted in face adaptation? The neural representations of expression in the human visual system. Brain Res 1127:80–89. https://doi.org/10.1016/j.brainres.2006.09.104 Ganel T, Goshen-Gottstein Y (2004) Effects of Familiarity on the Perceptual Integrality of the Identity and Expression of Faces: The Parallel-Route Hypothesis Revisited. J Exp Psychol Hum Percept Perform 30(3):583–597 Gehdu BK, Press C, Gray KLH, Cook R (2024) Autistic adults have insight into their relative face recognition ability. Sci Rep 14(1):17802. https://doi.org/10.1038/s41598-024-67649-8 Greimel E, Schulte-Rüther M, Kamp-Becker I, Remschmidt H, Herpertz-Dahlmann B, Konrad K (2014) Impairment in face processing in autism spectrum disorder: A developmental perspective. J Neural Transm 121(9):1171–1181. https://doi.org/10.1007/s00702-014-1206-2 Happé FGE (1996) Studying weak central coherence at low levels: Children with autism do not succumb to visual illusions: A research note. Child Psychol Psychiatry Allied Disciplines 37(7):873–877. https://doi.org/10.1111/j.1469 7610.1996.tb01483.x Happé F, Frith U (2006) The Weak Coherence Account: Detail-focused Cognitive Style in Autism Spectrum Disorders. J Autism Dev Disord 36(1):5–25. https://doi.org/10.1007/s10803-005-0039-0 Haxby JV, Hoffman EA, Gobbini MI (2000) The distributed human neural system for face perception. Trends Cogn Sci 4(6):223–233. https://doi.org/10.1016/S1364-6613(00)01482-0 Høyland AL, Nærland T, Engstrøm M, Lydersen S, Andreassen OA (2017) The relation between face-emotion recognition and social function in adolescents with autism spectrum disorders: A case control study. PLoS ONE 12(10):e0186124. https://doi.org/10.1371/journal.pone.0186124 Jemel B, Mottron L, Dawson M (2006) Impaired Face Processing in Autism: Fact or Artifact? J Autism Dev Disord 36(1):91–106. https://doi.org/10.1007/s10803-005-0050-5 Johnson MH, Senju A, Tomalski P (2015) Neurosci Biobehavioral Reviews 50:169–179. https://doi.org/10.1016/j.neubiorev.2014.10.009 . The two-process theory of face processing: Modifications based on two decades of data from infants and adults Johnson SA, Blaha LM, Houpt JW, Townsend JT (2010) Systems Factorial Technology provides new insights on global–local information processing in autism spectrum disorders. J Math Psychol 54(1):53–72. https://doi.org/10.1016/j.jmp.2009.06.006 Just MA, Keller TA, Malave VL, Kana RK, Varma S (2012) Autism as a neural systems disorder: A theory of frontal-posterior underconnectivity. Neurosci Biobehavioral Reviews 36(4):1292–1313. https://doi.org/10.1016/j.neubiorev.2012.02.007 Kay M, Wobbrock J (2016) ARTool: Aligned Rank Transform for Nonparametric Factorial ANOVAs . R package version 0.10.4. https://github.com/mjskay/ARTool Khorrami A, Tehrani-Doost M, Esteky H (2013) Comparison between Face and Object Processing in Youths with Autism Spectrum Disorder. An event related potentials study Kleinhans NM, Richards T, Sterling L, Stegbauer KC, Mahurin R, Johnson LC, Greenson J, Dawson G, Aylward E (2008) Abnormal functional connectivity in autism spectrum disorders during face processing. Brain 131(4):1000–1012. https://doi.org/10.1093/brain/awm334 Kleinhans N, Larsen SF, Estes A, Aylward E (2025) Intrinsic Functional Connectivity Alterations of the Fusiform Face Area in Autism Spectrum Disorder. NeuroSci 6(2):29. https://doi.org/10.3390/neurosci6020029 Li Y, Zhang M, Liu S, Luo W (2022) EEG decoding of multidimensional information from emotional faces. NeuroImage 258:119374. https://doi.org/10.1016/j.neuroimage.2022.119374 Mansouri H, Paige RL, Surles JG (2004) Aligned Rank Transform Techniques for Analysis of Variance and Multiple Comparisons. Commun Stat - Theory Methods 33(9):2217–2232. https://doi.org/10.1081/STA-200026599 Marascuilo LA, Levin JR (1970) Appropriate posthoc comparisons for interaction and nested hypotheses in analysis of variance designs: The elimination of Type IV errors. Am Educ Res J 7:397–421 Miller J (1986) Timecourse of coactivation in bimodal divided attention. Percept Psychophys 40(5):331–343 Moseley RL, Ypma RJF, Holt RJ, Floris D, Chura LR, Spencer MD, Baron- Cohen S, Suckling J, Bullmore E, Rubinov M (2015) Whole-brain functional hypoconnectivity as an endophenotype of autism in adolescents. NeuroImage: Clin 9:140–152. https://doi.org/10.1016/j.nicl.2015.07.015 Nayar K, Voyles AC, Kiorpes L, Di Martino A (2017) Global and local visual processing in autism: An objective assessment approach. Autism Res 10(8):1392–1404. https://doi.org/10.1002/aur.1782 Neufeld RWJ, Townsend JT, Jetté J (2007) Quantitative Response Time Technology for Measuring Cognitive-Processing Capacity in Clinical Studies. In: Neufeld RWJ (ed) Advances in clinical cognitive science: Formal modeling of processes and symptoms. American Psychological Association, pp 207–238. https://doi.org/10.1037/11556-007 Nomi JS, Uddin LQ (2015) Face processing in autism spectrum disorders: From brain regions to brain networks. Neuropsychologia 71:201–216. https://doi.org/10.1016/j.neuropsychologia.2015.03.029 Ozonoff S, Heung K, Byrd R, Hansen R, Hertz-Picciotto I (2008) The onset of autism: Patterns of symptom emergence in the first years of life. Autism Res 1(6):320–328. https://doi.org/10.1002/aur.53 Proff I, Williams GL, Quadt L, Garfinkel SN (2022) Sensory processing in autism across exteroceptive and interoceptive domains. Psychol Neurosci 15(2):105–130. https://doi.org/10.1037/pne0000262 Redfern AS, Benton CP (2017) Expression Dependence in the Perception of Facial Identity. I-Perception 8(3):2041669517710663. https://doi.org/10.1177/2041669517710663 Schiltz C, Sorger B, Caldara R, Ahmed F, Mayer E, Goebel R, Rossion B (2006) Impaired Face Discrimination in Acquired Prosopagnosia Is Associated with Abnormal Response to Individual Faces in the Right Middle Fusiform Gyrus. Cereb Cortex 16(4):574–586. https://doi.org/10.1093/cercor/bhj005 Schultz RT (2005) Developmental deficits in social perception in autism: The role of the amygdala and fusiform face area. Int J Dev Neurosci 23(2–3):125–141. https://doi.org/10.1016/j.ijdevneu.2004.12.012 Scofield DG (2015) nestedRanksTest: Mann-Whitney-Wilcoxon Test for Nested Ranks. R package version 0.2 Stantic M, Brewer R, Duchaine B, Banissy MJ, Bate S, Susilo T, Catmur C, Bird G (2021) The Oxford Face Matching Test: A non-biased test of the full range of individual differences in face perception. Behav Res Methods 54(1):158–173. https://doi.org/10.3758/s13428-021-01609-2 Taylor GJ, Bagby RM (2000) An overview of the alexithymia construct. In: Bar-On R, Parker JDA (eds) The handbook of emotional intelligence: Theory, development, assessment, and application at home, school, and in the workplace. Jossey Bass/Wiley, pp 40–67 Tottenham N, Tanaka JW, Leon AC, McCarry T, Nurse M, Hare TA, Marcus DJ, Westerlund A, Casey B, Nelson C (2009) The NimStim set of facial expressions: Judgments from untrained research participants. Psychiatry Res 168(3):242–249. https://doi.org/10.1016/j.psychres.2008.05.006 Towler JR, Jackson MC, Tree JJ (2025) Impairments for faces but not for abstract shapes in developmental prosopagnosia: Evidence from visual working memory tasks. Cognit Neuropsychol 1–22. https://doi.org/10.1080/02643294.2025.2498154 Townsend JT, Eidels A (2011) Workload capacity spaces: A unified methodology for response time measures of efficiency as workload is varied. Psychon Bull Rev 18(4):659–681. https://doi.org/10.3758/s13423-011-0106-9 Townsend JT, Fific M, Neufeld RWJ (2007) Assessment of mental architecture in clinical/cognitive research. In: Treat RRBTA, Baker TB (eds) Psychological clinical science: papers in honor of richard m. mcfall. Psychology, New York, pp 223–258 Townsend JT, Wenger MJ (2004) A Theory of Interactive Parallel Processing: New Capacity Measures and Predictions for a Response Time Inequality Series. Psychol Rev 111(4):1003–1035. https://doi.org/10.1037/0033 295X.111.4.1003 Uljarevic M, Hamilton A (2013) Recognition of Emotions in Autism: A Formal Meta- Analysis. J Autism Dev Disord 43(7):1517–1526. https://doi.org/10.1007/s10803-012-1695-5 Van Der Velde J, Van Tol M-J, Goerlich-Dobre KS, Gromann PM, Swart M, De Haan L, Wiersma D, Bruggeman R, Krabbendam L, Aleman A (2014) Dissociable morphometric profiles of the affective and cognitive dimensions of alexithymia. Cortex 54:190–199. https://doi.org/10.1016/j.cortex.2014.02.017 Ventura M, Innamorato F, Palmisano A, Cicinelli G, Nobile E, Manippa V, Keller R, Rivolta D (2023) Investigating the impact of disposable surgical face-masks on face identity and emotion recognition in adults with autism spectrum disorder. Autism Res 16(5):1063–1077. https://doi.org/10.1002/aur.2922 Vrancken L, Vermeulen E, Germeys F, Verfaillie K, Attention (2019) Percept Psychophysics, 81(1), 217–236. https://doi.org/10.3758/s13414-018-1603-y Weigelt S, Koldewyn K, Kanwisher N (2012) Face identity recognition in autism spectrum disorders: A review of behavioral studies. Neurosci Biobehavioral Reviews 36(3):1060–1084. https://doi.org/10.1016/j.neubiorev.2011.12.008 Weigelt S, Koldewyn K, Kanwisher N (2013) Face Recognition Deficits in Autism Spectrum Disorders Are Both Domain Specific and Process Specific. PLoS ONE 8(9):e74541. https://doi.org/10.1371/journal.pone.0074541 Wenger MJ, Townsend JT (2000) Basic Response Time Tools for Studying General Processing Capacity in Attention, Perception, and Cognition. J Gen Psychol 127(1):67–99. https://doi.org/10.1080/00221300009598571 Wechsler D (1997) Wechsler Adult Intelligence Scale–Third Edition (WAIS-III) [Database record]. APA PsycTests. https://doi.org/10.1037/t49755-000 Wobbrock JO, Findlater L, Gergle D, Higgins JJ (2011) The aligned rank transform for nonparametric factorial analyses using only anova procedures. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 143–146. https://doi.org/10.1145/1978942.1978963 Yankouskaya A, Booth DA, Humphreys G, Attention (2012) Percept Psychophysics, 74(8), 1692–1711. https://doi.org/10.3758/s13414-012-0345-5 Yankouskaya A, Humphreys GW, Rotshtein P (2014) The processing of facial identity and expression is interactive, but dependent on task and experience. Front Hum Neurosci 8. https://doi.org/10.3389/fnhum.2014.00920 Yankouskaya A, Rotshtein P, Humphreys GW (2014) Interactions between Identity and Emotional Expression in Face Processing across the Lifespan: Evidence from Redundancy Gains. Journal of Aging Research, 2014, 1–12. https://doi.org/10.1155/2014/136073 Yankouskaya A, Stolte M, Moradi Z, Rotshtein P, Humphreys G (2017) Integration of identity and emotion information in faces: fMRI evidence. Brain Cogn 116:29–39. https://doi.org/10.1016/j.bandc.2017.05.004 Yeung MK (2022) A systematic review and meta-analysis of facial emotion recognition in autism spectrum disorder: The specificity of deficits and the role of task characteristics. Neuroscience and Biobehavioral Reviews Yuan Y, Johnson VE (2008) Bayesian hypothesis tests using nonparametric statistics. Statistica Sinica 18:1185–1200 Zhao H-C, Lv R, Zhang G-Y, He L-M, Cai X-T, Sun Q, Yan C-Y, Bao X-Y, Lv X-Y, Fu B (2022) Alterations of Prefrontal-Posterior Information Processing Patterns in Autism Spectrum Disorders. Front NeuroSci 15:768219. https://doi.org/10.3389/fnins.2021.768219 Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryInformation.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7612091","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":514793452,"identity":"56d117b7-67d8-44f7-9bcf-ba83a43047d8","order_by":0,"name":"Marianna Constantinou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIiWNgGAWjYFADZhBRcSBBAkTzEFSeAFQE1nKGJC0gmrGNCC267YePPfz5wy7Pnp334GPeeXfyJGckMD5428Yg29+AXYvZmbR0Y56E5GIeZr5kY95tz4qlJRKYDee2MRjPOIBDyw0eM2mGBObEHmYgg3fb4cR5Egls0rxtDIkNeLRI/kioh2qZA9bC/hukZT4eLRI8CYehWhoOJ84G2sIM0rIBl5YzaWnSPGnHE3sO8xgbzjl2uFiy52Gz5JxzEsYbcWk5fviY5A+b6sT2/jOGD97UHM6TOJ588MObMhvZeTi0YAOMDUBCAkySBsjQMgpGwSgYBcMUAABGCFmB90pd2QAAAABJRU5ErkJggg==","orcid":"","institution":"Hamad Bin Khalifa University","correspondingAuthor":true,"prefix":"","firstName":"Marianna","middleName":"","lastName":"Constantinou","suffix":""},{"id":514793453,"identity":"40671742-9dce-4745-8daa-25a1c44334c4","order_by":1,"name":"Dena Al-Thani","email":"","orcid":"","institution":"Hamad Bin Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Dena","middleName":"","lastName":"Al-Thani","suffix":""},{"id":514793454,"identity":"d5daa4b6-b39c-4cab-8636-0f8e7827070c","order_by":2,"name":"Marwa Qaraqe","email":"","orcid":"","institution":"Hamad Bin Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Marwa","middleName":"","lastName":"Qaraqe","suffix":""},{"id":514793455,"identity":"aeea614b-82be-44e0-a8fd-2a7e96b21fc1","order_by":3,"name":"Ala Yankouskaya","email":"","orcid":"","institution":"Bournemouth University","correspondingAuthor":false,"prefix":"","firstName":"Ala","middleName":"","lastName":"Yankouskaya","suffix":""}],"badges":[],"createdAt":"2025-09-14 10:48:41","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7612091/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7612091/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91515216,"identity":"1d313106-9f47-4f9a-9b42-4e3fb83d5ee2","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":311434,"visible":true,"origin":"","legend":"\u003cp\u003eAn example of a stimulus set. Panel A (face experiment): IE – a face containing both the target identity and the target emotional expression; I – a face containing the target identity only, and not the target emotional expression; E – a face containing the target emotional expression only, and not the target identity; NT1-NT3 – faces containing neither the target identity nor the target emotion. Due to publication restrictions, we present other faces here (Ekman \u0026amp; O'Sullivan, 1988) as examples only. Panel B (object experiment): SC – an object containing both the target shape and the target colour; S – an object containing the target shape only, and not the target colour; C – an object containing the target colour only, and not the target shape; NT1-NT3 – objects containing neither the target shape nor the target colour.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/ddfad7124dde925e6f6d2b27.png"},{"id":91515219,"identity":"513ae383-30cd-4436-9b4b-3399ab4eead2","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":265072,"visible":true,"origin":"","legend":"\u003cp\u003eExploring interaction between Group (ASD and Control) and Stimuli (I, E, IE) for RTs. Panel A shows that the difference between ASD and Control group (depicted in dashed red line) for faces containing target identity alone (I) and target emotion alone (E) (depicted in vertical blue lines) are similar in terms of the magnitude, but different in direction. Panel B demonstrates significantly larger differences between ASD and Control for stimuli containing both the identity target and the emotion target (IE) compared to target identity alone (I). Panel C depicted significantly larger differences between ASD and Control for stimuli containing both the identity target and the emotion target (IE) compared to target emotion alone (E).\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/f4d81945b7713ce360762d3b.png"},{"id":91515223,"identity":"c77cf5cb-767a-4350-a7f1-672ca5888ae5","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":937682,"visible":true,"origin":"","legend":"\u003cp\u003eThe effects of Stimuli in the face and object experiments for the ASD and Control groups. Individual plots in each panel demonstrate individual’s experiment performance with black crosses mapping averaged medians for a stimulus in each experiment. The Differences section in each panel represents between-stimulus differences with error bars indicating variability outside the upper and lower quartiles and Post Hoc p-values adjusted for multiple comparisons. Significant differences are highlighted in black.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/bf89b55d7c7f52c93a311a83.png"},{"id":91515908,"identity":"ca6c2cb9-7bcb-4e82-b2c5-427155cb9e05","added_by":"auto","created_at":"2025-09-17 09:20:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":463102,"visible":true,"origin":"","legend":"\u003cp\u003eIndividual capacity functions in the face experiment in the ASD and Control groups. The black lines represent capacity coefficients derived from pooled[1] data across individuals. The horizontal black line depicts the predictions of UCIP models (unlimited capacity, C(t)= 1).\u003c/p\u003e\n\u003cp\u003e[1] To obtain the pooled lines, capacity coefficients were calculated based on distributions for three conditions (I, E, IE) across entire group of participants by, first, averaging the CDFs for each condition. Then we convert these into survival functions and took their negative natural log to create integrated hazards. The capacity coefficient for each group was then generated by creating a ratio of these average integrated hazards at each time bin.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/87bbe30bacfaeb6348755a11.png"},{"id":91515221,"identity":"2129d8f6-4c0c-4aa4-b707-88eb2575fef8","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":401689,"visible":true,"origin":"","legend":"\u003cp\u003eIndividual capacity functions in the object experiment in the ASD and Control groups. The black lines represent capacity coefficients derived from pooled data across individuals. The horizontal black line depicts the predictions of UCIP models (unlimited capacity, C(t)= 1).\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/86f716b6bf02055a223eca78.png"},{"id":91515225,"identity":"33894c89-b49b-4971-b978-34182c052530","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":72268,"visible":true,"origin":"","legend":"\u003cp\u003eGroups’ overall capacity in the face and object experiments.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/ea69d86e010424cffcfd3788.png"},{"id":91515224,"identity":"9597654f-da83-4e8c-b38e-66c94b869a90","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":752426,"visible":true,"origin":"","legend":"\u003cp\u003eA visualisation of the mean capacity distribution for each group, for faces (x) and objects (y), paired as a single data point (xy). The clouds’ spread represents variability and dispersion of the data points. The cloud for the ASD group is positioned mainly below 1 on both dimensions, whereas the CTRL group cloud clusters entirely above 1, reflecting higher general capacity.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/d661df4df763231e80e8eb1e.png"},{"id":91517506,"identity":"7b7ad748-8385-4433-960c-150db321de9b","added_by":"auto","created_at":"2025-09-17 09:36:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4275276,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/6734579a-151c-40ce-a6bb-737e44d7146c.pdf"},{"id":91515240,"identity":"e9bafd89-fa67-489a-b9bf-4bb1c239183f","added_by":"auto","created_at":"2025-09-17 09:12:22","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":30787490,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7612091/v1/4cf365fff656cc08c382b9b1.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eIntegrating identity and emotion information in faces in ASD – capacity limits of information processing\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAutism spectrum disorder (ASD) is a neurodevelopmental condition associated with a broad range of expressions, some of which include persistent deficits in social communication and social interaction (American Psychiatric Association, 2013). Social exchanges depend heavily on the processing of facial cues, making face perception a key process in contributing to the social challenges in ASD. Even though evidence reveals that individuals with ASD show impairments in face perception (Nomi \u0026amp; Uddin, 2015; Schultz, 2005; Stantic et al., 2021) the specific nature of these impairments remains unclear. The ambiguity is related to the dynamic information expressed in faces, which simultaneously carry both information about one’s identity and one’s emotional expressions. Therefore, distinguishing between these aspects, and whether they independently or cooperatively affect individuals with ASD, is necessary for identifying their specific perceptual challenges. Clarifying whether difficulties stem from impaired processing of facial identity, emotional expression, or the integration of both can inform how such deficits contribute to broader challenges in social development and inform appropriate and targeted interventions. Existing evidence shows that ASD is linked to a reduced sensitivity in perceiving either facial identity (Blair, 2002; Jemel et al., 2006), or processing emotional expressions\u0026nbsp;(Høyland et al., 2017; Uljarevic \u0026amp; Hamilton, 2013). However, it could also be the case that both dimensions are affected in ASD\u0026nbsp;(Greimel et al., 2014; Ventura et al., 2023; Weigelt et al., 2012).\u0026nbsp;Ventura et al. (2023)\u0026nbsp;used surgical face masks to disrupt recognition of identity and emotion in typically developing adults and individuals with ASD. Regarding identity, only typically developing participants showed a context congruence effect – better recognition for faces when learned and tested with masks. Individuals with ASD had poorer recognition regardless of the whether the testing condition matched the learning condition. Regarding emotion, while masks impaired recognition in both groups, the specific emotions affected varied between groups\u0026nbsp;(Ventura et al., 2023). Such findings indicate that both identity and emotion, when tested separately, are affected in ASD in the context of face processing. What remains to be examined is whether, when identity and emotion interact, typically developing individuals and individuals with ASD differ – and if so, how these differences are expressed.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe theoretical framework justifying the above experimental work proposes the existence of two pathways in face processing: one responsible for identity-related processing and another for emotion-related processing (Bruce \u0026amp; Young, 1986). These pathways are structurally and functionally separate, with identity processing being associated with activity in the fusiform gyrus and inferior occipital gyri, while emotion processing being linked to activity in the limbic system (Haxby et al., 2000). This distinction is further supported by neuropsychological evidence with impairments in the former being linked to prosopagnosia (i.e., selective inability to recognise faces; Farah et al., 1995; Schiltz et al., 2006), while impairments in the latter to alexithymia (i.e., deficits in recognising one’s emotions; Taylor \u0026amp; Bagby, 2000; Van Der Velde et al., 2014). Notably, developmental prosopagnosia does not necessarily co-occur with alexithymia (Towler et al., 2025), a separation also reported in individuals with autism (Gehdu et al., 2024). Even though early models treated identity and emotion as separate systems, more recent behavioural, electrophysiological, and neural evidence recognises their interconnectivity, challenging the idea of strict independence, at least in neurotypical populations\u0026nbsp;(Bach et al., 2014; D’Argembeau \u0026amp; Van Der Linden, 2011; Dobs et al., 2018; Alguacil et al., 2017; Fisher et al., 2016; Li et al., 2022; Yankouskaya et al., 2012). Despite this, experimentally capturing the interaction is challenging. Specifically, isolating identity-related information from emotion-related information is difficult, as emotional expressions involve both identity-independent and identity-dependent features\u0026nbsp;(Fox \u0026amp; Barton, 2007). Moreover, identity recognition can also be influenced by emotional expressions as affective states can impact how identity is perceived (Chen et al., 2015; Redfern \u0026amp; Benton, 2017).\u0026nbsp;Therefore, attempts to control one factor might interfere the processing of the other.\u003c/p\u003e\n\u003cp\u003eRather than supporting a modular system, the observed interconnectivity between identity and emotion processing suggests a functional interdependence. In typically developing individuals, these two factors are processed interactively. Critically, the integration of facial identity and facial expressions happens only when facial expressions are emotionally valanced (Vrancken et al., 2019; Yankouskaya et al., 2012), a mechanism which is supported by functional connections between the orbitofrontal cortex and the inferior occipital gyri (Yankouskaya et al., 2017). This suggests that executive mechanisms may allow for the integration between identity and emotion, enhancing perceptual processing. This would explain why performance is faster in response to faces which contain both identity and emotion target information, rather than each factor alone (Yankouskaya, et al., 2014). Such interactions are independent of cognitive ageing as they are found across adulthood (Yankouskaya, et al., 2014) and even older adults can sometimes perform better than younger adults, which could be explained by a related level of experience and familiarity with facial stimuli (Ganel \u0026amp; Goshen-Gottstein, 2004; Yankouskaya, et al., 2014). Therefore, the interactive nature of these factors is consistently embedded in typically developing individuals, possibly enhancing perceptual efficiency. Interestingly, the effect also generalises to object processing, suggesting that redundant target effects happen with responses being faster and more accurate when two target signals are presented simultaneously, rather than each target alone (e.g., shape and colour; see Miller, 1986). In terms of social context, integration has been found to also facilitate economic decision-making (Alguacil e al., 2017). However, to assess the beneficial nature of this interaction, it is first necessary to examine populations that may not engage this mechanism in the same way. This comparison can reveal whether the observed effects are related to the interaction itself, or to other differences in face processing strategies, such as processing features serially (one after the other, in sequence) or in parallel (independently but at the same time), rather than through an integrated process.\u003c/p\u003e\n\u003cp\u003eThis study aims to re-assess face processing abilities in adults with ASD. Specifically, the objectives of the current work were to address whether: i) individuals with ASD exhibit the ability to integrate information of facial identity and facial emotional expressions; and ii) whether this ability is exclusively linked to face processing, or rather reflects a general ability to process information in an integrative manner. To achieve these objectives, a divided attention task was used where participants were required to monitor the presence of two targets and respond quickly and accurately when they identified any target information (‘target present’). Alternatively, they were instructed to indicate the absence of a target (‘target absent’). Two experiments were utilised. In the ‘Face experiment’ face stimuli were used containing one stimulus displaying the target identity, one stimulus displaying the target emotional expression, one stimulus displaying both targets, and three stimuli containing no target information. In the ‘Object experiment’ object stimuli were utilised, manipulating the colour and shape to include one stimulus displaying the target colour, one stimulus displaying the target shape, one stimulus displaying both targets, and three stimuli with no target attributes. The modelling and comparison of response time distributions for single versus double targets, for each experiment, allowed for the estimation of capacity processing when the number of targets increased. Capacity processing, as developed in a human information processing framework – the System Factorial Technology (Townsend \u0026amp; Eidels, 2011; Townsend \u0026amp; Wenger, 2004; Wenger \u0026amp; Townsend, 2000), refers to how efficiently individuals handle increased information load. Here, reduced response times for double-target trials indicate more efficient information integration. By examining these effects in ASD, the current work investigates whether integrative processing is preserved in ASD, and whether it is unique to face processing.\u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003eParticipants\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNineteen adults with high functioning ASD were recruited via learning support committees at Bournemouth University, Aspens Charity Brighton, Liverpool Hope University, University of Liverpool, Liverpool John Moores University. Nineteen typically developing adults were recruited via email distribution lists across Bournemouth University, three universities in Liverpool, and the general public. Participants between groups were matched in age, sex, and level of education (see Supplementary Information, Table S1). All participants were over 18 years of age (age range 19-52 years), right-handed and were not on any medication at the time of the study. All participants in the ASD group had confirmed diagnosis of autism spectrum disorder.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExclusion criteria across groups included self-report of major neurological disorders, presence of systemic diseases, lack of normal or corrected-to-normal vision, and presence of colour blindness. All participants completed the Wechsler Abbreviated Scale of Intelligence (WAIS-III; Wechsler, 1999) and the Autism Spectrum Quotient (AQ; Baron-Cohen et al., 2001) to estimate intellectual abilities and autism symptom severity respectively. No group differences were found in intellectual abilities (see Supplementary Information, Fig. S1, A). Regarding the AQ scores, the control group scored lower (mean = 12.0, SD = 2.95, min = 7, max = 15.7), compared to the ASD group (mean = 33.0, SD = 5.0, min = 23, max = 42; see Supplementary Information, Fig. S1, B), with the scores of the latter being in line with previously reported literature (mean AQ score = 35.8, SD = 6.5; see Baron-Cohen et al., 2001).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEthical approval has been cleared by the Faculty of Research Ethics Committee at Liverpool Hope University, and Bournemouth University and the Access Review Group of Autism Together.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExperiments and Stimuli\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTwo experiments were used \u0026ndash; the face experiment and the object experiment. In each experiment participants performed a divided attention task which required the simultaneous monitoring of two targets. Participants had to respond with either \u0026lsquo;target present\u0026rsquo; when any target appeared on the screen, or \u0026lsquo;target absent\u0026rsquo; when a trial displayed no target information. Participants were instructed to respond as fast and as accurately as possible. The experiments differed only in stimuli and stimulus onset parameters. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the face experiment, the targets were identity and emotional expression. Three sets of stimuli were randomly assigned to participants in each group. Each set included six images of faces: a face containing target identity only (I), a face containing target emotion only (E), a face containing both target identity and target emotion (IE), and three faces containing no target information (NT1-NT3). All face stimuli were sourced from The NimStim Face Stimuli Set (Tottenham et al., 2009). Due to restricted permission to publish some face images, identification of actors and posed emotional expressions can be found in Supplementary Information (Table S2). The procedure of stimuli selection is described in detail in\u0026nbsp;Yankouskaya et al. (2012). To eliminate possible target judgments based on hairstyle, stimuli were cropped around the hairline. Background colour was adjusted to black. For an example of a stimulus set see Figure 1(A).\u003c/p\u003e\n\u003cp\u003eIn the object experiment, the targets were geometrical shape and colour. Three sets of stimuli were randomly assigned to participants. Each set consisted of six images of objects: an object containing target shape only (S), an object containing target colour only (C), an object containing target shape and target colour (SC), and three objects containing no target information (NT1-NT3). A list of the used stimuli sets is presented in Supplementary Information (Table S2). For an example of a stimulus set see Figure 1(B).\u003c/p\u003e\n\u003cp\u003eFace stimuli were 8 x 11 cm and object stimuli were 6 x 6 cm when displayed on a 15.5-inch Dell monitor. All stimuli were displayed at a viewing distance of 0.8 m and the angular width subtended by the stimulus was approximately 10\u0026deg;. The presentation of the stimuli and data acquisition were controlled using Cogent 2000 and Cogent Graphics developed by the Cogent 2000 team at the FIL and ICN (http://www.vislab.ucl.ac.uk/cogent_2000.php).\u003c/p\u003e\n\u003cp\u003eFor each experiment, stimuli were presented in a random order. A single trial began with a fixation cross at the centre of the screen (100 ms). Immediately after, the fixation cross was replaced by one of six images which remained for 300 ms in the face experiment, and 200 ms in the object experiment. A blank screen followed by a stimulus offset until a response was made. A jittered (100-600 ms) inter-trial blank display was presented after each trial. If the response time exceeded 2000 ms, the feedback message \u0026lsquo;Missed response\u0026rsquo; appeared, and the trail was repeated at the end of the experimental block.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEach experiment consisted of 420 trials (70 trials per stimulus). Half of these trials required a \u0026lsquo;target present\u0026rsquo; response. Participants were asked to have a short break (2 minutes) after completing 1/3 and 2/3 of each experiment.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eProcedure\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eEach participant took part in two separate testing sessions. In session one, participants completed the Wechsler Abbreviated Scale of Intelligence and a short demographic questionnaire. Session two started with either the face experiment or object experiment (the order was randomly assigned to participants in the ASD group; the same order was kept for matching participants in the Control group).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eParticipants began the experimental session with a written instruction for the first experiment. The instruction was repeated verbally by the investigator to ensure that the written instruction was understood correctly. A practice block of 36 trials followed (6 trials per stimulus) and feedback on accuracy and response speed was provided after each trial during the practice. If a participant made more that 50% of incorrect responses in practice blocks, the instruction was repeated, and a new data set was launched. Three participants (C1, C8, C16 Supplementary Information, Table S1) in the Control group and two participants in the ASD group (A8, A18, Supplementary Information, Table S1) required to repeat practice trials in the face experiment. All participants performed practice trials for the object experiment with an accuracy above 79%.\u003c/p\u003e\n\u003cp\u003eParticipants completed the face, and object experiments in a random order, with the Autism Spectrum Quotient being administered between the two experiments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData Analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe core analysis of the current study involved the assessment of capacity processing across the face experiment and object experiment in the ASD and control groups. Before that, to inform the reader of group-level accuracy and response time performance, and to demonstrate how the traditional approach supports capacity analysis, first non-parametric modelling of the group data is reported.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNon-parametric modelling of the group data\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFirst, to gain an idea about overall accuracy performance, the proportion of correct responses for each group and each experiment was calculated. A nested ranks test (implemented via the nestedRanksTest package; Scofield, 2015) was used to perform a non-parametric comparison between groups. This method extends the Mann-Whitney-Wilcoxon test to a mixed-model framework, enabling two analyses: i) treating group as a fixed effect, and Subject as a random effect, and ii) treating task as a fixed effect, and Subject as a random effect. Test statistics are expressed as z-scores based on rank comparisons between groups or tasks. Significance was determined by comparing the observed z-score to a null distribution generated through 10,000 bootstrap iterations. (Scofield, 2015). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSecond, a non-parametric mixed analysis of variance (ANOVA) was carried out to examine the effects of group on response times for stimuli of the face and object experiments, using the aligned ranks transformation (ART) approach implemented in the ARTool R package (Kay \u0026amp; Wobbrock, 2016). This approach involves computing residuals via a regression model, adding the effect of interest, and ranking the resulting sum prior to performing a parametric factorial ANOVA (Mansouri et al., 2004). The model included two fixed effects \u0026ndash; Group (ASD, Control) and Stimuli (images containing target information) \u0026ndash; the interaction between Group and Stimuli, and a random effect for Subject. Prior to model testing, the appropriateness of the ART procedure was confirmed by verifying that an ANOVA on the ranked data yielded values close to F = 0.00 and p = 1.00 for all effects except the one corresponding to the effect of interest (Wobbrock et al., 2011). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt should be noted that, unlike linear model tests which provide expected mean differences between factor level combinations, the ART approach interprets interactions as \u0026ldquo;differences of differences\u0026rdquo; (Levin \u0026amp; Marascuilo, 1971). In the present study this means assessing whether the difference between face stimuli in the ASD group significantly differs from that in the control group, and whether the difference between object stimuli in the ASD group significantly differs from that in the control group. Interaction effects were evaluated using the \u003cem\u003etestInteractions\u003c/em\u003e function form the phia-package (https://cran.r-project.org/web/packages/phia/phia.pdf), applying the Hommel-adjusted step-up modification of the Bonferroni method to correct for multiple comparisons (Blakesley et al., 2009).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMain effects were tested using Friedman\u0026rsquo;s test with Post Hoc analysis (Wilcoxon signed rank test) and Tukey HSD correction for multiple comparisons. Given the relatively small sample size, exact p-values were computed for the Wilcoxon signed rank based on the true distribution, providing conservative estimates that help control Type I error at the nominal significance level. For each comparison, the Hodges-Lehmann estimator was used to quantify the shift in location (i.e., estimated difference between medians of two distributions), along with 95% confidence intervals using a permutation approach. An effect size for the Wilcoxon signed rank test was calculated by dividing the absolute (positive) standardized test statistic Z by the square root of the number of pairs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo quantify evidence in favour of the null hypothesis (H0), Bayes factors (BFs) were calculated for all contrasts using the non-parametric approach proposed by (Yuan \u0026amp; Johnson, 2008), as implemented in JASP software (JASP Team, 2025, Version 0.95). Unlike standard Bayesian methods that rely on the sampling distribution of original data, this approach uses asymptotic approximations to the distributions of non-parametric test statistics.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eComputing capacity processing\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo test changes in efficiency of processing for dual target displays compared to single target displays, capacity coefficients were calculated for each participant in each experiment. The coefficient represents a ratio between RT performance for a stimulus containing two targets and a joined distribution for two single-target stimuli at a time interval (t) of 10 msec. The RT performance for each condition containing target information (I, E, IE in the face experiment, and S, C, SC in the object experiment) was estimated using the cumulative hazard function. The hazard function, representing a conditional probability that a response will occur in the next interval (i.e., \u003cem\u003et +\u0026nbsp;\u003c/em\u003eD\u003cem\u003et\u003c/em\u003e), captures the instantaneous and time-invariant likelihood of terminating an ongoing process at each unit of time. Computing the total amount of accumulated evidence for a response represents, therefore, the amount of work completed by time \u003cem\u003et\u003c/em\u003e and can be considered as a measure of efficiency (Townsend \u0026amp; Eidels, 2011). In this sense, the ratio between the cumulative hazard function for a stimulus containing two targets and the sum of cumulative hazard functions for single targets indicates whether, and how, adding a target affects the overall efficiency (capacity) of the system at each time interval:\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; H\u003csub\u003e12\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e)\u003c/p\u003e\n\u003cp\u003eC(t) = ------------------------ ,\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;H\u003csub\u003e1\u003c/sub\u003e(\u003cem\u003et\u003c/em\u003e) + H\u003csub\u003e2\u003c/sub\u003e(\u003cem\u003et\u003c/em\u003e)\u003c/p\u003e\n\u003cp\u003ewhere C(t) - capacity coefficient at time \u003cem\u003et\u003c/em\u003e;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eH\u003csub\u003e12\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e) \u0026ndash; the cumulative hazard function for a stimulus containing two targets (denoted as 1 \u0026ndash; target 1, 2 \u0026ndash; target 2);\u003c/p\u003e\n\u003cp\u003eH\u003csub\u003e1\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e) - the cumulative hazard function for a stimulus containing target 1\u003c/p\u003e\n\u003cp\u003eH\u003csub\u003e2\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e) - the cumulative hazard function for a stimulus containing target 2\u003c/p\u003e\n\u003cp\u003eA benchmark in an interpretation of capacity coefficient is the prediction derived from a standard parallel model with unlimited capacity (UCIP). In an experiment where participants are required to respond to either target, the standard parallel model predicts C(t) = 1 (i.e., the amount of work done on two units of information is equal to the sum of work done on every single unit). In the case when a parallel model has two items to process and the processing of both of them speeds up compared to when only one is processed, the system is said to exhibit super capacity (C(t) \u0026gt;1). If increasing a workload slows down processing time, the system will show limited capacity (C(t) \u0026lt; 1), indicating that processing of two-targets is less efficient compared to the baseline UCIP model (Neufeld et al., 2007). A special case of interest occurs if the system operates at fixed capacity (i.e., divides up a fixed resource across two targets); or serially processes two units of information (for example, working first on the identity target, and then on target emotion or vice versa). Although in both cases capacity will be hovering around C(t) = \u0026frac12;, the former assumes equal distribution parameters with H\u003csub\u003e12\u003c/sub\u003e(\u003cem\u003et\u003c/em\u003e) = H\u003csub\u003e1\u003c/sub\u003e(\u003cem\u003et\u003c/em\u003e) = H\u003csub\u003e2\u003c/sub\u003e(\u003cem\u003et\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eFinally, to estimate the population-level distributions of mean capacity ability, bootstrapping was conducted. Bootstrapping is a statistical method used for estimating the sampling distribution of a statistic, here the mean capacity ability, by resampling a dataset with replacement to create multiple simulated samples (Efron \u0026amp; Tibshirani, 1994). 2000 bootstrap samples were simulated by randomly sampling data points from the original dataset with replacement. For each of these resampled datasets the mean capacity for faces and objects was calculated per group.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e\n\u003ch2\u003eAccuracy performance\u003c/h2\u003e\n\u003cp\u003eParticipants in the ASD and Control groups were more accurate in the object experiment (Mdn\u0026thinsp;=\u0026thinsp;92.0, S.D. = 6.1 in ASD group; Mdn\u0026thinsp;=\u0026thinsp;92.0, S.D. = 2.7 in Control group) compared to face experiment (Mdn\u0026thinsp;=\u0026thinsp;84.5, S.D. = 7.4 in ASD group; Mdn\u0026thinsp;=\u0026thinsp;86.0, S.D. = 7.9 in Control group; see details in Supplementary Information, Fig. S2).\u003c/p\u003e\n\u003cp\u003eAccuracy performance was too high to allow for meaningful comparisons across the stimuli in each experiment. Therefore, analyses were conducted on i) differences between experiments in each group, ii) group differences in accuracy performance for each experiment, and iii) differences between stimuli for each experiment containing target information and stimuli for each experiment with no targets, in each group. The results of nested ranks tests indicate significantly higher accuracy in the object experiment compared to the face experiment in each group (Z\u0026thinsp;=\u0026thinsp;0.499, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.0001; Z\u0026thinsp;=\u0026thinsp;0.42, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.0001 respectively for ASD and Control). No significant differences were found between groups in accuracy for the face experiment (Z\u0026thinsp;=\u0026thinsp;0.091, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.12) and for the object experiment (Z\u0026thinsp;=\u0026thinsp;0.061, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.21). There were also no significant differences in accuracy performance between targets and non-targets in ASD (Z=-0.009, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.5) and Control (Z=-0.06, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.8) in both experiments (Z\u0026thinsp;=\u0026thinsp;0.003, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.68; Z\u0026thinsp;=\u0026thinsp;0.02, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.52 in the face and object experiment respectively).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n\u003ch2\u003eResponse time performance\u003c/h2\u003e\n\u003cp\u003eResponse time performance is displayed in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab1\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eMedian response time (*[95% CI]) for correct responses to **stimuli in face and object experiments in the ASD and Control groups.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth colspan=\"7\" align=\"left\"\u003e\n\u003cp\u003eFace experiment\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\"\u003e\n\u003cp\u003egroups\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eE\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eIE\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eNT1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNT2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNT3\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eASD\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e725.7 [607, 786]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e784.8\u003c/p\u003e\n\u003cp\u003e[641, 838]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e670.0\u003c/p\u003e\n\u003cp\u003e[625, 801]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e769.6\u003c/p\u003e\n\u003cp\u003e[678, 878]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e742.2\u003c/p\u003e\n\u003cp\u003e[687, 811]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e781.0\u003c/p\u003e\n\u003cp\u003e[679, 862]\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e696.3\u003c/p\u003e\n\u003cp\u003e[593, 866]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e696.6\u003c/p\u003e\n\u003cp\u003e[580, 838]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e609.9\u003c/p\u003e\n\u003cp\u003e[461,838]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e647.3\u003c/p\u003e\n\u003cp\u003e[504, 747]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e709.5\u003c/p\u003e\n\u003cp\u003e[623, 796]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e655.9\u003c/p\u003e\n\u003cp\u003e[572, 801]\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd colspan=\"7\" align=\"left\"\u003e\n\u003cp\u003eObject experiment\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eS\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eC\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003eSC\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNTo1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNTo2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eNTo3\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eASD\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e721.1\u003c/p\u003e\n\u003cp\u003e[579, 888]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e707.7\u003c/p\u003e\n\u003cp\u003e[600, 793]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e619.7\u003c/p\u003e\n\u003cp\u003e[538,685]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e634.3\u003c/p\u003e\n\u003cp\u003e[600, 788]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e663.1\u003c/p\u003e\n\u003cp\u003e[605, 796]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e655.9\u003c/p\u003e\n\u003cp\u003e[598, 768]\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eControl\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e601.2\u003c/p\u003e\n\u003cp\u003e[565, 642]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e576.4\u003c/p\u003e\n\u003cp\u003e[538, 646]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd colspan=\"2\" align=\"left\"\u003e\n\u003cp\u003e506.4\u003c/p\u003e\n\u003cp\u003e[471, 572]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e611.5\u003c/p\u003e\n\u003cp\u003e[546, 677]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e632.4\u003c/p\u003e\n\u003cp\u003e[552, 667]\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e621.5\u003c/p\u003e\n\u003cp\u003e[549, 655]\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\u003e* Confidence Interval (95% CI) was calculated using the bootstrap percentile method with 10,000 iterations.\u003c/p\u003e\n\u003cp\u003e** Stimuli in face experiment: I - target identity alone, E - target identity alone, IE \u0026ndash; both target identity and target emotion, NT1-NT3 \u0026ndash; three faces containing no target information. Stimuli in object experiment: S \u0026ndash; target shape alone, C \u0026ndash; target colour alone, SC \u0026ndash; both target shape and target colour, NTo1-NTo3 \u0026ndash; displays containing no target information.\u003c/p\u003e\n\u003cp\u003eNested ranks tests showed no significant differences in RT performance between face stimuli containing target (I, E, IE) and non-target (NT1, NT2, NT3) information in ASD (Z=-0.133, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.88) and Control (Z=-0.054, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.68) groups. Similar results were obtained in the object experiment where ASD and Control groups responded to stimuli containing targets (S, C, SC) as fast as to stimuli containing no targets (NTo1-NTo3) (Z=-0.043, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.65; Z=-0.191, n\u0026thinsp;=\u0026thinsp;21, p\u0026thinsp;=\u0026thinsp;.95 for ASD and Control groups respectively). The main interest of the current study is responses to stimuli containing target information. Therefore, all consequent analyses will focus on stimuli required responses \u0026lsquo;target present\u0026rsquo;.\u003c/p\u003e\n\u003cp\u003eTo examine the effects of group on response times, analysis of variance of aligned rank transformed data was performed separately for the face and object experiments. The sums of aligned responses for main effects and F values of ANOVAs on aligned responses of no interest showed values closer to zero in both experiments indicating that data were suitable for the aligned ranks transformation. A linear mixed-effects model with Group (ASD, Control), Stimuli (I, E, IE in the face experiment and S, C, SC in the object experiment) and the interaction between Group and Stimuli as fixed effects and subjects as a random effect was fitted to aligned ranks data. The results of both analyses are summarised in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n\u003cp\u003eAs it can be seen in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, the results of the ANOVA in the face experiment showed a significant interaction between Group and Stimuli and a significant main effect of Stimuli, but a main effect of group did not reach significance. Post Hoc analysis indicated that this interaction stems from significant differences between the ASD and Control groups across all stimuli (see Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab2\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eAnalysis of variance of aligned ranks transformed RT data and Post Hoc comparisons in the face and object experiments\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth colspan=\"3\" align=\"left\"\u003e\n\u003cp\u003eFace experiment\u003c/p\u003e\n\u003c/th\u003e\n\u003cth colspan=\"3\" align=\"left\"\u003e\n\u003cp\u003eObject experiment\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\"\u003e\n\u003cp\u003eMain effects\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e*F\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e*F\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eGroup\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.30\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e.58\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e4.71\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e.04\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eStimuli\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e26.99\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e37.2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eInteraction Group*Stimuli\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e18.31\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.38\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e.68\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003ePost Hoc for Group*Stimuli\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003e\u0026chi;\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eASD-Control:I-E\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e9.07\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e.005\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eASD-Control:I-IE\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e36.62\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eASD-Control:E-IE\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8.76\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e.005\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003ctfoot\u003e\n\u003ctr\u003e\n\u003ctd colspan=\"7\"\u003e*Type III Wald F-tests using the Kenward-Roger coefficient covariance matrix and Satterthwaite degrees of freedom (\u003cem\u003edf\u003c/em\u003e)\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tfoot\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u0026nbsp;In the object experiment, the results of the ANOVA did not reveal a significant interaction between Group and Stimuli.\u003c/p\u003e\n\u003cp\u003eNext, main effects in each experiment were explored using Friedman\u0026rsquo;s test as a non-parametric version of a one-way repeated measures ANOVA with Post Hoc analysis (Wilcoxon signed rank test and Tukey HSD correction for multiple comparisons). In the face experiment, the ASD group (\u003cem\u003e\u0026chi;\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e (2)\u0026thinsp;=\u0026thinsp;9.36, p\u0026thinsp;=\u0026thinsp;.009) was faster in responding to target identity alone compared to target emotional expression alone (p\u0026thinsp;=\u0026thinsp;.009). The differences between faces containing both targets and either single target were non-significant (p\u0026thinsp;=\u0026thinsp;.74; BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;3.19 for contrast between I and IE; p\u0026thinsp;=\u0026thinsp;.08, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;1.36 for contrast between E and IE). In contrast, Control group (\u003cem\u003e\u0026chi;\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e (2)\u0026thinsp;=\u0026thinsp;28.74, p\u0026thinsp;\u0026lt;\u0026thinsp;.001) responded significantly faster to faces containing both targets compared to either single target (p\u0026thinsp;\u0026lt;\u0026thinsp;.001; BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for contrast between I and IE; p\u0026thinsp;\u0026lt;\u0026thinsp;.001, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for contrast between E and IE), but the difference between the two single targets was non-significant (p\u0026thinsp;=\u0026thinsp;.87, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;1.9) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eIn the object experiment, ASD (\u003cem\u003e\u0026chi;\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e (2)\u0026thinsp;=\u0026thinsp;9.79, p\u0026thinsp;=\u0026thinsp;.007) and Control (\u003cem\u003e\u0026chi;\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e (2)\u0026thinsp;=\u0026thinsp;24.0, p\u0026thinsp;\u0026lt;\u0026thinsp;.001) groups responded faster to stimuli containing dual-targets (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e) compared to either single target display (contrast [SC-S] (ASD group: p\u0026thinsp;=\u0026thinsp;.001, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.05; Control group: p\u0026thinsp;\u0026lt;\u0026thinsp;.001, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), contrast [SC-C] (ASD group: p\u0026thinsp;\u0026lt;\u0026thinsp;.001, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.044; Control group: p\u0026thinsp;\u0026lt;\u0026thinsp;.001, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.008).\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eCapacity analysis\u003c/h3\u003e\n\u003cp\u003eIndividual capacity coefficients were computed in the face and object experiments and plotted as a function of response time divided into 10msec time bins (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e shows that face processing capacity, in fifteen out of nineteen control individuals (see Supplementary Information Table S3), ranges from a value (C(t)\u0026thinsp;=\u0026thinsp;1.5 to 4.5) at the fastest response times to somewhere just close to C(t)\u0026thinsp;=\u0026thinsp;1 for the slowest RTs. In contrast, all ASD participants showed limited capacity with curves places well below C(t)\u0026thinsp;=\u0026thinsp;1.0. Figure\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e shows that in the object experiment, the average capacity curve in ASD group was around C(t)\u0026thinsp;=\u0026thinsp;1 which is in line with the prediction of parallel processing model. For the control group, faster responses were accompanied by C(t) values above 1. To inform the reader about stability of the estimated capacity function, we demonstrated examples of individual capacity plots for ASD and Control participants (Supplementary Information, Fig. S3-S6).\u003c/p\u003e\n\u003cp\u003eNext, the overall capacity coefficients (averaged across all time bins) were calculated for each participant. Although the overall capacity value does not arbitrate between processing architectures for the ASD and Control groups, it provides our reader with an idea of whether increasing workload affects overall efficiency of the system (individual overall capacity coefficients are presented in Supplementary Information, Table S3).\u003c/p\u003e\n\u003cp\u003eTo test the effects of groups and experiments on overall capacity coefficients, a 2 (Group: ASD, Control) x 2 (Experiment: face, object) analysis of variance was conducted of aligned rank transformed overall capacity coefficients. The sums of aligned responses for main effects and F values of ANOVAs on aligned responses not of interest showed values close to zero in both experiments indicating that the data were suitable for the aligned ranks transformation. A linear mixed-effects model with Group, Experiment and the interaction between Group and Experiment as fixed effects and Subjects as a random effect was fitted to aligned ranks data. There was no significant interaction between Group and Experiment (F(1,54)\u0026thinsp;=\u0026thinsp;0.76, p\u0026thinsp;=\u0026thinsp;.38) but two main effects were significant. A main effect of Group (F(1,54)\u0026thinsp;=\u0026thinsp;75.9, p\u0026thinsp;\u0026lt;\u0026thinsp;.001) showed significantly lower overall capacity in ASD group compared to Control (W\u0026thinsp;=\u0026thinsp;10.0, p\u0026thinsp;\u0026lt;\u0026thinsp;.001, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.002). A main effect of Experiment ((F(1,54)\u0026thinsp;=\u0026thinsp;4.37, p\u0026thinsp;=\u0026thinsp;.04) indicates that overall capacity for the object experiment was higher compared to faces (V\u0026thinsp;=\u0026thinsp;35, p\u0026thinsp;=\u0026thinsp;.014), however, Bayes Factor provides moderate evidence for the hull hypothesis (BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.158).\u003c/p\u003e\n\u003cp\u003eFurther analyses to explore the main effects showed that in the Control group, the difference between overall processing capacity for faces (Mdn\u0026thinsp;=\u0026thinsp;1.298) and object attributes (Mdn\u0026thinsp;=\u0026thinsp;1.43) was non-significant (V\u0026thinsp;=\u0026thinsp;74.5, p\u0026thinsp;=\u0026thinsp;.42, r\u0026thinsp;=\u0026thinsp;0.21, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2.48). In contrast, the ASD group showed significantly lower overall processing capacity for faces (Mdn\u0026thinsp;=\u0026thinsp;0.579) compared to object attributes (Mdn\u0026thinsp;=\u0026thinsp;0.836) (V\u0026thinsp;=\u0026thinsp;35, p\u0026thinsp;=\u0026thinsp;.014, BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.166). Compared to controls, individuals with ASD exhibit significantly lower processing capacity for both object attributes (W\u0026thinsp;=\u0026thinsp;63.0; p\u0026thinsp;\u0026lt;\u0026thinsp;.001; BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.016) and face processing (W\u0026thinsp;=\u0026thinsp;3.0; p\u0026thinsp;\u0026lt;\u0026thinsp;.001 BF\u003csub\u003e01\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;0.00078) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eBootstrapped population-level distributions of the mean capacity are plotted for each group in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e, by pairing face and object capacities. Results revealed a clear separation between the two groups. Specifically, the ASD capacity values were positioned predominantly below 1 for both faces and objects, whereas the mean general capacity of the control group was above 1. In addition, the ASD group distribution was more tightly clustered, compared to the control group, indicating less heterogeneity. The complete separation between ASD and CTRL clouds highlights their distinction in general capacity ability. The bootstrapped 95% CI were: ASD-faces [0.54\u0026ndash;0.55], ASD-objects [0.89\u0026ndash;0.90], CTRL-faces [1.41\u0026ndash;1.41], ASD-objects [1.42\u0026ndash;1.54].\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study aimed to provide insights into two questions: i) whether individuals with ASD exhibit the ability to integrate information of facial identity and facial emotional expressions, and ii) whether this ability is linked to face processing only, or instead reflects a general ability to process information in an integrative manner. Using a divided attention task to assess facilitation effects when two types of information are presented, and the System Factorial Technology to estimate processing capacity of the system (Townsend \u0026amp; Eidels, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), results demonstrate differences between the ASD group and control group in the ability to integrate identity and emotional information in faces. Before discussing the findings, it is important to note that across groups, participants demonstrated high accuracy in detecting target information in both the face and object experiments. These high accuracy rates highlight that participants successfully identified the relevant stimuli, suggesting that group differences in response speed were not influenced by poor signal discrimination.\u003c/p\u003e\n\u003ch3\u003eLimited ability to integrate identity and emotion in faces in ASD\u003c/h3\u003e\n\u003cp\u003eThe current findings indicated no overall group differences in overall speed when responding in the face experiment. However, a distinct pattern emerged when examining responses to faces containing target information. Control participants responded significantly faster to faces with both targets compared to those with a single target, suggesting a processing advantage for dual-target faces. In contrast, participants with ASD did not show an advantage for dual targets. This then raises the question: what underlies these contrasting effects in response to dual versus single face targets?\u003c/p\u003e\u003cp\u003eSuch between-group differences may be related to difficulties in emotion processing commonly reported in ASD. The current findings show that participants with ASD were slower to respond to faces containing only emotional targets, compared to those containing only identity targets. This is consistent with previous research suggesting emotion processing difficulties (Uljarevic \u0026amp; Hamilton, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Yeung, \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). However, the current work goes a step further by offering a new observation: if impaired emotion processing was the main contributing factor, then responses to emotional targets alone should be significantly slower compared to responses for faces containing both emotional and identity targets. Yet, the non-parametric analyses did not support this, suggesting that emotion processing difficulties may not fully explain the absence of a dual-target advantage in ASD individuals. Still, the results should be interpreted with caution as a Bayes factor does not provide strong evidence for the null hypothesis here.\u003c/p\u003e\u003cp\u003eA second possible explanation emerges when considering the relationship between identity and emotion processing using estimates of processing efficiency. The capacity analysis in typically developing adults suggests that enhanced performance for dual targets reflects increased processing efficiency. Indeed, they showed higher capacity than predicted by a standard parallel model, which assumes independent processing of identity and emotions. Super capacity processing was evident when individual capacity coefficients were averaged across time bins (Table S3, Supplementary Information). The increase in capacity with higher task demands (i.e., dual versus single targets) suggests that target identity cues may facilitate the processing of emotional cues and vice versa, leading to an interaction between the two prior the \u0026ldquo;target present\u0026rdquo; decision is made. This finding aligns with previous studies reporting enhanced detection of one type of cue when accompanied by the other (Chen et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Fitousi, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; M. H. Johnson et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Yankouskaya et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). In contrast, the relationship between identity and emotion processing in ASD participants yielded limited processing capacity. Limited capacity indicates that processing of identity and emotions slows down as either channel is engaged. Such a slowdown may be caused by an inhibitory mechanism between identity and emotion (negative cross-talk; Wenger \u0026amp; Townsend, \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2000\u003c/span\u003e), where one type of information may actively interfere with the processing of the other type, making both slower. Alternatively, parallel processing with fixed, evenly shared capacity might occur, where both identity and emotion are processed at the same time, but the resources are split equally between them, so neither gets processed as efficiently as when processed individually. Instead of being processed in parallel, identity and emotion may be processed serially, which would also contribute to slower overall processing. Both parallel, and serial architectures predict C\u0026cong; 0.5, and, potentially, may explain capacity curves in our ASD participants (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Clarifying the precise mechanism is essential for advancing how identity and emotion are processed. Although this remains open for future research, the current findings provide evidence of restricted resources in the ASD group in face processing.\u003c/p\u003e\n\u003ch3\u003eIs the limited ability to integrate information face-specific in ASD?\u003c/h3\u003e\n\u003cp\u003eThe literature presents mixed findings, with some results supporting a face-specific processing mechanism that appears to be affected in ASD (Ewing et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Khorrami et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Weigelt et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), while other studies propose a more general processing deficit, not unique to faces (Bathelt et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Kleinhans et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). The current findings on medians of response time support the former account by demonstrating that ASD participants were faster (i) in the object experiment compared to the face experiment; and (ii) for double-target displays compared to either single-target displays in the object experiment, but not in the face experiment.\u003c/p\u003e\u003cp\u003eHowever, a closer examination of the processing architecture using a fine-grained scale of analysis revealed group differences in the object experiment. SFT is specifically designed for analysing individual participant data to uncover underlying cognitive architectures, and recent research continues to extend SFT\u0026rsquo;s application in studies with small numbers of participants (e.g., Fan et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This demonstrates that even small sample sizes can yield robust individual-level insights with the appropriate analytical approach. Here, individuals with ASD showed capacity estimates consistent with a prediction by a standard parallel model, indicating no processing advantage. In contrast, typically developing individuals showed evidence for super capacity, suggesting more efficient integration of shape and colour when processing objects. These results suggest that the ASD group process colour and shape information simultaneously, but the presence of both attributes does not benefit processing rates. In contrast, control participants showed a facilitatory interaction, where the presence of target colour enhanced the processing of target shape, and the presence of target shape enhanced the processing of target colour. Could these results be attributed to a diminished ability of the ASD group to generate an interactive type of processing? Our data suggest that we should not exclude this possibility. The ASD group had significantly lower overall capacity processing compared to control participants. This limited processing capacity could restrict the interaction between simple object attributes, and prompt reliance on a parallel type of processing. In the context of more complex information factors, such as identity and emotions, this limitation may lead to even greater difficulty, resulting in a more restricted and less efficient processing style.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eImplications of current findings\u003c/h2\u003e\u003cp\u003eThe central question across competing theories of information processing in ASD is whether characteristic patterns in task performance reflect weak integrative processing, strong local processing, or both (Happ\u0026eacute; et al., 1996; Happ\u0026eacute; \u0026amp; Frith, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Nayar et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Proff et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ventura et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). We suggest that the inconsistency in findings may arise from the loose operationalisation of experimental analyses. Our findings show that using the framework of System Factorial Technology offers a quantifiable way to assess integrative processing in the context of divided attention tasks. Specifically, the results indicate that individuals with ASD were able to detect and process identity and emotion cues when processing faces, with their accuracy close to controls. However, they did not benefit from integrating these cues \u0026ndash; pointing towards an impairment in interactive processing. Furthermore, this reduced integrative ability extended to simple, non-social stimuli as ASD participants showed limited capacity to integrate features of colour and shape when processing objects. These results support the core prediction of the weak central coherence theory. Therefore, while local processing remains intact, the ability to integrate information at a global level is impaired.\u003c/p\u003e\u003cp\u003eImplications of the current findings point to insights for understanding face processing difficulties in ASD. First, individuals with ASD showed a limited ability to integrate identity and emotion when processing faces. Could this deficit contribute to the broader difficulties in face processing observed in autism? While people vary in their ability to recognise identity and emotional cues depending on experience and familiarity (Yankouskaya, et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), all typically developing participants in the current work were more efficient when integrating these factors. This highlights the reliance on integrative processing and suggests an underlying mechanism to support this integration. Notably, supporting evidence comes from functional Magnetic Resonance Imaging (fMRI) findings showing that the interaction between identity and emotion during face processing is supported by synchronised activity between the right orbitofrontal cortex (OFC) and the inferior occipital gyrus (iOG), while also OFC activity is positively associated with processing capacity (Yankouskaya et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The whole-brain functional hypoconnectivity in autism (Moseley et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) with the reduced synchronised activity between frontal and posterior regions (Just et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Zhao et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) might explain difficulties in the integration of identity and emotion in individuals with autism. Recently, Kleinhans et al. (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) showed that even during resting state fMRI, functional connectivity of the fusiform gyrus was altered in individuals with ASD, compared to neurotypically developing adults, and these differences predicted later memory performance for face identity. This suggests that intrinsic neural organisation related to identity processing is already altered in ASD. In this study, the lack of integrative processes in the ASD group \u0026ndash; despite similar accuracy performance with the control group \u0026ndash; points towards a disruption in integration between these facial dimensions. What remains to be investigated is which specific mechanism explains this difficulty.\u003c/p\u003e\u003cp\u003eSecond, the present work demonstrates how advances in human information processing developed in a System Factorial Technology (Johnson et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Townsend \u0026amp; Eidels, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Townsend et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) framework can be applied to quantify the efficiency with which ASD participants (and others) process multiple signals when presented simultaneously. The capacity coefficient measures how efficiently a system speeds up across different stages of processing in a given task. Our individual and group level analyses showed that the capacity coefficients yield additional information than mean RT about system performance. Although initial analyses suggested comparable performance, the ASD group showed reduced integration efficiency, highlighting the limitations of solely relying on standard analyses and the value of relevant approaches when examining underlying cognitive mechanisms. Importantly, while it is possible that the average pattern of responses observed at the group level in ASD will also be observed at the individual level, this cannot be assumed without testing at the individual level (widely acknowledged but overlooked issue in face processing research). Studies frequently report large variability between individuals in the degree of these deficits, with subtle impairments and mixed features (see for discussion Ozonoff et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Therefore, group inferences can provide more meaningful information by assessing the dynamics of within-person processes as was demonstrated in our study (see also study by Johnson et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The System Factorial Technology offers \u0026lsquo;ready to use\u0026rsquo; tools for assessing individual characteristics of information processing which can be implemented in clinical settings for evaluating face processing deficits in ASD.\u003c/p\u003e\u003cp\u003eFinally, the current findings also raise important implications for learning contexts and educational practices. The limited capacity to simultaneously process and integrate, both social and non-social information may hinder how individuals with ASD engage with instructional environments. Effective learning often depends on integrating multimodal inputs \u0026ndash; such as facial expressions, gestures, tone of voice, visual materials \u0026ndash; which guide attention and support comprehension. Integrative processing deficits suggest that individuals with ASD may struggle to fully benefit from such environments. Furthermore, because the difficulties extend beyond social cues to non-social features, like shape and colour, these integrative challenges may affect how learners with ASD process complex information across academic domains, where simultaneous integration of multiple information is often required. Therefore, one future direction is to investigate how autistic learners perform when information is presented sequentially. This could inform the development of assistive technologies, aiming to reduce integrative load by presenting elements in sequence (e.g., flashcards showing words and images separately, rather than combining the two elements). Such adjustments could be implemented via personalised learning, aligning with the cognitive processing profiles of autistic learners.\u003c/p\u003e\u003cp\u003eTo conclude, Fifić and Townsend (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) suggested that an interactive dependency between features is the critical concept for modelling face perception. Our findings demonstrate that the ability to integrate personal and emotional information in faces is impaired in people with ASD and this, possibly, reflects a limited general ability of interactive processing. While supporting the main idea of the weak coherence hypothesis, our study, however, illustrates the need for further research into cognitive and neural mechanisms underlying the impairments in capacity processing. Such insights will underscore the importance of aligning educational approaches with cognitive processing profiles, to better support learning outcomes.\u003c/p\u003e\u003c/div\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlguacil S, Madrid E, Esp\u0026iacute;n AM, Ruz M (2017) Facial identity and emotional expression as predictors during economic decisions. Cogn Affect Behav Neurosci 17:315\u0026ndash;329. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13415-016-0481-9\u003c/span\u003e\u003cspan address=\"10.3758/s13415-016-0481-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmerican Psychiatric Association (2013) \u003cem\u003eDiagnostic and statistical manual of mental disorders\u003c/em\u003e (5th ed.). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1176/appi.books.9780890425596\u003c/span\u003e\u003cspan address=\"10.1176/appi.books.9780890425596\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBach DR, Schmidt-Daffy M, Dolan RJ (2014) Facial expression influences face identity recognition during the attentional blink. Emotion 14(6):1007\u0026ndash;1013. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/a0037945\u003c/span\u003e\u003cspan address=\"10.1037/a0037945\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBaron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E (2001) The Autism Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism. Malesand Females, Scientists and Mathematicians\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBathelt J, Koolschijn PCM, Geurts HM (2022) Atypically slow processing of faces and non-faces in older autistic adults. Autism 26(7):1737\u0026ndash;1751. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/13623613211065297\u003c/span\u003e\u003cspan address=\"10.1177/13623613211065297\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBlair R (2002) Fractionation of visual memory: Agency detection and its impairment in autism. Neuropsychologia 40(1):108\u0026ndash;118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0028-\u003c/span\u003e\u003cspan address=\"10.1016/S0028-\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e3932(01)00069\u0026thinsp;\u0026ndash;\u0026thinsp;0\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBlakesley RE, Mazumdar S, Dew MA, Houck PR, Tang G, Reynolds CF, Butters MA (2009) Comparisons of methods for multiple hypothesis testing in neuropsychological research. Neuropsychology 23(2):255\u0026ndash;264. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/a0012850\u003c/span\u003e\u003cspan address=\"10.1037/a0012850\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBruce V, Young A (1986) Understanding face recognition. Br J Psychol 77(3):305\u0026ndash;327. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.2044-8295.1986.tb02199.x\u003c/span\u003e\u003cspan address=\"10.1111/j.2044-8295.1986.tb02199.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen W, Liu CH, Li H, Tong K, Ren N, Fu X (2015) Facial expression at retrieval affects recognition of facial identity. Front Psychol. 6\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2015.00780\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2015.00780\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD\u0026rsquo;Argembeau A, Van Der Linden M (2011) Influence of facial expression on memory for facial identity: Effects of visual features or emotional meaning? Emotion 11(1):199\u0026ndash;202. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/a0022592\u003c/span\u003e\u003cspan address=\"10.1037/a0022592\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDobs K, Schultz J, B\u0026uuml;lthoff I, Gardner JL (2018) Task-dependent enhancement of facial expression and identity representations in human cortex. NeuroImage 172:689\u0026ndash;702. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroimage.2018.02.013\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2018.02.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEfron B, Tibshirani RJ (1994) An introduction to the bootstrap, 1st edn. Chapman and Hall/CRC. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1201/9780429246593\u003c/span\u003e\u003cspan address=\"10.1201/9780429246593\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEkman P, O'Sullivan M (1988) The role of context in interpreting facial expression: comment on Russell and Fehr (1987). \u003cem\u003eJournal of Experimental Psychology: General, 117\u003c/em\u003e(1), 86\u0026ndash;90\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEwing L, Pellicano E, Rhodes G (2013) Reevaluating the selectivity of face-processing difficulties in children and adolescents with autism. J Exp Child Psychol 115(2):342\u0026ndash;355. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jecp.2013.01.009\u003c/span\u003e\u003cspan address=\"10.1016/j.jecp.2013.01.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFan G, Collins AJ, Thomas RD (2025) Visual Processing Within and Across Hemispheres: A Systems Factorial Technology Analysis. Comput Brain Behav. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s42113-025-00249-5\u003c/span\u003e\u003cspan address=\"10.1007/s42113-025-00249-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFarah MJ, Levinson KL, Klein KL (1995) Face perception and within-category discrimination in prosopagnosia. Neuropsychologia 33(6):661\u0026ndash;674. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0028-3932(95)00002-K\u003c/span\u003e\u003cspan address=\"10.1016/0028-3932(95)00002-K\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFifić M, Townsend JT (2010) Information-processing alternatives to holistic perception: Identifying the mechanisms of secondary-level holism within a categorization paradigm. J Experimental Psychology: Learn Memory Cognition 36(5):1290\u0026ndash;1313. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/a0020123\u003c/span\u003e\u003cspan address=\"10.1037/a0020123\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFisher K, Towler J, Eimer M (2016) Facial identity and facial expression are initially integrated at visual perceptual stages of face processing. Neuropsychologia 80:115\u0026ndash;125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuropsychologia.2015.11.011\u003c/span\u003e\u003cspan address=\"10.1016/j.neuropsychologia.2015.11.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFitousi D (2017) What\u0026rsquo;s in a face file? Feature binding with facial identity, emotion, and gaze direction. Psychol Res 81(4):777\u0026ndash;794. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00426-016-0783-0\u003c/span\u003e\u003cspan address=\"10.1007/s00426-016-0783-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFox CJ, Barton JJS (2007) What is adapted in face adaptation? The neural representations of expression in the human visual system. Brain Res 1127:80\u0026ndash;89. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.brainres.2006.09.104\u003c/span\u003e\u003cspan address=\"10.1016/j.brainres.2006.09.104\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGanel T, Goshen-Gottstein Y (2004) Effects of Familiarity on the Perceptual Integrality of the Identity and Expression of Faces: The Parallel-Route Hypothesis Revisited. J Exp Psychol Hum Percept Perform 30(3):583\u0026ndash;597\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGehdu BK, Press C, Gray KLH, Cook R (2024) Autistic adults have insight into their relative face recognition ability. Sci Rep 14(1):17802. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-024-67649-8\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-67649-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGreimel E, Schulte-R\u0026uuml;ther M, Kamp-Becker I, Remschmidt H, Herpertz-Dahlmann B, Konrad K (2014) Impairment in face processing in autism spectrum disorder: A developmental perspective. J Neural Transm 121(9):1171\u0026ndash;1181. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00702-014-1206-2\u003c/span\u003e\u003cspan address=\"10.1007/s00702-014-1206-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHapp\u0026eacute; FGE (1996) Studying weak central coherence at low levels: Children with autism do not succumb to visual illusions: A research note. Child Psychol Psychiatry Allied Disciplines 37(7):873\u0026ndash;877. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1469 7610.1996.tb01483.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1469 7610.1996.tb01483.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHapp\u0026eacute; F, Frith U (2006) The Weak Coherence Account: Detail-focused Cognitive Style in Autism Spectrum Disorders. J Autism Dev Disord 36(1):5\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10803-005-0039-0\u003c/span\u003e\u003cspan address=\"10.1007/s10803-005-0039-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaxby JV, Hoffman EA, Gobbini MI (2000) The distributed human neural system for face perception. Trends Cogn Sci 4(6):223\u0026ndash;233. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S1364-6613(00)01482-0\u003c/span\u003e\u003cspan address=\"10.1016/S1364-6613(00)01482-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eH\u0026oslash;yland AL, N\u0026aelig;rland T, Engstr\u0026oslash;m M, Lydersen S, Andreassen OA (2017) The relation between face-emotion recognition and social function in adolescents with autism spectrum disorders: A case control study. PLoS ONE 12(10):e0186124. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0186124\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0186124\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJemel B, Mottron L, Dawson M (2006) Impaired Face Processing in Autism: Fact or Artifact? J Autism Dev Disord 36(1):91\u0026ndash;106. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10803-005-0050-5\u003c/span\u003e\u003cspan address=\"10.1007/s10803-005-0050-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJohnson MH, Senju A, Tomalski P (2015) Neurosci Biobehavioral Reviews 50:169\u0026ndash;179. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neubiorev.2014.10.009\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2014.10.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The two-process theory of face processing: Modifications based on two decades of data from infants and adults\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJohnson SA, Blaha LM, Houpt JW, Townsend JT (2010) Systems Factorial Technology provides new insights on global\u0026ndash;local information processing in autism spectrum disorders. J Math Psychol 54(1):53\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmp.2009.06.006\u003c/span\u003e\u003cspan address=\"10.1016/j.jmp.2009.06.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJust MA, Keller TA, Malave VL, Kana RK, Varma S (2012) Autism as a neural systems disorder: A theory of frontal-posterior underconnectivity. Neurosci Biobehavioral Reviews 36(4):1292\u0026ndash;1313. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neubiorev.2012.02.007\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2012.02.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKay M, Wobbrock J (2016) \u003cem\u003eARTool: Aligned Rank Transform for Nonparametric Factorial ANOVAs\u003c/em\u003e. R package version 0.10.4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/mjskay/ARTool\u003c/span\u003e\u003cspan address=\"https://github.com/mjskay/ARTool\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhorrami A, Tehrani-Doost M, Esteky H (2013) Comparison between Face and Object Processing in Youths with Autism Spectrum Disorder. An event related potentials study\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKleinhans NM, Richards T, Sterling L, Stegbauer KC, Mahurin R, Johnson LC, Greenson J, Dawson G, Aylward E (2008) Abnormal functional connectivity in autism spectrum disorders during face processing. Brain 131(4):1000\u0026ndash;1012. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/brain/awm334\u003c/span\u003e\u003cspan address=\"10.1093/brain/awm334\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKleinhans N, Larsen SF, Estes A, Aylward E (2025) Intrinsic Functional Connectivity Alterations of the Fusiform Face Area in Autism Spectrum Disorder. NeuroSci 6(2):29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/neurosci6020029\u003c/span\u003e\u003cspan address=\"10.3390/neurosci6020029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi Y, Zhang M, Liu S, Luo W (2022) EEG decoding of multidimensional information from emotional faces. NeuroImage 258:119374. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroimage.2022.119374\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2022.119374\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMansouri H, Paige RL, Surles JG (2004) Aligned Rank Transform Techniques for Analysis of Variance and Multiple Comparisons. Commun Stat - Theory Methods 33(9):2217\u0026ndash;2232. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1081/STA-200026599\u003c/span\u003e\u003cspan address=\"10.1081/STA-200026599\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMarascuilo LA, Levin JR (1970) Appropriate posthoc comparisons for interaction and nested hypotheses in analysis of variance designs: The elimination of Type IV errors. Am Educ Res J 7:397\u0026ndash;421\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiller J (1986) Timecourse of coactivation in bimodal divided attention. Percept Psychophys 40(5):331\u0026ndash;343\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoseley RL, Ypma RJF, Holt RJ, Floris D, Chura LR, Spencer MD, Baron- Cohen S, Suckling J, Bullmore E, Rubinov M (2015) Whole-brain functional hypoconnectivity as an endophenotype of autism in adolescents. NeuroImage: Clin 9:140\u0026ndash;152. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.nicl.2015.07.015\u003c/span\u003e\u003cspan address=\"10.1016/j.nicl.2015.07.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNayar K, Voyles AC, Kiorpes L, Di Martino A (2017) Global and local visual processing in autism: An objective assessment approach. Autism Res 10(8):1392\u0026ndash;1404. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/aur.1782\u003c/span\u003e\u003cspan address=\"10.1002/aur.1782\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNeufeld RWJ, Townsend JT, Jett\u0026eacute; J (2007) Quantitative Response Time Technology for Measuring Cognitive-Processing Capacity in Clinical Studies. In: Neufeld RWJ (ed) Advances in clinical cognitive science: Formal modeling of processes and symptoms. American Psychological Association, pp 207\u0026ndash;238. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/11556-007\u003c/span\u003e\u003cspan address=\"10.1037/11556-007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNomi JS, Uddin LQ (2015) Face processing in autism spectrum disorders: From brain regions to brain networks. Neuropsychologia 71:201\u0026ndash;216. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuropsychologia.2015.03.029\u003c/span\u003e\u003cspan address=\"10.1016/j.neuropsychologia.2015.03.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOzonoff S, Heung K, Byrd R, Hansen R, Hertz-Picciotto I (2008) The onset of autism: Patterns of symptom emergence in the first years of life. Autism Res 1(6):320\u0026ndash;328. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/aur.53\u003c/span\u003e\u003cspan address=\"10.1002/aur.53\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eProff I, Williams GL, Quadt L, Garfinkel SN (2022) Sensory processing in autism across exteroceptive and interoceptive domains. Psychol Neurosci 15(2):105\u0026ndash;130. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/pne0000262\u003c/span\u003e\u003cspan address=\"10.1037/pne0000262\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRedfern AS, Benton CP (2017) Expression Dependence in the Perception of Facial Identity. I-Perception 8(3):2041669517710663. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/2041669517710663\u003c/span\u003e\u003cspan address=\"10.1177/2041669517710663\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchiltz C, Sorger B, Caldara R, Ahmed F, Mayer E, Goebel R, Rossion B (2006) Impaired Face Discrimination in Acquired Prosopagnosia Is Associated with Abnormal Response to Individual Faces in the Right Middle Fusiform Gyrus. Cereb Cortex 16(4):574\u0026ndash;586. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cercor/bhj005\u003c/span\u003e\u003cspan address=\"10.1093/cercor/bhj005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchultz RT (2005) Developmental deficits in social perception in autism: The role of the amygdala and fusiform face area. Int J Dev Neurosci 23(2\u0026ndash;3):125\u0026ndash;141. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ijdevneu.2004.12.012\u003c/span\u003e\u003cspan address=\"10.1016/j.ijdevneu.2004.12.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eScofield DG (2015) nestedRanksTest: Mann-Whitney-Wilcoxon Test for Nested Ranks. R package version 0.2\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStantic M, Brewer R, Duchaine B, Banissy MJ, Bate S, Susilo T, Catmur C, Bird G (2021) The Oxford Face Matching Test: A non-biased test of the full range of individual differences in face perception. Behav Res Methods 54(1):158\u0026ndash;173. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13428-021-01609-2\u003c/span\u003e\u003cspan address=\"10.3758/s13428-021-01609-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTaylor GJ, Bagby RM (2000) An overview of the alexithymia construct. In: Bar-On R, Parker JDA (eds) The handbook of emotional intelligence: Theory, development, assessment, and application at home, school, and in the workplace. Jossey Bass/Wiley, pp 40\u0026ndash;67\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTottenham N, Tanaka JW, Leon AC, McCarry T, Nurse M, Hare TA, Marcus DJ, Westerlund A, Casey B, Nelson C (2009) The NimStim set of facial expressions: Judgments from untrained research participants. Psychiatry Res 168(3):242\u0026ndash;249. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.psychres.2008.05.006\u003c/span\u003e\u003cspan address=\"10.1016/j.psychres.2008.05.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTowler JR, Jackson MC, Tree JJ (2025) Impairments for faces but not for abstract shapes in developmental prosopagnosia: Evidence from visual working memory tasks. Cognit Neuropsychol 1\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/02643294.2025.2498154\u003c/span\u003e\u003cspan address=\"10.1080/02643294.2025.2498154\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTownsend JT, Eidels A (2011) Workload capacity spaces: A unified methodology for response time measures of efficiency as workload is varied. Psychon Bull Rev 18(4):659\u0026ndash;681. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13423-011-0106-9\u003c/span\u003e\u003cspan address=\"10.3758/s13423-011-0106-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTownsend JT, Fific M, Neufeld RWJ (2007) Assessment of mental architecture in clinical/cognitive research. In: Treat RRBTA, Baker TB (eds) Psychological clinical science: papers in honor of richard m. mcfall. Psychology, New York, pp 223\u0026ndash;258\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTownsend JT, Wenger MJ (2004) A Theory of Interactive Parallel Processing: New Capacity Measures and Predictions for a Response Time Inequality Series. Psychol Rev 111(4):1003\u0026ndash;1035. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/0033 295X.111.4.1003\u003c/span\u003e\u003cspan address=\"10.1037/0033 295X.111.4.1003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUljarevic M, Hamilton A (2013) Recognition of Emotions in Autism: A Formal Meta- Analysis. J Autism Dev Disord 43(7):1517\u0026ndash;1526. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10803-012-1695-5\u003c/span\u003e\u003cspan address=\"10.1007/s10803-012-1695-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVan Der Velde J, Van Tol M-J, Goerlich-Dobre KS, Gromann PM, Swart M, De Haan L, Wiersma D, Bruggeman R, Krabbendam L, Aleman A (2014) Dissociable morphometric profiles of the affective and cognitive dimensions of alexithymia. Cortex 54:190\u0026ndash;199. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cortex.2014.02.017\u003c/span\u003e\u003cspan address=\"10.1016/j.cortex.2014.02.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVentura M, Innamorato F, Palmisano A, Cicinelli G, Nobile E, Manippa V, Keller R, Rivolta D (2023) Investigating the impact of disposable surgical face-masks on face identity and emotion recognition in adults with autism spectrum disorder. Autism Res 16(5):1063\u0026ndash;1077. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/aur.2922\u003c/span\u003e\u003cspan address=\"10.1002/aur.2922\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVrancken L, Vermeulen E, Germeys F, Verfaillie K, Attention (2019) Percept Psychophysics, 81(1), 217\u0026ndash;236. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13414-018-1603-y\u003c/span\u003e\u003cspan address=\"10.3758/s13414-018-1603-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWeigelt S, Koldewyn K, Kanwisher N (2012) Face identity recognition in autism spectrum disorders: A review of behavioral studies. Neurosci Biobehavioral Reviews 36(3):1060\u0026ndash;1084. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neubiorev.2011.12.008\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2011.12.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWeigelt S, Koldewyn K, Kanwisher N (2013) Face Recognition Deficits in Autism Spectrum Disorders Are Both Domain Specific and Process Specific. PLoS ONE 8(9):e74541. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0074541\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0074541\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWenger MJ, Townsend JT (2000) Basic Response Time Tools for Studying General Processing Capacity in Attention, Perception, and Cognition. J Gen Psychol 127(1):67\u0026ndash;99. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00221300009598571\u003c/span\u003e\u003cspan address=\"10.1080/00221300009598571\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWechsler D (1997) \u003cem\u003eWechsler Adult Intelligence Scale\u0026ndash;Third Edition (WAIS-III)\u003c/em\u003e [Database record]. APA PsycTests. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/t49755-000\u003c/span\u003e\u003cspan address=\"10.1037/t49755-000\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWobbrock JO, Findlater L, Gergle D, Higgins JJ (2011) The aligned rank transform for nonparametric factorial analyses using only anova procedures. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 143\u0026ndash;146. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/1978942.1978963\u003c/span\u003e\u003cspan address=\"10.1145/1978942.1978963\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYankouskaya A, Booth DA, Humphreys G, Attention (2012) Percept Psychophysics, 74(8), 1692\u0026ndash;1711. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3758/s13414-012-0345-5\u003c/span\u003e\u003cspan address=\"10.3758/s13414-012-0345-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYankouskaya A, Humphreys GW, Rotshtein P (2014) The processing of facial identity and expression is interactive, but dependent on task and experience. Front Hum Neurosci 8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fnhum.2014.00920\u003c/span\u003e\u003cspan address=\"10.3389/fnhum.2014.00920\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYankouskaya A, Rotshtein P, Humphreys GW (2014) Interactions between Identity and Emotional Expression in Face Processing across the Lifespan: Evidence from Redundancy Gains. Journal of Aging Research, 2014, 1\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1155/2014/136073\u003c/span\u003e\u003cspan address=\"10.1155/2014/136073\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYankouskaya A, Stolte M, Moradi Z, Rotshtein P, Humphreys G (2017) Integration of identity and emotion information in faces: fMRI evidence. Brain Cogn 116:29\u0026ndash;39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.bandc.2017.05.004\u003c/span\u003e\u003cspan address=\"10.1016/j.bandc.2017.05.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYeung MK (2022) A systematic review and meta-analysis of facial emotion recognition in autism spectrum disorder: The specificity of deficits and the role of task characteristics. Neuroscience and Biobehavioral Reviews\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYuan Y, Johnson VE (2008) Bayesian hypothesis tests using nonparametric statistics. Statistica Sinica 18:1185\u0026ndash;1200\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao H-C, Lv R, Zhang G-Y, He L-M, Cai X-T, Sun Q, Yan C-Y, Bao X-Y, Lv X-Y, Fu B (2022) Alterations of Prefrontal-Posterior Information Processing Patterns in Autism Spectrum Disorders. Front NeuroSci 15:768219. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fnins.2021.768219\u003c/span\u003e\u003cspan address=\"10.3389/fnins.2021.768219\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Bournemouth University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"ASD, capacity processing, identity, emotional expression, integration","lastPublishedDoi":"10.21203/rs.3.rs-7612091/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7612091/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe ability to combine identity and emotional information when viewing faces allows typically developing individuals to process facial information efficiently. The purpose of this study was to examine whether adults with autism spectrum disorder (ASD) share this ability, and test if it reflects a broader pattern of integrative information processing. Nineteen adults with ASD and nineteen typically developing adults completed a divided attention task where they had to indicate when a target was present. Each participant completed two separate experiments: i) the face experiment, which required detecting targets based on identity and emotional expression; and ii) the object experiment, which required detecting targets based on colour and shape. Analyses assessed whether, and how, participants responded more efficiently in the presence of both targets, compared to either single target. By employing mathematical modelling tools developed in the System Factorial Technology (SFT) framework, results showed that ASD individuals process faces at the rate of control individuals, however, failed to exhibit integrative properties. Contrary to controls, adults with ASD showed limited capacity to engage in integrative processing, even for object attributes, suggesting a domain-general integrative processing difficulty. Implications of these findings generalise to personalisation practices in education and can be used to inform the development of assistive technologies in ASD.\u003c/p\u003e","manuscriptTitle":"Integrating identity and emotion information in faces in ASD – capacity limits of information processing","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-17 09:12:17","doi":"10.21203/rs.3.rs-7612091/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f51f6a5e-5694-4198-ae8e-c8fceb3a392a","owner":[],"postedDate":"September 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":54684096,"name":"Psychology"}],"tags":[],"updatedAt":"2025-09-17T09:12:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-17 09:12:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7612091","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7612091","identity":"rs-7612091","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}