Edited by: Laura Stoppelbein, University of Alabama, United States
Reviewed by: Geraldine Kechid, Centre Hospitalier Regional et Universitaire de Lille, France; Eric Bizet, Université Louis-Pasteur, France
This article was submitted to Child and Adolescent Psychiatry, a section of the journal Frontiers in Psychiatry
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social functioning and restricted, repetitive patterns of behavior, interests, or activities (
Individuals with ASD experience daily challenges with social function, which frequently impact negatively their relationships and access to education or employment. Typically, children and adults with ASD encounter difficulties to understand other's emotional states or intentions. This may lead to inappropriate behaviors in social situations. Thus, despite good intellectual abilities, high functioning adults with ASD may ask inappropriate questions, act oddly and are therefore vulnerable to bullying and isolation. These difficulties have been linked to well-documented deficits in social cognition in ASD (
Research on social cognition in ASD has essentially focused on emotional processing (i.e., identification of emotions from different cues, including facial expression) and theory of mind (i.e., ToM; ability to mentalize other's mental states which allow us to make inferences on their intentions, beliefs, and knowledge), showing alterations in both processes (
An influential theory considering social dysfunction in ASD puts forward an alteration of ToM [Mindblindness theory, (
Moreover, social cognitive deficits in ASD could be a consequence of an atypical sensory processing. Indeed, a great number of studies using eye-tracking have shown different visual exploration in this population, with fewer gaze directed toward social comparatively to non-social elements [(
The alterations in the perception and understanding of social situations may lead to misinterpretations and in some adults with ASD to attribution biases. Research in this domain is scarce. Nevertheless, the few studies conducted showed no self-blaming pattern but rather healthy and normal self-serving attributions (
Most tools developed for neuropsychological evaluation of Emotional Processing present static photographs of highly expressive emotional expressions (e.g., Ekman Face Test) and often fail to show any impairment in ASD adults without intellectual disability (
Concerning Theory of Mind, the most classical tasks (i.e., false belief tasks; strange stories tasks…) do not sufficiently target subtle impairments encountered by adults with ASD without intellectual disability and especially in those with good verbal skills (
To our knowledge there are very few standardized evaluation tools allowing the assessment of social perception deficits in ASD, and none of these have been standardized in French language. For instance, the Social Perception subtest from the Advanced Clinical Solution has been used to reveal such deficits in adolescents and adults with ASD (
Similarly, only few standardized questionnaires have been developed in order to evaluate attribution biases (
Deficits in social cognition are observed in several psychiatric conditions other than ASD (schizophrenia, ADHD, bipolar disorder, anorexia…), and have been linked to functional outcome [e.g., (
Only few validated and standardized tools are available in French language for the clinical assessment of social cognition in adults (
The aim of the present study was to examine the relevance of the “ClaCoS” battery for the assessment of social cognition in adults with ASD without intellectual disability compared to control typical developing subjects. Based on the existing literature, we expected the ASD group to be less efficient than the controls on the neuropsychological evaluation of all dimensions of social cognition, and in particular emotional perception, social perception and theory of mind. Concerning attributional style, this hypothesis could be more uncertain, as previous studies showed no difference between ASD and controls (
The study was carried out in accordance with the Declaration of Helsinki and approved by the local Ethics Committee (CPP Lyon-Sud Est IV, no. 15/041; ANSM, no. 2015-A00580-49). Written informed consent to take part in the study was obtained from all participants. The control subjects were paid 30 euros for their participation.
All participants were enrolled in a multisite study assessing social cognition in adults with autism and schizophrenia with the “ClaCoS” Battery. This study was conducted in three sites in France: Child Psychiatry Department specialized in autism, University Hospital of Tours, in Tours; Hospital Le Vinatier in Lyon and Groupe Hospitalier Universitaire Paris, Psychiatry and Neuroscience in Paris.
Participants with ASD without intellectual disability (
Typical developing adults (
Exclusion criteria for both groups included: [1] the presence or history of neurological disorders affecting brain function, [2] the presence of severe visual or hearing impairments interfering with assessment, [3] the absence of French language proficiency or important reading difficulties and [4] an abuse of substance in the past month (tobacco excluded).
The control participants were selected from a larger dataset (
Participants characteristics.
Age (years) | 45 | 27.7 (7.9) | 27.6 (7.8) | 0.968 |
Gender (F:M) | 45 | 10:35 | 9:36 | 0.796 |
Education (years) | 45 | 13.0 (2.4) | 13.3 (1.7) | 0.693 |
ADOS-2 (social interactions +communication) | 22 | 12.6 (4.6) | – | NA |
ADI (social interactions) | 42 | 18.8 (7.9) | – | NA |
ADI (verbal communication) | 42 | 13.1 (5.3) | – | NA |
Verbal IQ | 45 | 117 ( |
– | NA |
Performance IQ | 45 | 103 ( |
– | NA |
40 | 24.7 (10.5) | 39.1 (10.4) | <0.001 | |
Time (ms) | 44 | 9.8 (3.9) | 7.9 (2.5) | 0.004 |
Number of errors | 44 | 1.0 (1.8) | 0.6 (1.1) | 0.691 |
Time (ms) | 44 | 101.9 (47.2) | 78.5 (24.8) | 0.005 |
Number of errors | 44 | 0.5 (0.9) | 0.2 (0.5) | 0.224 |
All participants were tested in a silent room by an experienced neuropsychologist and placed, for computerized tests, at 23 inches from a 15-inch computer screen. They underwent a full assessment with the “ClaCoS” battery, developed by the multicentric Research group in psychiatry GDR3557 [for a more detailed presentation of each test, see (
Tests constituting the “ClaCoS” battery.
The ACSo (
The TREF (
The PerSo measures the perception of social situations, using pictures taken from the material “ColorCards—Social behavior.” Participants completed 3 successive tasks. First, they were asked to list all the elements perceived in the picture, providing a global “fluency score.” Then, they were instructed to explain the social situation freely, leading to a “non-indexed interpretation score.” Indexed questions were then proposed if some of the expected elements were missing (main character; location; interactions), resulting in an “indexed interpretation score.” A “total interpretation score” was then computed by adding the “non-indexed” and “indexed” interpretation scores. Finally, the participants were asked to extract a social rule that could be related to the card, producing a “social knowledge score.”
The MASC test [(
AIHQ-S measures attributional biases from ambiguous social situations [AIHQ-S, Adapted From Combs et al. (
In addition, empathy was assessed using the Empathy Quotient questionnaire [EQ, (
Statistical analyses were performed using STATISTICA v13.3 software (TIBICO®). For a better uniformity across the different analyses, non-parametric statistics were chosen, as there was a violation of the assumption of homogeneity of variance on some measures, according to the Levene's test. The Mann-Whitney non parametric
The ASD and control groups did not differ in terms of age, gender or education level (
The total score as well as all four sub-scores (emotional perception, social perception and knowledge, theory of mind and attributional style) were higher in adults with ASD compared to controls (all
Scores from the “ClaCoS” battery in adults with ASD and controls.
Total score | 24.24 (8.0) | 9.6 (5.1) | 158.5 | <0.001 | <0.001 |
Emotional perception | 3.5 (1.7) | 1.3 (1.2) | 293.0 | <0.001 | <0.001 |
Social perception and knowledge | 6.4 (2.7) | 2.5 (1.9) | 262.0 | <0.001 | <0.001 |
Theory of mind | 6.8 (2.4) | 2.8 (1.7) | 187.5 | <0.001 | <0.001 |
Attributional style | 4.9 (2.8) | 1.6 (1.4) | 295.5 | <0.001 | <0.001 |
% Of correct recognition | 65.7 (10.2) | 70.6 (6.4) | 699.5 | 0.011 | 0.077 |
Happiness | 87.2 (11.5) | 91.1 (8.7) | 925.0 | 0.468 | 3.276 |
Anger | 61.1 (27.4) | 68.1 (18.9) | 909.5 | 0.406 | 2.842 |
Sadness | 69.1 (19.6) | 72.1 (19.2) | 969.0 | 0.727 | 5.089 |
Fear | 78.7 (15.7) | 83.7 (13.5) | 888.5 | 0.311 | 2.177 |
Disgust | 57.7 (14.9) | 62.2 (12.2) | 933.5 | 0.522 | 3.654 |
Contempt | 40.3 (21.9) | 46.4 (17.9) | 788.5 | 0.070 | 0.490 |
Recognition threshold | 53.3 (9.9) | 48.3 (6.6) | 662.0 | 0.005 | 0.035 |
Happiness | 32.3 (11.6) | 29.4 (10.0) | 886.5 | 0.291 | 2.037 |
Anger | 58.6 (25.0) | 50.2 (17.9) | 800.5 | 0.086 | 0.602 |
Sadness | 53.0 (18.9) | 48.8 (16.8) | 878.5 | 0.278 | 1.946 |
Fear | 43.1 (16.0) | 36.9 (14.6) | 764.0 | 0.045 | 0.315 |
Disgust | 58.1 (12.8) | 55.0 (11.5) | 858.5 | 0.206 | 1.442 |
Contempt | 72.4 (17.8) | 69.8 (13.6) | 888.0 | 0.313 | 2.191 |
Fluency score | 80.4 (36.9) | 106.0 (31.7) | 595.0 | <0.001 | 0.004 |
Interpretation (total score) | 19.2 (3.2) | 20.9 (2.5) | 703.0 | 0.012 | 0.060 |
Non-indexed interpretation | 9.0 (2.0) | 9.9 (1.7) | 774.0 | 0.052 | 0.260 |
Indexed interpretation | 10.2 (1.4) | 11.0 (1.1) | 660.5 | 0.003 | 0.015 |
Social knowledge score | 4.8 (2.0) | 5.6 (1.8) | 770.5 | 0.048 | 0.240 |
Total score | 26.3 (5.4) | 31.8 (3.6) | 507.5 | <0.001 | <0.001 |
Excessive ToM | 7.2 (2.8) | 5.1 (2.5) | 688.0 | 0.008 | 0.032 |
Less ToM | 7.7 (3.4) | 6.1 (3.4) | 808.5 | 0.098 | 0.392 |
No ToM | 3.7 (2.1) | 2.0 (1.7) | 683.0 | 0.007 | 0.028 |
Hostility bias | 1.9 (0.7) | 1.7 (0.6) | 864.5 | 0.230 | 1.382 |
Attribution of responsibility score | 2.8 (0.7) | 2.4 (0.6) | 694.0 | 0.010 | 0.062 |
Intentionality score-IS | 3.1 (1.0) | 2.6 (0.7) | 659.5 | 0.004 | 0.026 |
Anger score- AS | 2.3 (0.8) | 2.1 (0.6) | 811.5 | 0.104 | 0.626 |
Blame score- BS | 2.8 (0.9) | 2.5 (0.7) | 771.0 | 0.051 | 0.307 |
Agression bias-AB | 1.4 (0.4) | 1.6 (0.4) | 658.5 | 0.004 | 0.023 |
Adults with ASD required a higher threshold in order to correctly recognize emotional facial expressions from photographs (adjusted
Adults with ASD were less efficient than control participants on the assessments of both the fluency and the interpretation of the social situation (
The total score (correct “ToM” answers) was higher in controls than in adults with ASD (adjusted
No difference was found between adults with ASD and controls for the hostility bias (
All differences previously reported between the ASD and controls remained significant after controlling for visuospatial discrimination and attentional abilities (all F > 6.20;
Correlations were performed between the scores of each dimension of the ACSo and the results on the test designed to assess social cognitive impairments on the corresponding dimension of social cognition. This allowed us to examine the relationship between the participant's subjective complaints in specific domains of social cognition and the actual impairments measured through the neuropsychological assessment (
Main correlations between subjective complaints and social cognition assessments for each of the four domains. Results of Spearman correlations are reported (statistical values r and adjusted p after Bonferonni correction for multiple analysis).
Subjective complaint concerning emotional perception (ACSo) was positively correlated with the emotion recognition threshold (TREF threshold;
Subjective complaint concerning social perception and knowledge (ACSo) was negatively correlated with the interpretation score (PerSo indexed interpretation score:
Subjective complaint concerning theory of mind (ACSo) was negatively correlated with the total score (MASC;
Subjective complaint concerning attributional style (ACSo) was positively correlated with the intentionality score (
The main goal of our study was to evaluate the relevance of the “ClaCoS” battery for the assessment of social cognition impairments in adults with ASD without intellectual disability. We further examined the links between subjective complaints and objectively measured impairments on the different components of social cognition. Overall, adults with ASD reported greater subjective complaints than controls in each of the four areas explored by the ACSo. They also showed deficits on all of the neuropsychological tests from the “ClaCoS” battery, which explored the same four main components of social cognition: emotional perception, social perception and knowledge, theory of mind and attributional style (
Adults with ASD were impaired in all four domains of social cognition assessed by the “ClaCoS” battery. Deficits in emotional perception and theory of mind are classically reported in ASD, while social perception and attributional style are less frequently explored. In particular, deficits in emotional perception and theory of mind constitute core features of ASD and are crucial clinical signs examined in the process of diagnosing ASD (
Using the TREF, in the ASD group compared to the controls, we observed a higher global recognition threshold and a trend for a lesser overall percentage of correct emotion recognition. Thus, the recognition threshold seems to be a more relevant measure in order to uncover impairments on emotional recognition in this population. Our result is consistent with an extensive amount of researches showing impairments on emotional processing in ASD [(
Adults with ASD showed lower scores at the PerSo compared to controls, suggesting impaired social perception. More precisely, they reported fewer visual elements from the social scenes (fluency score) and were less accurate in their interpretation of the depicted social situations. The difference between groups on the number of visual elements reported does not seem to be exclusively linked to slower overall visuospatial or attentional abilities in ASD. Indeed, it remained significant when controlling for such effect of visual attention. It could rather reflect slower overall cognitive processing or more specifically slower verbal elaboration required in this task. It could also reflect a well-described lack of generation of new ideas (
As expected, we observed an impaired theory of mind (ToM), in line with previous studies showing that the MASC test is sensitive to reveal subtle ToM alterations in adults with ASD, without intellectual disability (
To date, there are very few studies focusing on attribution biases in ASD. This component of social cognition is rarely explored in clinical practice. The studies mainly focused on causal attribution (internal vs. external) and showed no differences on attributional style between ASD adolescents and adults compared to neurotypical individuals (
An interesting feature of the “ClaCoS” is the assessment of patients' subjective complaints in daily life in the field of social cognition. To our knowledge, the ACSo is the first transnosographic scale examining this question. Our results suggest that adults with ASD without intellectual disability are able to express complaints from a self-report questionnaire. This short scale seems to constitute an interesting media which may help ASD adults to express the challenges they encounter in social situations, especially for those having difficulties in verbal expression or initiation. Interestingly, there was an association between subjective complaints and objective measures obtained from the neuropsychological assessment in all four domains of social cognition. These results tend to support the relevance of each neuropsychological test of the “ClaCoS” battery to assess the specific domain of social cognition which impacts the participant's everyday social functioning. They further suggest that ASD adults are aware of their social difficulties in different areas, in line with their good insight and metacognitive abilities (
It seems unlikely that our effects were mediated by socio-demographic factors, as the groups were matched according to age, gender, and educational level. Moreover, the ASD group had good intellectual and very efficient language abilities, as revealed by their IQ measures, suggesting that the deficits observed in the evaluation of social cognition are unlikely to be linked to poorer intellectual or verbal abilities. Note however that we could not fully control the impact of verbal and performance IQ, as these were not recorded in control participants. Although the ASD participants were slower on both visuo-spatial and attentional tests, they correctly processed the visual information. They were sufficiently engaged in the task but needed more processing time, consistently with previous observations (
Our results should nevertheless be interpreted with several limitations. Although our population was larger than most studies considering the assessment of social cognition in ASD, it remains of a moderate sample size. Future studies including larger groups are needed in order to replicate these findings. In particular, the correlations reported here were of medium effect size (Cohen criterion) and thus have to be taken with caution. Moreover, the ASD participants had average or over-average Intellectual Quotients. Their verbal IQ was in average higher than the performance IQ. Future studies are required to evaluate whether these results can be generalized to ASD individuals with lower verbal or general intellectual abilities. More generally, the links between neurocognitive and social cognitive performance remains to be explored. The addition of other validated tools for social cognition evaluation could also be interesting as external validation of the tests constituting the “ClaCoS” battery. Furthermore, the use of clinical tools allowing the assessment of hyper- and hypo-mentalization in ASD could allow the identification of different clinical profiles, in line with the transnosographic view of psychiatric illnesses. Further studies including larger sample size and different clinical profiles are needed in order to confirm and extend the present results.
The “ClaCoS” is a functional evaluation battery of the four main dimensions of social cognition which may be altered in different psychiatric conditions, consistent with a transnosographic perspective. To our knowledge, “ClaCoS” is the only existing social cognitive battery including a subjective evaluation of the individual's impairment, as well as an assessment of attributional style. As a whole, the current study suggests that performance on the “ClaCoS” battery accurately reflects everyday life difficulties of adults with ASD. It seems to be a well-suited tool to help uncover alterations in specific domains of social cognition in this population. This allows the selection of the most appropriate therapeutic program according to each patient's functional profile: from the most basic perceptual processes to higher level metacognition abilities. Our study shows that ASD adults without intellectual disability have a good self-awareness of their impairments in different domains of social cognition which can lead to the specific challenges they encounter with social functioning in everyday life situation. Self-assessment helps to involve and motivate patients to participate in cognitive remediation therapy (
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by CPP Lyon Sud Est IV, No. 15/041; ANSM, No. 2015-A00580-49. The patients/participants provided their written informed consent to participate in this study.
SM-K, AT, LB, CD-A, EP, ZP, IC-M, BG, IA, JG, and EH-D made substantial contributions to conception and design. SM-K, AT, LB, CD-A, EP, ZP, IA, JG, and EH-D made substantial contributions to acquisition of data. SM-K, AT, SR, JG, and EH-D made substantial contributions to analysis and interpretation of data. SM-K, AT, LB, CD-A, EP, ZP, SR, IA, JG, EH-D, IC-M, BG, M-OK, NF, and FB-B participated in drafting the article. All authors contributed to, have approved the final manuscript and agree to be accountable for the content of the work.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
The authors would like to thank all the people involved in the ClaCoS project, and particularly Veronique Vial who monitored the study. They also thank Serge Berthier for assistance in preparing this manuscript and Nicolas Jousselain for help with English editing.