Savant syndrome is a phenomenon whereby individuals with cognitive impairments have one or more outstanding abilities, inconsistent with their general intellectual functioning. Approximately 50% of savant individuals have autism spectrum disorder (ASD), and 10–30% of people with ASD have savant skills. To shed additional light on this considerable overlap, we compared autistic traits as measured by the Social-Responsiveness-Scale (SRS) between 712 children with at least one reported savant skill, as determined by designated questions from the ADI-R questionnaire (savant group), and 2,032 non-savant children from the Simons-Simplex-Collection (SSC) database. We also examined SRS scores of the parents of these children and compared parent-child differences in SRS scores between the savant and non-savant groups. Savant children had significantly lower SRS scores (less deficiencies) compared to non-savant children (P < 0.05), while no such differences were observed among their parents. Further intra-familial analyses revealed weak pairwise-correlations (r = −0.015–0.141) between SRS scores of parents and their children, and significantly larger parent-child differences in standardized SRS scores within savant families (P < 0.05). These findings suggest that the less severe autistic traits among savant children with ASD compared to other people with ASD is not likely to be a familial trait.
Savant syndrome is a fascinating phenomenon whereby individuals with cognitive impairments have one or more outstanding abilities, inconsistent with their general intellectual functioning1. The special skills typically occur in five areas - music, art, calendar calculating, mathematics or mechanical/visual-spatial skills, and are usually associated with outstanding memory that is restricted to the area of expertise2,3,4,5. In most cases, the condition is evident from early childhood5,6,7, however few cases of acquired savant skills have been documented following a CNS injury or disease8,9.
Interestingly, there is a high concordance between savant syndrome and autism spectrum disorder (ASD) where approximately 50% of savant individuals have also ASD5,10,11, and 10–30% of persons with ASD have exceptional abilities4,12,13. Another similarity between these two conditions is that both have a similar male-to-female ratio of 4-6:15,14. In the general population, a relation between special abilities and ASD-like-traits has been observed, where children with special abilities show elevated ASD-like traits compared to those without such abilities15,16,17,18. In contrast, within ASD populations, Individuals with special abilities tend to have less ASD-like traits, or – elevated social skills13,19.
The considerable concordance between ASD and savant syndrome could stem from a shared etiology of these two conditions, and/or from a mechanism by which one condition predisposes the emergence of the other. Theories supporting the first option include the “Enhanced Perceptual Functioning” model that suggests that the over-functioning of brain regions involved in perceptual functions among people with ASD may be also responsible for the high prevalence of special abilities in these people20,21. Veridical mapping has been further proposed as the specific mechanism that accounts for the role of perception in the manifestation of exceptional skills among people with ASD22,23,24. A few rare genetic syndromes support these theories where people with these syndromes have both autistic features and specific enhanced skills25,26,27. Alternatively, the “Empathizing–Systemizing” theory, suggests that the attention to detail observed in ASD individuals occurs in the service of achieving an ultimate understanding of a system, and may predispose the emergence of savant skills28. Similarly, in their “Fractionable Triad Proposal”, Happé and Vital showed that the greater number of ASD-like traits found in savant-autistic children was particularly due to increased restricted and repetitive behaviors and interests (RRBIs), while their social and communication impairments were somewhat reduced17. Therefore, they suggested that the autistic characteristic of a detail focused cognitive style (weak coherence) predisposes to talent, and that the presence of savant skills may aid social competence. Other studies found no indication that elevated RRBIs predict the existence of savant skills, although they did find that autistic savants have better cognition and social skills than people with ASD without any exceptional talent4,13,29.
Today, the high concordance between the savant syndrome and ASD is still an enigma. The main goal of this study was to examine whether children with ASD and exceptional skills (i.e. autistic savants) have different autistic characteristics, than other children with ASD, and whether such differences are familial or not. To address this question, we compared various autistic traits as measured by the Social Responsiveness Scale (SRS) questionnaire30 between children with ASD and their parents. These analyses allowed us to assess whether the elevated social skills observed among individuals with ASD and savant skills may have a familial origin.
Data and Methods
The data for this study was derived from the Simons Simplex Collection (SSC; version 15)31 which includes a wide variety of clinical and genetic information from 2,852 families in which only one child has a diagnosis of ASD (simplex families). Presence of savant skills was determined using five designated questions in the Autism Diagnostic Interview-Revised (ADI-R) questionnaire32, as defined in previous, large scale studies4,12,13,29,33. A subject was defined as currently having an exceptional skill that is “definitely above the subject’s general level of ability and above the general population’s normal level” if he/she had a score of 2 (“isolated skill, not used functionally/meaningfully”) or 7 (“isolated skill, used meaningfully”) in at least one of the following special skill items: Visuospatial ability (item 88); musical ability (item 90); drawing skill (item 91); reading ability (item 92); and computational ability (item 93).
Next, we used data from the Social Responsiveness Scale (SRS) questionnaire30 filled by the parents of children with ASD to identify the presence and extent of autistic traits in this sample. We also used data from the adult version of the SRS questionnaire that was filled by both child’s parents on each other, to assess the parental social responsiveness in an equivalent manner. Each item in the SRS questionnaire is rated on a scale of 0 (not true) to 3 (almost always true), and the total score serves as an index for the severity of autistic traits where a higher total score indicates a more severe social deficit. We used these raw SRS scores (of both child and adult versions) throughout this study. Further, the SRS questionnaire can be divided into five subscales according to the domain of autistic trait (social awareness, social communication, social cognition, autistic mannerisms and social motivation), and each of these subscales can be assessed separately.
First, we compared demographic features and primary ASD diagnosis measures from the ADOS and ADI-R tools using chi-square (for nominal variables) and Mann-Whitney U tests (for continuous variables). We further used independent samples t-tests or Mann-Whitney U tests (where a violation of the normal distribution assumption occurred) to compare the total SRS scores and scores of the different SRS domains between the savant and non-savant groups, as well as between specific savant subtypes, and the non-savant group. Significant differences in SRS scores were reevaluated in multivariate logistic regression models that controlled for the children’s sex and age.
Finally, we aimed at evaluating the familial contribution to the child’s autistic characteristics. Hence, we examined the correlations between SRS scores of children and their parents. In addition, we computed SRS z-scores for children (standardized by the other children SRS scores) and their parents (standardized by the same gender parent SRS scores) to bring the SRS scores of all family members to the same scale. Then, we applied a two-way ANOVA of repeated measures to these SRS z-scores followed by paired t-tests to assess the effects of the parent autistic characteristics and the child’s savant status as well as their interaction on the child’s SRS score. Statistically significant results were determined at the P < 0.05 level, with a false-discovery rate (FDR) correction for multiple testing34 when necessary.
Compliance with ethical standards
All procedures and experimental protocols performed in this study involving human participants were approved by the institutional research committee at Ben-Gurion University. The study was performed in accordance with the guidelines and regulations of the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from a parent and/or legal guardian of all participants in this study.
Of the 2,852 children with autism at the SSC database, 2,744 (96.2%) had data from both SRS and ADI-R questionnaires and hence were included in our study. Of these, 712 children (25.9%) had at least one reported exceptional talent (see methods) and were consequently included in the savant group in this study. There were no statistically significant differences in gender nor race between the savant and non-savant groups. However, savant children were, on average, six months older than non-savant children at the time of ADOS administration (112.9 ± 42.7 vs. 106.7 ± 42.7 respectively; Table 1). In addition, savant children had significantly higher IQ, nonverbal communication, and social abilities compared to non-savant children (Table 1). Notably, these differences, as well as all the findings that are presented in further analyses did not change after accounting for the age difference between the groups using regression models (Supplementary Table S1).
To study differences in social responsiveness between savant and non-savant children with ASD, we compared the SRS scores between these two groups (Fig. 1, gray columns, and Supplementary Fig. S1). There were significant differences between these two groups in the total SRS score (P = 0.015, Cohen’s d = 0.103), as well as in the scores of the awareness domain (P = 0.012, Cohen’s d = 0.106), the cognition domain (P < 0.001, Cohen’s d = 0.151), and in the communication domain (P = 0.034, Cohen’s d = 0.090) with savants having on average lower scores (less deficiencies) than non-savants in all of these domains. There were non-significant differences between the savant and non-savant groups in the mannerism and motivation domain scores. We further examined the SRS scores of children with specific reported talents (Fig. 1, white columns). In general, SRS scores of all the different savant subgroups were lower than the scores of the non-savants (less deficiencies). Specifically, statistically significant differences after correcting for multiple testing included lower total SRS score in the computational, reading, and drawing subgroups (PFDR = 0.04, Cohen’s d = 0.200; PFDR = 0.001, Cohen’s d = 0.161; and PFDR = 0.04, Cohen’s d = 0.220; respectively), lower awareness scores in the drawing and musical subgroups (PFDR = 0.04, Cohen’s d = 0.214; PFDR = 0.04, Cohen’s d = 0.180; respectively), lower cognition scores in the computational, reading, drawing, and visuospatial subgroups (PFDR = 0.001, Cohen’s d = 0.273; PFDR = 0.001, Cohen’s d = 0.200; PFDR = 0.04, Cohen’s d = 0.242; and PFDR = 0.03, Cohen’s d = 0.205; respectively), and lower communication scores in the drawing and musical subgroups (PFDR = 0.04, Cohen’s d = 0.207; PFDR = 0.04, Cohen’s d = 0.172; respectively).
Next, we examined the adult SRS scores of both mothers and fathers of children in this study. No significant differences in autistic traits were seen between parents of the savant and non-savant groups (P > 0.06, Cohen’s d < 0.012; Fig. 2). However, significant gender differences (P < 0.05, Cohen’s d > 0.050) were observed in all five domains of the SRS questionnaire where fathers performed better (lower SRS scores) in the cognition and mannerisms domains, whereas mothers had better awareness, communication and motivation skills. These differences were consistent among parents of both savant and non-savant children with ASD except in the motivation domain where the gender difference in the savant group was only marginally significant (P = 0.057, Cohen’s d = 0.069). Interestingly, no such gender differences in SRS scores were seen among the children of these parents (Supplementary Fig. S2).
Finally, we examined the relationships between SRS scores of each child and their parent(s). First, we evaluated the parent-child pairwise correlations in SRS scores. Overall, parents had SRS scores that were more similar to each other than to their children across all SRS subscales and in both savant and non-savant groups (supplementary Table S2). Next, we applied a repeated measures ANOVA to these data followed by paired t-test analyses (See Methods) to explicitly scrutinize the possible reasons for the differences in SRS scores between child-parent pairs, which is not possible when comparing group means. The results of this analysis highlight three notable trends (Fig. 3): (1) children with autism were more similar to their mothers than to their fathers (smaller pairwise differences, with the only exception in the autistic-mannerisms domain); (2) there were larger pairwise differences in the SRS scores among family trios of the savant groups than among the non-savant families; and (3) pairwise differences of the savant and non-savant groups were consistently in different directions (i.e. positive differences for the savant pairs and negative differences in the non-savant pairs or vice versa). These trends resulted in a significant parent (mother/father) X savant status (savant/non-savant) interaction (P < 0.05) in the total SRS score and all its subdomains except the communication domain (P = 0.11) suggesting that parent-child differences in SRS scores are remarkably larger in families of autistic savants. The child–father and child–mother differences in raw SRS scores are depicted in supplementary Table S3.
The main goal of this study was to examine whether children with ASD and exceptional skills (i.e. autistic savants) have different autistic characteristics, as measured by the SRS, than other children with ASD, and whether such differences are familial or not. We show that parents of children with and without savant skills do not have different social responsiveness characteristics. Yet, in line with previous studies13,19, children with autism and savant skills have significantly lower SRS scores (less deficiencies) in a range of social impairment domains, as well as having higher IQ, and lower scores in the ADOS and ADI-R tests than non-savant autistic children. The observed differences in age and IQ between these two groups could contribute to the reported differences is social skills. Alternatively, the less apparent autistic features among the savant children with ASD could lead to a delayed diagnosis in these children. Of note, these differences tend to be larger among children with specific skill types (Fig. 1, white columns), suggesting that individuals with different savant skills may have distinct patterns of autistic traits.
Our findings of no significant difference in the autistic mannerisms domain of the SRS between the savant and non-savant groups join other findings (e.g.4,29) that partially negate the “Fractionable Triad Proposal” which suggests that the autistic characteristic of a detail focused cognitive style (weak coherence, RRBIs) predisposes to talent17,21,35,36. In contrast, the observation that the savant group is characterized by better social awareness in our study is compatible with the suggestion that the presence of special skills may aid social competence17,37. Importantly, similar results were seen when we compared each of the five savant subgroups (defined by their talents) to the non-savant group. This may also support the premise that the presence of exceptional talent among people with ASD enhances their social skills, regardless of the type of talent.
Interestingly, there were significant gender differences in all domains of the adult version of the SRS questionnaire, however no such differences were seen in the child questionnaires. A possible explanation of such observation is a rater bias, due to the different gender ratio of the raters of these questionnaires; Child questionnaires are more commonly filled by mothers than by fathers38 whereas adult questionnaires are always filled by one parent, on the second parent. Thus, the observed differences in SRS scores between the mothers and fathers in this study most likely reflect differences in the rater’s judgment in the different rater groups. Yet, the possibility that these differences reflect true differences in social behavior is still valid. Unfortunately, the lack of information regarding the gender of the child SRS rater in our dataset disallowed us from testing this hypothesis.
Finally, we observed larger intra-familial differences in standardized SRS scores among the savant families than in the non-savant families. This finding implies that the differences in autistic traits between the savant and non-savant children with ASD are unlikely driven by familial factors. We therefore suggest that the relationship between talent and social behavior observed in this study and others13,17 is more likely a result of one trait nurturing the other. Nevertheless, our data cannot indicate whether the existence of exceptional talent improves social competence, or whether the predisposition for better social skills promotes talent. In addition, a mutual synergism between these two traits whereby the existence of one enhances the other and vice-versa is also a valid possibility. Further empirical studies of talented children with ASD are thus warranted to gain better understanding of the mechanisms underlying talent in these children. Results of such studies will have remarkable social and clinical implications to people with ASD as well as to other human beings.
The strengths of this study include its large sample size, especially the relatively large number of female subjects which is a recurring limitation among studies of autistic populations. The large sample size allowed us to examine gender-based differences, which was not possible in previous, smaller samples. Nevertheless, our findings should be interpreted in the context of several limitations. First, the savant definition was based on parent report, and was not validated by empirical tests. Using the questions from the ADI-R to determine savant skills in children with ASD may result in either underestimation or overestimation of savant skills in the study. Multiple other large-scale studies (e.g.4,12,13,29,33 used the parental reports to determine exceptional talents among children with ASD. Some of these studies, which used both parental reports and empirical tests of some of the reported skills4,39, suggested that parental reports provide a good proxy for their child’s exceptional talent. Nevertheless, this approach is certainly imperfect, and may cause inclusion of empirically non-talented children in the savant group. Second, despite the relatively large sample size of this study, the classification of the savant group into talent-specific subgroups resulted in small sample sizes, which together with the need of multiple testing correction reduced the statistical power to draw robust conclusions about differences between these groups. Third, the Simons simplex sample is not a genuine representation of the autistic population in the US or elsewhere. Thus, caution should be taken in generalization of the conclusions of this study. Fourth, this is a cross-sectional study whereby single measurements of both the SRS and savant skills were taken at the same time. Therefore, we could not determine which of these traits predispose the other one. Finally, we did not have data regarding the occurrence of exceptional talents among the parents of the children in our dataset and hence could not assess whether the talents reported in the children are inherited or acquired.
Our results suggest that the better social responsiveness observed among people with ASD and savant skills compared to other people with ASD is not likely to be a familial trait. Further studies are required to examine whether any of these distinct autistic features predisposes the other, or whether they co-develop simultaneously.
Down, J. L. On some of the mental affections of childhood and youth: Being the Lettsomian lectures delivered before the Medical society of London in 1887, together with other papers. (J. & A. Churchill, 1887).
Treffert, D. A. The idiot savant: a review of the syndrome. The American journal of psychiatry 145, 563–572, https://doi.org/10.1176/ajp.145.5.563 (1988).
Kehrer, H. E. Savant capabilities of autistic persons. Acta Paedopsychiatr 55, 151–155 (1992).
Howlin, P., Goode, S., Hutton, J. & Rutter, M. Savant skills in autism: psychometric approaches and parental reports. Philosophical transactions of the Royal Society of London. Series B, Biological sciences 364, 1359–1367, https://doi.org/10.1098/rstb.2008.0328 (2009).
Treffert, D. A. Savant syndrome: realities, myths and misconceptions. J Autism Dev Disord 44, 564–571, https://doi.org/10.1007/s10803-013-1906-8 (2014).
Howlin, P. Understanding savant skills in autism. Dev Med Child Neurol 54, 484, https://doi.org/10.1111/j.1469-8749.2012.04244.x (2012).
Hughes, J. R. A review of Savant Syndrome and its possible relationship to epilepsy. Epilepsy & behavior: E&B 17, 147–152, https://doi.org/10.1016/j.yebeh.2009.12.014 (2010).
Jackson, M. & Warrington, E. K. Arithmetic skills in patients with unilateral cerebral lesions. Cortex; a journal devoted to the study of the nervous system and behavior 22, 611–620 (1986).
Treffert, D. A. & Tammet, D. Islands of Genius: The Bountiful Mind of the Autistic, Acquired, and Sudden Savant. (Jessica Kingsley, 2011).
O’Connor, N. & Hermelin, B. Low intelligence and special abilities. J Child Psychol Psychiatry 29, 391–396 (1988).
Miller, L. K. Defining the savant syndrome. Journal of Developmental and Physical Disabilities 10, 73–85 (1998).
Bolte, S. & Poustka, F. Comparing the intelligence profiles of savant and nonsavant individuals with autistic disorder. Intelligence 32, 121–131, https://doi.org/10.1016/j.intell.2003.11.002 (2004).
Ben-Itzchak, E., Aviva, B. & Zachor, D. A. Are special abilities in autism spectrum disorder associated with a distinct clinical presentation? Research in Autism Spectrum Disorders 7, 1122–1128, https://doi.org/10.1016/j.rasd.2013.05.003 (2013).
Fombonne, E. Epidemiological trends in rates of autism. Mol Psychiatry 7(Suppl 2), S4–6, https://doi.org/10.1038/sj.mp.4001162 (2002).
Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J. & Clubley, E. The autism-spectrum quotient (AQ): evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. J Autism Dev Disord 31, 5–17 (2001).
Brown, W. A. et al. Autism-related language, personality, and cognition in people with absolute pitch: results of a preliminary study. J Autism Dev Disord 33, 163-167; discussion 169 (2003).
Vital, P. M., Ronald, A., Wallace, G. L. & Happe, F. Relationship between special abilities and autistic-like traits in a large population-based sample of 8-year-olds. J Child Psychol Psychiatry 50, 1093–1101, https://doi.org/10.1111/j.1469-7610.2009.02076.x (2009).
Baron-Cohen, S., Wheelwright, S., Burtenshaw, A. & Hobson, E. Mathematical Talent is Linked to Autism. Human nature (Hawthorne, N.Y.) 18, 125–131, https://doi.org/10.1007/s12110-007-9014-0 (2007).
Hu, V. W. & Steinberg, M. E. Novel clustering of items from the Autism Diagnostic Interview-Revised to define phenotypes within autism spectrum disorders. Autism Res 2, 67–77, https://doi.org/10.1002/aur.72 (2009).
Hagmann, C. E. et al. Children with Autism Detect Targets at Very Rapid Presentation Rates with Similar Accuracy as Adults. J Autism Dev Disord 46, 1762–1772, https://doi.org/10.1007/s10803-016-2705-9 (2016).
Mottron, L., Dawson, M., Soulieres, I., Hubert, B. & Burack, J. Enhanced perceptual functioning in autism: an update, and eight principles of autistic perception. J Autism Dev Disord 36, 27–43, https://doi.org/10.1007/s10803-005-0040-7 (2006).
Mottron, L., Dawson, M. & Soulieres, I. Enhanced perception in savant syndrome: patterns, structure and creativity. Philosophical transactions of the Royal Society of London. Series B, Biological sciences 364, 1385–1391, https://doi.org/10.1098/rstb.2008.0333 (2009).
Mottron, L. et al. Veridical mapping in the development of exceptional autistic abilities. Neuroscience and biobehavioral reviews 37, 209–228, https://doi.org/10.1016/j.neubiorev.2012.11.016 (2013).
Bouvet, L. et al. Veridical mapping in savant abilities, absolute pitch, and synesthesia: an autism case study. Frontiers in psychology 5, 106, https://doi.org/10.3389/fpsyg.2014.00106 (2014).
Dykens, E. M., Sutcliffe, J. S. & Levitt, P. Autism and 15q11-q13 disorders: behavioral, genetic, and pathophysiological issues. Mental retardation and developmental disabilities research reviews 10, 284–291, https://doi.org/10.1002/mrdd.20042 (2004).
Levitin, D. J. et al. Characterizing the musical phenotype in individuals with Williams Syndrome. Child neuropsychology: a journal on normal and abnormal development in childhood and adolescence 10, 223–247, https://doi.org/10.1080/09297040490909288 (2004).
Milner, K. M. et al. Prader-Willi syndrome: intellectual abilities and behavioural features by genetic subtype. J Child Psychol Psychiatry 46, 1089–1096, https://doi.org/10.1111/j.1469-7610.2005.01520.x (2005).
Baron-Cohen, S. Autism: the empathizing-systemizing (E-S) theory. Annals of the New York Academy of Sciences 1156, 68–80, https://doi.org/10.1111/j.1749-6632.2009.04467.x (2009).
Bennett, E. & Heaton, P. Is talent in autism spectrum disorders associated with a specific cognitive and behavioural phenotype? J Autism Dev Disord 42, 2739–2753, https://doi.org/10.1007/s10803-012-1533-9 (2012).
Constantino, J. N. & Gruber, C. P. Social responsiveness scale (SRS). (Western Psychological Services Los Angeles, CA, 2007).
Fischbach, G. D. & Lord, C. The Simons Simplex Collection: a resource for identification of autism genetic risk factors. Neuron 68, 192–195, https://doi.org/10.1016/j.neuron.2010.10.006 (2010).
Lord, C., Rutter, M. & Le Couteur, A. Autism Diagnostic Interview-Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. J Autism Dev Disord 24, 659–685 (1994).
Meilleur, A. A., Jelenic, P. & Mottron, L. Prevalence of clinically and empirically defined talents and strengths in autism. J Autism Dev Disord 45, 1354–1367, https://doi.org/10.1007/s10803-014-2296-2 (2015).
Benjamini, Y. & Y., H. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society 57, 289–300 (1995).
Pring, L., Hermelin, B. & Heavey, L. Savants, segments, art and autism. J Child Psychol Psychiatry 36, 1065–1076 (1995).
Happé, F. & Frith, U. The weak coherence account: detail-focused cognitive style in autism spectrum disorders. J Autism Dev Disord 36, 5–25, https://doi.org/10.1007/s10803-005-0039-0 (2006).
Happé, F. & Vital, P. What aspects of autism predispose to talent? Philosophical transactions of the Royal Society of London. Series B, Biological sciences 364, 1369–1375, https://doi.org/10.1098/rstb.2008.0332 (2009).
Constantino, J. N. & Todd, R. D. Autistic traits in the general population: a twin study. Arch Gen Psychiatry 60, 524–530, https://doi.org/10.1001/archpsyc.60.5.524 (2003).
Meilleur, A. A., Berthiaume, C., Bertone, A. & Mottron, L. Autism-specific covariation in perceptual performances: “g” or “p” factor? PLoS One 9, e103781, https://doi.org/10.1371/journal.pone.0103781 (2014).
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Daniel, E., Menashe, I. Exploring the familial role of social responsiveness differences between savant and non-savant children with autism. Sci Rep 10, 2255 (2020). https://doi.org/10.1038/s41598-020-59209-7
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.