Introduction

Autism is defined as a complex neurodevelopmental disorder that is characterized by a range of challenges associated with social communication, as well as restricted and repetitive behaviors or interests. Autism is often diagnosed early in life but can vary significantly in severity and presentation among individuals1,2,3. An autism diagnosis is made based on standardized tests that include behavioral assessments, clinical opinion or a combination of these procedures4,5. Autism is a diverse and multifaceted condition, shaped by genetic, biological, environmental and social factors6,7. In turn, autism encapsulates a spectrum of behaviors and experiences, thus incorporating the recognition of neurodiversity and the validity of the autistic identity itself8,9.

Individuals on the spectrum can faces challenges with social cognition, or the ability to perceive, interpret, and respond appropriately to the social world. Initial studies predominantly characterized autism as a disorder characterized by deficits in social cognition, particularly through the lens of Theory of Mind (ToM) deficits, which refer to challenges with understanding the mental states of others10,11,12. However, more recent developments have focused on the heterogeneity of social cognition within the autism spectrum, leading to an increased focus on individual differences10. This conceptual shift has prompted researchers to investigate how intrinsic motivational factors including of allostasis-driven learning mechanisms interplay with cognitive processes, contrasting with conventional approaches that emphasized cognitive impairments as primary deficits13.

This Special Collection brings together a series of studies that reflect current progress in emerging research themes in social cognition in autism. Below are short summaries of the articles presented in the collection which have been organized along 5 broad themes: Social cognition; Social perception; Eye contact; Emotional regulation and Interpersonal synchrony.

Theme 1: social cognition

Autistic individuals may process social cognitive information differently, especially in the areas of empathy, mentalizing, social concept processing and self-generated thought. In a text-based empathy task, autistic adults demonstrated lower cognitive and affective empathy compared to non-autistic adults, along with reduced emotional reactivity. Importantly, emotional reactivity mediated the association between autism diagnosis and empathy levels, pointing to a mechanistic link14. Social cognitive processing of abstract information also appears to differ in individuals with high autistic traits. Birba and colleagues found that adults with higher autistic traits had difficulty recalling and interpreting information from socially relevant stories. Social cognition may also differ between autistic men and women15. Del Bianco et al. reported sex-specific neural response patterns during face processing tasks. While perceptual responses to faces were similar, differences emerged in cognitive processing stages and these differences were more pronounced in autistic participants than neurotypical controls16.

In the domain of mentalizing, a study by Wu et al. showed that autistic mothers displayed improved implicit mentalizing in evaluative contexts but reported poorer mental health17. Strawson et al. used experience sampling during a working memory task to examine the content and modality of participants’ thoughts18.

Finally, a systematic review of 52 social cognition interventions found that while social training programs were generally effective, the degree of skill transfer depended on factors such as age, symptom severity, delivery method, and context19.

In sum, these studies highlight that social cognitive differences in autism can be complex, context-dependent and influenced by emotional reactivity, cognitive styles and gender.

Theme 2: social perception

Differences in social information processing in autism may arise from atypical sensory processing or integration of multisensory and contextual cues. In the area of emotion recognition, Keating and colleagues found that autistic participants had intact emotion recognition and more precise inner representations of what speed of movement corresponded to happy, angry and sad facial expressions in point light videos which they—in contrast to non-autistic peers—did however not use for emotion recognition20. Further, Bögl and colleagues could show that adults with higher autistic traits used acoustic heart rate feedback less for rating own and other’s emotions compared to adults with lower levels of autistic traits, potentially reflecting alterations in how internal and external sensory information is weighted against each other during emotion recognition21. Similarly, Folz and colleagues could show that higher autistic traits were linked to a reduced ability to recognize specific emotions, despite intact levels of automatic facial mimicry, potentially suggesting a reduced integration of own physiological signals when interpreting others’ emotions22. In an intervention study, Scherf et al. found that improvements in the usage of eye gaze cues in autistic adolescents after a serious game intervention in only the treatment group while an increase of overall visual attention to faces was similarly found both in the treatment and in the non-treatment group23.

In the area of auditory social perception, Cai et al. found that autistic people reported using laughter similarly often in social contexts as neurotypical controls, but interpreted fake and sincere laughter as more similar in meaning24. Tsuji and colleagues found that higher autistic traits were not associated with greater difficulties in identifying words in non-vocal vs vocal noise in non-autistic females, potentially indicating that the previously reported attentional bias towards non-vocal sounds might be specific to autistic populations25.

Theme 3: eye gaze behaviors

Eye gaze behaviors may be different for some people with autism. Face gaze patterns and directed attention to other social cues can also vary widely in autism. Thorsson et al. reported that fluctuations in eye gaze behaviors can vary according to dyadic roles during a social interaction26. A controlled experiment by et al. Jording et al. compliment these findings. In this study, autistic individuals had to communicate with a partner using eye gaze cues whether or not to take a joint action27.

Theme 4: arousal/emotion regulation

Early autism research often characterized emotional responses as blunted. However, more recent work covered in this collection reflects a more nuanced understanding whereby autistic individuals may regulate emotions differently, not less effectively, especially when supported with context-sensitive strategies. Gaziel-Guttman et al. examined whether young adults with autism could use cognitive reappraisal to manage shame. While autistic participants reported lower initial shame than their neurotypical peers, both groups benefited equally from reappraisal28. Yet, those with autism used more neutral and fewer positive or realistic reinterpretations. Similarly, research on arousal is evolving beyond phenotypic characterizations. Galazka et al. investigated pupillary contagion an involuntary response to others’ pupil sizes as an indicator of emotional arousal29.

Collectively these studies suggest that emotion regulation and arousal in autism are context-dependent and potentially more intense, which may inform future interventions that recognize and leverage these differences.

Theme 5: interpersonal synchrony

Interpersonal synchrony is a complex social process that promotes prosociality within interactions, such as by improving social bonding, and it has also been linked to learning through social observation of others, which may also be reduced in autism30. Like many features of autism, interpersonal synchrony is present but is often atypical in the autistic population31. Donati et al. suggests that one possible explanation for atypical interpersonal synchrony in autism may be differences in lateral brain organization32. Krishnan-Barman et al., reported higher activation in areas related to processing motor and visual information (superior temporal sulcus) when autistic participants were being watched while completing an imitation task33. Koehler et al. proposed that motion synchrony may be a useful clinical marker autism for diagnosis34.

Conclusions

Research on social cognition in autism has evolved from early, small-scale studies of ToM and gaze behavior in children to a broader, lifespan perspective capturing variability and strengths across individuals. The field now recognizes emotional and sensory influences on social understanding and increasingly employs tools such as hyperscanning, real-time interaction paradigms and machine learning for more ecologically valid insights. Future work should integrate lived experience, leverage innovative methods and translate findings into interventions that enhance social functioning in daily life.