Sarah Joanne White

Autistic people experience disproportionate challenges in noisy environments than non-autistic people. To capture a naturalistic representation of listening effort in adverse condition, an autistic (n=27) and non-autistic group (n=22) were presented with an audiobook and different speech masking noises while their pupils were measured with an eye tracker. The audiobook task consisted of a target-only condition and three noise conditions: 1-speaker, 2-speaker, and speech-shaped noise. The task was entirely passive with no behavioral or subjective measures. We measured the mean and variance of the evoked pupillary response—a reliable measure of cognitive effort. Primary findings revealed effects of the speech-masking conditions and condition occurrence on temporal dynamics of pupillary behavior. Specifically, the autistic group exhibited a general increase of mean pupil dilation over time, whereas the non-autistic group showed no difference in pupil size over time. Both groups showed increased variance in pupil size over time in all conditions throughout the task. These findings demonstrate that an audiobook-in-noise task can provide insight into temporal dynamics of pupil behavior during naturalistic speech perception in noise (SPiN). These results also suggest increased listening effort required in the autistic sample compared to the non-autistic controls.

What does laughter mean, and why does laughter make things funnier? Using a novel implicit measure of laughter processing, we added genuine or posed laughter to various forms of comedic stimuli to examine the similarities and differences in the modulatory effects of laughter between autistic and non-autistic people. In non-autistic adults, adding laughter increased the perceived funniness of humorous stimuli, including jokes, burps and slapstick videos, with genuine laughter leading to funnier ratings than posed laughter. However, these laughter modulation effects were not consistently observed in autistic adults. While the laughter modulation effect was present for jokes, it was less pronounced than in non-autistic adults and was absent when modulating the funniness of burps and slapstick videos. Overall, our findings highlight the distinct ways autistic people experience the implicit processing of laughter, which may impact social communication in daily interactions. This study offers a novel perspective on how the interplay between humour and laughter contribute to everyday social experiences.

Social discounting describes the tendency to give fewer resources to those we feel less close to. Previous work suggests autistic individuals show a flatter decline in generosity towards socially distant others compared with non-autistic participants. We investigated whether this enhanced prosociality towards socially distant others in autism was driven by genuinely higher prosociality or instead a preference for repetitive responding. We measured prosocial preferences in 37 autistic and 38 non-autistic adults using the social value orientation questionnaire, where participants allocated money between themselves and people at six different social distances (e.g. friend vs stranger). We replicated previous findings by showing that autistic adults were more prosocial than non-autistic adults towards more socially distant others. Crucially, these effects were not driven by more repetitive responding in autism, nor by differences in attitudes to money. While autistic people often face challenges navigating their social worlds, our findings reinforce the view that autism is also associated with more prosociality. We argue that differences in fairness norms could drive more prosociality in autism, but this remains to be tested in future work. Lay Sharing and giving to others are important for our social relationships. Previous studies show that when given opportunities to share money, autistic and non-autistic people give the same amount of money to people they feel close to, like their friends. However, compared with non-autistic people, autistic people give more money to people they feel less close to, like strangers. In this study, we replicated this finding. Compared with non-autistic participants, autistic participants were more generous to people they did not feel close to. We also found that this increased generosity in autism was not the result of autistic participants responding more repetitively in the task. Autistic and non-autistic participants also showed similar attitudes towards money. We propose that some autistic people could be more generous because they show differences in how they think about fairness. But future studies will need to look at this more closely. We hope that our results can help to change the way people think about social behaviour in autism. While autistic people often face challenges navigating their social worlds, autism can also be associated with more generosity.

Intergroup bias modulates both Theory of Mind (ToM) during social judgments, yet behavioural measures alone cannot fully elucidate the underlying mechanisms. Multimodal approaches are therefore needed to uncover the corresponding neural correlates. Here, we used functional near-infrared spectroscopy (fNIRS) to investigate how intergroup bias shapes neural responses during smile discrimination–a proxy for ToM. Thirty-three adults viewed videos of ingroup or outgroup members displaying genuine or posed smiles, judging smile authenticity (ToM) and smiler gender (Gender), with facial expressions recorded. Behaviourally, no intergroup bias emerged; however, neuroimaging data revealed distinct pathways: ingroup smiles engaged the middle frontal gyrus (MFG) and dorsolateral prefrontal cortex (dlPFC), whereas outgroup smiles activated the inferior frontal gyrus (IFG). ToM conditions elicited stronger IFG and superior temporal cortex (STC) responses for outgroup members compared with Gender conditions. Although facial mimicry was absent, integrated fNIRS and facial analyses demonstrated that intergroup bias differentially recruits neural systems during ToM-driven smile processing. These findings indicate that social affiliation and cognitive control networks are sensitive to ingroup cues, while outgroup cues elicit greater mentalizing, potentially reflecting perceptions of outgroup members as less trustworthy and requiring more effortful interpretation.

What does laughter mean, and why does laughter make things funnier? Using a novel implicit measure of laughter processing, we added genuine or posed laughter to various forms of comedic stimuli to examine the similarities and differences in the modulatory effects of laughter between autistic and non-autistic people. In non- autistic adults, adding laughter increased the perceived funniness of humorous stimuli, including jokes, burps, and slapstick videos, with genuine laughter leading to funnier ratings than posed laughter. However, these laughter modulation effects were not consistently observed in autistic adults. While the laughter modulation effect was present for jokes, it was less pronounced than in non-autistic adults and was absent when modulating the funniness of burps and slapstick videos. Overall, our findings highlight the distinct ways autistic people experience the implicit processing of laughter, which may impact social communication in daily interactions. This study offers a novel perspective on how the interplay between humour and laughter contribute to everyday social experiences.

In everyday life, listeners spontaneously and rapidly form first impressions from others' voices. Previous research shows that, compared to non‐autistic people, autistic people show similarities and differences in how they evaluate others based on their faces. However, it remains unclear whether autistic people form first impressions from voices in the same way as non‐autistic people. We asked both autistic and non‐autistic listeners to rate an inferred characteristic (trustworthiness) and an apparent characteristic (age) from voice recordings to establish how they form first impressions from voices. Non‐autistic and autistic listeners formed first impressions for age and trustworthiness in similar ways. Specifically, both groups showed comparable overall ratings of age and trustworthiness of voices. Further, both autistic and non‐autistic listeners required similar amounts of information to form an impression. Finally, when comparing trait impressions within‐ and across‐groups, we again found no systematic evidence of impression formation differing between autistic and non‐autistic people. These findings indicate that first impression formation is potentially a relative social strength in autism. This suggests that the social challenges encountered by autistic people may be confined to specific areas of social perception rather than being universally pervasive.

It has been suggested that mentalizing abilities underlie the distinct profiles of autism characteristics observed in Cornelia de Lange (CdLS) and fragile X syndromes (FXS). However, traditional explicit mentalizing tasks have high language demands that may mask mentalizing abilities. We compared performance on traditional explicit tasks and an implicit anticipatory looking mentalizing task in children with CdLS (N=9), boys with FXS (N=9), autistic (N=22) and neurotypical (N=34) children. The groups showed divergent patterns of performance. Neurotypical children had higher explicit mentalizing scores than all other groups. However, neurotypical, FXS and CdLS groups showed better implicit mentalizing performance than autistic children. Both chronological age and receptive language ability correlated with explicit mentalizing scores in neurotypical children. In autistic children, there was an association between explicit mentalizing score and receptive language ability but not chronological age. Explicit mentalizing score was not associated with receptive language ability or chronological age in the CdLS and FXS groups. Neither chronological age nor receptive language ability correlated with implicit mentalizing task performance in any group. Despite having intellectual disability and autistic characteristics, findings demonstrate that children with CdLS and FXS are skilled at spontaneously mentalizing.

Evidence supporting the influential mentalizing account of autism (that autistic people struggle to spontaneously represent others’ beliefs) has traditionally required autistic people to mentalize about non-autistic people. Yet, we know that individuals mentalize better for people who are like them and autistic people are likely to see non-autistic people as different to themselves. Here, we tested whether an actor’s diagnosis affected the spontaneous mentalizing of autistic and non-autistic adults by recording eye-movements. We found that autistic adults correctly anticipated the actor’s action based on her beliefs only when she was autistic. Conversely, non-autistic adults correctly anticipated her action only when the actor was non-autistic. Thus, the predominantly non-autistic world we inhabit seems to underlie autistic people’s difficulty mentalizing, challenging the idea of a fundamental impairment.

In everyday life, listeners spontaneously and rapidly form first impressions from others' voices: Are they young? Are they trustworthy? Previous research shows that, compared to non-autistic people, autistic people show similarities and differences in how they evaluate others based on their faces. However, it remains unclear whether autistic people form first impressions from voices in the same way as non-autistic people. We asked both autistic and non-autistic listeners to rate a social characteristic (Trustworthiness) and a physical characteristic (Age) from voice recordings that lasted 50, 200, or 800 ms to establish how they form first impressions from voices. We found that non-autistic and autistic listeners formed first impressions for Age and Trustworthiness in similar ways. Specifically, both groups showed comparable overall ratings of Age and Trustworthiness of voices. In terms of how much voice information is needed to form impressions, both autistic and non-autistic listeners also exhibited similar temporal trajectories. Finally, when comparing trait impressions within- and across- groups, we again found no systematic evidence of impression formation differing between autistic and non-autistic people. These findings highlight that social perception can be remarkably similar in autistic and non-autistic people. This, therefore, contextualises the often-reported social challenges encountered by autistic people, suggesting that these may be confined to specific individuals or to specific domains of social perception rather than being universally pervasive.

To overcome the criticisms of earlier work using anticipatory-looking paradigms to investigate implicit mentalizing in autism, Wu et al. (2024) recently presented a multi-trial paradigm with matched true-belief and false-belief conditions. This study is a replication of Wu et al.’s work using a novel set of stimuli and a separate group of participants, all of whom were naïve to this paradigm. Fifty-four (26 autistic) participants completed an implicit mentalizing task with both high- and low-demand false-belief conditions with two corresponding true-belief control conditions, alongside an explicit mentalizing task. In accordance with Wu et al. (2024), our findings support the presence of spontaneous mentalizing difficulties in autistic adults, despite strong explicit mentalizing abilities. Therefore, we provide further evidence that whilst some autistic adults with high-IQs may have developed compensatory strategies that mask mentalizing difficulties on explicit tasks, implicit tasks reveal differences in the underlying spontaneous mechanism.