Although Autism Spectrum Conditions (ASC) and Schizophrenia Spectrum Conditions (SSC)
are recognized as distinct diagnostic categories with independent features, they share
a history of clinical entanglement stemming from overlapping symptomatology, particularly
in the area of social behavior (1). Examination of cognition and the neurobiology
of ASC and SSC, both in the conditions themselves and within their subclinical manifestations,
offers potential for illuminating the shared and unique mechanisms underlying their
social characteristics. In recent years, both direct comparisons of ASC and SSC and
continuous explorations of the autism-psychosis spectrum have begun to show promise
for producing more precise segmentation and greater clinical utility than traditional
comparisons to non-clinical controls. Such studies ultimately may help to inform etiological
understanding, improve screening and diagnosis, and provide more targeted support.
The goal of this Frontiers Research Topic is to showcase innovative new work examining
cognitive and neurobiological features of the autism-psychosis spectrum. Consistent
with the primary focus of research activity in this area, included articles can be
organized under two primary subsections: Cognition and Neurobiology.
Collectively, the articles on cognition suggest that both ASC and SSC are often (but
not always) characterized by neuro- and social cognitive differences compared to non-affected
controls that relate in both direct and indirect ways to the broader functional and
social disabilities associated with the conditions. For instance, Sijtsma et al. demonstrate
that adolescents with more autistic-like experiences are less likely to be nominated
as friends by their peers despite no relation found between autistic-like experiences
and social cognitive performance or self-reported rates of friendships. This suggests
that traditional social cognitive measures may not always capture the social differences
associated with subclinical autistic characteristics affecting peer relationships.
Relatedly, Larson et al. found that autistic people who have experienced psychosis
had higher rates of schizotypy and emotional difficulties than both neurotypical controls
and autistic people with no psychosis history despite little to no difference between
the groups on measures of perspective taking. Their findings also highlight a previously
under-recognized clinical characteristic associated with schizotypal traits: high
affective lability, which refers to elevated shifting between different emotional
states. Likewise, the work by van der Linden et al. also suggests that the full impact
of psychotic experiences (PE) among autistic individuals are not well-understood.
Their paper indicates that autistic people do not differ from controls in lifetime
PE but do report more frequent momentary PE and greater distress associated with their
PE. They conclude that stress may serve as an important risk factor for PE among autistic
individuals. Meanwhile, the papers by Maat et al. and Abu-Akel et al. demonstrate
that the presence of cross-diagnostic symptoms along the autism-psychosis spectrum
may produce novel cognitive effects in autistic people not seen in autism alone. In
Maat et al., response time latency to social and non-social stimuli was increased
among autistic adolescents who present with features of psychosis. Specifically, those
with attenuated psychosis syndrome, a condition defined by subclinical positive symptoms
associated with risk of subsequent psychosis, did not differ from controls on measures
of pattern, face, or emotion recognition, but exhibited slower response times to stimuli.
Similarly, Abu-Akel et al. found that autistic people with co-occurring schizotypal
personality disorder (SPD) and higher positive psychotic symptoms exhibited more sustained
(but not inhibitory) attention relative to individuals with autism or SPD alone. Such
findings suggest that autistic and positive symptoms may diametrically influence sustained
attention and, more broadly, highlight the need for better understanding the relationship
between autism and SPD and the effects of dual diagnosis.
The paper by Deste et al. also reports combinatory effects, but unlike the Larson
et al. and Maat et al. papers that examined autistic people with features of psychosis,
their study focuses on autistic symptoms among patients with schizophrenia and their
association with relationship outcomes. In line with a recent large-scale study in
psychotic patients and their siblings (2), they find that higher autism symptoms constitute
a significant predictor of poor social relationships in schizophrenia, even after
controlling for other relevant demographic and clinical variables. Autism and schizophrenia,
however, are not only characterized by social difficulties but also multisensory processing
difficulties, and the paper by Noel et al. offers a direct comparison of visual-tactile
spatial multisensory processing in the two conditions. They find that both groups
do not differ from non-affected controls on a cross-modal congruency, but autistic
adults exhibit a smaller and more restricted peri-personal space (i.e., the space
immediately around the body) than both controls and adults with schizophrenia. Additionally,
they find an association between smaller peri-personal space and social symptoms,
suggesting a link between some aspects of multisensory processing and social-emotional
functioning. Next, Kuo and Eack provide the first systematic review and meta-analytic
comparison of non-social cognitive abilities in autism and schizophrenia. In contrast
to studies of social cognition that do not show many performance differences between
the groups (3–5), their comparison revealed superior performance in autism on working
memory, visuospatial memory and learning, language, and comparable performance on
processing speed, attention, and verbal comprehension. These distinguishable cognitive
profiles extended from adolescence to middle adulthood and collectively indicate important
neurocognitive differences that may serve as distinguishing mechanisms of their overlapping
reductions in social cognitive performance.
As many of these studies highlight, clinical differentiation across the autism-psychosis
spectrum can be difficult among those presenting with co-occurring mental health conditions
and shared social characteristics. One paper in the collection (Demetriou et al.)
examines the utility of using machine learning to differentiate autism, early psychosis,
and social anxiety disorder based on a comprehensive battery of neurocognitive, social
cognitive, and mood assessments. Social cognition, visuospatial memory, and mood (e.g.,
depression, anxiety, and stress) differentiated the clinical groups from a neurotypical
control group and the social anxiety group from the autism and early psychosis groups.
The autism and early psychosis groups were more difficult to differentiate, with only
psychomotor speed and stress distinguishing them.
The cognitive differences characterizing the autism-psychosis spectrum are supported
by an underlying neurobiology that five papers in this collection seek to better understand.
First, Barlati et al. offer an incisive review of the shared and divergent neuroanatomical,
neurofunctional, and molecular markers of social cognition in autism and schizophrenia.
Consistent with the goals of this Frontiers Research Topic, their review is organized
in accordance with a Research Domain Criteria perspective that seeks to identify behavioral
and biological signatures of clinical symptoms spanning diagnostic conditions, and
as such serves as a useful summary and guide for future research initiatives. Relatedly,
Nair et al. provide a comprehensive narrative review of functional connectivity studies
of social cognition in autistic adolescents and those with early-onset psychosis and
contextualize these findings within a developmental framework. They conclude that
disruptions in default mode network connectivity are associated with social difficulties
in both groups but are less relevant to other features of each condition. Meanwhile,
Samaey et al. provide a more focused review of the neural underpinnings of one social
cognitive ability in particular, facial expression processing, and conclude that altered
connectivity and activation in the fusiform gyrus and amygdala among autistic individuals
and those with primary psychosis may be influenced by adverse childhood events. They
argue that more integrative studies across clinical conditions, framed within a developmental
context, are needed to fulfill the promise of identifying selective biomarkers.
Conclusions from these three reviews are supported by two empirical papers in the
collection. Brady et al. examined associations between brain connectivity and social
cognition in sample of people with psychosis and neurotypical controls and found evidence
across both samples that implicated a cerebellar-parietal circuit strongly linked
with social cognitive ability. They highlight this circuit as a potential trans-diagnostic
biomarker of reduced social cognitive performance. Meanwhile, Foss-Feig et al. leveraged
a longitudinal sample to examine the P300 ERP component as a potential indicator of
conversion to psychosis among individuals at clinical high risk (CHR) with a history
of autism. Unlike previous findings suggesting that P300 amplitude reductions predict
psychosis in CHR populations without autism, they found enhanced neural responses
during attentional orienting tasks in their small sample of individuals with combined
CHR and autism. Such a result is consistent with other papers in this Research Topic
indicating that the combination of symptoms across the autism-psychosis spectrum may
interact in unpredicted ways and not follow a simple main effect framework.
Author Contributions
NS wrote the first draft of the manuscript. AP and TZ contributed intellectually to
the manuscript and helped revise it. All authors have read and approved the submitted
version.
Conflict of Interest
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.