Sensory perception is often characterized by altered calibration to contextual regularities, but the basis of this difference remains unclear. We tested whether temporal magnitude perception in autism relies on modality-specific or amodal calibration using an adapted range-of-standards paradigm. Autistic and non-autistic adults completed separate auditory and visual duration-discrimination tasks. Across blocks, we manipulated the contextual duration range and whether the standards establishing that range were presented in the same modality as, or a different modality from the target stimulus. Non-autistic participants showed an auditory advantage in duration discrimination and stronger context effects within than across modalities, consistent with modality-specific temporal coding. In contrast, autistic adults exhibited reduced auditory advantage and comparable context effects across modalities, suggesting broader and less differentiated temporal calibration. To examine whether these patterns reflected synesthetic cross-modal tendencies, we tested an additional group of adults with synesthesia. Synesthetes showed calibration patterns resembling the non-autistic group, indicating that the cross-modal effects observed in autism cannot be attributed to synesthetic mechanisms. The study included only adults, limiting conclusions about developmental trajectories and whether atypical temporal calibration emerges early or reflects lifelong adaptation. Autistic individuals demonstrate enhanced cross-modal context effects that exceed those observed in synesthesia, consistent with a more amodal temporal magnitude code in autism.
Emotion dysregulation and sleep difficulties are highly prevalent transdiagnostic features in autism and Attention Deficit Hyperactivity Disorder (ADHD). Yet the differential mechanisms underlying these dimensions remain poorly understood, particularly in adults. This study investigates the specific profiles of emotion dysregulation (ED) and sleep difficulties in autistic adults, adults with ADHD, adults with co-occurring autism and ADHD, and controls, and examines two reciprocal mediation models linking neurodevelopmental features, ED, and insomnia symptom severity. 858 adults (autism, n = 163; ADHD, n = 235; co-occurring autism and ADHD, n = 186; neurotypical controls, n = 274) completed validated self-report questionnaires assessing core neurodevelopmental symptoms, alexithymia, ED, sleep quality, insomnia severity, and chronotype. Between-group comparisons were conducted using General Linear Models controlling for age and gender. Within-group mediation analyses were performed using bias-corrected bootstrap methods. Autistic adults and adults with co-occurring autism and ADHD displayed the highest overall ED, driven primarily by alexithymia (particularly difficulties identifying and describing feelings) and reduced use of cognitive reappraisal. Sleep efficiency was specifically reduced in autistic adults and in those with co-occurring autism and ADHD. Adults with ADHD showed a distinct profile characterized by affective lability, emotional over-reactivity, prolonged sleep latency, and evening chronotype. Adults with co-occurring autism and ADHD showed the most severe profiles across most domains. Mediation analyses revealed that across all clinical groups, ED mediated the relationship between core neurodevelopmental features and insomnia severity, with alexithymia emerging as an additional indirect predictor in autism and in the co-occurring group. All measures relied on self-report. The sample was predominantly female, which limits generalizability to men and gender-diverse individuals. The cross-sectional design prevents causal inference, and longitudinal studies are needed to confirm the directionality of these relationships. These findings identify ED as a transdiagnostic mediating pathway between neurodevelopmental features and sleep difficulties, while highlighting condition-specific mechanisms. Specifically, in ADHD, core symptoms may foster emotional hyperarousal and lability that interferes with sleep initiation, whereas in autism, alexithymia-related difficulties in attending to and appraising internal signals may sustain poorly integrated interoceptive states incompatible with consolidated sleep. The study was approved by the Ethics Committee of the University of Strasbourg (CER#2023–37).
Autism spectrum disorder (ASD) is characterized by marked difficulties in social interaction, with atypical eye-gaze behavior as a core feature. Eye-tracking studies have consistently reported reduced attention to social information in ASD across a large range of paradigms, suggesting that social perception abnormalities span multiple levels of processing, from initial orienting toward socially salient stimuli to fine-grained allocation of attention to social cues. Most studies have relied on group-level comparisons, and how these levels relate at the individual level remains unclear, particularly regarding inter-individual variability and whether these processes rely on shared mechanisms across development. We collected eye-tracking data from 150 participants with ASD and typical development (TD), from 2.6 to 26.7 years old, using two paradigms: (i) a preferential looking task contrasting human motion versus geometric motion to assess social orienting, and (ii) a naturalistic paradigm with complex social scenes to measure attention to socially informative regions (eyes). Inter-individual variability was characterized using Shapiro–Wilk and Levene’s tests. Group differences and associations between paradigms were analyzed using general linear models controlling for age and sex. Individuals with ASD looked significantly less to human motion (p = 1.4 × 10⁻⁵) and to the eyes of characters (p = 4.6 × 10⁻⁴) compared to TD participants, across the full age range. Both groups exhibited substantial inter-individual variability in gaze behavior, with continuous and normally distributed patterns. A significant positive association was found between social orienting and attention to the eyes across all participants (p = 0.004), independent of age, sex, and diagnosis. These findings show that initial social orienting and later fine-grained social attention are linked at the individual level, supporting an integrated multi-level model of social perception. Preferential looking differences in ASD extend beyond early childhood, and the results reveal within-subject coherence across levels of social processing. Substantial inter-individual variability further supports a dimensional organization of social perception across development. Even though the cross-sectional design precludes conclusions about developmental trajectories and heterogeneity in cognitive and language abilities was not accounted for, together, these findings provide a framework for linking behavioral markers of social perception to underlying neurodevelopmental mechanisms.
Aggressive behaviors represent a common and clinically challenging comorbidity in individuals with autism spectrum disorder (ASD), yet the underlying neural mechanisms remain poorly understood. This study aimed to investigate the circuit-level pathology driving elevated aggression in Neuroligin-3 R451C knock-in mice (NL3R451C or KI mice). Aggressive behaviors in adult male mice were assessed using the resident-intruder and self-initiated aggression tests. Following the R-I test, c-Fos brain mapping was performed. Fiber photometry recording was used to capture neuronal activity in selected brain regions during attack bouts. Chemogenetic inhibition was applied to establish the causal role of specific neuronal subtypes or circuits in elevated aggression. Additive NL3 expression was used to examine the genetic contribution of distinct neuronal subpopulations to maladaptive aggression. NL3R451C mice exhibited heightened aggression and aggression-seeking behavior, accompanied by increased c-Fos expression in multiple brain regions, including the lateral hypothalamus (LH), prelimbic cortex (PrL), and ventral tegmental area (VTA). During attacks, KI mice showed enhanced calcium transients in LH GABAergic neurons and in PrL→LH glutamatergic neurons, but not in LH glutamatergic neurons. Critically, chemogenetic suppression of LH GABAergic neurons, or of LH neurons projecting to the VTA or periaqueductal gray (PAG), attenuated excessive aggression in KI mice, whereas inhibition of PrL→LH neurons had no effect. Moreover, additive NL3 expression specifically in LH GABAergic neurons, but not in PrL neurons, was sufficient to ameliorate the aggressive phenotype. First, the synaptic mechanisms underlying LH GABAergic dysregulation remain to be defined by direct electrophysiological characterization. Second, LH GABAergic neurons are heterogeneous, and our approach did not identify which specific subtype mediates excessive aggression. Third, our study used a male‑only design, reflecting the X‑linked clinical presentation of the NL3 R451C mutation. Future studies in female mutants could provide insights into sexual dimorphism and the underlying mechanisms. Fourth, NL3 rescue was achieved via viral‑mediated adult additive expression rather than developmental precisely genetic correcting. We conclude that hyperactive LH GABAergic neurons, caused by NL3 loss, is a core pathological mechanism underlying abnormal aggression in NL3R451C mouse model of ASD.
