Broader autism phenotype, the default mode network, and brain entropy: A network-level hypothesis.

Autism spectrum disorder (ASD) is highly heritable, and autism-related traits extend beyond diagnostic thresholds into the broader autism phenotype (BAP), a set of subclinical social-communication, cognitive, sensory, and personality-related characteristics observed in some relatives of autistic individuals and in the general population. The BAP literature is heterogeneous: many family and twin studies report elevated trait levels among first-degree relatives, whereas others report null, domain-specific, or subgroup-specific effects. We therefore frame BAP as a probabilistic marker of shared liability rather than a universal familial profile or a missed diagnosis. Here, we integrate behavioral, cognitive, genetic, and neuroimaging evidence and propose, rather than establish, a network-level model in which atypical default mode network (DMN) organization and altered coupling with salience and central executive networks may help explain links among sensory integration, executive rigidity, and socio-affective/mentalizing traits. We also operationalize brain entropy as a family of neuroimaging metrics that quantify the irregularity, variability, or complexity of neural signals (e.g., sample entropy, multiscale entropy, wavelet-based regularity, and Lempel-Ziv complexity). Entropy may provide a testable bridge between neural dynamics and BAP-related behavioral heterogeneity; however, direct evidence in BAP remains limited. We conclude by outlining testable predictions, limitations, and future directions for multimodal, developmentally informed studies.