From heterogeneity to idiosyncrasy in the autistic brain
摘要
The profound heterogeneity of the autism ‘spectrum’ has hindered the search for unifying mechanisms, leaving the promise of subtyping unfulfilled. Here we argue this stems from focusing on eliminating, rather than explaining, variability. We propose a fundamental reframing, elevating idiosyncrasy—stable, person-specific neural, cognitive and behavioral patterns—from statistical noise to the core signal to be understood. We introduce a dynamical systems perspective where autistic neurodevelopment follows an altered stochastic trajectory. This process generates a unique distribution of outcomes—featuring both a recognizable central tendency (the prototype) and broad, heavy tails (idiosyncrasy). This structured variability implies embedded threshold effects rather than discrete subtypes or simple Gaussian-like variation (a dimensional continuum). This framework explains why group-average approaches are often inconclusive and necessitates a shift toward individual-level, non-Gaussian analytical tools. Embracing idiosyncrasy as a hallmark of more rigorously defined autism can resolve long-standing paradoxes, advance research, foster personalized support, and bridge the gap between biomedical science and the neurodiversity movement.