Abstract
<p>This reflection critically examines McChristie and Wilder's (2026) study, in which autistic children with expressive language support needs were repeatedly instructed to undress and redress by a familiar adult therapist as part of an abuse-prevention intervention. While acknowledging the researchers' protective intentions, we argue the study reflects a longstanding tendency in applied behavior analysis to position autistic children as objects of intervention rather than agents in their own safety. Using Causal Agency Theory and Milton's double empathy problem, we identify three underlying assumptions: abuse as a discrete rather than relational event, the remediation of compliance through further compliance-based instruction, and children as subjects rather than self-determined agents. We argue these assumptions risk reproducing the very dynamics the intervention sought to prevent. True protection requires centering autistic children's self-knowledge, communication, and agency rather than shaping their compliance to adult-defined rules.</p>