A systematic review of collaborative interventions between physical products and digital applications for children with autism spectrum disorder from an evidence-based perspective — Evidence accumulation characteristics and implications for design review

Integrated physical product-digital application systems support autistic children, but component collaboration and evidence generation remain underexamined. Following PRISMA 2020, six databases were searched (2015-2025); two reviewers screened studies involving autistic participants aged ≤ 18 years and appraised artifact and evaluation reporting using DSR-oriented appraisal. Of 7,245 records, 20 studies were included. Social interaction was the most common target, followed by emotional skills and attention. Studies were concentrated among school-aged children and in structured educational or clinical settings. Applications most frequently served as content or training hubs, followed by personalizers or decision engines; other roles were uncommon. DSR-oriented reporting adequacy was high in five studies, acceptable in four, and low in eleven. More complete evidence chains generally involved clearer component roles, stronger links between system functions and outcomes, and more transparent evaluation procedures. Research themes expanded over time but remained parallel and heterogeneous rather than forming a unified cumulative pathway. The findings suggest that progress may depend less on technological variety than on stable links among user requirements, system architecture, implementation conditions, and long-term outcome assessment. We propose an evidence-oriented heuristic framework covering deployment controllability, outcome verifiability and evaluation-chain complexity, which supports cumulative evidence generation for tangible-digital autism interventions.