Video-Based Analysis of Language and Pre-Pragmatic Skills for Identifying Autism Risk in Infancy: A Case-Observation...

Early childhood is a critical period for optimizing developmental outcomes. Although early social-communication differences may be observable before a formal diagnosis of autism spectrum disorder (ASD), clinical diagnosis often occurs around age three. This study compared retrospective home-video indicators of pre-pragmatic, receptive-language, and expressive-language behaviors in infants aged 0-12 months who were later diagnosed with ASD and typically developing (TD) peers. The aim was to identify group-level candidate early behavioral indicators rather than to evaluate diagnostic or screening accuracy.

A single-center retrospective case-control design was used. Home videos recorded from birth to 12 months were analyzed for 30 children (ASD, n = 15; TD, n = 15). Communication behaviors were coded as present or absent using structured checklists across three domains: receptive language (61 items), expressive language (41 items), and pre-pragmatic skills (19 items). Item-level between-group differences were tested using two-sided Fisher's exact tests; effect sizes were estimated as odds ratios (ORs) with 95% confidence intervals (CIs). Benjamini-Hochberg false discovery rate (FDR) correction was applied within each domain.

Of the 121 assembled outcomes, 120 were analyzable. Overall, 63 of 120 items (52.5%) differed between groups at p < .05, and 55 of 120 (45.8%) remained significant after FDR correction. The highest proportion of FDR-significant differences was observed in receptive language (35/61, 57.4%), followed by expressive language (15/41, 36.6%) and pre-pragmatic skills (5/18, 27.8%). Directionality was highly consistent, with 54 of 55 FDR-significant items showing lower endorsement in the ASD group.

Infants later diagnosed with ASD showed lower endorsement across a broad range of pre-pragmatic and early language behaviors, particularly in receptive language. These findings identify candidate early behavioral indicators and support the potential adjunctive value of retrospective home-video review in developmental history taking. They do not validate a screening or diagnostic method.