This paper investigates the potential of generative artificial intelligence to emulate theoretically grounded patterns in parenting behavior across parent gender and child developmental status. Employing ChatGPT-4, we generated synthetic data modeled on the Parenting Authority Questionnaire–Revised (PAQ-R), guided by a comprehensive synthesis of the developmental and parenting literature. The dataset comprised 600 simulated parent profiles, encompassing mothers and fathers of children classified as typically developing, diagnosed with ADHD, or with autism spectrum disorder (ASD). A two-way multivariate analysis of variance (MANOVA) was conducted to evaluate hypothesized main and interaction effects. Results revealed statistically significant patterns that corresponded with established theoretical frameworks: authoritative parenting scores were highest among mothers of typically developing children, with notable reductions observed in neurodivergent child groups, while authoritarian parenting was disproportionately elevated among fathers of children with ADHD. These outcomes demonstrate that large language models can generate structured, theory-consistent datasets, underscoring their potential utility in developmental and family science for simulation-based hypothesis testing and scalable model validation.

Keywords: Parenting styles, ASD, ADHD, AI, Children.