Joint attention in autism: A narrative review of assessment techniques from behavioral observation to artificial intelligence.

Joint attention is when two people look at the same thing together, like when you and your child both look at a toy. This skill is often challenging for autistic children but is important for communication and learning. This review looked at different ways researchers measure joint attention, from watching children directly to using computer programs and AI. Understanding how to better assess joint attention can help with earlier diagnosis and better support for autistic children.

This narrative review examines the evolution of joint attention assessment methods for children with autism spectrum disorder from 2002-2024. Joint attention, the ability to share focus with others on objects or events, is crucial for social communication and language development but is frequently impaired in autism. The review categorizes assessment approaches into human-mediated techniques and technology-driven methods, including emerging artificial intelligence applications. The authors analyze various methods' target populations, data collection processes, and validation strategies, identifying strengths and limitations of existing approaches to inform future research directions and early intervention strategies for children with ASD.

The review identifies assessment approaches that may inform clinical practice for early autism diagnosis and intervention planning. Technology-assisted methods, including AI, may offer more objective joint attention assessment tools, potentially improving diagnostic accuracy and intervention targeting for children with ASD.

As a narrative review, this study does not provide quantitative synthesis of results or systematic quality assessment of included studies. The review methodology and specific findings from individual studies are not detailed in the abstract, limiting evaluation of evidence quality.

Joint attention (JA), the shared focus between two individuals on an object or event, plays a pivotal role in social communication, cognitive development, and language acquisition during early childhood. However, JA is frequently impaired in children with autism spectrum disorder (ASD), highlighting the need for precise assessment to support early diagnosis and intervention. This narrative review explores the evolution of JA assessment methods, tracing the shift from human-mediated techniques to technology-driven approaches, including artificial intelligence (AI). The study analyzes research indexed in major bibliographic databases between 2002 and 2024, categorizing findings into human-mediated and technology-assisted methods.

Key aspects such as target populations, data collection processes, and validation strategies are examined. By highlighting the strengths and limitations of existing approaches, the review identifies future research directions that can advance JA assessment and inform early intervention strategies, ultimately benefiting children with ASD and their families.