Project Details
Projekt
Determining potential behavioural markers for diagnostics in autism through virtual reality (VIDA)
Subject Area
Personality Psychology, Clinical and Medical Psychology, Methodology
Term
since 2025
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 544704001
The VIDA project is a basic research study addressing the diagnostic challenges associated with autism by using virtual reality (VR) technology and machine learning (ML). For Autism, a lifelong neurodevelopmental condition, we lack biomarkers, making diagnosis reliant on observational and developmental information. The absence of effective diagnostic tools for adults and the rising prevalence of autism highlight the need for innovative approaches. VIDA utilizes a controlled VR social interaction scenario, capturing individual nonverbal behaviors such as eye gaze, head and body movements. These behaviors, central to autism symptomatology, will be analyzed for their diagnostic classification potential. Importantly, the project includes a clinical comparison group of individuals with Borderline Personality Disorder (BPD) to assess the specificity of identified markers. ML, specifically recurrent neural networks, will be employed to classify autism, BPD, and typically developed comparison groups based on the collected data. The ML model will be externally validated on a new sample of individuals with autism at a different site. The research plan involves: (1) Collecting nonverbal data from the autism, BPD, and comparison sample groups, (2) training ML models for classification and validating results against existing assessments, (3) external validation of the ML model using a separate sample and (4) dissemination of results for future clinical-translational research and engagement with stakeholders. The significance of VIDA lies in its potential to offer objective and reliable markers for autism diagnosis from a very controlled VR environment allowing for in-depth analysis. The precisely determined patterns of individual markers in this basic research study will contribute to a deeper understanding of autism communication dynamics and will enable future clinical-translational research to employ and test the established markers in either VR or natural interaction settings. The project's non-invasive methods, coupled with the expertise of the research team and collaborations, ensure the feasibility and reliability of the proposed study. The outcomes of VIDA could significantly advance our understanding of autism, contribute to diagnostic improvements, and influence broader applications in clinical psychology and psychiatry.
DFG Programme
Research Grants
Co-Investigators
Professor Dr. Daniel Roth; Professor Dr. Kai Vogeley
