Identification of markers for autism-spectrum disorder traits

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder that is characterized by severe social and functional deficits. Importantly, individuals with psychiatric disorders express elevated levels of ASD exacerbating social isolation, however, ASD is often not recognized and thus, deficits are not treated adequately. Evidence points towards the diagnostic relevance of different ASD markers including voice, gaze, attention to social-emotional cues as well as information processing. However, these indicators remain fragmented. Accordingly, there is a lack of combining multidimensional markers of ASD to assist objective identification of ASD-like traits in a comprehensive approach. In the planned project, we aim to assess the effectiveness of measures that can be captured with widely accessible digital devices as objective markers to screen the expression of ASD traits in a cohort of individuals with psychiatric disorders. In addition, we will combine multimodal inputs with voice samples of the participants, post process the obtained data, and enter it in a convolutional neural network to generate probabilities of belonging on the autism spectrum. Soft- and hardware for the computerized testing will be provided by the company Vitafluence.