We investigate the capabilities of automatic nonverbal behavior descriptors to identify indicators of psychological disorders such as depression, anxiety, and post-traumatic stress disorder. We seek to confirm and enrich present state of the art, predominantly based on qualitative manual annotations, with automatic quantitative behavior descriptors. We propose nonverbal behavior descriptors that can be automatically estimated from audiovisual signals, including facial expressions, gestures, and acoustic characteristics.

We introduce a new dataset called the Distress Assessment Interview Corpus (DAIC) which comprises 100+ dyadic interactions between a confederate interviewer and a paid participant as well as 100+ interactions between a virtual human designed for psychological screening and a participant. Our evaluation on this dataset shows correlation of automatic behavior descriptors with specific psychological disorders as well as a generic distress measure. We investigate the capability of automatic algorithms to classify distressed behaviors in speaker-independent experiments. Additionally, we examine the impact of the posed questions' affective polarity, as motivated by findings in the literature on positive stimulus attenuation and negative stimulus potentiation in emotional reactivity of psychologically distressed participants.