Proceedings of the LREC 2020 Workshop on: Resources and Processing of Linguistic, Para-linguistic and Extra-linguistic Data from People with Various Forms of Cognitive/Psychiatric/Developmental Impairments (RaPID-3)
RaPID-3 aims to be an interdisciplinary forum for researchers to share information, findings, methods, models and experience on the collection and processing of data produced by people with various forms of mental, cognitive, neuropsychiatric, or neurodegenerative impairments, such as aphasia, dementia, autism, bipolar disorder, Parkinson's disease or schizophrenia. Particularly, the workshop's focus is on creation, processing and application of data resources from individuals at various stages of these impairments and with varying degrees of severity. Creation of resources includes e.g. annotation, description, analysis and interpretation of linguistic, paralinguistc and extra-linguistic data (such as spontaneous spoken language, transcripts, eyetracking measurements, wearable and sensor data, etc). Processing is done to identify, extract, correlate, evaluate and disseminate various linguistic or multimodal phenotypes and measurements, which then can be applied to aid diagnosis, monitor the progression or predict individuals at risk.
Disorganized, or incoherent, speech is one of the important criteria for diagnosing schizophrenia. However, there is still a lack of a rather quick objective method of measuring speech coherence. Automated discourse analysis is a possible solution to this problem. We analyzed discourse coherence in a set of spoken narratives by people with schizophrenia and neurotypical speakers of Russian. All narratives were manually rated for violations of completeness, local, global and dimensional coherence. A number of automated vector semantics methods were used for approximation of the manual rating scores. The metrics used proved to be a good approximation for manual scoring, and a combination of them was efficient for classification narratives in schizophrenia and neurotypical groups.