Resources and ProcessIng of linguistic, para-linguistic and extra-linguistic Data from people with various forms of cognitive/psychiatric/developmental impairments
RaPID-6@MENTAL.ai, half-day event, XXXday XX of May 2026, Palau de Congressos de Palma, Palma, Mallorca, Spain
Healthcare professionals and clinicians are increasingly inclined to employ non-invasive, cost-effective, and easily measurable techniques as a complement to existing medical and clinical evaluations for the early diagnosis and monitoring of brain and mental disorders.
Although many causes of cognitive and neuropsychiatric impairments are difficult to anticipate or predict accurately, physicians and clinicians consider a broad range of potentially contributing factors, including traumatic brain injuries, genetic predispositions, medication side effects, and congenital anomalies. Emerging evidence suggests that the collection and analysis of human language data (e.g., spontaneous storytelling) together with extra-linguistic and production-based measures (e.g., eye-tracking, wearable devices, or sensor data) can serve as a valuable complement to clinical diagnosis and may provide the basis for developing future objective criteria for identifying progressive decline or degeneration in mental and brain functioning.
An important area of research in computational linguistics and Natural Language Processing (NLP) focuses on the processing, analysis, and interpretation of such data. Current work in this field, based on linguistically oriented analyses of text and speech produced by clinical populations in comparison to healthy adults, has yielded promising results. These include early diagnosis and risk prediction, differentiating individuals across varying degrees of severity, sub-typing clinical groups, and monitoring disease progression through longitudinal analyses of language samples and para- or extra-linguistic measurements from multiple modalities. Moreover, the development of robust computational tools for linguistic analysis relies heavily on solid infrastructures that support efficient processing and storage of the language data involved. Such infrastructures not only facilitate advances in NLP and AI but also play a crucial role in enabling collaborative research by supporting the seamless exchange of methods and findings across diverse linguistic datasets and analytical frameworks.
Nevertheless, substantial work remains to achieve more accurate predictive estimates and fine-grained classification frameworks. Further research is needed to ensure that such methods can reliably complement the existing battery of medical and clinical examinations used for the early diagnosis and monitoring of neurodegenerative and other brain and mental disorders. In turn, this line of work may contribute to the development of large-scale, non-invasive, time- and cost-effective, and objective future clinical tests in neurology, psychology, and psychiatry.
In collaboration with the MENTAL.ai project (part of the Caen Strategy for Accelerating Research – CaeSAR initiative), the RaPID-6@MENTAL.ai@LREC workshop underscores the central role of AI-driven methods in advancing research and clinical applications for mental health and cognitive impairments. MENTAL.ai brings together a multidisciplinary team to push forward AI-powered 6P medicine—enhancing early detection, monitoring, diagnosis, and treatment of psychiatric and neurodegenerative disorders—while developing multimodal datasets, digital phenotyping tools, generative clinical-note systems, and ethical frameworks for trustworthy, human-centered neuropsychiatric care.
RaPID-6@MENTAL.ai & LREC-2026 is supported by:
