arXiv:2502.09487v3 Announce Type: replace-cross Abstract: Characterising how we verbalise our feelings is central to psychological assessment and intervention, yet the mapping between narrative and affective state remains poorly understood. Across two large studies (n=1257), we parameterised the structure and dynamics of depressive states by quantifying participants' internal narratives through large-language-model representations and their subspaces. In Study 1, we found verbal descriptions of symptom-specific thoughts captured granular information predictive of standardised, self-reported de
The proliferation of advanced large language models allows for unprecedented quantification and analysis of complex human verbal data, creating new avenues for psychological research.
This research suggests a scientific, scalable method for understanding and measuring affective states, potentially revolutionizing mental health assessment and intervention.
The ability to 'parameterize' internal narratives with LLMs transforms subjective psychological assessment into a more objective, data-driven science, offering new diagnostic and therapeutic tools.
- · Mental healthcare providers
- · AI/ML researchers in psychology
- · Pharmaceutical companies developing mental health treatments
- · Patients seeking personalised mental health care
- · Traditional, purely qualitative psychological assessment methods
- · Mental health diagnostic models lacking computational rigor
Individualized, data-driven mental health interventions become more feasible and effective.
Mental health diagnostics could integrate directly into AI-powered personal assistants or smart devices, offering continuous monitoring.
Ethical frameworks for AI in mental health become paramount, given the intimate access to individuals' internal states and narratives.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG