We introduce a transcript-based alexithymia profile: the Alexithymia Language Index (ALI). It uses interpretable language markers aligned with PAQ’s DIF, DDF, and EOT structure, focusing on auditable evidence rather than numeric scores.

A guide describing consent choices, data handling, audit logging, and proposed study modes for US-based research collaborations on Feelpath.

A study proposal asking one question: do personalized psychoeducation tools (AI topic+emotion summaries, a dynamic emotion wheel, emotion analytics, and emotion annotation) reduce alexithymia symptoms?

A multi-site validation program to evaluate convergent validity, reliability, and sensitivity-to-change of ALI-derived evidence markers and facet labels against PAQ-24 and human coding.

A study proposal to validate a transcript-based estimate of emotion skills (ELI) against PERS, PES, EBQ, and PERCI, with an auditable, quote-backed approach and a small human-coding audit.

A study proposal to evaluate conversation-derived psychoeducation (session reviews, emotion clarity, and guided annotation) introduced under therapist control, prioritizing feasibility, trust, and mechanism signals before any condition-specific outcome claims.