Natural Language Processing (NLP) in mental health has largely focused on social media data or classification problems, often shifting focus from high caseloads or domain-specific needs of real-world practitioners. This study utilizes a dataset of 644 participants, including those with Bipolar Disorder, Schizophrenia, and Healthy Controls, who completed tasks from a standardized mental health instrument. Clinical annotators were used to label this dataset on five clinical variables. Expert annotations across five clinical variables demonstrated that contempo- rary language models, particularly smaller, fine-tuned models, can enhance data collection and annotation with greater accuracy and trust than larger commercial models. We show that these models can effectively capture nuanced clinical variables, offering a powerful tool for advancing mental health research. We also show that for clinically advanced tasks such as domain-specific annotation LLMs provide wrong labels as compared to a fine-tuned smaller model.

Contemporary NLP has rapidly progressed from feature-based classification to fine-tuning and prompt-based techniques leveraging large language models. Many of these techniques remain understudied in the context of real-world, clinically enriched spontaneous dialogue. We fill this gap by systematically testing the efficacy and overall performance of a wide variety of NLP techniques ranging from feature-based to in-context learning on transcribed speech collected from patients with bipolar disorder, schizophrenia, and healthy controls taking a focused, clinically-validated language test. We observe impressive utility of a range of feature-based and language modeling techniques, finding that these approaches may provide a plethora of information capable of upholding clinical truths about these subjects. Building upon this, we establish pathways for future research directions in automated detection and understanding of psychiatric conditions.

NLP offers a myriad of opportunities to support mental health research. However, prior work has almost exclusively focused on social media data, for which diagnoses are difficult or impossible to validate. We present a first-of-its-kind dataset of manually transcribed interactions with people clinically diagnosed with bipolar disorder and schizophrenia, as well as healthy controls. Data was collected through validated clinical tasks and paired with diagnostic measures. We extract 100+ temporal, sentiment, psycholinguistic, emotion, and lexical features from the data and establish classification validity using a variety of models to study language differences between diagnostic groups. Our models achieve strong classification performance (maximum F1=0.93-0.96), and lead to the discovery of interesting associations between linguistic features and diagnostic class. It is our hope that this dataset will offer high value to clinical and NLP researchers, with potential for widespread broader impacts.