Explaining Models of Mental Health via Clinically Grounded Auxiliary Tasks

Abstract

Models of mental health based on natural language processing can uncover latent signals of mental health from language. Models that indicate whether an individual is depressed, or has other mental health conditions, can aid in diagnosis and treatment. A critical aspect of integration of these models into the clinical setting relies on explaining their behavior to domain experts. In the case of mental health diagnosis, clinicians already rely on an assessment framework to make these decisions; that framework can help a model generate meaningful explanations.In this work we propose to use PHQ-9 categories as an auxiliary task to explaining a social media based model of depression. We develop a multi-task learning framework that predicts both depression and PHQ-9 categories as auxiliary tasks. We compare the quality of explanations generated based on the depression task only, versus those that use the predicted PHQ-9 categories. We find that by relying on clinically meaningful auxiliary tasks, we produce more meaningful explanations.

A review of the explainability and safety of conversational agents for mental health to identify avenues for improvement

“…In addition, explainability saw an uptake of 23% in terms of safety across the same language models. Similar results were noticed when PHQ-9 was used in explainable training of language models (Zirikly and Dredze, 2022 ). Given these circumstances, VMHAs can efficiently integrate with clinical practice guidelines such as PHQ-9 and GAD-7, utilizing reinforcement learning.…”

“…Adherence can be thought of as aligning the question generation and response shaping process in a VMHA to external clinical knowledge such as PHQ-9. For instance, Roy et al and Zirikly et al demonstrated that under the influence of datasets grounded in clinical knowledge, the generative model of VMHA can provide clinician-friendly explanations (Zirikly and Dredze, 2022 ; Roy et al, 2023 ). Another form of adherence is in the form of regulating medication adherence in users.…”