Measuring activation during behavioral activation therapy: a proof-of-concept study using smartphone sensors and LLM-derived ratings in adolescents with anhedonia

TL;DR

This study shows that smartphones and AI can track therapy progress in teens with depression by measuring daily activity and mood changes.

Contribution

First proof-of-concept showing LLMs and smartphone sensors can validly measure behavioral activation in adolescent depression therapy.

Findings

GPT-4o ratings of free-text entries correlated with smartphone mobility data and self-reported activation.

Higher GPT-rated activation predicted better daily mood and reduced depressive symptoms.

Passive sensing features forecasted weekly improvements in anhedonia and depression.

Abstract

Adolescent depression remains a major public health concern, and Behavioral Activation (BA), a brief therapeutic intervention designed to reduce depression-related avoidance and boost engagement in rewarding activities, has shown encouraging results. Still, few studies directly measure the hypothesized mechanism of “activation” in daily life, especially using low-burden, ecologically valid methods. This proof-of-concept study evaluates the validity of two technology-based approaches to measuring activation in adolescents receiving BA: smartphone-based mobility sensing and large language model (LLM) ratings of free-response text. Adolescents (n = 38, ages 13–18) receiving 12-week BA therapy for anhedonia completed daily ecological momentary assessment (EMA) reporting on positive and negative affect. GPT-4o was used to rate behavioral activation from EMA free-text entries. A subsample (n = 13) contributed passive smartphone sensing data (e.g., accelerometer activity, GPS-derived mobility). Activation and symptoms were assessed weekly via self-report. GPT-derived activation ratings correlated positively with passive sensing indicators (number of places visited, time away from home) and self-reported activation. Within-person increases in GPT-rated activation were associated with higher daily positive affect and lower negative affect. Passive sensing features also forecasted weekly improvements in anhedonia and depressive symptoms. Associations emerged primarily at the within-person level, suggesting that changes in activation relative to one’s own baseline are clinically meaningful. This study demonstrates the feasibility and validity of passively measuring behavioral activation in adolescents’ daily lives using smartphone data and LLMs. These tools hold promise for advancing data-informed psychotherapy by tracking therapeutic processes in real time, reducing reliance on self-report, and enabling personalized, adaptive interventions. Clinical Trial Registry: NCT02498925.

Behavioral Activation therapy helps depressed teens by encouraging them to engage in more rewarding activities. We tested whether smartphones and LLM could automatically track this 'activation' in daily life. Using movement data from phones and LLM-based ratings from 38 teens in therapy, we found these digital tools accurately detected when teens were more active and higher activation predicted better mood and fewer depression symptoms. This technology could help therapists monitor progress in real-time and personalize treatment more effectively.