The main aim of the present study is to investigate the effects of a Motivational Interviewing-based artificial intelligence chatbot on social incentive processing in college students with elevated levels of depression and anhedonia by combining a randomized active-control intervention design with pre- and post-intervention Social Incentive Delay Task assessments during fMRI.
Anhedonia represents a core characteristic of depression and is characterized by reduced experience of pleasure. It is closely related to decreased motivation, altered reward processing, changes in affective responsiveness, and alterations in intrinsic brain network function. Anhedonia is not specifically targeted by currently available pharmacological interventions. Initial evidence indicates that an increased willingness to change and implementation of change in daily life can alleviate anhedonia. The present study aims to examine whether a Motivational Interviewing-based AI chatbot can lead to changes in social incentive processing in college students with elevated anhedonia and depressive symptoms. Social incentive processing is included because social approval, social feedback, and interpersonal reward are important sources of motivation in daily life and may be altered in individuals with elevated anhedonia. The Social Incentive Delay Task allows the study to examine behavioral and neural responses during the anticipation and receipt of social incentives. To this end, eligible participants with a total score of 22 or higher on the Snaith-Hamilton Pleasure Scale and a score of 14 or higher on the Beck Depression Inventory will undergo a randomized, between-subjects, active-control intervention study. Participants will be assigned to either a Motivational Interviewing-based chatbot group or an active control chatbot group for 1 week. Pre- and post-intervention assessments will include self-report questionnaires and the Social Incentive Delay Task during functional magnetic resonance imaging to examine psychological, behavioral, and neural effects of the intervention.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
DOUBLE
Enrollment
80
The experimental chatbot is designed to use principles of Motivational Interviewing to support participants in exploring their personal values, motivation for change, and daily behavioral goals related to pleasure, engagement, and reward-seeking. During the intervention period, participants will interact with the chatbot regularly through brief text-based conversations. The chatbot will provide empathic, non-judgmental responses, encourage reflection on current difficulties, and help participants identify small, feasible actions that may increase daily engagement and positive experiences. It will not provide diagnosis, crisis counseling, or medical treatment.
Participants will interact with a chatbot matched in format and frequency of use. This chatbot will provide neutral nature-related stories or general natural history content. It will be designed to maintain participant engagement while avoiding therapeutic techniques, motivational interviewing strategies, behavioral activation guidance, or personalized mental health advice. This active control condition will help control for nonspecific effects of chatbot interaction, attention, expectancy, and digital engagement.
University of Electronic Science and Technology of China
Chengdu, Sichuan, China
RECRUITINGNeural Activity During Social Incentive Anticipation
Participants will undergo task-based BOLD fMRI while completing the Social Incentive Delay Task. Neural responses during the anticipation of social incentive cues will be compared with responses to neutral cues using a general linear model. Pre-to-post intervention changes will be evaluated from the resulting beta contrast estimates.
Time frame: Baseline before the first chatbot interaction and Week 1 after completion of the chatbot intervention.
Neural Activity During Monetary Incentive Anticipation
Task-based BOLD fMRI data will be acquired while participants perform the Monetary Incentive Delay Task. Neural responses associated with the anticipation of monetary reward and punishment will be estimated using a general linear model. Beta contrast estimates will be calculated for monetary reward versus neutral and monetary punishment versus neutral conditions. These estimates will be used to assess changes between baseline and post-intervention.
Time frame: Baseline before the first chatbot interaction and Week 1 after completion of the chatbot intervention.
Behavioral Responsiveness to Social Incentive Cues
Reaction time, accuracy, and hit rate will be calculated separately for social incentive cue conditions and neutral cue conditions in the Social Incentive Delay Task before and after the intervention. Behavioral differences between social incentive and neutral cue conditions will also be derived and compared from baseline to post-intervention.
Time frame: Baseline before the first chatbot interaction and Week 1 after completion of the chatbot intervention.
Behavioral responsiveness to monetary incentives
The Monetary Incentive Delay Task will be administered before and after the intervention. Reaction time, accuracy, and hit rate will be extracted separately for monetary incentive cue conditions and neutral cue conditions, and changes in behavioral differences between monetary incentive and neutral cue conditions will be calculated.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Time frame: Baseline before the first chatbot interaction and Week 1 after completion of the chatbot intervention.