Technical Program

Paper Detail

Paper: PS-1B.13
Session: Poster Session 1B
Location: Symphony/Overture
Session Time: Thursday, September 6, 18:45 - 20:45
Presentation Time:Thursday, September 6, 18:45 - 20:45
Presentation: Poster
Publication: 2018 Conference on Cognitive Computational Neuroscience, 5-8 September 2018, Philadelphia, Pennsylvania
Paper Title: Temporal structure of learning to regulate ventral tegmental area using real-time fMRI neurofeedback
Manuscript:  Click here to view manuscript
Authors: Shabnam Hakimi, Duke University, United States; Jeffrey MacInnes, University of Washington, United States; Kathryn Dickerson, Alison Adcock, Duke University, United States
Abstract: The ventral tegmental area (VTA) and its dopaminergic projections are central to volitional behavior. Previous research from our group demonstrated that individuals can use real-time neurofeedback training to learn to reliably self-activate the VTA using self-generated motivational imagery (MacInnes & Dickerson et al., 2016). The mechanism of learning, however, is not yet known. Here, we investigated how the temporal structure of neurofeedback training impacts successful transfer of VTA self-activation. We analyzed veridical VTA neurofeedback during self-activation trials in one of three temporal contexts (individual trial, scanning run, or full training session) to test the extent to which slope of VTA response over time during each context explains change in self-activation ability. A comparison of the model evidence suggested that, relative to trial and run, the full session best explained the magnitude of transfer from pre- to post-training (p < 0.01). These preliminary data suggest that the overall training context may be a better predictor of learning from VTA neurofeedback than individual training episodes—regardless of their success.