research: real-time fMRI

overview of real-time fMRI neurofeedback

Real-time fMRI

Unlike standard MR imaging, real-time fMRI techniques allow ongoing changes in brain activation to be monitored and used throughout an experiment. Potential applications include QA monitoring, neurofeedback, and dynamic experimental control.

My work focuses on developing comprehensive, yet flexible real-time fMRI software tools, and using these methods to ask questions about trained self-regulation of neuromodulatory regions.

cognitive neurostimulation of the VTA

Research Overview

Can healthy individuals learn to generate and sustain activation in the dopaminergic midbrain? To test this question we provided participants with real-time neurofeedback from the Ventral Tegmental Area (VTA), and asked them to generate effective strategies for maximizing activation. Participants who received accurate VTA neurofeedback increased activation during training, as well as afterward, without feedback. The novel demonstration that humans can volitionally activate neuromodulatory regions holds promise for new educational and clinical interventions.


MacInnes JJ*, Dickerson KC*, Chen N, Adcock RA. (2016) Cognitive Neurostimulation: Learning to Volitionally Sustain Ventral Tegmental Area Activation. Neuron, 89(6); [*co-first authors]

video abstract

Pyneal: rt-fMRI software development

Pyneal: Real-time fMRI analysis software

Pyneal is a set of python-based tools I developed for accessing and analyzing fMRI data in real-time. It offers pre-built analysis routines, as well as the option of writing custom statistical models (using a provided template) that can be executed on each incoming volume. The output of the analyis is hosted via its own server, allowing experimental presentation software to access the results as needed.


  • Python-based GUI and tools for interfacing with the scanner and analyzing incoming fMRI data
  • Built-in analysis options including incremental GLM and summary statistics on specified ROIs
  • Custom stats module, allowing users to write their own python-based analysis routines (template and sample scripts included)
  • Matlab/PsychToolbox functions for retrieving data throughout task presentation
  • Sample PsychToolbox code for feedback displays
  • Scan simulator for testing and debugging