This project will examine whether individuals can be trained, using real-time feedback about brain function during neuroimaging, to effectively remove thoughts from mind by providing them with a sense of what it feels like to successfully remove a thought.
On the first day, an anatomical MRI scan will be followed by a 12-min resting state scan, a 30-min item localizer task in which participants view a set of 80 scene stimuli 3 times each, and a 10-min memory test outside the scanner. The brain data from the localizer will be used to identify baseline measures of item-specific brain activity patterns. On 3 subsequent days (within the next 2 weeks), participants will receive neurofeedback training. In each 1-hr session, an anatomical scan will be followed by 8 runs (5 min each) of 20 trials, half with each operation, randomly ordered, and a 10-min behavioral memory test. In the scanner, participants will perform a working memory task in which they encode a scene image and then are instructed to either maintain that image in mind for a few seconds or to suppress that image from mind. At the end of each trial, neurofeedback will be provided to the participant visually to denote their "level of engagement" in the instructed operation on that trial. The neurofeedback is derived from the MRI classifier fit for the instructed operation (maintain or suppress) from a working memory operation classifier that is pre-trained on prior data. Participants (N=40) will be randomly assigned to either the positive or negative valence condition. They will encounter stimuli from their assigned valence only during the neuroimaging task, but from both valences during the behavioral tests, so that the investigators can evaluate whether any learning generalizes across emotional valence.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
NONE
Enrollment
40
fMRI neurofeedback is a technique that allows individuals to observe real-time feedback of their brain activity as measured by fMRI scans. During a session, the participant lies inside an MRI scanner while their whole brain is scanned. This data is processed in real-time by a computer and then presented to the participant in the form of visual feedback. This feedback might be presented as a graph, a color-coded representation, or some other easily interpretable format. The participant is instructed to modulate their brain activity based on the feedback they receive. This can involve various mental strategies, such as focusing attention on specific thoughts, images, or sensations. Over repeated sessions, participants learn to consciously influence their brain activity based on the feedback they receive. Through trial and error, they discover which mental strategies are most effective for achieving their desired changes in brain activity.
Neurofeedback learning
Visual fMRI neurofeedback (e.g., the diameter of a circle) will be provided to the participant at the end of each trial denoting the "level of engagement" in the instructed operation on that trial. The neurofeedback score is derived from the fMRI pattern classifier using regularized logistic regression for the instructed operation from a pre-trained working memory operation classifier. Scores range from 0 to 1, with higher scores indicating better fits to the target pattern of activity. The investigators will quantify neurofeedback learning by computing changes in feedback scores within each session and across the sessions.
Time frame: Within 2 weeks after the first day of the experiment, 3 neurofeedback training sessions lasting about 1 hour will occur at least 24 hours apart. The outcome measure will be computed at the end of all 3 sessions.
Neural consequences of control
Proactive interference for the encoding of trial N, separately following maintain trials and suppress trials, will be assessed using within-subject item-specific activity pattern matching between the localizer data and the main task data. Proactive interference would be reflected in a reduced correspondence (lower scores) between the localizer representation for an item and the representation of that item when it is encoded on trial N. Higher scores would reflect a release from proactive interference.
Time frame: Within 2 weeks after the first day of the experiment, 3 neurofeedback training sessions lasting about 1 hour will occur at least 24 hours apart. The outcome measure will be computed across all trials in each run, and for each of the 3 sessions.
Behavioral consequences of control
After each session, participants will be evaluated with behavioral memory tests on both positive and negative information. Two pictures (of the same valence) are presented for memorization on each trial, an instruction is given to manipulate the memory of one of those images, then a probe image appears and participants are to respond, as quickly and accurately as then can, YES if that probe image was one of the images presented at the beginning of the trial (even if the participant was asked to suppress that image) or NO otherwise. Half the probes contain new images. The accuracies and response times for this memory test will be statistically analyzed. The investigators will compare differences in these measures across condition (maintain vs. suppress) and cue type (manipulated vs. non-manipulated item). For the 3 neurofeedback sessions, the investigators will also relate a participant's neurofeedback scores with their behavioral performance on this memory test.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Time frame: This outcome measure will be collected at the end of the first day of the experiment and at the end of each of the 3 subsequent neurofeedback sessions occurring at least 24 hours apart within the next 2 weeks.
Training transfer across valence
The investigators will evaluate whether neurofeedback training transfers across emotional valence. Half of the participants will be trained on positive images, and half will be trained on negative images. The behavioral memory test at the end of each session includes both positive and negative images, and the outcomes measures from this test (as described in Primary Outcome 3) will be compared between these two groups of participants. Data will be statistically evaluated for whether the emotional valence on which a participant is trained has an effect that is selective (showing greater changes in accuracy and/or response times for trials of that same valence) or non-selective (no differences in performance metrics for trials with positive or negative valence).
Time frame: Within 2 weeks after the first day of the experiment, 3 neurofeedback training sessions lasting about 1 hour will occur at least 24 hours apart. This outcome measure will be collected at the end of each of the 3 sessions.