The role of superstition of cognitive control during neurofeedback training
Background: Real-time fMRI neurofeedback is growing in reputation as a means to alter brain activity patterns and alleviate psychiatric symptoms. Activity in ventral striatum structures is considered an index of training efficacy. fMRI response in these brain regions indicates neurofeedback-driven associative learning. Here we investigated the impact of mere superstition of control as observed during neurofeedback training on patterns of fMRI activation. Methods: We examined the brain activations of a large sample of young participants (n = 97, 50 female, age range 18-54yrs) in a simple fMRI task. Participants saw a display similar to that typically used for real-time fMRI. They were instructed to watch the bars' movements or to control them with their own brain activity. Bar movements were not connected with brain activity of participants in any way and perceptions of control were superstitious. After the pretended control condition, they rated how well they were able to control the bars' movements. Results: Strong activation in the basal ganglia and ventral striatum as well as in large portions of the anterior insula, supplementary motor area, and the middle frontal gyrus due to the superstition of brain control. Conclusions: The superstition of control over one's own brain activity in a pretended neurofeedback training session activates the same neural networks as neurofeedback-driven learning. Therefore, activity in the basal ganglia and ventral striatum cannot be taken as evidence for neurofeedback-driven associative learning unless its effects are proven to supersede those elicited by appropriate sham conditions.