I am broadly interested in how experience alters memory and perception. I investigate this topic using fMRI studies, computational modeling, and behavioral studies.
Under what circumstances does experience cause memories to differentiate?
Competition plays a key role in driving learning. However, the literature describes contradictory results as it investigates how competition effects memories-- in some cases, learning further integrates linked stimuli, and in others, it differentiates them. So, at times, learning seems to increase the neural overlap of the things being learned, and at others, learning decreases that neural overlap. I'm using neural net models to illustrate how similar memories become less similar in the brain under certain conditions. This will hopefully enhance our understanding of the seemingly contradictory ways in which competition effects memory.
How can real-time neurofeedback train novel visual representations of categories?
We developed methods to decode categories in the brain during an fMRI scan and display the results to the subject in real-time as the scan proceeds. We used this technique to train novel visual categories in the brain, by rewarding subjects with positive feedback based on their neural activity. We started out with several stimuli that created diffuse neural activity, and by using live feedback reinforcement, trained individual stimuli to trigger pre-selected regions of brain activity.
How does our ability to extract statistical structures from our environment impact the storage of memories?
We use statistical regularities we discover in our environment as a heuristic to speed up our thinking, judgement, and ability to predict behaviors. We do this on a number of dimensions: length, speed, orientation, position, etc. This perceptual averaging in turn can affect our memories of several examples within a group, as we recall an item as being closer to the mean of the group to which it belongs. I studied how time perception compared with visual perception in regard to perceptual averaging.