Guiding Philosophy

The goal of cognitive psychology and neuroscience is to understand the mind and brain, and in practice this means studying specific components of cognition, such as attention, perception, learning, and memory. Often, though, these components are studied in isolation, and we risk missing the forest for the trees. The overarching theme of our research is that cognitive processes are inherently dynamic and interactive, and that exploring their behavioral and neural interactions can be an especially effective way to understand how they work. Below are some of our research programs that highlight this philosophy.

Statistical Learning

We repeatedly encounter the same people, places, and things, and over time they tend to show up in the same configurations and sequences. The process by which we automatically detect and represent these regularities is known as statistical learning. When encountering a new environment, statistical learning can help learn the relative locations of objects in a room, the boundaries between words in a language, and the sequence of landmarks on the way home.

We take a multidisciplinary approach to studying statistical learning, including behavioral tasks, fMRI in healthy adults, computational modeling with neural networks, and studies of patients with brain lesions or implants. These complementary techniques have led to progress in understanding how statistical learning works, how it is supported in the brain, and how it influences other cognitive processes. For example:

Schapiro, A. C., Turk-Browne, N. B., Botvinick, M. M., & Norman, K. A. (2017). Complementary learning systems within the hippocampus: A neural network modeling approach to reconciling episodic memory with statistical learning. Philosophical Transactions of the Royal Society B, 372, 20160049.

Graves, K. N., Antony, J. W., & Turk-Browne, N. B. (2020). Finding the pattern: Online extraction of spatial structure during virtual navigation. Psychological Science, 31, 1183–1190.

Predictive Coding

What is the purpose of statistical learning and other forms of learning and memory? Intuitively, they help us understand the world and remember the past. But no two situations ever repeat exactly, so their real value is in allowing us to build on past experience to generate predictions about the future. These predictions not only help facilitate behavior, as our brain gets a head start on what is about to happen, but also help refine knowledge by testing out and revising mental hypotheses.

Prediction is typically studied in the context of perception, how expectations influence sensory processing. We have studied where these predictions come from. Treating prediction as an interface between perception and memory, we have shown with many of the techniques described above how past episodes and regularities are retrieved and reinstated in the brain. For example:

Kok, P., & Turk-Browne, N. B. (2018). Associative prediction of visual shape in the hippocampus. Journal of Neuroscience, 38, 6888-6899.

Hindy, N. C., Ng, F. Y., & Turk-Browne, N. B. (2016). Linking pattern completion in the hippocampus to predictive coding in visual cortex. Nature Neuroscience, 19, 665-667.

Infant Development

The last couple of decades have led to a revolution in our understanding of the mind and brain in mature adults and in development from childhood through adolescence. These advances were enabled by the availability of new tools in cognitive neuroscience that safely record activity in the healthy human brain, particularly the advent of fMRI. However, because of the challenges of using these tools in infants and toddlers, much less is known about how their brains work.

We have been particularly interested in questions of learning and memory, given that infancy is an especially remarkable period. This is when we acquire language, learn how to walk, build social bonds, and develop preferences and feelings. This has required a broad effort to make cutting-edge fMRI techniques for data acquisition and analysis feasible in infants who are awake and performing cognitive tasks. This long-term project is in early stages, but we have released theoretical and methodological perspectives (along with some datasets and software to help build a community):

Ellis, C. T., & Turk-Browne, N. B. (2018). Infant fMRI: A model system for cognitive neuroscience. Trends in Cognitive Sciences, 22, 375-387.

Ellis, C. T., Skalaban, L. J., Yates, T. S., Bejjanki, V. R., Córdova, N. I., & Turk-Browne, N. B. (2020). Re-imagining fMRI for awake behaving infants. Nature Communications, 11, 4523.

Cognitive Training

Outside of developmental research, cognition is often treated as static. We are able to see and hear certain things and not others, pay attention for only so long, remember trivial facts but forget vital information, and make both good and regrettable decisions. Are we stuck like this in adulthood, or is there room for self-improvement? The challenge in trying to train cognition is that, if easy, it would have already happened — we have countless hours perceiving, attending, remembering, etc. Practice alone, of the everyday variety, does not help much. There has been recent success in training cognition through adaptive behavioral interventions, including action video game playing.

We take a different approach of training cognition by more directly changing the brain. This involves a different kind of fMRI study, known as real-time fMRI, where brain data are measured and analyzed immediately. The results of the analysis are used to update the ongoing task, often referred to as neurofeedback. In this way, good and bad brain states can be rewarded and penalized, respectively. This research is ongoing, but there are some promising indications that attention and memory might be malleable:

deBettencourt, M. T., Cohen, J. D., Lee, R. F., Norman, K. A., & Turk-Browne, N. B. (2015). Closed-loop training of attention with real-time brain imaging. Nature Neuroscience, 18, 470-475.

deBettencourt, M. T., Turk-Browne, N. B., & Norman, K. A. (2019). Neurofeedback helps to reveal a relationship between context reinstatement and memory. NeuroImage, 200, 292-301.

Research

Guiding Philosophy

Statistical Learning

Predictive Coding

Infant Development

Cognitive Training

Who has sponsored our research?

Where has our research been covered?