Neurophysiology of Visual Perception

National Institute Of Mental Health

Investigators

Linked publications, trials & patents

Abstract

Vision holds reign over many aspects of human cognition, allowing us to make detailed observations of the world, its objects, spatial layout, and one another. Visual cognition infuses our thought processes, our memory, and our language. This human exceptionalism rests atop an impressive visual repertoire shared by many other mammalian species, who use vision for important aspects of their natural life. Our research investigates the neural mechanisms involved in basic perception, including that of objects, scenes, and individuals. Much of what we understand about this process comes from experiments in which isolated stimuli are presented onto a screen. This simplified stimulation method allows for the evaluation of responses across the brain. Over several decades, this approach has led researchers to conceive of vision as series of processing stages in the brain, whereby the stimulation of more and more specialized neurons lead to the subjective elements of our visual experience. While this has been our view as well, our recent foray into more naturalistic visual approaches, summarized in our recent review (Leopold and Park, 2020), has led us into new experimental domains. In one project, we have investigated whether there is a fundamental difference between continual visual experience, as for example experienced while watching a video, and the discrete stimulation of the brain that forms the basis of most visual neuroscience experiments. In our experiments, we have allowed subjects to freely view natural movies that play out over several minutes, record their gaze and brain activity during this time. In one study, which is now nearing submission, we determined that the continuity of vision itself strongly influenced the responses of the brain. For example, when the response selectivity of neurons to elements of a video was compared to the randomized presentation of those elements in isolation, many neurons showed minimal correspondence. This was particularly true for the onset response of the brief presentations, which had a near-zero correlation with the same neurons responses observed during continuous viewing. In another project, we are examining widespread components of the fMRI signal that are entrained by video viewing. This signal, elicited consistently by repeated viewings of the same videos, is expressed simultaneously across most visually responsive regions of the brain. This signal is not one that would be predicted by standard models of visual processing. Moreover, we have shown that it is synchronized with the activity of certain high-level visual neurons. This work is closely related to our other studies using the method of single-unit fMRI mapping, as well as to those studying the particular role of the basal forebrain in regulating network activity across the brain. In a third project, from which a manuscript is currently in preparation, we have begun to investigate the way that the high-level visual areas within the brain encode spatial structure. Until now, the theories of object vision have taken the brain's representation of objects to be nearly independent of space and put aside attributions such as whether the brain is tuned for the absolute size of an object or its distance from an observer. Our initial results suggest that these variables are strongly represented in the responses of object-selective neurons in the brain. Interestingly, however, our results suggest that neurons are not tuned for the expected absolute size of objects, but rather for extreme absolute sizes. This surprising finding is currently under further investigation in the laboratory, including its possibly relationships to the brains embedding of norms for perception.