LN UCNI Banner
The neurophysiology of visual perception

Goals and Objectives
The long-term goal of the Unit on Cognitive Neurophysiology and Imaging (UCNI) is to understand the large-scale organization of visual processing in the brain as it relates to our perceptual experience. Investigating how neural signals shape to the contents of our perception is a great challenge in systems neuroscience that necessarily combines concepts from perceptual psychology and sensory physiology.

Studying the neural basis of visual perception is inherently challenging because of the dissimilar nature of psychological and biological measurements. For the psychologist, it is possible to speak precisely about subjective quantities in vision such as brightness or depth, yet psychophysical methods are limited in their capacity to deliver insights into specific brain mechanisms. Sensory physiologists, on the other hand, can be quantitative about the magnitude or duration of a neural response, but often need to speculate about the relation of such responses to perception. Thus studying neural processes underlying perception requires the marriage of experimental tools having distinct historical origins, a process that continues to evolve.

We have taken the approach of using illusions that challenge perception by destabilizing vision. Ambiguous or inherently conflicting visual stimuli provoke the brain into initiating a sequence of perceptual alternations, where an unchanging visual pattern is seen to reverse its structure every few seconds. Such bistable stimuli raise a poignant question for the sensory physiologist: where in the brain do neurons respond strictly based on external visual input, and where do they reflect internally determined perceptual states? And more generally, how does the brain interpret its inherently ambiguous sensory input to construct a meaningful and stable perceptual world?

Visual Perception Images
Figure 1. Some patterns that challenge perception. The Dalmatian dog on the left, once seen, is permanently visible. Viewing the second pattern reveals alternation between horizontal green stripes and vertical red ones, with intervening periods of seeing of seeing only the textured overlay. The third pattern, a variant of the Necker cube, demonstrates that 3D shape can be supported by illusory contours, and in this case can even reverse perspective. The last image is M.C. Escher’s "Sun and Moon".

We have previously developed a number of such paradigms that are useful for testing nonhuman primates [1-6]. Recently we have been focusing on one paradigm in particular, termed generalized flash suppression (GFS). In GFS, a salient visual target is presented on a screen, and appears to suddenly vanish when a number of surrounding dots are presented in the periphery (This illusion can be approximated by viewing the movie below with red/green anaglyph glasses). When the target disappears, it can remain invisible for several seconds, and there are a number of stimulus parameters govern the probability that the target disappears on any particular trial. For some stimulus settings, the target is seen to disappear on roughly 50% of the trials, but remain visible on the other 50%. In this case, what processes in the brain determine whether the target is seen on any particular trial?

GFS Image - No Audio\  Figure 2. Generalized Flash Suppression (GFS). This stimulus, which combines principles from Bonneh's motion induced blindness and Wolfe's binocular rivalry flash suppression, permits the prolonged disappearance of a salient perceptual target by the addition of a dynamic surrounding pattern. Unlike visual masking, the target can remain invisible for several seconds. Unlike binocular rivalry, there is no direct spatial or interocular contrast. The effect is approximated here upon prolonged fixation of the center cross while wearing red/green anaglyph glasses. Under laboratory conditions, perceptual disappearance is all-or-none, and cannot be discriminated from physical removal of the target.

Many studies in the past decade have investigated this type of question, employing either microelectrode recordings in monkeys in functional MRI (fMRI) studies in humans. While these approaches agree on many fronts, they have led to flatly contradictory conclusions regarding the role of the primary visual cortex in determining the visibility of a stimulus. Human imaging results have repeatedly shown that the primary visual cortex (V1) is strongly modulated according the visibility of a stimulus. In contrast, monkey neurophysiology studies have shown nearly the opposite, that it is the physical stimulus rather than the percept that is signaled by neurons in that area.

V1-3D Brain Image - No Audio\  In an effort to address this conundrum, and to further understand how activity in V1 contributes to determining the contents of perception, we performed both electrophysiological and fMRI studies in the same nonhuman primate subjects. We found that the previous contradictory results could not be ascribed to either species differences or differences in the paradigm. Instead, there is a real and consistent difference between the fMRI responses and the neural response. By monitoring the fMRI signal, it is possible to determine whether a stimulus is visible or invisible to the monkey, while it is impossible to determine this by measuring the responses of individual neurons. This work has great and unexpected implications for the use of fMRI at studying brain processes, a point that we elaborate upon in a different section of this report. alternative.

In a related subproject, we are examining the role of visual thalamic structures in supporting a visual percept. In that study we are employing multielectrode neurophysiological techniques to examine the pulvinar nucleus and the lateral geniculate nucleus, each of which is heavily interconnected with the visual cortex. We found that responses in the lateral geniculate nucleus were not affected by the visibility of the target (similar to the primary visual cortex). In contrast, we found that responses in the visual pulvinar nucleus were modulated significantly according to whether the monkey reported seeing a target on a given trial or not. These findings are helping us to how understand disparate by heavily interconnected parts of the brain act together as a sort of "perceptual circuit".

Figure 3. Effects of physical stimulation, physical removal, and perceptual suppression on the neural and BOLD fMRI signals. Example session for which data was collected in a block design for both signal types. Note that the two signals are in good agreement in the first four columns, corresponding to the physical presentation and removal of the target stimulus. In contrast, the two signals diverge in the rightmost column, during perceptual suppression (SUPP). Under this condition, only the BOLD signal, but not the spiking responses, show decreased activity compared to the visible (ON) conditions.

We are presently extending this work by combining local pharmacological approaches with both visualization in MRI, as well as behavior, to further elucidate the role of the pulvinar in perception. In a novel paradigm, we are able to inject an inhibitory agent, along with an MR-visible contrast agent, into the brain of awake nonhuman primates as they perform a visual task. Using fMRI, we track how inactivating certain portions of their brain interferes with processing in various neural circuits, and over what time course. Through such reversible inactivation experiments, we are able to test hypotheses about the critical elements involved in such perceptual circuits as mentioned above.
 GFS Image - No Audio\
Finally, there is a suspicious coincidence between the effects observed during perceptual suppression, and those observed in the absence of spatial attention. Like perceptual suppression, spatial attention is reflected in the BOLD signal in V1, but not in the firing of individual neurons. Might the modulation observed in the two paradigms be actually reflecting the same process? The precise relationship between attention and perception has long been sought. We are beginning to address this question on neural terms, asking whether the modulatory effects of attention and perceptual suppression are additive or singular.

Return to UCNI Projects
  1. Maier,A., Logothetis,N.K., & Leopold,D.A. Global competition dictates local suppression in pattern rivalry. J. Vis. 5, 668-677 (2005).
  2. Wilke,M., Logothetis,N.K., & Leopold,D.A. Generalized flash suppression of salient visual targets. Neuron 39, 1043-1052 (2003).
  3. Maier,A., Wilke,M., Logothetis,N.K., & Leopold,D.A. Perception of temporally interleaved ambiguous patterns. Curr. Biol. 13, 1076-1085 (2003).
  4. Leopold,D.A., Maier,A., & Logothetis,N.K. Measuring subjective visual perception in the nonhuman primate. Journal of Consciousness Studies 10, 115-130 (2003).
  5. Leopold,D.A., Wilke,M., Maier,A., & Logothetis,N.K. Stable perception of visually ambiguous patterns. Nat. Neurosci. 5, 605-609 (2002).
  6. Leopold,D.A. & Logothetis,N.K. Multistable phenomena: changing views in perception. Trends Cogn Sci. 3, 254-264 (1999).

This page was last updated May 6, 2008

 The Laboratory of Neuropsychology is part of the Division of Intramural Research Programs within the National Institute of Mental Health (NIMH), which is a part the National Institutes of Health (NIH), and is a component of the U.S. Department of Health and Human Services.