The visual system has developed robust mechanisms to solve the complex computational problem of object recognition. The responses of neurons at the highest level of the macaque visual processing system, anterior inferotemporal cortex (AIT), are selective for complex objects and are critical for object identification. The responses of AIT neurons depend not only on the stimuli encountered in any given moment, but are also strongly influenced by the recent history of stimulation (adaptation). The overall aim of this proposal is a systematic understanding of factors that influence the adaptive state of AIT neurons and to determine the effects of adaptation on the coding of visual objects. Our inquiries into adaptation in AIT are two-fold. First, we propose to use adaptation as a tool to probe stimulus selectivity in AIT. At earlier stages of the visual system (such as V1 and MT), adaptation to prolonged visual stimulation results in shifted tuning functions and these shifts are correlated with shifts in perception. These adaptive shifts are thought to arise as a consequence of disrupting the circuitry that acts to mold tuning in these areas, and have provided insight into the mechanisms underlying stimulus selectivity. In AIT, adaptation is strong and is selective for shape.
Our aim i s to determine whether and how adaptation changes the stimulus selectivity of AIT neurons. Specifically, we propose to explore the effect of adaptation on a variety of AIT 'tuning' dimensions to gain insight into the mechanisms underlying these complex and little-understood receptive fields. Second, our aim is to understand how adaptation impacts the encoding of objects during natural vision. On the timescales of natural vision, adaptation in AIT is robust and shape-selective. This suggests that adaptation alters object coding across the AIT population. In particular, natural temporal image sequences are likely to produce strong adaptation effects and may thus provide important clues into the role of adaptation in object coding in the ventral visual stream. Understanding the neural coding of objects is critical to a deep understanding of human visual perception and long-term memory, and is needed to meaningfully repair the disruption of these brain processes or to create prosthetics that may stand in for such disruption. ? ? ?

Agency
National Institute of Health (NIH)
Institute
National Eye Institute (NEI)
Type
Postdoctoral Individual National Research Service Award (F32)
Project #
5F32EY018063-02
Application #
7388213
Study Section
Special Emphasis Panel (ZRG1-F02B-G (20))
Program Officer
Oberdorfer, Michael
Project Start
2007-03-15
Project End
2010-03-14
Budget Start
2008-03-15
Budget End
2009-03-14
Support Year
2
Fiscal Year
2008
Total Cost
$51,278
Indirect Cost
Name
Massachusetts Institute of Technology
Department
Other Basic Sciences
Type
Schools of Arts and Sciences
DUNS #
001425594
City
Cambridge
State
MA
Country
United States
Zip Code
02139