de.mpg.escidoc.pubman.appbase.FacesBean
English
 
Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Poster

Generalization of neural selectivity across spatial scale in monkey extrastriate area V4

MPS-Authors
http://pubman.mpdl.mpg.de/cone/persons/resource/persons84041

Lee,  H
Department Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons84063

Logothetis,  NK
Department Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons84154

Rainer,  G
Department Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available
Citation

Lee, H., Logothetis, N., & Rainer, G. (2004). Generalization of neural selectivity across spatial scale in monkey extrastriate area V4. Poster presented at 34th Annual Meeting of the Society for Neuroscience (Neuroscience 2004), San Diego, CA, USA.


Cite as: http://hdl.handle.net/11858/00-001M-0000-0013-D7B7-0
Abstract
Lesions of area V4 lead to impairments in size discrimination. To study the involvement of single V4 neurons in generalization across spatial scale, we trained monkeys on a modified delayed matching-to-sample task. On each trial, one of five novel natural scenes was briefly shown at the center of gaze at one of three sizes (small: 6°×6°, medium 10°×10° and large 15°×15°). After a delay of 1 second a probe object (always at medium size) was presented and monkeys had to report if it matched the sample. We found behavioral evidence for partial generalization across spatial scale. Monkeys’ performance was best for medium size samples (92 correct). Paired t-tests (evaluated at P<.05) showed that performance was significantly lower for both small (73 correct) and large (80 correct), but greater than chance performance (50 correct). We have preliminary data from 34 neurons recorded in one monkey. Quantitative receptive field spot mapping revealed that the receptive field centers of most neurons were within 4 degrees of the fixation point. Most (25/34 or 74) neurons distinguished in terms of their sample evoked activity among the sample stimuli at one or more of the three spatial scales (ANOVAs evaluated at P<0.05). We used a correlation analysis to evaluate how well neurons generalized across spatial scale, by comparing tuning functions for small and large stimuli to the tuning functions for the standard medium size. We found that on average neurons appeared to generalize to both larger and smaller stimuli (mean correlation coefficients: 0.62 for larger size, 0.38 for smaller size). For about a quarter of these neurons (6/25 or 24) correlation coefficients were greater than 0.8 for both size increases and decreases suggesting robust generalization to both smaller and larger sizes. The results reveal extent and limits of generalization across image size at the level of single neurons in extrastriate area V4.