Purchase this article with an account.
Jay Hegde, Donatello Arienzo; Neural Substrates of Camouflage-Breaking . Journal of Vision 2016;16(12):521. doi: https://doi.org/10.1167/16.12.521.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Recognizing an object camouflaged against its background, or camouflage-breaking, can be a matter of life or death in nature. It is also fundamentally important in warfare and game hunting. The neural mechanisms of camouflage-breaking remain unclear. We measured the neural responses of human subjects using functional magnetic resonance imaging (fMRI) while the subjects (N = 22) freely searched the camouflage scene and reported whether or not a camouflaged target object, such as a human face, was present in the scene. Using an analysis of variance (ANOVA) design, we identified 29 cortical and sub-cortical regions of interest (ROIs) that showed highly significant differential responses during the camouflage-breaking task (p 10-9, corrected for multiple comparisons). The differential neural responses were not attributable to eye movements in any ROI (partial F tests; F [1,12] less than 0.01, p 0.05 in all cases). Neural activity in 6 of the ROIs in the occipito-temporal pathway were collectively diagnostic of successful camouflage-breaking on a trial-to-trial basis, as determined by multi-voxel pattern analysis (MVPA; class accuracy, 0.73; area under the ROC curve, 0.74; p 0.05). Multivariate Granger Causality analysis of the network connectivity identified 129 highly significant pairwise, directional connections during 'hit' trials wherein the subjects correctly reported that the scene contained a camouflaged target (p 10-9, corrected). The pattern of network connectivity was significantly different when the subjects failed to detect the target ('miss' trials; quadratic assignment procedure [QAP], p 10-6). A separate control experiment (N = 4) showed that the aforementioned network connectivity patterns during camouflaged-breaking were significantly different than those during conventional visual search, in which the subjects reported whether a non-camouflaged 'odd-man-out' face target was present among non-camouflaged distractor faces (QAP, p 10-6). Together, our results characterize, for the first time, the brain regions that subserve camouflage-breaking.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only