FMRI for Functional Localization and Task Difficulty Assessment During Visual Search for Military Vehicles

摘要

Past and current U.S. Army computational vision models designed to determine the difficulty of visual detection of camouflage for military vehicles are extremely limited in the sense that they do not encompass much of the brain outside the retina and visual cortex, and within those areas, do so to a very limited degree. A method and preliminary experiments to obtain the raw data to construct better and more representative models of human vision and cognition is presented. The inclusion of specific neurocircuitry in the computational model as opposed to the 'black box' standard used in psychophysics is now possible. The combination of psychophysics and fMRI has the potential to give a more complete view of the neural systems that are relied on for different perception tasks related to camouflage and deception.