Arun and Olson (2010) conducted a visual search experiment where human
subjects were asked to identify, as quickly as possible, an oddball
image embedded among multiple distractor images. The reciprocal of the
search times for identifying the oddball (in humans) and an ad hoc
neuronal dissimilarity index, computed from measured neuronal
responses to component images (in macaques), showed a remarkable
correlation. In this talk, I will describe a model, an active
sequential hypothesis testing model, for visual search. The analysis
of this model will suggest a natural alternative neuronal
dissimilarity index. The correlation between the reciprocal of the
search times and the new dissimilarity index continues to be equally
high, and has the advantage of being firmly grounded in decision
theory. I will end the talk by discussing the many gaps and challenges
in our modeling and statistical analysis of visual search. The talk
will be based on ongoing work with Nidhin Koshy Vaidhiyan (ECE) and S.
P. Arun (CNS).