Near - optimal visual search : behavior and neural basis

1 Theory .......................................................................................................................... 1 1.1 Global log likelihood ratio ................................................................................... 1 1.2 Local log likelihood ratio (neural form) ............................................................... 3 1.3 Local log likelihood ratio (behavioral form) ........................................................ 5 1.4 Relation of optimal model to max and sum models ............................................. 7 1.5 Non-optimal local decision variables ................................................................... 8 2 Behavioral experiments ............................................................................................... 8 2.1 Experiments 1a and 2a ......................................................................................... 8 2.2 Individual-subject ROCs ...................................................................................... 9 2.3 Bayesian model comparison ................................................................................. 9 3 Neural implementation .............................................................................................. 10 3.1 Intuition behind a nonlinearity ........................................................................... 10 3.2 Network performance ......................................................................................... 10 3.3 Effect of non-Poisson-like statistics ................................................................... 11 Supplementary Figures are at the end.