A Biophysical Model of Decision Making in an Antisaccade Task Through Variable Climbing Activity

Abstract

We present a biophysical model of saccade initiation based on competitive integration of planned and reactive cortical saccade decision signals in the intermediate layer of the superior colliculus. In the model, the variable slopes of the climbing activities of the input cortical decision signals are produced from variability in the conductances of Na + , K + , Ca 2+ activated K + , NMDA and GABA currents. These cortical decision signals are integrated in the activities of buildup neurons in the intermediate layer of the superior colliculus, whose activities grow nonlinearly towards a preset criterion level. When the level is crossed, a movement is initiated. The resultant model reproduces the unimodal distributions of saccade reaction times (SRTs) for correct antisaccades and erroneous prosaccades as well as the variability of SRTs (ranging from 80ms to 600ms) and the overall 25% of erroneous prosaccade responses in a large sample of 2006 young men performing an antisaccade task.