Abstract
Understanding visual decision-making entails understanding how speed-accuracy tradeoff (SAT) is accomplished. Investigations of human SAT with non-invasive measures have demonstrated variation of multiple processes and brain regions. Investigations of monkey SAT with neurophysiological measures have also demonstrated variation of multiple types of neurons in diverse structures. Neural signals in both species indicate network-wide gain changes underlying SAT. Previous work with humans showed that pupil diameter is a reliable index of global neural gain during speed-accuracy tradeoff. To assess the ubiquity of SAT effect on neural gain, we investigated changes in pupil dilation and constriction during visual search. Four macaque monkeys performed visual search to locate a target (T/L) presented amongst seven distractors (L/T) with overall luminance equated across array configurations. Trials began when monkeys fixated a central stimulus, the color of which cued emphasis on response speed (green for Fast condition) or accuracy (red for Accurate condition). For each monkey, response time was shorter, and choice error rate, higher, in the Fast relative to the Accurate condition. We found robust pupil dilation during the baseline period before array appearance. The rate of dilation was greater in the Fast relative to the Accurate condition. Notably, pupil dilation did not predict occurrence of response errors in choice or timing. These new observations replicate previous reports of larger pupil size when speed is emphasized and reinforce the hypothesis that pupil size is an overt marker of covert gain changes that accomplish SAT.