Human express saccades: effects of randomization and daily practice

Publisher Website
Google Scholar
Abstract
When human subjects are asked to execute saccades from a fixation point to a peripheral target, if the fixation point is turned off some time (200 ms) before the target is turned on, the distribution of the saccadic reaction times is bimodal. The first peak occurs at about 100 ms and represents the population of express saccades. If the target location is kept constant the express saccades have reaction times of about 100 ms. If the target location is randomized between right and left (distance from fixation point constant at 4 deg) the reaction times of the express saccades are increased by about 15 ms. If the target location is randomized between 4 deg and 8 deg (direction constant to the right) no increase of the reaction time is observed. The proportion of express saccades increases with daily practice and their reaction times decrease slightly from 105 ms to 98 ms. If an anticipatory saccade was made after reaction times below 75 ms, it frequently undershot the target by more than 20% and was followed by a corrective saccade. The corrections could be executed at times where usually an express saccade would have occurred such that all of these corrections began at about the same time, i.e. 100 ms after target onset, implying intersaccadic intervals between 100 ms and zero (!)