The unidirectional prosaccade switch-cost: Correct and error antisaccades differentially influence the planning times for subsequent prosaccades

Abstract

Antisaccades produce longer reaction times (RT) than their prosaccade counterparts and this latency increase has been linked to an oculomotor 'pre-setting' that prevents the evocation of a stimulus-driven prosaccade. Moreover, a consequence of oculomotor pre-setting is a lengthening of the RTs associated with a subsequent prosaccade. The goal of the present study was to determine whether the constituent elements associated with planning a correct antisaccade (i.e., response suppression and vector inversion) imparts a residual delay that inhibits the planning of a subsequent prosaccade. To that end, participants alternated between pro- and antisaccades in a pseudo-randomized task-switching schedule (e.g., AABBAAB…) and responses were cued via a paradigm that was designed to evoke frequent error antisaccades (i.e., a saccade initially, and incorrectly, planned to the target stimulus). Results showed that RTs for correct antisaccades were longer than error antisaccades and that prosaccades preceded by the former, but not the latter, trial-type were associated with a reliable increase in RT (i.e., prosaccade switch-cost). In other words, error antisaccades were associated with a failure to withhold a stimulus-driven prosaccade and did not delay the planning of a subsequent prosaccade. Based on these findings we propose that the prosaccade switch-cost is not related to an explicit awareness of task goals; rather, our results are consistent with the assertion that a consequence of response suppression and vector inversion is a residual inhibition of stimulus-driven oculomotor planning networks.

Oculomotor task switching: alternating from a nonstandard to a standard response yields the unidirectional prosaccade switch-cost

“…Third, the RTs of classic prosaccades were lengthened by preceding classic antisaccades, whereas the RTs of classic antisaccades were not influenced by preceding classic prosaccades. This finding is consistent with previous work showing that the classic cuing condition elicits a unidirectional prosaccade switch-cost (Chan and DeSouza 2013; DeSimone et al 2014;Heath 2012a,b, 2014;Weiler et al 2014a). Task-set inertia explains the unidirectional prosaccade switch-cost.…”

“…Moreover, such an explanation also accounts for the results reported by DeSimone et al (2014), who demonstrated that prosaccade RTs are not lengthened when preceded by a directionally incorrect antisaccade. Recall that the successful completion of a nonstandard task requires a cognitively mediated task set.…”

“…Thus the completion of an antisaccade does not selectively inhibit the oculomotor networks involved in the planning of stimulus-driven prosaccades; rather, the antisaccade task set delays the planning of any subsequent response entailing direct SR mapping. This finding is an important addition to the oculomotor task-switching literature as previous work has selectively evaluated the expression of the unidirectional prosaccade switch-cost via a classic cuing paradigm (e.g., Chan and DeSouza 2013;DeSimone et al 2014;Heath 2012a,b, 2014;Weiler et al 2014a).…”

“…As well, it is important to recognize that during antisaccade trials participants occasionally fail to engage the processes of response suppression and/or vector inversion and execute a saccade to the veridical target location instead of the instructed mirror-symmetrical location (i.e., an antisaccade directional error; see Fischer and Weber 1996;Mokler and Fischer 1999). DeSimone et al (2014) demonstrated that antisaccade directional error trials, unlike their correctly executed counterparts, do not lengthen the RT of a subsequent prosaccade. Based on these results, it was proposed that the unidirectional prosaccade switch-cost is attributed to one, or both, of the constituent processes associated with the evocation of a directional correct antisaccade (i.e., response suppression and vector inversion).…”

“…These behavioral "costs" have been attributed to the top-down and two-component processes of 1) suppressing a prepotent prosaccade (i.e., response suppression), and 2) the visual remapping of the target's spatial location to its mirrorsymmetrical position (180°spatial transformation; i.e., vector inversion) (Zhang and Barrash 2000; for review see Munoz and Everling 2004). Moreover, neuroimaging and electrophysiology studies have shown that antisaccades produce an increased activity in the classic cortical saccade network (i.e., frontal eye field, supplementary eye field, and lateral intraparietal area), dorsolateral prefrontal cortex, and anterior cingulate cortex (Brown et al 2007; see also Curtis and D'Esposito 2003;DeSouza et al 2003;Ford et al 2005) as well as a respective increase and decrease in the activity of collicular fixation and saccade build-up neurons before target onset (Everling et al 1998(Everling et al , 1999. Such oculomotor modulations have been linked to a preparatory set that decreases the likelihood of evoking a prepotent prosaccade, and provides the requisite top-down sensorimotor transformation necessary to support a directionally correct antisaccade (Brown et al 2007).…”