Brain oscillatory activity associated with switch and mixing costs during reactive control

Abstract

Although this ancient Latin quote was not coined with reference to task-switching, a common finding in the cognitive control literature is that switching from one task to another incurs a significant "switch cost" as compared to repeating the same task. Yet, there is also a "mixing cost" when contrasting performance on the all-repeat trials from the single-task blocks and the repeat trials from the mixed blocks (Monsell, 2003; Rubin & Meiran, 2005). The switch cost is generally taken as an index of a phasic task-set reconfiguration process, which operates on a trial-by-trial basis to flexibly switch between different tasks and which itself includes multiple sub-processes such as retrieval of current rules and appropriate stimulus-response mappings from memory, resolution of carryover interference from the previous trial, suppression of the no longer relevant task-set and activation of the new relevant one (Monsell, 2003). Conversely, the mixing cost is thought to reflect higher memory demands

Cognitive flexibility in post-traumatic stress disorder: Sustained interference associated with altered modulation of cortical oscillatory activity during task-switching

“…During the preparatory (cue-test stimuli) interval, these patterns of modulation of cortical oscillations are more elevated for switch compared to repeat trials. Similar results have also been reported in a task using simultaneous cues and test stimuli ( Capizzi et al, 2020 ). The higher increase in theta band activity following the cue onset in switch trials is believed to reflect neuronal activity associated with proactive goal-updating processes ( Cooper et al, 2017 ) and increased cognitive control ( Cavanagh and Frank, 2014 ) required by shifts of attention.…”

“…The task used in our study requires the execution of a motor response according to a stimulus–response contingency that depends on the attended stimulus feature, while inhibiting the neuronal representation of the conflicting stimulus–response contingency used in previous trials. The stronger suppression of beta band activity in switch versus repeat trials observed in our study and others ( Cunillera et al, 2012 , Foxe et al, 2014 , Capizzi et al, 2020 ) may be explained based on two premises. First, beta band oscillations in prefrontal regions may reflect activity in competitive neuronal ensembles encoding the stimulus–response contingencies corresponding to alternative rules ( Buschman et al, 2012 ).…”

“…This interpretation is based on observations that anticipatory allocation of attention is linked to suppression of alpha band activity in sensory areas that process the attended sensory inputs, while enhancement of alpha oscillations in other sensory areas signals the disengagement of other sensory systems ( Worden et al, 2000 , Gould et al, 2011 , Haegens et al, 2011 ). Finally, the suppression of beta band activity has been linked to interference control (suppression) that is manifested in multiple brain regions and predominantly in regions involved in motor response selection ( Proskovec et al, 2019 , Capizzi et al, 2020 ). This interpretation receives support from findings of stronger suppression of beta band activity for incongruent compared to congruent trials in Stroop tasks ( Zhao et al, 2015 , Tafuro et al, 2019 ) particularly in the presence of interference that affects the process of motor response selection ( Zhao et al, 2015 ).…”

Temporally and functionally distinct large-scale brain network dynamics supporting task switching

“…Further studies involving iEEG oscillations other than high-gamma activity are warranted to determine the network dynamics supporting cognitive flexibility. Prior scalp EEG studies reported that the theta amplitude and coherence across frontal and parietal regions differed between switch and repeat trials ( Cooper et al, 2015 ; Capizzi et al, 2020 ; McKewen et al, 2021 ).…”

Fronto-parietal homotopy in resting-state functional connectivity predicts task-switching performance

“…We analyzed mixing and switching costs as markers of sustained and phasic control processes, respectively. We also manipulated cue-to-target interval (CTI: 300 vs. 1200 ms) on a trial-by-trial basis to differentially modulate cognitive demands in this paradigm (e.g., Meiran 1996 ; Cooper et al 2015 ; Capizzi et al 2020 ).…”