Neural Mechanisms of Strategic Adaptation in Attentional Flexibility

Abstract

Individuals are able to adjust their readiness to shift spatial attention, referred to as “attentional flexibility,” according to the changing demands of the environment, but the neural mechanisms underlying learned adjustments in flexibility are unknown. In the current study, we used fMRI to identify the brain structures responsible for learning shift likelihood. Participants were cued to covertly hold or shift attention among continuous streams of alphanumeric characters and to indicate the parity of target stimuli. Unbeknown to the participants, the stream locations were predictive of the likelihood of having to shift (or hold) attention. Participants adapted their attentional flexibility according to contextual demands, such that the RT cost associated with shifting attention was smallest when shift cues were most likely. Learning model-derived shift prediction error scaled positively with activity within dorsal and ventral frontoparietal regions, documenting that these regions track and update shift likelihood. A complementary inverted encoding model analysis revealed that the pretrial difference in attentional selection strength between to-be-attended and to-be-ignored locations did not change with increasing shift likelihood. The behavioral improvement associated with learned flexibility may primarily arise from a speeding of the shift process rather than from preparatory broadening of attentional selection.

Analogous computations in working memory input, output and motor gating: Electrophysiological and computational modeling evidence

“…In biologically inspired models of PBWM [ 11 , 12 , 52 ], parallel processing can occur when PFC projections to lower level striatum are strong, resulting in faster accumulation of evidence across levels of the hierarchy to favor a single response. The evidence for parallel processing across levels of information is also well documented in task switching literature when task sets and responses both switch [ 53 – 55 ] but also when context and items in WM both switch [ 56 ]. Parallel processing in these cases is signified by under-additive interactions reflecting that the cost of switching two levels of decision was less than the sum of their individual durations.…”

“…Third, we tested the prediction that input, output and response gating are processed in parallel. Converging evidence from neural network modeling of hierarchical rule learning [11,12,52], and experimental work with cognitive-control tasks (e.g., [53][54][55]) suggest that WM and action selection are managed by independent mechanisms, operating mostly in parallel on different levels of information. In biologically inspired models of PBWM [11,12,52], parallel processing can occur when PFC projections to lower level striatum are strong, resulting in faster accumulation of evidence across levels of the hierarchy to favor a single response.…”

Analogous computations in working memory input, output and motor gating: Electrophysiological and computational modeling evidence

“…Furthermore, converging evidence from modeling [12, 64], and experimental work (e.g., [65-66]) demonstrated under-additive interactions when more than one order of switching was needed (e.g., switching task-sets and responses or switching context and items in WM). These results suggest that action selection is partially parallel, such that decisions about a response are processed to some degree even while the identity of the cognitive rule is uncertain.…”

“…In biologically inspired models of PBWM [11,12,51], parallel processing can occur when PFC projections to lower level striatum are strong, resulting in faster accumulation of evidence across levels of the hierarchy to favor a single response. The evidence for parallel processing across levels of information is also well documented in task switching literature when task sets and responses both switch [51–53] but also when context and items in WM both switch [54]. Parallel processing in these cases is signified by under-additive interactions reflecting that the cost of switching two levels of decision was less than the sum of their individual durations.…”

The Location Independence of Learned Attentional Flexibility

“…Although individuals differ in their overall levels of cognitive flexibility (Cools, 2008), they also fluctuate over time such that at some moments they are more ready to execute switches than at others (Leber, 2010;Leber et al, 2008;Sali et al, 2016). In the domain of spatial attention, recent work has identified the importance of learning in shaping these moment-by-moment changes in shift readiness (Sali et al, 2015(Sali et al, , 2020, referred to here as attentional flexibility.Imagine that an individual is learning to operate a new piece of machinery that requires monitoring several gauges. Through trial and error, she may learn that she needs to frequently shift attention between two gauges to track the machine's performance.…”