Transfer of learned cognitive control settings within and between tasks

Merve Ileri-Tayar,Caroline Moss,Julie M Bugg
DOI: https://doi.org/10.1016/j.nlm.2022.107689
Abstract:Cognitive control is modulated based on learned associations between conflict probability and stimulus features such as color. We investigated whether such learning-guided control transfers to novel stimuli and/or a novel task. In Experiments 1 and 2, participants experienced an item-specific proportion congruence (ISPC) manipulation in a Stroop (Experiment 1) or Flanker (Experiment 2) task with mostly congruent (MC) and mostly incongruent (MI) colors in training blocks. During a transfer block, participants performed the same task and encountered novel transfer stimuli paired with MC or MI colors. Evidencing within-task transfer, in both experiments, responses were faster to incongruent transfer stimuli comprising an MI color compared with an MC color. In Experiment 3, we investigated between-task transfer from Stroop to Flanker. After training with an ISPC manipulation in the Stroop task, a Flanker task was completed with the same colors but without an ISPC manipulation (i.e., 50% congruent). Responses were faster to incongruent transfer stimuli paired with the previously-MI colors compared with the previously-MC colors. Additionally, transfer was evident in the first half of the Flanker task but not the second half. The evidence for within-task transfer, in combination with the novel evidence for between-task transfer, suggests learned control settings are flexibly retrieved and executed when predictive cues signaling these control settings are encountered in novel stimuli or a novel task. Theoretical implications are discussed alongside potential neural mechanisms mediating transfer of learning-guided control.
What problem does this paper attempt to address?