Strategic attention to memory: Proportion congruent effects in the episodic flanker task

The episodic flanker task is a memory analog of the perceptual flanker task, designed to assess parallels between attention to memory and perception. Subjects receive lists of six letters followed by probes that contain a cued target and flanking letters, which are the same as or different from the list (list: STVPDR, same probe: STVPDR, different probe: BNYPKL; cued position is underlined). Their task is to say whether the probed letter occupied the cued position in the list. The compatibility effect is the interaction between response and flankers: for "yes" responses, performance is better with same flankers than with different flankers; for "no" responses, performance is better with different flankers than with same flankers. We report another parallel between memory and perception, demonstrating strategic modulation of the compatibility effect by manipulating the proportion of congruent trials. In Experiments 1 and 2, 80% of the trials were congruent or 20% were congruent. Experiments 3 and 4 began with 50% congruent trials and switched to 80% (Experiment 3) or 20% (Experiment 4). Experiment 5 was a perceptual version of Experiment 2. Each experiment showed proportion congruent effects in response time and error rate: the compatibility effect was largest with 80% congruent trials and smallest with 20% congruent. Novel lists were used on each trial to rule out item-specific proportion congruent effects. There was little evidence of conflict adaptation in any experiment. These results support the attentional account of the proportion congruent effect and strengthen the parallels between attention in memory and perception.