Cross-Modal Facilitation of Episodic Memory by Sequential Action Execution

Abstract

Throughout our lives, the actions we produce are often highly familiar and repetitive (e.g., commuting to work). However, layered upon these routine actions are novel, episodic experiences. Substantial research has shown that prior knowledge can facilitate learning of conceptually related new information. But despite the central role our behavior plays in real-world experience, it remains unclear how engagement in a familiar sequence of actions influences memory for unrelated, nonmotor information coincident with those actions. To investigate this, we had healthy young adults encode novel items while simultaneously following a sequence of actions (key presses) that was either predictable and well-learned or random. Across three experiments (N = 80 each), we found that temporal order memory, but not item memory, was significantly enhanced for novel items encoded while participants executed predictable compared with random action sequences. These results suggest that engaging in familiar behaviors during novel learning scaffolds within-event temporal memory, an essential feature of episodic experiences.

Keywords: episodic memory; motor actions; open data; open materials; prior knowledge; sequence learning; temporal memory.

Fig. 1.

Experiment design. The study procedure for all three experiments, including an example encoding sequence (left) and memory-test trials (middle and right), is shown in (a). During the encoding phase in Experiments 1 and 3, participants had to rely on memory when making responses during predictable events but were provided with cues during random events. In Experiment 2, all aisle responses were cued. The relationship between item and action sequences is illustrated in (b). The learned action sequence was hypothesized to provide a scaffold for novel items seen throughout a single errand/event.

Fig. 2.

Pretraining design and behavior. The pretraining procedure for all experiments is shown in (a). Participants completed three cycles of study and test periods to allow for robust learning of the predictable (“Pred”) store’s aisle sequence. Pretraining performance is plotted separately for Experiments 1 (b), 2 (c), and 3 (d). The plots on the left show mean reaction times (RT) for predictable and random (“Rand”) aisle responses made during the study period of the third (“intermixed”) block of pretraining. The plots on the right show accuracy across sequence test repetitions. Error bars indicate within-subjects standard errors, and dots represent individual participants. Asterisks indicate significant between-condition differences (***p < .001).

Fig. 3.

Order reconstruction memory results for Experiment 1. The bar plot in (a) shows ordinal accuracy for novel items from predictable (“Pred”) and random (“Rand”) events. The bar plot in (b) shows the proportion of events in which zero to six items were selected in the correct ordinal position. (Note that the near-zero proportion of events in which five items were remembered stems from the fact that getting exactly five out of six responses correct required either missing one response or selecting the same item twice in different ordinal positions, both of which rarely occurred.) The plot in (c) shows ordinal accuracy as a function of sequence position within the event. Error bars indicate within-subjects standard errors, and dots represent individual participants. Dashed lines indicate chance performance (1/6 = .17). Asterisks indicate significant between-condition differences, corrected for multiple comparisons (*p < .05, **p < .01, ***p < .001).

Fig. 4.

Spatial memory test performance in Experiments 1 (top row) and 2 (bottom row). Panels (a) and (c) show mean accuracy as a function of store condition (predictable [“Pred”] and random [“Rand”]), whereas panels (b) and (d) show accuracy in each condition as a function of the sequence position in which an item was encountered during encoding (with significance values false-discovery-rate corrected for multiple comparisons). Error bars indicate within-subjects standard errors, and dots represent individual participants. Dotted lines indicate chance performance (1/4 = .25). Asterisks indicate significant between-condition differences (*p < .05).

Fig. 5.

Order reconstruction memory results for Experiments 2 (left column) and 3 (right column). Plots in (a) and (d) show order memory performance as mean ordinal accuracy for predictable (“Pred”) and random (“Rand”) events. Plots in (b) and (e) show the proportion of tested events in which zero to six items were selected in the correct ordinal position. Plots in (c) and (f) show memory performance within each condition as a function of sequence position (1–6). Error bars indicate within-subjects standard errors, and dots represent individual participants. Dashed lines indicate chance performance (1/6 = .17). Symbols indicate significant between-condition differences (~p < .05, uncorrected; *p < .05, **p < .01, corrected for multiple comparisons).

Fig. 6.

Item recognition memory test results for Experiment 3. The bar plot (a) shows the difference in d′ (a measure of recognition accuracy) for items from the predictable (“Pred”) and random (“Rand”) store. Panel (b) shows the rate of each response type during the recognition test, split by condition (FA = false alarm, CR = correct rejection). Error bars indicate within-subjects standard errors, and dots represent individual participants. Examples of “old” images and corresponding “new” lures are shown in (c).