Factors Affecting the Updating of Human Memory Scripts

Abstract

Memory updating involves modifying prior knowledge based on new information. Most of our knowledge about this process in humans comes from the study of word lists. Little is known about the dynamics of updating action sequences - scripts. This dissertation explores the dynamics of script memory updating, examining how newly learned memory scripts influence preexisting ones, and, conversely, how established scripts affect the learning of new scripts.We explored the effect of newly learned scripts on preexisting ones in two experiments. In Experiment 1, participants learned two similar scripts in succession (Script 1 and Script 2). Each script involved selecting an object from a pair of similar objects and placing it into one of six boxes. Four or six object pairs from Script 1 were repeated in Script 2, but the correct object choices and box placements differed. The following day, participants were asked to recall Script 1. Results indicated that learning Script 2 led to changes in the memory of Script 1, affecting not only the elements related to object memory and box location memory but also the action script per se (defined as the linkage between object and box location). These results demonstrated that new learning can update a previously acquired script. In Experiment 2, we examined whether this updating was a stable trait or influenced by temporary states by testing participants twice. The results were inconclusive, perhaps because cognitive tasks designed to minimize confounds may not effectively capture individual variability. Experiments 3 and 4 were conducted to assess how established scripts affect newly learned ones (i.e., how memory of Script 1 affects memory of Script 2), by testing the later learned script, Script 2, on Day 2. Experiment 3 followed the same protocol as Experiments 1 and 2, while Experiment 4 introduced a one-day interval between the two script learning phases to align with the three-day timeline used in previous human memory reconsolidation studies. On Day 3, participants were randomly assigned to recall either Script 1 or Script 2. The term “intrusion” was used here to characterize memory modifications - both when new information influenced an older memory or vice versa. Results showed bidirectional intrusions across object, box location, and linked script memories for both the Two-Day and Three-Day protocols, with greater influence from newly learned scripts on older ones for object and box location memories. Overall, the findings reveal bidirectional intrusions between preexisting and newly learned scripts across different time intervals, challenging the simplistic binary view that reconsolidation is an all-or-nothing process where memories are either entirely updated or remain unchanged. Instead, reactivation produces varying degrees of change across different memories, in either content or strength, or both, allowing old and new traces to coexist. This coexistence introduces the possibility of competitive or integrative dynamics between preexisting and newly acquired memory representations.