Observed actions and their enduring cognitive imprint beyond correction

Observing others' actions can blur the boundary between self and other, leading to false memories of self-performance. Although previous research using the observation-inflation paradigm has examined the mechanisms underlying this phenomenon, little is known about its persistence and the cognitive processes that sustain it. In Experiment 1, participants performed or read action phrases and later observed videos of others performing actions. During the five days before the memory test, they received corrective feedback distinguishing actions they had performed from those merely observed. Corrective feedback reduced false self-performance reports, yet participants still showed longer response times when judging previously observed action phrases than unobserved ones, suggesting that prior observation continued to influence retrieval processing even after correction. Experiment 2 employed event-related potentials to differentiate familiarity- and recollection-based processes. Larger LPC amplitudes for read-and-observed phrases, together with the absence of a reliable FN400 difference, suggested a greater contribution of recollection-related than familiarity-related processes. Together, the findings show that even when false beliefs are corrected, observed actions may continue to exert a lasting influence on subsequent memory judgments.