THE BEST CUES EVOLVE WITH MEMORY: A NEW CHALLENGE FOR THE ENCODING SPECIFICITY HYPOTHESIS

The Encoding Specificity Hypothesis, a cornerstone in memory research, posits that the successful recollection of an event is cue dependent and relies on the degree of match between cues present at retrieval and encoding conditions. Here, we challenge this principle by proposing that, since neural memory traces and their psychological experience change over time, the optimal cue should change accordingly, matching the current state of the encoded memory. We examined this hypothesis using a paired-associate paradigm, where participants encoded pairs of objects and recalled one of them upon being presented with the other encoded object (i.e., identical cue), a similar version of it or a different version of it. We tracked the current state of the memory by means of a recognition task on the cues, which could follow (Experiment 1) or precede (Experiment 2) the retrieval of the paired associate. In Experiment 1, participants were faster retrieving a target memory when presented with cues recognised as identical, followed by recognised as similar and then different. This effect was stronger than the one of the actual similarity of the cues. Furthermore, non-encoded cues recognised as identical led to faster response times (RTs) compared to encoded identical cues, contrasting with the Encoding Specificity Hypothesis. No significant difference was observed for the accuracy at retrieval and Experiment 2 failed to replicate the effect on RT, likely due to the early recognition phase reactivating the event memory. We discuss the results as preliminary evidence of a potentially necessary perspective change: from encoding-retrieval match, to cue updating based on memory evolution.