On the Search for a Selective and Retroactive Strengthening of Memory: Is There Evidence for Category-Specific Behavioral Tagging?

Storing motivationally salient experiences preferentially in long-term memory is generally adaptive. Although such relevant experiences are often immediately obvious, a problem arises when the relevance of initially ambiguous events becomes evident sometime after encoding. Is there a mechanism that enables the retroactive enhancement of specific memories? Recent evidence suggests the existence of such a mechanism that selectively strengthens weak memories for neutral stimuli from one category when their respective category gains motivational significance later. Although such a selective retroactive memory enhancement has considerable implications for adaptive memory, evidence for this phenomenon is based on only few studies. Here, we report data from four attempts to replicate category-specific retroactive memory enhancements for neutral stimuli from a category that was later predictive of aversive electric shocks. Although our data showed enhanced memory for the arousing stimuli themselves as well as related subsequent stimuli, none of our experiments provided any evidence for category-specific retroactive memory enhancement when strictly replicating the analysis strategy from the original study. In an additional analysis focusing on high confidence memory only, one of four experiments indicated a significant retroactive memory effect but only in corrected recognition and not in d ' based on signal detection theory. In an analysis pooled across all experiments, we found a small but significant retroactive memory effect again solely for high-confidence corrected recognition, although the corresponding Bayesian analysis indicated even substantial evidence for the null hypothesis. Overall, our data cast doubt on the reliability and generalizability of the proposed selective retroactive enhancement of initially weak memory.