Event boundaries are steppingstones for memory retrieval

Personal experience extends over time. When we recall memories from our lives, we often scan extended episodes to remember, for instance, where we placed our car keys. Here we investigate the hypothesis that high level structure, marked by boundaries between events, guides us through this retrieval process. We present a model of memory search in which humans can replay past experience on a fine-grained temporal scale, but also skip ahead to the beginning of a new event in order to speed up the retrieval process. In a naturalistic interview paradigm, we provide evidence in human response times for such a skipping mechanism: When participants scanned extended segments from a movie in their memory, their reaction times were better explained by a model incorporating the number of event boundaries in the segment and the distance of the target to the previous event boundary (with both predictors explaining significant variance), compared to a model incorporating the actual duration of a segment. This supports the idea that, in scanning their memory, humans can skip to the beginning of a new event if they decide that the target is not present in that event; this has the effect of decoupling their memory scanning reaction times from the actual duration of the segment. This conclusion is further supported by the results of a second study where participants were asked to do detailed mental simulation instead of memory scanning. Participants took substantially longer to perform such mental simulation compared to participants that scanned their memory. Those mental simulation times were still explained better by a model where participants skip, but -- compared to memory scanning -- the threshold for skipping was much higher.

An Investigation of the Slope Parameters of Reaction Times and P3 Latencies in the Sternberg Memory Scanning Task – A Fixed-Links Model Approach

The speed of retrieving information from short-term memory is thought to be captured in the slope of the linear function of reaction times (RTs) regressed on set size in the Sternberg memory scanning task (SMST). Individual differences in the slope parameter have been hypothesized to correlate with general intelligence (g). However, this correlation can usually not be found. This present study chose a fixed-links model (FLM) approach to re-evaluate the RT slope parameter on a latent level in a sample of 98 participants aged 18 to 61 years. The same was tried for P3 latencies to investigate whether or not both parameters measure the same cognitive processes in the SMST, and to assess the usability of both slopes to predict g. For RTs, a linear increase with set size was found. The RT slope correlated with g on a manifest level already. The FLM approach could significantly increase the relationship between the slope and g. For P3 latencies, no evidence for a linear increase was found. This indicates that RTs and P3 latencies do not measure the same cognitive processes in the SMST. The FLM proved suitable to investigate the association between the speed of short-term memory scanning and intelligence.

Falsas memórias: Processos paralelos ou seriais?

Este trabalho teve como objetivo investigar processos paralelos e seriais nas falsas memórias (FM). Em um delineamento 3 x 3 x 3, foram manipulados os fatores carga de memória (3, 5 e 7 itens), tipo de item (alvo, distrator crítico e distrator não relacionado) e tempo de exposição dos itens (750, 1.500 e 3.000 milissegundos). O tempo de reação aumentou linearmente com o aumento da carga de memória, indicando que processos seriais atuam no processamento de alvos e de falsas memórias. Houve um aumento da percentagem de erros para alvos e distratores críticos com o aumento do tempo de exposição. Os dados foram discutidos em termos das teorias de memory scanning e de ativação/monitoramento.

The diffusion model of reaction time for recent negative probes

The study investigates the problem of information displacement from short-term memory. In two experiments, reaction times for recent negative probes were analyzed in the Sternberg’s memory scanning task. The diffusion model of reaction times was used with parameters estimated with the fast-dm software. It was found (experiment 1) that recent negative probes are characterized by a reduction in the speed of information accumulation (drift rate). This suggests residual activation of irrelevant cognitive representation in memory after they have been displaced from short-term memory. It was also found (experiment 2) that negative probes semantically related to items in a preceding target set (semantic recent negative probes) are characterized by a similar decrease in the drift rate. This suggests activation spreading from irrelevant cognitive representations displaced from short-term memory along semantic connections and identifies activated long-term memory as the target of information displacement from short-term memory. Additional mechanisms of short-term memory scanning (negative priming and dynamic decision thresholds) are discussed.

When Group Means Fail: Can One Size Fit All?

The present study examined whether a cognitive process model that is inferred based on group data holds, and is meaningful, at the level of the individual person. Investigation of this issue is tantamount to questioning that the same set and configuration of cognitive processes is present within all individuals, a usually untested assumption in standard group-based experiments. Search from memory as assessed with the Sternberg memory scanning paradigm is among the most widely studied phenomena in cognitive psychology. According to the original memory scanning model, search is serial and exhaustive. Here we critically examined the validity of this model across individuals and practice. 32 younger adults completed 1488 trials of the Sternberg task distributed over eight sessions. In the first session, group data followed the pattern predicted by the original model, replicating earlier findings. However, data from the first session were not sufficiently reliable for identifying whether each individual complied with the serial exhaustive search model. In sessions six to eight, when participants performed near asymptotic levels of performance, between-person differences were reliable, group data deviated substantially from the original memory search model, and the model fit only 13 of the 32 participants’ data. Our findings challenge the proposition that one general memory search process exists within a group of healthy younger adults, and questions the testability of this proposition at the individual level in single-session experiments. Implications for cognitive psychology and cognitive neuroscience are discussed with reference to earlier work emphasizing the explicit consideration of potentially existent individual differences.Author NoteThis study was conducted within the project “Cognitive and Neural Dynamics of Memory Across the Lifespan (CONMEM) ” at the Max Planck Institute for Human Development, Berlin, Germany. The work was supported by the Max Planck Society and the Gottfried Wilhelm Leibniz Prize 2011 of the German Research Foundation awarded to UL. MW-B’s work is supported by a grant from the German Research Foundation (DFG, WE 4269/3-1) as well as an Early Career Research Fellowship 2017 – 2019 awarded by the Jacobs Foundation.We thank Lene-Marie Gassner and all research assistants involved in data collection.

Temporal Prediction Errors Affect Short-Term Memory Scanning Response Time

The Sternberg short-term memory scanning task has been used to unveil cognitive operations involved in time perception. Participants produce time intervals during the task, and the researcher explores how task performance affects interval production – where time estimation error is the dependent variable of interest. The perspective of predictive behavior regards time estimation error as a temporal prediction error (PE), an independent variable that controls cognition, behavior, and learning. Based on this perspective, we investigated whether temporal PEs affect short-term memory scanning. Participants performed temporal predictions while they maintained information in memory. Model inference revealed that PEs affected memory scanning response time independently of the memory-set size effect. We discuss the results within the context of formal and mechanistic models of short-term memory scanning and predictive coding, a Bayes-based theory of brain function. We state the hypothesis that our finding could be associated with weak frontostriatal connections and weak striatal activity.