Explicit instructions and consolidation promote rewiring of automatic behaviors in the human mind

Abstract

One major challenge in human behavior and brain sciences is to understand how we can rewire already existing perceptual, motor, cognitive, and social skills or habits. Here we aimed to characterize one aspect of rewiring, namely, how we can update our knowledge of sequential/statistical regularities when they change. The dynamics of rewiring was explored from learning to consolidation using a unique experimental design which is suitable to capture the effect of implicit and explicit processing and the proactive and retroactive interference. Our results indicate that humans can rewire their knowledge of such regularities incidentally, and consolidation has a critical role in this process. Moreover, old and new knowledge can coexist, leading to effective adaptivity of the human mind in the changing environment, although the execution of the recently acquired knowledge may be more fluent than the execution of the previously learned one. These findings can contribute to a better understanding of the cognitive processes underlying behavior change, and can provide insights into how we can boost behavior change in various contexts, such as sports, educational settings or psychotherapy.

Deconstructing Procedural Memory: Different Learning Trajectories and Consolidation of Sequence and Statistical Learning

“…Rapid statistical learning has also been reported before: for instance, in the ASRT study of Szegedi-Hallgató et al (2017), statistical learning was apparent already in the first epoch in the Explicit group but seemed to have larger individual differences in the Implicit groups as only one of the two Implicit groups exhibited significant statistical learning in the first epoch (see Supplementary results and figures in Szegedi-Hallgató et al, 2017). Similarly, in Kóbor et al (2018) study, statistical learning was observed in the first epoch of the explicit version of the ASRT task, along with a significant sequence learning as well.…”

“…Consequently, a possible explanation for the very rapid statistical learning is that, in an explicit condition, the instructions and motivation to learn can have an overarching effect, providing a cognitive state, in which not only the instructed sequential but also the uninstructed statistical regularities can be learned quickly. Although this was not in the primary focus of these previous studies, if we take a closer look at the learning trajectories, it appears that statistical regularities are extracted very early and no (or very little) further gains may be observed during training if explicit instructions are given for the sequential information (Szegedi-Hallgató et al, 2017; Kóbor et al, 2018). In contrast, in the implicit conditions, statistical learning may undergo further improvements during training (Szegedi-Hallgató et al, 2017), above and beyond the strengthening of the acquired knowledge as suggested in the previous paragraph.…”

Deconstructing Procedural Memory: Different Learning Trajectories and Consolidation of Sequence and Statistical Learning

“…Consequently, a possible explanation for the very rapid statistical learning is that, in an explicit condition, the instructions and motivation to learn can have an overarching effect, providing a cognitive state, in which not only the instructed sequential but also the uninstructed statistical regularities can be learned quickly. Although this was not in the primary focus of these previous studies, if we take a closer look at the learning trajectories, it appears that statistical regularities are extracted very early and no (or very little) further gains may be observed during training if explicit instructions are given for the sequential information (Szegedi- Hallgató et al, 2017;Kóbor et al, 2018). In contrast, in the implicit conditions, statistical learning may undergo further improvements during training (Szegedi- Hallgató et al, 2017), above and beyond the strengthening of the acquired knowledge as suggested in the previous paragraph.…”

“…Although this was not in the primary focus of these previous studies, if we take a closer look at the learning trajectories, it appears that statistical regularities are extracted very early and no (or very little) further gains may be observed during training if explicit instructions are given for the sequential information (Szegedi- Hallgató et al, 2017;Kóbor et al, 2018). In contrast, in the implicit conditions, statistical learning may undergo further improvements during training (Szegedi- Hallgató et al, 2017), above and beyond the strengthening of the acquired knowledge as suggested in the previous paragraph. These observations support the interpretation that explicit instructions and the motivation to learn can have an overarching effect in that not only the instructed sequential but also the uninstructed statistical regularities can be learned more quickly.…”

“…Namely, both are statistical learning in a broader sense. When acquiring frequency information (statistical learning in the narrow sense), a 2nd order probabilistic sequence should be learned, in which there are always one probable continuation and some less probable continuations for the first two elements of a given three-element stimulus chunk (Szegedi-Hallgató et al, 2017;Kóbor et al, 2018). When acquiring order information (sequence learning), the 2nd order transitional probability is equal to one; namely, Deconstructing procedural memory 39 consecutive elements in the sequence could be predicted with 100% certainty from the previous sequence element (Kóbor et al, 2018).…”