Preregistration: Freeing capacity in WM through the use of LTM representations
Peter Shepherdson, Lea Maria Bartsch · OSF Preprints (OSF Preprints) · 2019
Background and Research Question In the first two experiments of this preregistered study (see OSF) we aimed to investigate whether the presence and use of LTM representations frees capacity for maintaining additional information in WM. In the first phase of our first experiment, we presented participants with word pairs for them to encode into LTM (LTM learning phase). Subsequently, they completed trials of a WM task, also involving word pairs. Crucially, the pairs presented in each WM trial consisted of varying numbers of new pairs (LTM unavailable) and the previously learned LTM pairs. The results of this first experiment provided evidence that cued recall performance in the WM task was unaffected when memory set size increased through the addition of LTM-available pairs but deteriorated when set size increased through adding LTM-unavailable (new) pairs. In the second Experiment we not only varied WM loads across two levels (2, and 4 pairs), but also varied LTM loads across three levels of LTM load (0, 2, and 4 pairs). Recall performance deteriorated with LTM load for low WM load (2 pairs) but remained superior to performance with the same set size of only LTM-unavailable (new) pairs. In contrast, LTM load did not affect recall performance at higher WM load (4 pairs). This added to the evidence of the first experiment suggesting that individuals can outsource workload to LTM to optimize performance, but also speaks towards a WM system with a flexible gate to LTM that can be opened or closed depending on the current cognitive need. So far, our experiments have focused on the effects of having a LTM representation available when having to maintain information over a short period of time. However, WM is typically described as a system for both storing and processing information on a short-term basis. Further, these two functions are often assumed to trade off, with memory load reducing processing efficiency and processing load reducing memory accuracy (e.g., Oberauer, 2009; Unsworth et al., 2009; Vergauwe et al., 2014). Thus, in Experiment 4, we aim to explore how the availability of LTM influences the ability to process recently encoded information. We therefore plan to adapt the task used in the first and second experiments to allow us to better investigate processing: The memoranda will consist of word-number pairs (e.g. pie = 12), which will be presented at varying set sizes to the participants sequentially. At test, participants will be presented with mathematical equations incorporating a word from that trial, and asked to determine whether the equations are correct (e.g.: pie + 11 = 23 ?). Critically, the pairs presented in each WM trial consist of varying numbers of new pairs (LTM unavailable) and again previously learned LTM pairs. In short, here we aim to extend the findings of Experiment 1 and 2, which hinted at a flexible gate in WM, opening to LTM when representations of memoranda are available. Specifically, we plan to investigate whether the LTM advantage found in those previous experiments persists in cases where the memoranda have to be processed upon recall, or whether the necessity for their retrieval into WM for processing abolishes this advantage. This will help us to understand the scope of the utility that LTM for specific contents might provide to WM functions that involve manipulating those contents, which could have implications for areas as diverse as reading, problem-solving, reasoning, and decision-making. Methods Participants We will first collect data from 30 participants who did not complete Experiment 1, 2 or 3, online (recruitment via Prolific academic). We will then run a first analysis and check the evidence (indexed by the Bayes Factor, BF) for differences between our conditions of interest. Our goal is to report BF ≥ 3 for or against interaction effects of LTM load with WMload, and all main effects (WMload, LTMload). If we do not reach the targeted BF after the initial data collection, we will add bouts of 10 participants and re-run the analyses. We will stop data collection once the BF ≥ 3, or once we have collected usable data from 100 participants. Participants will be replaced (i.e., their data considered unusable) if: (a) their overall response accuracy across all conditions is greater than 2 standard deviations below the overall mean; (b) their recall accuracy in the LTM test for the second recall instance of each pair is below 70% correct (at this point, they will have had two chances to encode and one testing instance to learn the pairs) (c) they do not complete all experimental conditions; or (d) if they do not comply with the instructions. In order to assess the latter, data from participants who spend less than 10 seconds reading the instructions will be excluded from the analysis. We will also exclude data from any trials with response times 3 are regarded as providing substantial evidence for one hypothesis over the other. We will use an MCMC algorithm (implemented in Stan; Carpenter et al., 2017) that estimates the posteriors by sampling parameter values proportional to the product of prior and likelihood. These samples are generated through 4 independent Markov chains, with 1000 warmup samples each, followed by 50000 samples drawn from the posterior distribution which are retained for analysis. Following Gelman and colleagues (2013), we will confirm that the 4 chains converge to the same posterior distribution by verifying that the rhat statistic – reflecting the ratio of between-chain variance to within-chain variance – is < 1.05 for all parameters, and we will visually ins