The role of the working memory storage component in a random-like series generation

Biesaga, Mikołaj; Nowak, Andrzej · 2024 · OpenAlex-citations

DOI: 10.1371/journal.pone.0296731

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the cognitive mechanisms underlying human attempts to generate random-like series, specifically focusing on the role of working memory. Humans lack an innate random generator and instead attempt to simulate random processes, a task often biased by the "gambler’s fallacy" and limited by working memory capacity. The authors test a theoretical model proposing that random series generation involves both the storage component (maintaining recent choices) and the processing component (evaluating randomness) of working memory. The research aims to determine how reducing the load on the storage component affects performance and how individual differences in working memory capacity and randomness judgment influence the algorithmic complexity of generated series. The researchers conducted two studies using algorithmic complexity, grounded in Kolmogorov-Chaitin theory, as a robust measure of randomness. Study 1 employed a between-subjects experimental design where participants generated a binary series of 120 elements. The key manipulation was the visibility of the last seven generated elements: in the "visible" condition, past choices were displayed, reducing the cognitive load on working memory storage; in the "invisible" condition, participants had to rely solely on internal memory. Participants also completed a randomness judgment task. Study 2 utilized a correlational design to examine the relationship between individual working memory capacity and the complexity of generated series. The results supported the hypothesis that working memory storage constraints significantly impact random series generation. In Study 1, participants in the visible condition, who did not have to actively maintain past choices in their working memory, were able to sustain high-quality, random-like performance for a longer duration compared to those in the invisible condition. This indicates that offloading the storage component mitigates the typical decay in randomness over time. Additionally, Study 1 found a positive correlation between the ability to distinguish complex patterns in the judgment task and the algorithmic complexity of the series generated, suggesting that better randomness judgment facilitates better generation. Study 2 revealed that individuals with higher working memory capacity produced series with greater algorithmic complexity. These findings highlight that the quality of human-generated random series is not merely a result of flawed heuristics but is fundamentally constrained by working memory resources. The study demonstrates that the storage component of working memory plays a critical role in maintaining the context necessary for evaluating randomness. By showing that reducing storage load prolongs high-quality performance and that higher capacity correlates with more complex outputs, the research provides empirical support for models that distinguish between storage and processing components in this cognitive task. This offers insights into the dynamic nature of random-like series production and the specific cognitive bottlenecks involved.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• working memory