Charting the trajectory of forgetting: Insights from a working memory period paradigm

Towse, John N.; Hitch, Graham J.; Horton, Neil · 2019 · Memory & Cognition

DOI: 10.3758/s13421-019-00916-6

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 "working memory period," a measure of working memory endurance defined by the duration of processing episodes a participant can tolerate while retaining a fixed number of items. Unlike traditional working memory span, which measures capacity (item count), period assesses the longevity of memory representations during concurrent processing. The authors aimed to replicate previous findings that working memory period predicts educational attainment and to test the generality of "card-order effects"—performance variations based on the sequence of short and long processing episodes. The research sought to determine if subtle within-trial sequence configurations significantly influence forgetting rates and whether these effects generalize across different trial structures. The study involved 184 children aged 7 to 11 years. Participants completed variants of an operation period task across two test sessions separated by approximately one week. In this task, children performed four self-paced arithmetic sums per trial, recalling the answers in serial order. Trials consisted of fixed and variable processing episodes, with variable episodes increasing in duration across levels to determine the maximum tolerable retention period. The researchers manipulated the order of short and long processing episodes in four permutations: at the end positions, central positions, the first half of the sequence, and the second half of the sequence. Additionally, participants completed standardized tests of scholastic attainment, specifically word reading and number skills. The results demonstrated that individual differences in working memory period were reliable across test sessions and significantly predictive of reading and number skills. The study replicated previous card-order effects, finding a significant advantage for sequences where short processing episodes followed long ones (short-late advantage) when these episodes were positioned at the end or in the center of the trial. A similar advantage was found when short episodes followed long ones in the first half of the sequence. However, this effect did not generalize to the second half of the sequence, where no significant advantage was observed. Furthermore, analysis of recall accuracy revealed that items associated with long processing episodes were recalled less accurately than those from short episodes, and items from variable episodes were recalled less accurately than fixed ones. A robust practice effect was also observed, with performance improving significantly between the first and second test sessions. The findings suggest that working memory period is a valid and reliable construct that captures individual differences in memory endurance distinct from capacity. The partial generalization of card-order effects indicates that working memory performance is sensitive to specific within-trial sequence configurations, challenging simplistic models that treat retention as a unitary process. The results support theoretical frameworks emphasizing rapid forgetting during processing intervals, such as the time-based resource-sharing model. By demonstrating that working memory period predicts scholastic attainment and is influenced by subtle temporal dynamics, the study argues for a broader view of working memory constraints that incorporates endurance and forgetting rates alongside traditional capacity measures.

Key finding

Working memory period is a reliable predictor of children's scholastic attainment and is systematically affected by the specific sequence order of processing episodes, though not all order manipulations yield the same performance advantages.

Methodology

lab_experiment

Sample size: 184

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. Discovered via author_sweep_intake on 2026-05-28.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

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

• working memory
• prospective memory interruptions