Chunking, boosting, or offloading? Using serial position to investigate long-term memory's enhancement of verbal working memory performance

Bartsch, Lea M.; Shepherdson, Peter · 2022 · Crossref

DOI: 10.3758/s13414-022-02625-w

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 how episodic long-term memory (LTM) enhances performance in verbal working memory (WM) tasks, specifically examining whether the serial position of pre-learned information within a memory list affects this benefit. The research aims to distinguish between three theoretical accounts of LTM-WM interaction: the chunking account, which posits that LTM items free WM capacity for subsequent items (predicting a position-dependent benefit); the trace strength account, which suggests LTM facilitates retrieval regardless of position; and the offloading account, which proposes that LTM items are not encoded into WM at all, freeing capacity independently of their position. The authors conducted two experiments using an online paradigm. Participants first learned 42 word pairs to establish episodic LTM traces. They then performed a WM task involving lists of four word pairs, where some pairs were previously learned (LTM-available) and others were novel. Crucially, the position of the LTM pairs within the four-item list was manipulated: beginning (positions 1–2), middle (positions 2–3), or end (positions 3–4). Performance was assessed via a four-alternative forced-choice test. Experiment 2 replicated this design but included a distractor task during the retention interval to test the role of the focus of attention and the stability of offloaded information. Results from Experiment 1 showed that including LTM pairs improved overall WM performance compared to lists with only new pairs, but this benefit did not vary based on the serial position of the LTM pairs. This finding contradicts the standard chunking account, which predicts greater benefits when capacity is freed early in the list. Instead, the data were initially consistent with the trace strength account. However, Experiment 2 revealed that a distractor task impaired memory for new pairs but largely protected memory for LTM pairs. This dissociation supports the offloading account: LTM pairs are offloaded to long-term storage without occupying WM resources, thereby freeing capacity for new items. The distractor interfered with the maintenance of new items in WM but did not disrupt the offloaded LTM representations. The study concludes that the enhancement of WM performance by episodic LTM is best explained by an offloading mechanism rather than chunking or trace strength alone. Access to reliable LTM representations allows individuals to bypass WM encoding for those items, reducing cognitive load. This implies that WM and LTM interact dynamically, with the WM system negotiating resource allocation based on the availability of prior knowledge. These findings refine theoretical models of memory by demonstrating that LTM can serve as a functional substitute for WM storage, independent of the temporal position of the information within a trial.

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 2026-06-26; verification: verified.

Topics

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

• working memory