Set size effects on working memory precision are not due to an averaging of slots

Pratte, Michael S. · 2020 · OpenAlex-citations

DOI: 10.3758/s13414-019-01902-5

archive: archived pipeline: cataloged verified

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

Summary

This paper investigates the mechanisms underlying the decline in visual working memory precision as memory load (set size) increases. While discrete capacity theories posit that items are either stored or lost, empirical data show that the precision of successfully stored items also decreases with set size. The dominant explanation for this phenomenon is the "slots-plus-averaging" model, which suggests that items are stored in multiple replicate slots when below capacity, and these representations are averaged during retrieval, causing precision to degrade until capacity limits are reached. The author challenges this account by testing whether set size effects on precision are specific to working memory storage or if they arise from earlier perceptual or encoding processes. To address this, the study employed three experiments using a delayed estimation task where participants reported the color of a probed item. Experiment 1 compared standard working memory conditions (1000 ms retention) with iconic memory conditions using very brief retention intervals (0 ms, 33 ms, and 50 ms). In iconic memory tasks, items can be retrieved from the sensory trace without requiring concurrent storage of multiple items in working memory. If the slots-plus-averaging model were correct, precision should not decline with set size in these conditions because the effective working memory load is one. Experiment 2 expanded on this by testing a wider range of set sizes (2 to 10) with a 100 ms retention interval to determine if precision plateaus at working memory capacity limits (~3–4 items) or continues to decline. Experiment 3 further distinguished between simultaneous and sequential presentation to rule out perceptual crowding or attentional limitations during encoding. The results consistently contradicted the slots-plus-averaging model. In Experiment 1, memory precision declined significantly with set size in both working memory and iconic memory conditions (33 ms and 50 ms), and the magnitude of this decline was statistically indistinguishable between the two. This indicates that the precision deficit is not unique to working memory storage. Furthermore, Experiment 2 revealed that precision declined monotonically across set sizes 2 through 10 in iconic memory conditions, failing to plateau at the typical working memory capacity limit of three or four items. Finally, Experiment 3 found that set size effects on precision persisted even when items were presented sequentially, ruling out simultaneous perceptual limitations as the cause. These findings imply that set size effects on memory precision do not stem from the averaging of memory slots or other working memory storage processes. Instead, the results suggest that precision declines are likely attributable to limitations in earlier stages of processing, such as encoding or attention, rather than the discrete storage mechanisms proposed by contemporary discrete capacity theories. This rejection of the slots-plus-averaging model necessitates a reevaluation of how discrete capacity theories are formulated and tested, highlighting that precision effects are not a reliable indicator of working memory storage limits.

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