Attentional limits in memory retrieval.

Carrier, L. Mark; Pashler, Harold · 1995 · Journal of Experimental Psychology Learning Memory and Cognition

DOI: 10.1037//0278-7393.21.5.1339

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 whether episodic memory retrieval can operate in parallel with other cognitive processes or if it is subject to a central processing bottleneck. The authors address conflicting conclusions in prior literature: some dual-task studies suggested memory retrieval is parallel, while psychological refractory period (PRP) studies indicated a bottleneck in response selection. The authors argue that previous dual-task studies confounded memory retrieval with response selection. To isolate retrieval, they employed a PRP design where participants performed a speeded auditory choice reaction task (Task 1) concurrently with a memory retrieval task (Task 2). In Experiment 1, participants performed cued recall of word pairs. Task 1 involved classifying tone pitch, and Task 2 required verbally recalling the associate of a presented cue word. The duration of the retrieval stage was manipulated by testing pairs either once (first-presentation) or twice (second-presentation), with the latter expected to yield faster retrieval. Stimulus onset asynchronies (SOAs) between Task 1 and Task 2 stimuli varied from 50 to 1,200 ms. Experiment 2 utilized a similar design but replaced cued recall with speeded yes-no recognition, manipulating retrieval duration by varying the number of prior presentations of target words (one vs. five). The results from both experiments demonstrated that reducing the SOA significantly increased reaction times for the memory task. Crucially, the slowing effect caused by reduced SOA was additive with the effects of variables that manipulated retrieval duration (presentation condition). According to PRP theory, an additive interaction indicates that the manipulated stage (memory retrieval) occurs at or after the bottleneck. If retrieval occurred in parallel with Task 1 processing, an underadditive interaction would be expected. The data showed no such underadditivity; instead, retrieval was delayed by the concurrent choice task. These findings support the "response-selection/retrieval bottleneck hypothesis," concluding that memory retrieval cannot occur in parallel with other cognitive processes requiring response selection. The authors argue that the central bottleneck responsible for dual-task interference encompasses memory retrieval, not just response selection. This challenges the view that retrieval is a parallel process and suggests that attentional limits apply to the retrieval of information from memory as well as the selection of motor responses.

Key finding

Memory retrieval is subject to a central processing bottleneck that prevents it from occurring in parallel with response selection in a concurrently performed task.

Methodology

lab_experiment

Sample size: 60

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 openalex_abstract on 2026-05-08.

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.

• dual task multitasking