Selective attention and serial processing in briefly presented visual displays

Eriksen, Charles W.; Colegate, Robert L. · 1971 · OpenAlex-citations

DOI: 10.3758/bf03207451

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 mechanisms of selective attention and serial processing in briefly presented visual displays. The research addresses how observers select specific items from a multi-item display when only a few can be reported accurately. Previous work suggested that selection occurs within a 200–300 millisecond window, potentially involving a serial encoding process where items are transferred from an iconic representation to short-term memory. The authors aimed to determine whether performance decrements in multi-item tasks result from the difficulty of locating multiple indicators or from the serial nature of encoding letters, during which the visual icon deteriorates. The researchers conducted two experiments using a tachistoscope to present circular arrays of eight capital letters for brief durations (mean 38 msec). Subjects reported letters designated by indicators (lines) presented either simultaneously with the display, 250 msec prior, or in other configurations. Experiment 1 involved five conditions: single indicator/single response, two simultaneous indicators/two responses, early and simultaneous indicators/two responses, single indicator/two responses (indicator between two letters), and opposite/single response. Experiment 2 replicated these conditions with a new subject group and added control conditions, including early and delayed indicators, to verify the timing effects and address potential discriminability issues with specific letters. The results demonstrated that accuracy varied significantly based on the experimental condition. In conditions requiring two responses, the second reported letter was often less accurate than the first, particularly when indicators were simultaneous or when one preceded the display. However, when the two designated letters were spatially separated, the accuracy of the first response increased while the second decreased, suggesting a "tradeoff" where resources allocated to the first encoded item impacted the second. Crucially, average accuracy across different two-response conditions was comparable, indicating that the primary limitation was not the number of indicators but the serial encoding of the letters themselves. The data supported the hypothesis that encoding is a serial process; as the first letter is encoded, the iconic representation of the display deteriorates, reducing the clarity available for encoding subsequent letters. The findings support a model of selective attention in which items are encoded serially from an iconic store. The study concludes that the capacity to report multiple items is constrained by the time required for serial encoding and the concurrent decay of the visual icon. This implies that selective attention involves a sequential processing mechanism rather than parallel processing of all displayed items, with performance determined by the order of encoding and the stability of the iconic representation during that process.

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.

• visual search