Timecourse of coactivation in bimodal divided attention

Miller, Jeff · 1986 · OpenAlex-citations

DOI: 10.3758/bf03203025

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 timecourse of response coactivation in bimodal divided attention, specifically addressing how visual and auditory signals interact to produce the redundant signals effect (RSE), where responses to simultaneous signals are faster than to single signals. The research aims to distinguish between "separate-decisions" (race) models, which posit that the response is triggered by whichever signal is detected first, and "coactivation" models, which propose that both signals jointly activate a single response process. Furthermore, the study seeks to differentiate between two classes of coactivation models: "exponential" models, which are history-free and rely on instantaneous signal states, and "accumulation" models, which integrate activation over time. The experiment utilized a go/no-go detection task with two practiced subjects. Participants monitored visual (plus signs) and auditory (780-Hz tones) channels, responding as quickly as possible to any signal. Redundant signals were presented asynchronously with signal onset asynchronies (SOAs) ranging from 0 to 167 msec. The authors derived and tested mathematical inequalities based on reaction time (RT) cumulative distribution functions. Inequality 2 was used to test race models, predicting that the proportion of fast responses to redundant signals cannot exceed the sum of proportions for single signals. Inequality 3 was derived to test exponential coactivation models, asserting that responses to asynchronous signals must be attributable entirely to either the initial single signal or the subsequent redundant state, but not both. Results significantly violated the predictions of race models, particularly when the visual signal preceded the auditory signal by 67–100 msec. This timing allowed the slower visual signal to contribute to the response before the faster auditory signal triggered it, creating more fast responses than race models permit. Additionally, the data violated Inequality 3, ruling out exponential coactivation models. The violations indicated that responses were influenced by both the initial single signal and the subsequent redundant configuration, a pattern inconsistent with history-free processes. Instead, the findings supported accumulation models, where activation from the first signal contributes to a process that continues to accumulate input from the second signal over time. The study concludes that the RSE in bimodal tasks is best explained by an accumulation model of coactivation. This implies that the cognitive system integrates evidence from multiple sensory channels over a considerable period, rather than relying on instantaneous, independent processing channels or history-free decision mechanisms. These findings refine the understanding of divided attention by demonstrating that response activation is a temporal integration process, challenging simpler statistical or instantaneous models of multisensory integration.

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.

• multisensory crossmodal