Presenting segmented images in a rapid serial visual presentation stream improves search accuracy

Diaz, Krystina; Becker, Mark W.; Peltier, Chad; Bolkhovsky, Jeffrey B. · 2025 · Crossref

DOI: 10.1186/s41235-025-00653-2

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 whether presenting segmented images in a rapid serial visual presentation (RSVP) stream improves target detection accuracy compared to viewing full images, addressing the high rate of missed targets in real-world visual search tasks like radiology and security screening. The authors hypothesize that RSVP mitigates "selection errors"—where observers fail to inspect the target region or terminate search prematurely—by forcing systematic inspection of the entire display within a fixed time. Five experiments were conducted to test this hypothesis and identify the underlying mechanisms, comparing Full-Image conditions against various RSVP conditions while keeping total search time constant. Experiment 1 replicated prior findings using simple Landolt C arrays, demonstrating that RSVP significantly increased hit rates (69.55%) compared to Full-Image search (62.95%). Experiment 2 extended this to complex "Where’s Waldo" scenes, finding that RSVP not only increased hit rates (65.15% vs. 49.56%) but also increased sensitivity (d-prime), despite a concurrent rise in false alarms indicating a liberal shift in decision criterion. Experiment 3 tested whether reduced peripheral clutter drove these gains by adding background clutter to RSVP segments; results suggested that clutter reduction was not the primary mechanism for improved performance. Experiments 4 and 5 examined the role of eye movements. Experiment 4 prompted eye movements via RSVP segment locations, while Experiment 5 limited non-search-related eye movements. These experiments indicated that lower performance in Full-Image conditions under time pressure is attributable to inefficient, irrelevant eye movements that waste time. The findings conclude that RSVP enhances search accuracy by maximizing the time available for inspecting and processing each segment, thereby preventing premature search termination and ensuring comprehensive coverage of the search area. The advantage arises because RSVP controls presentation timing and location, eliminating the observer’s ability to skip regions or quit early. The authors suggest that segmenting search displays and presenting them in an RSVP stream could significantly improve performance in high-stakes real-world applications, such as TSA baggage screening or medical imaging, where missing targets has severe consequences. This approach offers a viable strategy for enhancing visual search efficiency without increasing total search time.

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