Visual search and target cueing: A comparison of head-mounted versus hand-held displays on the allocation of visual attention

Wickens, C. D. · 1998 · openalex

DOI: 10.1037/e445012005-001

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 how specific augmented reality (AR) display features—target cueing, image realism, and interactivity—affect operator attention and trust biases. The research addresses the problem of "attentional tunneling," where operators focus excessively on automated guidance cues at the expense of monitoring the broader visual scene, and "trust bias," where operators over-rely on automation even when it is unreliable. The authors sought to determine whether these compelling display features induce miscalibrations in attention and trust that compromise performance in complex spatial environments. The experiment involved sixteen military personnel performing a terrain association task while searching for camouflaged targets in a simulated environment. The study manipulated three key variables: the reliability of target cueing (100% vs. 75%), the level of image realism (varying polygon count and texture detail), and interactivity (active navigation vs. passive viewing). Subjects used either head-mounted displays (HMDs) or hand-held displays. Crucially, the environment included high-priority, uncued targets to measure the degree of reliance on cueing information. Performance was assessed through target detection rates, false alarm rates, sensitivity measures, and subjective reports of inconsistencies between the simulation and paper maps. The results demonstrated that while cueing improved detection of expected targets, it significantly reduced the detection of unexpected, high-priority uncued targets, confirming the presence of attentional tunneling. This bias was exacerbated by high image realism, which increased reliance on cueing when the data was reliable. Regarding trust, subjects exhibited significant overtrust in the automation; even when cueing reliability was reduced to 75%, participants continued to rely on the cues more than warranted. This overtrust manifested as increased false alarms, decreased sensitivity, and a risky criterion shift. Notably, the loss of reliability in the cueing was so detrimental that it eliminated the performance benefits previously gained from cueing. Interactivity did not directly influence trust but increased resource demands, which modulated performance under unreliable cueing conditions. The study concludes that AR display features like cueing and high realism can induce dangerous attention and trust biases. Operators tend to cognitively "lazy" reliance on automation, failing to monitor uncued areas even when they know the system is imperfect. The findings imply that enhancing image realism or interactivity does not inherently improve trust calibration; rather, realism can dangerously amplify overreliance on flawed automation. For AR design, these results suggest that while cueing aids specific tasks, it imposes significant costs on situational awareness, particularly when the automation is not perfectly reliable. Designers must account for these biases, as the compelling nature of AR displays can lead to critical failures in detecting unexpected events.

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 2026-06-09; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• visual search
• hud ar windshield