Mixing it up: Intermixed and blocked visual search tasks produce similar results

Wolfe, Jeremy M.; Hong, Injae; Mitra, Ava A.; Objio, Eduard; Khalifa, Hula; Ali, Yousra · 2025 · Crossref

DOI: 10.3758/s13414-025-03077-8

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 the principles of visual search derived from laboratory experiments using blocked trials apply to real-world scenarios where search tasks are frequently intermixed. Traditional visual search research typically involves observers performing hundreds of trials of the same task in a block, allowing for the establishment of rules regarding reaction times and errors. However, real-world search often involves switching targets or distractors with each trial. The authors sought to determine if mixing different search tasks qualitatively alters search behavior compared to blocked conditions, and whether such mixing disrupts adaptive models of search termination, particularly on target-absent trials. The researchers conducted five experiments comparing mixed and blocked conditions. Experiment 1 tested four different feature searches (color, lighting direction, orientation, and Vernier offset) with 39 participants. Experiments 2 and 3 involved searching for the same target (a green "O") across four tasks defined by different distractor sets, ranging from easy color contrasts to difficult conjunction searches. Experiment 4 searched for different targets among constant distractors, and Experiment 5 allowed participants to choose tasks. In all experiments, participants performed tasks in both blocked sequences (100 trials per task) and mixed sequences (tasks randomly intermixed). Data were collected online, measuring reaction times and accuracy across varying set sizes (8, 16, and 24 items). The results demonstrated no qualitative change in search behavior between mixed and blocked conditions. In Experiment 1, reaction time slopes and error rates were similar across conditions, with mixed conditions sometimes showing faster performance, likely due to learning effects since blocked trials were administered first. Analysis of search termination on target-absent trials revealed that participants could maintain separate quitting thresholds for different tasks even when mixed, showing no significant disruption in their ability to terminate unsuccessful searches. Experiments 2 and 3 confirmed these findings across a wide range of task difficulties, showing that mixing tasks did not impair search efficiency or increase error rates. The data indicated that observers could switch between search rules with little to no cost. The findings support the generalizability of visual search rules established through blocked laboratory trials to more naturalistic, mixed-search environments. This suggests that decades of research on blocked trials remain relevant for understanding real-world visual attention. However, the results pose a challenge to simple adaptive models of search termination, which often assume that quitting thresholds are adjusted based on recent trial history within a single task type. The ability of observers to maintain distinct termination rules for different tasks in a mixed sequence implies a more complex mechanism for managing search termination than previously modeled.

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
• inattentional change blindness
• task switching