Classic Visual Search Effects in an Additional Singleton Task: An Open Dataset

Abstract

Visual search refers to our ability to find what we are looking for among many competing visual inputs. Here, we report the availability of a rich dataset that replicates key visual search effects and shows that these effects are robust to several changes to the experimental design. Experiment 1 replicates classic findings from an additional singleton visual search task. First, participants are captured by a salient but irrelevant color singleton, as indexed by slower response times when a color singleton distractor is present versus absent. Second, attentional capture by a color singleton is reduced when the visual search array contains heterogeneous shapes rather than homogenous shapes. Finally, attentional capture by a color singleton is reduced when the display colors are repeated rather than switched unpredictably from trial to trial. Experiment 2 demonstrates that these classic visual search effects are robust to small procedural changes such as task timing (i.e., a 2-8 second rather than ~1 second inter-trial interval). Experiment 3 demonstrates that these classic effects are likewise robust to changes to the distractor frequency (75% rather than 50%) and to fully blocking versus interleaving blocks of two task conditions. All told, this dataset includes 8 sub-experiments, 190 participants and >210,000 trials, and it will serve as a useful resource for power analyses and exploratory analyses of visual search behaviors.

Keywords: additional singleton task; attentional capture; distractor suppression; stimulus history.

Figure 1

Schematic of the task conditions. On each trial of the task, participants searched for the target (diamond shape) and reported the orientation of the line inside (horizontal or vertical). After a blank inter-trial interval, the next search display appeared. Non-target shapes could be heterogeneous (assorted shapes) or homogeneous (all circles). The display colors could vary randomly from trial to trial (color variable) or stay constant within the entire block (color constant). The four conditions in this figure were collected across-subjects in Experiments 1a–1d. The conditions in the top row (Exp 2A) and bottom row (Exp 2B) were collected within-subjects in Experiment 2. The conditions in the right column were collected within-subjects in Experiment 3A and 3B.

Figure 2

Response times as a function of condition in Experiment 1. We observed expected signatures of a singleton-detection strategy (homogeneous non-target shapes) versus feature search strategy (heterogeneous non-target shapes), and both singleton-detection and feature search were modulated by stimulus history (color constant vs. variable). Gray violins indicate “distractor absent” trials; colored violins indicate “distractor present” trials. Individual participants are shown as dots and transparent gray lines. Each subplot shows the response time data from a sub-experiment (1A–1D). Asterisks indicate uncorrected post-hoc comparisons between each adjacent pair of violins (distractor present vs. absent), n.s. p ≥ .05, ** p < .01, *** p <.001.

Figure 3

Response times as a function of condition in Experiment 2. Key visual search effects were preserved when a longer inter-trial interval was used (2–8 sec). Each subplot shows the response time data from a sub-experiment In Experiment 2, all displays were set size 4. Gray violins indicate “distractor absent” trials; colored violins indicate “distractor present” trials. Individual participants are shown as dots and transparent gray lines. Asterisks indicate uncorrected post-hoc comparisons between each adjacent set of bars, n.s. p ≥ .05, ** p < .01, *** p < .001.

Figure 4

Response times as a function of condition in Experiment 3. When participants are in singleton-detection mode, attentional capture by a salient distractor is attenuated when colors are repeated (Color Constant) compared to when colors randomly vary (Color Variable). This general pattern did not differ as a function of interleaving (3A) versus grouping (3B) blocks of the two stimulus history conditions. In Experiment 3, all displays were set size 4 and non-target items were always homogeneous (singleton-detection mode). Gray violins indicate “distractor absent” trials; colored violins indicate “distractor present” trials. Individual participants are shown as dots and transparent gray lines. Asterisks indicate uncorrected post-hoc comparisons between each adjacent set of bars, n.s. p ≥ .05, **p < .01, ***p < .001.