Abstract
Target and distractor templates play a pivotal role in guiding attentional control during visual search, with the former template facilitating target search and the latter template leading distractor suppression. We first investigated whether task-irrelevant colors could earn their value through color-target contingency in the training phase and bias attention when they became a distractor in search for a singleton shape during the test phase. Colors provided useful information for target selection, with high- and low-informational values, respectively, in Experiments 1 and 2. Experience-based attentional biases were observed in the first half of the former experiment, and null results were observed in the latter. Experiment 3 verified whether the null results were elicited because the response-relevant feature inside of the singleton shape was also a singleton. Colors were task defined in the training phase, and the test display was the same as that used in Experiment 2. Experience-based attentional biases were observed in the first half of the test phase. In Experiment 4, we tested whether decreasing the consistency of distractor processing can lengthen the duration of experience-based attentional biases by increasing the number of possible response-relevant features inside of the colored distractor. The results showed experience-based attentional biases throughout the test phase. The results highlight the ideas that the informational value provided by a feature dimension for facilitating target selection can modify a target template and that the consistency of rejecting a distractor feature can play a role in the formation of a distractor template.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
People conduct visual searches frequently in everyday life. Visual search, which requires attentional control, is modulated by the following three factors: top-down task goals, bottom-up stimulus salience, and selection history (Awh et al., 2012; Failing & Theeuwes, 2018; Luck et al., 2021). The influence of selection history on attentional control can be of short-term or long-term durations. The short-term effects of selection history are evident in intertrial priming, where a prior episode is reactivated for a period of approximately 5–8 subsequent trials (Kruijne et al., 2015; Maljkovic & Nakayama, 1994; Nakayama et al., 2004) or following 15 intervening trials (Thomson & Milliken, 2012a, 2012b, 2013). The long-term influence of selection history is observed across blocks (Bogaerts et al., 2022; Chun & Jiang, 1998; Kruijne & Meeter, 2015, 2016; B. Wang & Theeuwes, 2018; Yeh et al., 2014), after a training phase with hundreds of trials (Leber & Egeth, 2006; Sha et al., 2017a, 2017b) or across sessions occurring one day (Z. Huang & Li, 2022; Kim & Anderson, 2019; Sha & Jiang, 2016), one week (Kruijne & Meeter, 2016; Leber et al., 2009), or two weeks (Knight et al., 2016) later. Compared with the cross-trial priming, the long-term influence reflects the impact of memory traces accumulated over a longer period of prior experiences.
Long-term habitual biases are of interest, as the results from this line of work have important implications for designing clinical interventions or educational programs by understanding how habitual biases are developed over time and how such biases can be reduced through new learning. Among the experimental contexts that have been adopted in prior research, the results from the training-transfer paradigm demonstrated how habitual biases resist attentional control based on bottom-up salience. By using a training-transfer paradigm, Leber and colleagues (Leber & Egeth, 2006; Leber et al., 2009) asked participants to search for a target of a consistent color among heterogeneous colored distractors (feature-search group) or search for a unique color among all gray distractors (singleton-search group) in the training phase. In the test phase, the target was in a consistent color among gray distractors. The target color pops out so that a search can be accomplished via priority guidance (Liesefeld & Müller, 2020) in a singleton-search mode (Leber & Egeth, 2006). However, participants in the feature-search group continued using the same feature-search strategy that they learned in the training phase.
More interestingly, such habitual biases resist not only bottom-up salience but also a change in the task goal. Sha and Jiang (2016) asked their participants to search for a red or green circle among circles in heterogeneous colors and to assess the orientation of the line inside of the target circle in the training phase. The task was changed to search for a singleton shape and to assess the line orientation that was located inside in the test phase. The reaction was slower when a distractor was in a previously designated target color compared with a neutral condition, in which either designated color was absent. Despite the new task goal and bottom-up salience in the target shape, attention was biased based on prior experiences. It is likely that the adoption of the same search mode or strategy could provide the benefit of offloading attentional demands in adaptation to task demands in a dynamic environment (Anderson et al., 2021).
Strong experience-based biases may have arisen from attentional priming of a strong target template formed in the training phase. With the generation of an episodic trace each time a perceptual stimulus aligns with the target template (Logan, 1988), the priority of target features is amplified after each successful match. Target repetition strengthens the representation of the target template. The target template is maintained in working memory to guide visual search (Bundesen, 1990; Bundesen et al., 2005; Chelazzi et al., 1993, 1998; Desimone & Duncan, 1995; Duncan & Humphreys, 1989; Li, 1999, 2002; Wolfe, 2021; Zelinsky & Bisley, 2015). When a strong target template is activated, it facilitates target processing.
The target template can be task defined or manifested through implicit learning of the regularities in the visual environment. Moreover, a template can contain the probability distribution of a target or distractor feature (see Kristjánsson, 2023). Task-defined templates are formed by the task rule through instructions such as searching for a conjunction target (Berggren & Eimer, 2018; Kaptein et al., 1995; Kristjánsson et al., 2002; Sha et al., 2017a) or explicit cues to attend or ignore stimuli with a specific feature (see Carlisle, 2023). Templates can also be formed from implicit learning of regularities in the visual environment. The regularity may manifest in the association between a target location and the spatial configuration of the search display (Chun & Jiang, 1998), the predictable remapping of multiple target locations (Conci & Müller, 2012), and the priority of different target colors (Munneke et al., 2020). Apart from target regularity, regularity can manifest in a distractor property such as the color (Stilwell & Vecera, 2023; Thayer et al., 2023), location (Zhang et al., 2022), and the distribution shape of line orientations with its summary statistics (Chetverikov et al., 2016, 2017). Furthermore, regularity in both target and distractor locations could be retained simultaneously to influence visual search (C. Huang et al., 2022).
As a template can be formed through a top-down task rule or by learning the regularity in the visual environment, the finding that target-defined colors do not lead to durable experience-based attentional biases is of interest. To demonstrate that value-driven attentional biases are different from history-driven biases, Anderson and colleagues (Anderson & Halpern, 2017; Anderson et al., 2011) showed that two task-defined colors did not induce durable experience-based biases. In contrast, color-driven attentional biases were found when one of the two colors was associated with a high reward. Anderson et al. (2021) have suggested that a feature that is useful for target localization can lead to a history of the sought target and induce an enduring attentional bias even when the feature is no longer task relevant. The null results in the control experiments (Anderson & Halpern, 2017; Anderson et al., 2011) showed that usefulness alone may not be sufficient to induce durable attentional biases. Although useful for target localization, colors were not prioritized as a crucial feature within the target template.
The first two experiments of the present study investigated how task-irrelevant colors can induce enduring attentional biases through the use of a training-test-phase paradigm. Two colors consistently direct target localization with a high- and low-informational value for goal accomplishment in Experiment 1 and Experiment 2, respectively. We postulated that for a target-associated feature to gain importance within the target template, it must provide a high-informational value for achieving the behavioral goal to entice the learning mechanism to increase its priority weight. The search context must be difficult so that each successful guidance to the target reinforces the learning mechanism to adjust the priority weight of the feature in the target template. With the accumulation of successful guidance, the feature earns a high priority and becomes an important element of the target template.
Experiment 1
The objective of this experiment was to investigate the context for a task-irrelevant color feature to be upweighted in a target template to induce durable attentional biases. Prior studies (Anderson & Halpern, 2017; Anderson et al., 2011) have shown that a task-defined color did not elicit a durable attentional bias unless this color was associated with a high reward. In the search context of these control experiments, the task was to report the orientation of the only canonical line that was in the display. Even though colors were task defined, priority guidance (Liesefeld & Müller, 2020) by orientation in a singleton-search mode (Bacon & Egeth, 1994) could easily accomplish the behavioral goal. Colors were useful; however, the processing of the colors provides a low marginal utility beyond the processing of orientation features. Colors had a low-informational value so that the incentive was low for the learning mechanism to upweight the associated colors in the template. When color processing could provide a high level of informational value, the incentive to upgrade the priority weight would increase, and colors should induce durable experience-based attentional biases.
The search context in the current experiment contained six colored circles with a notched square shown within each circle. The participants were asked to search for the only horizontal notch and to assess the gap direction, while vertical notches were shown in other circles. Importantly, the target was always embedded in red or green so that each color could consistently guide attention toward the target. As the response was based on the gap direction, colors were task irrelevant. Given the low perceptual discriminability of the response-relevant feature dimension, target search requires scrutinizing across the stimuli in the display. Color-target contingency can guide attention toward the target and ease search difficulty. Moreover, episodic traces of successful color-guided target localization highlight the utility and increase the incentive for the learning mechanism to increase the priority weight of the colors in the target template. With the heightened priority based on the learned color association to target selection, durable experience-based attentional biases should be observed in the test phase.
Method
Participants
Thirty volunteers participated in the experiment for a monetary reward (NT $150). We pre-determined a sample size of 30 participants based on two criteria. First, a large effect size (i.e., Cohen's d = 0.88) of color-induced attentional biases was observed in the first experiment by Sha and Jiang (2016), where colors were necessary for target search. To achieve this effect size, a sample size of 19 participants reaches 95% power (two-tailed, alpha level = .05) based on the analysis using G*Power 3.10 (Faul et al., 2009). Second, sample sizes ranging from 24 to 40 participants were adopted in previous studies that addressed similar issues (Anderson & Halpern, 2017; Sha & Jiang, 2016; Wang & Theeuwes, 2018). All participants had normal or corrected-to-normal vision and were naïve regarding the purpose of the experiment. The National Taiwan Normal University Research Ethics Committee reviewed and approved this study (202005HS022). All participants read and signed an informed consent form.
Design
The experiment consisted of two phases—a training phase and a test phase. In the training phase, six heterogeneous colored circles appeared on the display. Each circle had a notched square inside. The participant’s task was to search for the only horizontal notched square and judge the direction of the gap. The horizontal notched square was always embedded inside red or green color. In the test phase, the task was to search for a singleton shape and judge the gap direction of the horizontal notched square inside the shape. In the invalid condition (50% of the trials), red or green was presented as a colored distractor. In the neutral condition (50% of the trials), red or green colors were absent. The two conditions occurred in random order.
Stimuli
Stimuli were presented on a black background (0.11 cd/m2) at a viewing distance of 75 cm. The fixation display contained a white fixation cross (0.5° × 0.5° visual angle) at the center of the screen. The search display in the training phase consisted of six colored circles (2.3° × 2.3°) equidistantly placed on the rim of an imaginary circle (a radius of 5° visual angle). A white horizontal notched square was always embedded inside red (26.02 cd/m2) or green (87.93 cd/m2) with equal frequency for each color assignment across trials. The top and bottom notched squares were randomly chosen to be embedded inside the other five colored circles: blue (10.78 cd/m2), cyan (95.86 cd/m2), pink (35.58 cd/m2), orange (43.49 cd/m2), yellow (114.30 cd/m2), and white (127.90 cd/m2). The search display in the test phase was similar to that presented in the training phase, with a white notched square inside the six colored shapes in the display. Half of the test trials showed five circles and one diamond; the other half showed five diamonds and one circle in random order. The notched square inside the unique shape was always horizontal.
Procedure
The experiment began with a training phase, with 50 practice trials and two blocks of 120 training trials each. This duration of the training phase is comparable to that used in the control experiment of the study by Anderson and Halpern (2017). Each trial began with a fixation presented for a random interval selected from 400, 500, and 600 ms. The search array was then presented and remained on the screen until the participant responded or 800 ms had elapsed. Participants were required to search for the only square with a horizontal notch and report whether the notch was on the left or right side as quickly and as accurately as possible (Fig. 1A). Additionally, they were asked to press the ‘‘m’’ key for a right gap or the ‘‘z’’ key for a left gap. The search display was followed by a blank screen for 1,000 ms. When the participant did not press any key within 800 ms, a “Too slow” feedback was shown on the screen for 1,500 ms, and a 1000-Hz tone was emitted for 500 ms. When the participant pressed the wrong key, an “Incorrect” feedback and the1000-Hz tone were presented. The intertrial interval was 1,000 ms (see Fig. 1A). Once the training phase was completed, the participants rested for at least 5 min and performed a singleton-shape search task in the test phase, with 20 practice trials and two blocks of 120 experimental trials each. The trial sequence was the same as in the training phase (see Fig. 1B), and a “Too slow” feedback was presented when the participant did not press any key within 1,200 ms. Participants were told to search for a singleton shape to report the gap direction of the horizontal notched square inside the singleton shape.
The trial sequences in the training phase (A) and the test phase (B) of Experiment 1. Participants were asked to search for the only square with a horizontal notch inside and to report the notch direction in the training phase. In the test phase, participants were told to search for a singleton shape and to report the direction of the horizontal notch inside the shape. (Color figure online)
Results and discussion
In all four experiments, data were trimmed with 3 standard deviations as the cutoff criterion. We focus on the results observed in the test phase. Both accuracy and reaction-time (RT) data in the training phase showed performance improvement as training proceeded. In addition to the traditional method, Bayesian analysis was conducted on RT data using JASP (JASP Team, 2022) to verify the strength of evidence for supporting a hypothesis that is of importance.
Data from one participant were excluded from the analysis because of poor performance in the test phase, with 3 standard deviations below the mean accuracy. Thus, the data of 29 participants were analyzed (66% female, mean at 21.48 ± 2.25 years of age).
The mean accuracy observed in the test phase was 91% (see Table 1). The results from a 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) repeated-measures analysis of variance (ANOVA) showed that only the main effect of block was significant, F(1, 28) = 44.08, p < .001, \({\upeta }_{p}^{2}\) = .62, BF10 = 46,406.59, with a higher accuracy in Block 2 (93%) than in Block 1 (88%). Neither the main effect of trial type, F(1, 28) = 2.09, p = .159, \({\upeta }_{p}^{2}\) = .07, BF10 = 0.36, nor its interaction with block, F(1, 28) = 0.76, p < .001, \({\upeta }_{p}^{2}\) < .01, BFexcl = 3.40, reached significance.
Reaction times exceeding 3 standard deviations in each condition of each participant were removed, thus excluding 0.52% of the total trials. The RT data are plotted in Fig. 2. A 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) repeated-measures ANOVA revealed a significant main effect of block, F(1, 28) = 5.26, p = .030, \({\upeta }_{p}^{2}\) = .16, BF10 = 2.52. Mean RT was faster in Block 2 (666 ms) than in Block 1 (679 ms). The main effect of trial type was not significant, F(1, 28) = 0.48, p = .496, \({\upeta }_{p}^{2}\) = .02, BF10 = 0.28. Importantly, the interaction between block and trial type was significant, F(1, 28) = 7.80, p = .009, \({\upeta }_{p}^{2}\) = .22, BFincl = 5.05. In Block 1, performance was faster in the neutral condition (675 ms) than in the invalid condition (684 ms), F(1, 28) = 6.37, p = .018, \({\upeta }_{p}^{2}\) = .19, BF10 = 2.89, whereas the simple effect was not significant in Block 2, F(1, 28) = 1.59, p = .219, \({\upeta }_{p}^{2}\) = .05, BF10 = 0.50.
The results of the test phase in Experiment 1: A durable bias was observed in the first half of the test phase, with longer reaction time in the invalid condition than in the neutral condition. Error bars show the standard error (SE) of the mean
The results showed that task-irrelevant colors induced experience-based attentional biases in the first half of the test phase. According to the conceptual model proposed by Liesefeld and Müller (2020), priority guidance is unlikely to accomplish the behavioral goal in the search context, as no item pops up on the display during the training phase. Clump scanning is needed to scrutinize the notched squares before localizing the target. As the perceptual discriminability of response-relevant features is low, color-target contingency can successfully guide attention toward the target for response selection. Each episodic trace of successful goal accomplishment can reinforce the value of color information. Across the training trials, color information earns its value, and its priority weight is increased to become an important feature of the target template. Predictive information becomes valuable only when it can effectively and efficiently accomplish the behavioral goal. However, the bias lasted for only half of the test phase.
Experiment 2
The results of Experiment 1 demonstrated experience-based attentional biases in the first half of the experiment. Visual search was difficult; thus, color-target contingency provides useful information with a high value. In contrast, the target was an orientation singleton in the control experiments of previous studies (Anderson & Halpern, 2017; Anderson et al., 2011). Therefore, priority guidance (Liesefeld & Müller, 2020) by orientation can achieve the behavioral goal efficiently. The color information provides a low marginal utility beyond the processing of orientation features. The value of color information is low; therefore, colors would not be upweighed in the target template. The aim of this experiment was to demonstrate that colors with a low-informational value could not lead to durable attentional biases. Moreover, explicit knowledge of the color-target contingency would not increase the incentive for the learning mechanism to adjust the weight of the target-associated colors in the target template. We expected to replicate the null results that were observed in the control experiments of previous studies (Anderson & Halpern, 2017; Anderson et al., 2011), regardless of whether the participants were informed of the color-target contingency or not.
Method
Participants
Fifty-two volunteers (54% female, 21.12 ± 2.59 years of age) participated in the experiment for either course credits or monetary rewards (NT$ 150). All participants had normal or corrected-to-normal vision and were naïve regarding the purpose of the experiment.
Design, stimuli, and procedure
All aspects of this experiment were the same as those adopted in Experiment 1, except for two changes. First, participants were randomly assigned to one of two conditions. In one condition, the participants were not informed of the color-target contingency (not-informed group), and the participants in the other condition were told about the contingency (informed group). Second, the horizontal notched square was replaced by a white canonical line while the vertical notched squares were replaced by tilted lines in both the training and test phases. The task in the training phase was to search for the canonical line and assess whether it was vertical or horizontal. The task in the test phase was to search for the singleton shape and judged the direction of the canonical line inside. The line orientation inside the other five colored circles was tilted, randomly selected from 45° to the left and 45° to the right. Figure 3 illustrates a trial procedure in the training phase (Fig. 3A) and the test phase (Fig. 3B).
The trial sequences in the training phase (A) and the test phase (B) of Experiment 2. In the training phase, participants were instructed to judge the direction of a canonical line. In the test phase, participants were asked to search for a singleton shape and judge the orientation of the canonical line inside this unique shape. (Color figure online)
Results and discussion
The mean accuracy observed in the test phase was 91% (see Table 1). A 2 (condition: not-informed vs. informed) × 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) mixed ANOVA showed no significant main effects of condition, F(1, 50) = 0.30, p = .585, \({\upeta }_{p}^{2}\) = .01, BF10 = 0.55, block, F(1, 50) = 0.71, p = .404, \({\upeta }_{p}^{2}\) = .01, BF10 = 0.33, and trial type, F(1, 50) = 1.29, p = .262, \({\upeta }_{p}^{2}\) = .03, BF10 = 0.29. The analysis also showed no significant interactions of Condition × Block, F(1, 50) = 0.20, p = .658, \({\upeta }_{p}^{2}\) < .01, BFexcl = 12.20, Condition × Trial Type, F(1, 50) = 2.77, p = .102, \({\upeta }_{p}^{2}\) = .05, BFexcl = 6.37, Block × Trial Type, F(1, 50) = 2.53, p = .118, \({\upeta }_{p}^{2}\) = .05, BFexcl = 8.70, and Condition × Block × Trial Type, F(1, 50) = 0.91, p = .345, \({\upeta }_{p}^{2}\) = .02, BFexcl = 40.00. We then excluded RTs that exceeded 3 standard deviations in each condition for every participant, resulting in the exclusion of 2.44% of the total trials. The RT data of the correct trials are plotted in Fig. 4. A 2 (group: not-informed group vs. informed group) × 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) mixed ANOVA also showed no significant main effects of condition, F(1, 50) = 0.58, p = .451, \({\upeta }_{p}^{2}\) = .01, BF10 = 154.56, block, F(1, 50) = 1.96, p = .168, \({\upeta }_{p}^{2}\) = .04, BF10 = 0.44, and trial type, F(1, 50) = 0.04, p = .835, \({\upeta }_{p}^{2}\) < .01, BF10 = 0.18. The analysis also showed no significant interactions of Condition × Block, F(1, 50) = 1.45, p = .234, \({\upeta }_{p}^{2}\) = .03, BFexcl = 9.80, Condition × Trial Type, F(1, 50) = 2.48, p = .122, \({\upeta }_{p}^{2}\) = .05, BFexcl = 12.99, Block × Trial Type, F(1, 50) = 0.09, p = .763, \({\upeta }_{p}^{2}\) < .01, BFexcl = 38.46, and Condition × Block × Trial Type, F(1, 50) = 0.33, p = .566, \({\upeta }_{p}^{2}\) < .01, BFexcl = 200.00.
The results of the test phase in Experiment 2: Reaction times did not differ between the neutral and invalid conditions in the test phase. Error bars show the standard error (SE) of the mean
In comparison with Experiment 1, the informational value of the colors was reduced by presenting orientation-singleton targets in Experiment 2. In this context, the task could be accomplished by priority guidance (Liesefeld & Müller, 2020) based on the processing of orientation features. Color information provides limited marginal utility and hence has a low value for goal accomplishment. As a results, although the target-associated color could still provide target-related information in the training phase, the priority of the target-associated color was not upweighted and could not lead to experience-based attention biases. This finding is similar to previous studies (Anderson & Halpern, 2017; Anderson et al., 2011), suggesting the importance of informational value.
Experiment 3
The results of Experiments 1 and 2 showed that color-target contingency only induced experience-based attentional biases in the first half of the test phase when the contingency provided useful information with a high value for target selection in the training phase. When the informational value of color-target contingency was low for goal accomplishment in the training phase, even the knowledge of the contingency did not lead to durable experience-based attentional biases. In addition to informational value, another reason for observing different results may be due to the search context during the test phase. Although the participants in both experiments were asked to evaluate the direction of the stimulus inside of a singleton shape in the test phase, the target search was much easier in Experiment 2 than in Experiment 1. The target was both a singleton in shape and also a singleton in orientation. The perceptual discriminability of the response-relevant feature dimension was high among the displayed stimuli. The target could easily win the race in the test phase of Experiment 2. In contrast, the perceptual discriminability between the target and distractors was low in the test phase of Experiment 1. The singleton shape may not win the race in every trial.
To clarify whether search difficulty in the test phase may have led to different observations, we conducted this experiment with color information that had a high value in the training phase and an easy search in the test phase. Following Sha and Jiang’s (2016) study, we adopted a search context in which color information is necessary during the training phase. The participants were asked to search for a red or green circle and to assess whether the line inside of the circle was vertical or horizontal. All of the lines inside of the colored circles were canonical. The search context in the test phase was identical to that used in Experiment 2, with the line inside of the singleton shape being the only canonical line in the display. The perceptual discriminability of the response-relevant feature dimension was high among the stimuli.
Method
Participants
Twenty-six volunteers participated in the experiment for course credits or monetary rewards (NT$ 150). An analysis of statistical power was conducted using Superpower (Lakens & Caldwell, 2021). Based on the interaction effect observed in Experiment 1, it was determined that a minimum sample size of 23 participants was necessary to achieve 80% power with an alpha level of 0.05.
Design, stimuli, and procedure
All aspects were similar to that used in Experiment 2, except for the search context in the training phase. Participants were instructed to search for a red or green circle and judge the orientation of the line inside the circle in the training phase. All white lines inside the colored circles were canonical, randomly chosen between vertical and horizontal orientations. In the test phase, participants were asked to ignore the color and judge the orientation of the line inside a singleton shape. All the lines inside the distractors were randomly chosen to tilt to the left or to the right as in Experiment 2. Figure 5 shows the trial sequence.
The trial sequences in the training phase (A), and in the test phase (B) of Experiment 3. Participants were asked to search for a red or green circle and to report the orientation of the line that was inside the circle in the training phase. In the test phase, participants were told to ignore the colors and to report the line orientation that was inside a unique shape (a diamond among circles or a circle among diamonds). All the lines inside the distractors were titled to the left or right. (Color figure online)
Results and discussion
The data from two participants were excluded from the analysis because their performance in the test phase was 3 standard deviations below the mean accuracy. Thus, the data of 24 participants were included in the analyses (42% female, 22.67 ± 2.62 years of age).
The mean accuracy observed in the test phase was 91% (see Table 1). The results from a 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) repeated-measures ANOVA showed that only the main effect of block was significant, F(1, 23) = 24.04, p < .001, \({\upeta }_{p}^{2}\) = .51, BF10 = 154.56, with a higher accuracy in Block 2 (94%) than in Block 1 (90%). Neither the main effect of trial type, F(1, 23) = 0.04, p = .850, \({\upeta }_{p}^{2}\) < .01, BF10 = 0.29, nor its interaction with block, F(1, 23) = 3.67, p = .068, \({\upeta }_{p}^{2}\) = .14, BFexcl = 0.63, reached significance.
Reaction times exceeding 3 standard deviations in each condition of each participant were removed, thus excluding 0.76% of the total trials. The RT data are plotted in Fig. 6. A 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) repeated-measures ANOVA revealed a significant main effect of trial type, F(1, 23) = 9.62, p = .005, \({\upeta }_{p}^{2}\) = .30, BF10 = 4.46; performance was faster in the neutral condition (655 ms) than in the invalid condition (668 ms). The main effect of block was not significant, F(1, 23) = 0.42, p = .525, \({\upeta }_{p}^{2}\) = .02, BF10 = 0.31. Importantly, the interaction between block and trial type was significant, F(1, 23) = 8.91, p = .007, \({\upeta }_{p}^{2}\) = .28, BFincl = 16.70. In Block 1, performance was faster in the neutral condition (651 ms) than in the invalid condition (675 ms), F(1, 23) = 13.16, p = .001, \({\upeta }_{p}^{2}\) = .36, BF10 = 20.61, whereas the simple effect was not significant in Block 2, F(1, 23) = 0.17, p = .685, \({\upeta }_{p}^{2}\) < .01, BF10 = 0.30.
The results of the test phase in Experiment 3: Bias was observed in the first half of the test phase with faster reaction time in the neutral condition than in the invalid condition. Error bars show the standard error (SE) of the mean
The results replicated the findings of Experiment 1, with experience-based attentional biases observed only in the first half of the test phase. Across three experiments, it is shown that color information must guide target localization for response selection with a high-informational value during the training phase to induce durable experience-based attentional biases in the test phase. Moreover, the biases lasted only for half of the test phase. In the test phase of both Experiments 1 and 3, the behavioral goal was to search for a singleton shape and to assess the gap direction or the line orientation inside of the shape. However, the target search was much easier in Experiment 3 than in Experiment 1 because of the high perceptual discriminability of the response-relevant feature dimension among the stimuli. The question then arises as to what could be the common factor that leads to the same result pattern. We postulated that one factor could be the consistency in distractor processing during the first half of the test phase. In Experiment 1, the gap direction of the notched square inside of a prior target color was always vertical with two possible alternatives. In Experiment 3, the line inside of a prior target color was always oriented in a tilted direction with two possible alternatives. It is plausible that the consistency in distractor processing in the first half of the test phase renders the formation of a distractor template for rejection. With the template formed, the interference caused by the distractor with a prior target color is minimized in the second half of the test phase.
Experiment 4
Prior research has shown that distractor templates play an important role in attentional control (Arita et al., 2012; Beck & Hollingworth, 2015; Beck et al., 2018; Carlisle, 2019, 2023; Chao, 2024; Cunningham & Egeth, 2016; Geng et al., 2019; Leber et al., 2016; Liesefeld & Müller, 2019; Moher & Egeth, 2012; Reeder et al., 2017; Sawaki & Luck, 2011; Vatterott & Vecera, 2012; Won & Geng, 2018). The consistency in distractor processing can influence attentional control. Visual search is more efficient when its spatial location (B. Wang et al., 2019), color (De Tommaso & Turatto, 2019; Failing et al., 2019; Geyer et al., 2008; Stilwell et al., 2019; Vatterott & Vecera, 2012), or stimulus onset (Sayim et al., 2010) is repeatedly experienced across trials. Distractor suppression is proactively engaged prior to stimulus onset (C. Huang et al., 2021), which results from the accumulation of repeated experiences (Bogaerts et al., 2022). The aim of this experiment was to investigate whether the consistency of distractor processing could modulate the duration of experience-based attentional biases. We modified the search context that was used in the test phase of Experiment 3 by increasing the number of possible line orientations inside of the distractor with a prior target color from two to four.
Method
Participants
Thirty volunteers participated in Experiment 4 in exchange for monetary rewards (NT$ 150). The sample size of Experiment 4 was estimated based on the interaction effect observed in Experiment 1 to provide a power greater than 0.80 (alpha = 0.05). All participants had normal or corrected-to-normal vision and were naïve regarding the purpose of the experiment.
Design, stimuli, and procedure
All aspects are like those of Experiment 3, except for changing the orientation inside the distractors during the test phase. A test display consisted of two canonical lines and four tilted lines. The tilted lines were randomly selected toward the left or right. In the neutral condition (50% of the trials), red or green colors were not present (Fig. 7A). In the invalid condition (50% of the trials), red and green distractors occurred with equal frequency. Inside the red or green shapes, half of the trials contained a canonical line with equal frequency represented between the vertical and horizontal directions (Fig. 7B), and the other half of the trials contained a tilted line with equal frequency between the left and right directions (Fig. 7C).
An illustration of the search display of the test phase in Experiment 4, with (A) neither a red shape nor a green shape is present in a neutral condition, (B) a green or red shape encloses a canonical line in an invalid condition, and (C) a green or red shape encloses a tilted line in an invalid condition. (Color figure online)
Results and discussion
Data from one participant were excluded from the analysis because performance in the test phase was 3 standard deviations below the mean accuracy. Thus, the data of 29 participants were included in the analyses (52% female, 23.97 ± 2.73 years of age).
The mean accuracy observed in test phase was 82% (see Table 1). A 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) repeated-measures ANOVA showed null results of block, F(1, 28) = 0.30, p = .590, \({\upeta }_{p}^{2}\) = .01, BF10 = 0.31, trial type, F(1, 28) = 0.05, p = .820, \({\upeta }_{p}^{2}\) < .01, BF10 = 0.27, and Block × Trial Type, F(1, 28) = 0.63, p = .436, \({\upeta }_{p}^{2}\) = .02, BFexcl = 14.08. Trials with correct responses in which RTs exceeded 3 standard deviations in each condition of each participant were trimmed, thus excluding 0.44% of the total correct trials. The RT data are plotted in Fig. 8. A 2 (block: 1 vs. 2) × 2 (trial type: neutral vs. invalid) repeated-measures ANOVA revealed a significant main effect of block, F(1, 28) = 12.54, p = .001, \({\upeta }_{p}^{2}\) = .31, BF10 = 26.52. The overall RTs decreased as the experiment progressed. The main effect of trial type was significant, F(1, 28) = 17.01, p < .001, \({\upeta }_{p}^{2}\) = .38, BF10 = 71.04, with better performance in the neutral condition (716 ms) than in the invalid condition (735 ms). The interaction between block and trial type was not significant, F(1, 28) = 0.03, p = .865, \({\upeta }_{p}^{2}\) = .001, BFexcl = 0.90.
The results of the test phase in Experiment 4: Reaction time in the invalid condition was longer than that in the neutral condition in both blocks. Error bars show the standard error (SE) of the mean
The results demonstrated experience-based attentional biases lasting throughout the entire test phase. In contrast with the results of Experiment 3, the findings support the idea that the consistency of distractor rejection could modulate the duration of experience-based attentional biases. When the line inside of a prior target color varied between two alternatives in Experiment 3, the bias disappeared after the first half of the test phase. When the line varied among the four alternatives, bias was still evident in the second half of the test. The effect of consistency could have arisen because a learning mechanism is more efficient with low variability than with high variance for abstracting the regularity of the visual environment (Kristjánsson, 2023).
General discussion
An understanding of the formation and maintenance of target and distractor templates is important, as templates guide attentional control (Kristjánsson, 2023). By using a training-test-phase paradigm, the current study first investigated whether the priority of task-irrelevant colors in the target template can be modified to induce durable experience-based attentional biases. We assumed that experience-based attentional biases toward a distractor with a prior target color only emerge in the test phase when the color feature has a high-priority weight after training. The results showed that the priority of task-irrelevant colors in a target template was amplified after repeated association with a target in the training phase only when the color feature provided a high-informational value for target localization and selection (Experiment 1). When the search can be accomplished efficiently via priority guidance on the response-relevant feature dimension in the training phase (Experiment 2), a task-irrelevant color feature that is useful for target localization did not lead to experience-based attention biases. Moreover, prior knowledge of the color-target contingency did not entice the learning mechanism to modify the target template when the search was easy.
Experiment 3 ruled out the possibility that the difference in the results arises from the difference in the search context during the test phase. Colors defined the target during the training phase; hence, it had a high-informational value for target selection. In the test phase, the target search was easy, as was observed in Experiment 2, with high-perceptual discriminability of response-relevant feature dimension observed among the displayed stimuli. The results replicate those of Experiment 1, with experience-based attentional biases observed in the first half of the test phase. In Experiment 4, we investigated whether the consistency in distractor processing could be a reason for observing relatively short-lived attentional biases in Experiments 1 and 3. The search context in the training phase was identical to that used in Experiment 3, whereas the distractors inside of a prior target color were more variable in the test phase. The results showed experience-based attentional biases throughout the test phase.
The rise of durable experience-based attentional bias
Our results are in accordance with prior research that showed that both the explicit task rule (Miranda & Palmer, 2014; Sha & Jiang, 2016; L. Wang et al., 2013) and implicit learning of regularity (Kyllingsbæk et al., 2001, 2014; Lin et al., 2016; Qu et al., 2017) can lead to durable experience-based attentional biases. When colors explicitly defined the target, a robust experience-based attentional bias was observed in the first half of the test phase (Experiment 3) or throughout the test phase (Experiment 4). In the search context of these two experiments, color information is necessary for guiding attention to the target among highly similar distractors. Thus, participants adopted a feature-search mode (Bacon & Egeth, 1994) or clump-scanning strategy (Liesefeld & Müller, 2020) based on colors to localize the target. Color information provides a high value for efficient target localization.
When colors did not define the target (Experiments 1 and 2), color features had little priority in the target template at the beginning of the training phase, as colors were not in the top-down task rule. However, color-target contingency can lead to experience-based attentional biases. More importantly, our findings highlight that the usefulness for target localization alone is not sufficient for inducing experience-based attentional biases. A task-irrelevant feature must provide a high-informational value for the learning mechanism to increase its priority in the target template. In Experiment 2, colors were highly predictive of the target location. However, null biases were observed in the test phase because the task in the training phase can be accomplished via priority guidance based on the orientation without using the color information. Colors are useful for target localization; however, the contingency has a low-informational value for efficient goal accomplishment. Thus, even prior knowledge of the color-target contingency still cannot influence the learning mechanism to increase the priority weight of the colors in the target template.
The effect of informational value is in line with the findings of a previous study. In Lin et al.’s (2016) study, the participants searched for a T among Ls inside of colored circles and reported its orientation while the target was always inside of a specific color. As fine discrimination is needed during clump scanning, the associated color provides information with a high value for localizing the target during the training phase. As the color repeatedly guides attention toward the target, the priority value of the color inside of the target template is increased over hundreds of trials in training. The target-associated color induces a durable attentional bias. Similarly, decreasing search difficulty diminishes the processing of predictive information (Conn et al., 2020) because the value of the information is reduced to achieve the task goal.
It has been noted that the stimulus-task context influences the informational value of a target-associated feature, as selection history is sensitive to contextual factors (Anderson et al., 2021). Moreover, different mechanisms may be engaged for different components of experience-driven attentional biases. Three different mechanisms, including stimulus–response learning and stimulus–outcome learning, may subserve history-based attentional control (Anderson et al., 2021). In Experiment 2, we replicated the findings of previous studies (Anderson & Halpern, 2017; Anderson et al., 2011) and showed that a history of sought targets cannot render enduring attentional biases when the utility of color contingency is low. In this stimulus-task context, rewards can boost the utility of color information so that stimulus–outcome associative learning increases the priority of the high-reward color to render reward-based attentional biases.
In the studies that manipulated the probability of one color relative to another color in their association with a singleton target during the training phase (Kruijne & Meeter, 2015; Sha et al., 2017a), durable biases were not observed once the frequency became comparable in the test phase. In this context, where priority guidance can accomplish the task goal, the utility of the feature frequency can be rapidly reset. In the context when a color can coincide with a distractor in a proportion of the training trials, the bias toward the high-frequency target color endures even when two colors were equally linked to the target and a distractor in the test phase (Conn et al., 2020; Kruijne & Meeter, 2015, 2016; Sha et al., 2017b). The diagnostic value (Sha et al., 2017b), which involves the probability of associating a color with the target relative to the probability of linking the color with a distractor, is important in this stimulus-task context. The importance may result from separating target and distractor representations maintained in working memory so that the influence of diagnostic value is sustained in the test phase. Further investigations are needed to demonstrate how various attentional control strategies and learning and memory mechanisms interact to develop different components of experience-driven attentional biases in a variety of stimulus-task contexts.
The fall of durable experience-based attentional bias
The influence of selection history can gradually disappear over test trials. Multiple factors may underlie the decay of habitual biases. We considered two factors that may conjointly influence the duration of experience-based biases in the present stimulus-task context. The first factor is the strength of the target template. It has been shown that the intertrial priming effect lasts longer for rare trials with contextual distinctiveness in color, spatial location, or stimulus configuration (Thomson & Milliken, 2012a, 2013). The reason could be due to the fact that distinctiveness enhances memory (Hunt & McDaniel, 1993). In a training-test-phase paradigm, the number of training trials influences the strength of the target template. According to the instance theory (Logan, 1988), a memory trace of processing is generated on each trial. The duration of the training phase influences how many episodic traces are formed to impact the memory strength of the past experiences and correspondingly affect the strength of the target template. Leber and Egeth (2006) demonstrated that a training phase of 40 trials did not result in the automatic use of a previous attentional set in the test phase, whereas 320 training trials induced the continued use of the same set. In the present study, the training phase comprised 240 trials; therefore, the priority of the color feature may not be overly robust and can be downweighted through consistent distractor rejection.
The second factor involves whether efficient control over distractors can be developed with the processing of instances accumulated in the first half of the test phase to form a distractor template for rejection. The experience-based attentional bias lasted for 120 trials in Experiments 1 and 3, whereas the bias lasted for 240 trials in Experiment 4. The number of possible alternatives inside of the distractor with a prior target color was two in the former two experiments and four in the latter experiment. As it is more difficult to abstract the regularity in a more variable context (Kristjánsson, 2023), increasing the number of possible response-relevant orientations inside of the distractor with a prior target color can lead to difficulty in forming a distractor template for rejection. Future research is necessary to explore whether the difficulty arises because participants pay more attention to contextual information with a large variability (Chao, 2009; Qu et al., 2017). It is likely that the priority weight of stimulus features is amplified every time attention is directed to a stimulus. The amplification induces a short-term habitual bias, with enhanced cross-trial priming with four distractor alternatives compared with two alternatives (Geyer et al., 2006). However, the short-term effect may undermine the influence of downweighting from successful rejection that can dissociate the colors from the response-relevant features. Additionally, whether distractor suppression is executed over all of the distractors or specifically to the distractor in a prior target color remains to be clarified. Finally, the time course for forming a distractor template as a function of the memory strength of the target template is of interest.
Search context underlying the experience-based attentional biases
The results of this study also highlight how the search context can influence the rise and fall of durable experience-based attentional biases. Color information has a high value in the training phase only when the search requires clump scanning. In this search context, the perceptual discriminability of the response-relevant feature dimension is low; thus, searching cannot be easily accomplished without fine discrimination via clump scanning (Liesefeld & Müller, 2020) in a feature-search mode (Bacon & Egeth, 1994). When searching in the training phase can be accomplished via priority guidance (Liesefeld & Müller, 2020) in a singleton-search mode (Bacon & Egeth, 1994), color information loses value to induce durable experience-based attentional biases. The prioritization of a feature associated with the target is increased solely when it consistently and effectively facilitates target localization for response selection during clump scanning.
In the test phase, the stimulus display shares high similarity with that used in the training phase. The participants may have continued the adoption of the same search strategy in the first half, even though the test target can be easily identified by being a singleton shape in the display. The adoption of the same attentional set (Leber & Egeth, 2006) highlights the difficulty of updating the priority weight for the unique shape when the response-relevant features remain the same. The priority weight of prior target colors was reduced only after repeated experiences of rejecting the distractor. In this search context, there was no other strong competitor in the processing of the response-relevant features. The test target gradually won the race by being the only singleton shape and the only horizontal notched square (Experiment 1) or the only canonical line (Experiment 3). Participants could shift toward the adoption of priority guidance in the second half, and the experience-based bias disappeared. In Experiment 4, the distractor with a prior target color could compete in the processing of line orientation in half of the test trials; therefore, priority guidance via orientation would not be the best strategy. Thus, experience-based attentional biases lasted longer. Future research is needed to clarify the involvement of different search strategies in different search contexts to modulate the duration of experience-based attentional biases. Further investigation is also needed to clarify whether it is the winning of the test target, the suppression of the distractor with a prior target color, or both that causes the decay of experience-based attentional biases.
It is noted that the adoption of the same search strategy during the test phase may have resulted from saliency rather than the alternation of color priority within the target template, as red and green were designated as target-associated colors for all participants. However, it is possible that the results did not solely stem from salience. Firstly, these two colors do not possess the highest luminance among those displayed. Secondly, they did not induce experience-based biases when possessing low-informational value. Nevertheless, future research should randomly assign target colors to each participant to confirm the role of informational value in the observation of experience-based attentional biases.
The role of working memory
Working memory plays an important role in the formation and maintenance of targets (Carlisle et al., 2011; Hollingworth & Beck, 2016; Kristjánsson & Kristjánsson, 2018; Wolfe, 2021) and distractor templates (Kristjánsson, 2023; Wolfe, 2021; Won & Geng, 2018), especially when the templates are task defined. Once the task rule is instructed, this information is maintained in working memory and activated in each trial of processing. The probabilistic representation that is abstracted via the processing of the regularity in the visual environment is also maintained in working memory (Kristjánsson, 2023), as the existing distribution must be updated by the new information gathered from each trial of processing. The demand on working memory may be quite high in the present experimental context, with a response deadline of 800 ms in the training phase and 1,200 ms in the test phase. Within this short duration, the participants needed to activate the existing target template during the training phase. As a result, the mean accuracy did not approach 90% in the first few trial bins of the training phase in all four experiments. In the test phase, participants also need to adjust the weight according to the new task rule and update the distractor information in a short period of time. Further research should investigate how the neural mechanisms of working memory interact with sensory areas to form and maintain target and distractor templates.
The underlying mechanisms of weight adjustment
The inference of upweighting target-associated colors within the target template is drawn from the observed delayed responses in the invalid condition when a prior target color is present, as opposed to the neutral condition when neither target color is present. Similarly, the deduction of a distractor template is made based on the duration of the distraction effect in the invalid condition. Future research is essential to furnish direct evidence elucidating the neural dynamics underlying various manifestations of attentional biases. Previous studies have demonstrated that top-down control, achieved through explicit cueing of target and distractor colors, engenders active alterations in neural dynamics before and after the search display (Chidharom & Carlisle, 2023). Conversely, inter-trial learned attentional bias towards a high-probability target location is governed by activity-silent mechanisms (Duncan et al., 2023). Whether such activity-silent mechanisms facilitate the adjustment of priority weights in other feature dimensions such as color and orientation remains to be validated. The outcomes of this line of inquiry may ultimately clarify the processes involved in the formation, modification, and maintenance of a template.
Conclusions
The results of the current study highlight the roles of informational value and consistent distractor rejection in the rise and fall of experience-based attentional biases. Information usefulness for target localization alone is insufficient for a target feature to be prioritized in the target template. When target search could be easily accomplished by being an orientation singleton in the race toward the behavioral goal, target-contingent colors did not induce experience-based attentional bias. Only when color information can efficiently localize and select a target could the priority weight of the colors be upweighted to bias attention in the test phase. Once the biases were developed, the duration of their impact could vary. In addition to the strength of past experiences, the variability of processing response-relevant features inside of the prior target colors could modulate the duration of experience-based attentional biases. With two possible response-relevant features inside of the prior target colors, attentional bias was observed only in the first half of the test. When the number of possible alternatives increased to four, bias was observed throughout the entire test phase. The results highlight how attentional control is influenced by the dynamic interplay among task goals, the memory strength of past experiences, stimulus saliency and the recent experiences of controlling over the distractors.
Data availability
The data and materials for this study are available upon request to hfchao@mx.nthu.edu.tw.
Code availability
Not applicable.
References
Anderson, B. A., & Halpern, M. (2017). On the value-dependence of value-driven attentional capture. Attention, Perception, & Psychophysics, 79, 1001–1011. https://doi.org/10.3758/s13414-017-1289-6
Anderson, B. A., Kim, H., Kim, A. J., Liao, M. R., Mrkonja, L., Clement, A., & Grégoire, L. (2021). The past, present, and future of selection history. Neuroscience & Biobehavioral Reviews, 130, 326–350. https://doi.org/10.1016/j.neubiorev.2021.09.004
Anderson, B. A., Laurent, P. A., & Yantis, S. (2011). Value-driven attentional capture. Proceedings of the National Academy of Sciences, 108, 10367–10371. https://doi.org/10.1073/pnas.1104047108
Arita, J. T., Carlisle, N. B., & Woodman, G. F. (2012). Templates for rejection: Configuring attention to ignore task-irrelevant features. Journal of Experimental Psychology: Human Perception and Performance, 38, 580–584. https://doi.org/10.1037/a0027885
Awh, E., Belopolsky, A. V., & Theeuwes, J. (2012). Top-down versus bottom-up attentional control: A failed theoretical dichotomy. Trends in Cognitive Sciences, 16, 437–443. https://doi.org/10.1016/j.tics.2012.06.010
Bacon, W. F., & Egeth, H. E. (1994). Overriding stimulus-driven attentional capture. Perception & Psychophysics, 55, 485–496. https://doi.org/10.3758/BF03205306
Beck, V. M., & Hollingworth, A. (2015). Evidence for negative feature guidance in visual search is explained by spatial recoding. Journal of Experimental Psychology: Human Perception and Performance, 41, 1190–1196. https://doi.org/10.1037/xhp0000109
Beck, V. M., Luck, S. J., & Hollingworth, A. (2018). Whatever you do, don’t look at the …: Evaluating guidance by an exclusionary attentional template. Journal of Experimental Psychology: Human Perception and Performance, 44, 645–662. https://doi.org/10.1037/xhp0000485
Berggren, N., & Eimer, M. (2018). Object-based target templates guide attention during visual search. Journal of Experimental Psychology: Human Perception and Performance, 44, 1368–1382. https://doi.org/10.1037/xhp0000541
Bogaerts, L., Siegelman, N., Christiansen, M. H., & Frost, R. (2022). Is there such a thing as a ‘good statistical learner’? Trends in Cognitive Sciences, 26, 25–37. https://doi.org/10.1016/j.tics.2021.10.012
Bundesen, C. (1990). A theory of visual attention. Psychological Review, 97, 523–547. https://doi.org/10.1037/0033-295X.97.4.523
Bundesen, C., Habekost, T., & Kyllingsbæk, S. (2005). A neural theory of visual attention: Bridging cognition and neurophysiology. Psychological Review, 112, 291–328. https://doi.org/10.1037/0033-295X.112.2.291
Carlisle, N. B. (2019). Flexibility in attentional control: Multiple sources and suppression. The Yale Journal of Biology and Medicine, 92, 103–113. https://doi.org/10.1167/19.10.53d
Carlisle, N. B. (2023). Negative and positive templates: Two forms of cued attentional control. Attention, Perception, & Psychophysics, 85, 585–595. https://doi.org/10.3758/s13414-022-02590-4
Carlisle, N. B., Arita, J. T., Pardo, D., & Woodman, G. F. (2011). Attentional templates in visual working memory. Journal of Neuroscience, 31, 9315–9322. https://doi.org/10.1523/JNEUROSCI.1097-11.2011
Chao, H.-F. (2009). Revisiting the prime–probe contextual similarity effect on negative priming: The impact of cue variability. European Journal of Cognitive Psychology, 21, 484–500. https://doi.org/10.1080/09541440802049051
Chao, H.-F. (2024). Attentional control and priming: The distractor set hypothesis. Journal of Experimental Psychology: Learning, Memory, and Cognition, 50(3), 347–361. https://doi.org/10.1037/xlm0001252
Chelazzi, L., Duncan, J., Miller, E. K., & Desimone, R. (1998). Responses of neurons in inferior temporal cortex during memory-guided visual search. Journal of Neurophysiology, 80, 2918–2940. https://doi.org/10.1152/jn.1998.80.6.2918
Chelazzi, L., Miller, E. K., Duncan, J., & Desimone, R. (1993). A neural basis for visual search in inferior temporal cortex. Nature, 363, 345–347. https://doi.org/10.1038/363345a0
Chetverikov, A., Campana, G., & Kristjánsson, Á. (2016). Building ensemble representations: How the shape of preceding distractor distributions affects visual search. Cognition, 153, 196–210. https://doi.org/10.1016/j.cognition.2016.04.018
Chetverikov, A., Campana, G., & Kristjánsson, Á. (2017). Representing color ensembles. Psychological Science, 28, 1510–1517. https://doi.org/10.1177/0956797617713787
Chun, M. M., & Jiang, Y. (1998). Contextual cueing: Implicit learning and memory of visual context guides spatial attention. Cognitive Psychology, 36, 28–71. https://doi.org/10.1006/cogp.1998.0681
Conci, M., & Müller, H. J. (2012). Contextual learning of multiple target locations in visual search. Visual Cognition, 20, 746–770. https://doi.org/10.1080/13506285.2012.694376
Conn, K. M., Becker, M. W., & Ravizza, S. M. (2020). Persistent guidance of attention in visual statistical learning. Journal of Experimental Psychology: Human Perception and Performance, 46, 681–696. https://doi.org/10.1037/xhp0000741
Cunningham, C. A., & Egeth, H. E. (2016). Taming the white bear: Initial costs and eventual benefits of distractor inhibition. Psychological Science, 27, 476–485. https://doi.org/10.1177/0956797615626564
De Tommaso, M., & Turatto, M. (2019). Learning to ignore salient distractors: Attentional set and habituation. Visual Cognition, 27, 214–226. https://doi.org/10.1080/13506285.2019.1583298
Desimone, R., & Duncan, J. (1995). Neural mechanisms of selective visual attention. Annual Review of Neuroscience, 18, 193–222. https://doi.org/10.1146/annurev.ne.18.030195.001205
Duncan, J., & Humphreys, G. W. (1989). Visual search and stimulus similarity. Psychological Review, 96, 433–458. https://doi.org/10.1037/0033-295X.96.3.433
Duncan, D. H., van Moorselaar, D., & Theeuwes, J. (2023). Pinging the brain to reveal the hidden attentional priority map using encephalography. Nature Communications, 14(1), 4749. https://doi.org/10.1038/s41467-023-40405-8
Failing, M., & Theeuwes, J. (2018). Selection history: How reward modulates selectivity of visual attention. Psychonomic Bulletin & Review, 25, 514–538. https://doi.org/10.3758/s13423-017-1380-y
Failing, M., Wang, B., & Theeuwes, J. (2019). Spatial suppression due to statistical regularities is driven by distractor suppression not by target activation. Attention, Perception, & Psychophysics, 81, 1405–1414. https://doi.org/10.3758/s13414-019-01704-9
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149–1160. https://doi.org/10.3758/BRM.41.4.1149
Geng, J., Won, B. Y., & Carlisle, N. (2019). Distractor ignoring: Strategies, learning, and passive filtering. Current Directions in Psychological Science, 28, 600–606. https://doi.org/10.1177/0963721419867099
Geyer, T., Müller, H. J., & Krummenacher, J. (2006). Cross-trial priming in visual search for singleton conjunction targets: Role of repeated target and distractor features. Perception & Psychophysics, 68, 736–749. https://doi.org/10.3758/BF03193697
Geyer, T., Müller, H. J., & Krummenacher, J. (2008). Expectancies modulate attentional capture by salient color singletons. Vision Research, 48, 1315–1326. https://doi.org/10.1016/j.visres.2008.02.006
Hollingworth, A., & Beck, V. M. (2016). Memory-based attention capture when multiple items are maintained in visual working memory. Journal of Experimental Psychology: Human Perception and Performance, 42, 911–917. https://doi.org/10.1037/xhp0000230
Huang, C., Donk, M., & Theeuwes, J. (2022). Proactive enhancement and suppression elicited by statistical regularities in visual search. Journal of Experimental Psychology: Human Perception and Performance, 48, 443–457. https://doi.org/10.1037/xhp0001002
Huang, C., Vilotijević, A., Theeuwes, J., & Donk, M. (2021). Proactive distractor suppression elicited by statistical regularities in visual search. Psychonomic Bulletin & Review, 28, 918–927. https://doi.org/10.3758/s13423-021-01891-3
Huang, Z., & Li, S. (2022). Reactivation of learned reward association reduces retroactive interference from new reward learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 48, 213–225. https://doi.org/10.1037/xlm0000987
Hunt, R. R., & McDaniel, M. A. (1993). The enigma of organization and distinctiveness. Journal of Memory and Language, 32, 421–445. https://doi.org/10.1006/jmla.1993.1023
JASP Team. (2022). JASP (Version 0.17.2.1) [Computer software].
Kaptein, N. A., Theeuwes, J., & van der Heijden, A. H. C. (1995). Search for a conjunctively defined target can be selectively limited to a color-defined subset of elements. Journal of Experimental Psychology: Human Perception and Performance, 21, 1053–1069. https://doi.org/10.1037/0096-1523.21.5.1053
Kim, H., & Anderson, B. A. (2019). Dissociable neural mechanisms underlie value-driven and selection-driven attentional capture. Brain Research, 1708, 109–115. https://doi.org/10.1016/j.brainres.2018.11.026
Knight, H. C., Smith, D. T., Knight, D. C., & Ellison, A. (2016). Altering attentional control settings causes persistent biases of visual attention. Quarterly Journal of Experimental Psychology, 69, 129–149. https://doi.org/10.1080/17470218.2015.1031144
Kristjánsson, Á. (2023). Priming of probabilistic attentional templates. Psychonomic Bulletin & Review, 30, 22–39. https://doi.org/10.3758/s13423-022-02125-w
Kristjánsson, Á., Wang, D., & Nakayama, K. (2002). The role of priming in conjunctive visual search. Cognition, 85, 37–52. https://doi.org/10.1016/S0010-0277(02)00074-4
Kristjánsson, T., & Kristjánsson, Á. (2018). Foraging through multiple target categories reveals the flexibility of visual working memory. Acta Psychologica, 183, 108–115. https://doi.org/10.1016/j.actpsy.2017.12.005
Kruijne, W., & Meeter, M. (2015). The long and the short of priming in visual search. Attention, Perception, & Psychophysics, 77, 1558–1573. https://doi.org/10.3758/s13414-015-0860-2
Kruijne, W., & Meeter, M. (2016). Long-term priming of visual search prevails against the passage of time and counteracting instructions. Journal of Experimental Psychology: Learning, Memory, and Cognition, 42, 1293–1303. https://doi.org/10.1037/xlm0000233
Kruijne, W., Brascamp, J. W., Kristjánsson, Á., & Meeter, M. (2015). Can a single short-term mechanism account for priming of pop-out? Vision Research, 115, 17–22. https://doi.org/10.1016/j.visres.2015.03.011
Kyllingsbæk, S., Schneider, W. X., & Bundesen, C. (2001). Automatic attraction of attention to former targets in visual displays of letters. Perception & Psychophysics, 63, 85–98. https://doi.org/10.3758/BF03200505
Kyllingsbæk, S., Van Lommel, S., Sørensen, T. A., & Bundesen, C. (2014). Automatic attraction of visual attention by supraletter features of former target strings. Frontiers in Psychology, 5, Article 1383. https://doi.org/10.3389/fpsyg.2014.01383
Lakens, D., & Caldwell, A. R. (2021). Simulation-based power analysis for factorial analysis of variance designs. Advances in Methods and Practices in Psychological Science, 4(1). https://doi.org/10.1177/2515245920951503
Leber, A. B., & Egeth, H. E. (2006). Attention on autopilot: Past experience and attentional set. Visual Cognition, 14, 565–583. https://doi.org/10.1080/13506280500193438
Leber, A. B., Gwinn, R. E., Hong, Y., & O’Toole, R. J. (2016). Implicitly learned suppression of irrelevant spatial locations. Psychonomic Bulletin & Review, 23, 1873–1881. https://doi.org/10.3758/s13423-016-1065-y
Leber, A. B., Kawahara, J. I., & Gabari, Y. (2009). Long-term abstract learning of attentional set. Journal of Experimental Psychology: Human Perception and Performance, 35, 1385–1397. https://doi.org/10.1037/a0016470
Li, Z. (1999). Contextual influences in V1 as a basis for pop out and asymmetry in visual search. Proceedings of the National Academy of Sciences of the United States of America, 96, 10530–10535. https://doi.org/10.1073/pnas.96.18.10530
Li, Z. (2002). A saliency map in primary visual cortex. Trends in Cognitive Sciences, 6, 9–16. https://doi.org/10.1016/S1364-6613(00)01817-9
Liesefeld, H. R., & Müller, H. J. (2019). Distractor handling via dimension weighting. Current Opinion in Psychology, 29, 160–167. https://doi.org/10.1016/j.copsyc.2019.03.003
Liesefeld, H. R., & Müller, H. J. (2020). A theoretical attempt to revive the serial/parallel-search dichotomy. Attention, Perception, & Psychophysics, 82, 228–245. https://doi.org/10.3758/s13414-019-01819-z
Lin, Z., Lu, Z. L., & He, S. (2016). Decomposing experience-driven attention: Opposite attentional effects of previously predictive cues. Attention, Perception, & Psychophysics, 78, 2185–2198. https://doi.org/10.3758/s13414-016-1101-z
Logan, G. D. (1988). Toward an instance theory of automatization. Psychological Review, 95, 492–527. https://doi.org/10.1037/0033-295X.95.4.492
Luck, S. J., Gaspelin, N., Folk, C. L., Remington, R. W., & Theeuwes, J. (2021). Progress toward resolving the attentional capture debate. Visual Cognition, 29, 1–21. https://doi.org/10.1080/13506285.2021.1978672
Maljkovic, V., & Nakayama, K. (1994). Priming of pop-out: I. role of features. Memory & Cognition, 22, 657–672. https://doi.org/10.3758/BF03209251
Miranda, A. T., & Palmer, E. M. (2014). Intrinsic motivation and attentional capture from gamelike features in a visual search task. Behavior Research Methods, 46, 159–172. https://doi.org/10.3758/s13428-013-0357-7
Moher, J., & Egeth, H. E. (2012). The ignoring paradox: Cueing distractor features leads first to selection, then to inhibition of to-be-ignored items. Attention, Perception, & Psychophysics, 74, 1590–1605. https://doi.org/10.3758/s13414-012-0358-0
Munneke, J. E., Corbett, J. E., & van der Burg, E. (2020). Learned prioritization yields attentional biases through selection history. Attention, Perception, & Psychophysics, 82, 2244–2256. https://doi.org/10.3758/s13414-020-01970-y
Nakayama, K., Maljkovic, V., & Kristjánsson, A. (2004). Short-term memory for the rapid deployment of visual attention. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 397–408). MIT Press.
Qu, Z., Hillyard, S. A., & Ding, Y. (2017). Perceptual learning induces persistent attentional capture by nonsalient shapes. Cerebral Cortex, 27, 1512–1523. https://doi.org/10.1093/cercor/bhv342
Reeder, R. R., Olivers, C. N., & Pollmann, S. (2017). Cortical evidence for negative search templates. Visual Cognition, 25, 278–290. https://doi.org/10.1080/13506285.2017.1339755
Sawaki, R., & Luck, S. J. (2011). Active suppression of distractors that match the contents of visual working memory. Visual Cognition, 19, 956–972. https://doi.org/10.1080/13506285.2011.603709
Sayim, B., Grubert, A., Herzog, M. H., & Krummenacher, J. (2010). Display probability modulates attentional capture by onset distractors. Journal of Vision, 10(3), 1–8. https://doi.org/10.1167/10.3.10
Sha, L. Z., & Jiang, Y. V. (2016). Components of reward-driven attentional capture. Attention, Perception, & Psychophysics, 78, 403–414. https://doi.org/10.3758/s13414-015-1038-7
Sha, L. Z., Remington, R. W., & Jiang, Y. V. (2017a). Short-term and long-term attentional biases to frequently encountered target features. Attention, Perception, & Psychophysics, 79, 1311–1322. https://doi.org/10.3758/s13414-017-1317-6
Sha, L. Z., Remington, R. W., & Jiang, Y. V. (2017b). Rapid and selective updating of the target template in visual search. Journal of Vision, 17(1), 1–18. 0.1167/17.1.36
Stilwell, B. T., & Vecera, S. P. (2023). Learned distractor rejection persists across target search in a different dimension. Attention, Perception, & Psychophysics, 85, 785–795. https://doi.org/10.3758/s13414-022-02559-3
Stilwell, B. T., Bahle, B., & Vecera, S. P. (2019). Feature-based statistical regularities of distractors modulate attentional capture. Journal of Experimental Psychology: Human Perception and Performance, 45, 419–433. https://doi.org/10.1037/xhp0000613
Thayer, D. D., Miller, M., Giesbrecht, B., & Sprague, T. C. (2023). Learned feature regularities enable suppression of spatially overlapping stimuli. Attention, Perception, & Psychophysics, 85, 769–784. https://doi.org/10.3758/s13414-022-02612-1
Thomson, D. R., & Milliken, B. (2012a). Perceptual distinctiveness produces long-lasting priming of pop-out. Psychonomic Bulletin & Review, 19, 170–176. https://doi.org/10.3758/s13423-011-0199-1
Thomson, D. R., & Milliken, B. (2012b). Revisiting the time course of inter-trial feature priming in singleton search. Psychological Research Psychologische Forschung, 77, 637–650. https://doi.org/10.1007/s00426-012-0455-7
Thomson, D. R., & Milliken, B. (2013). Contextual distinctiveness produces long-lasting priming of pop-out. Journal of Experimental Psychology: Human Perception and Performance, 39, 202–215. https://doi.org/10.1037/a0028069
Vatterott, D. B., & Vecera, S. P. (2012). Experience-dependent attentional tuning of distractor rejection. Psychonomic Bulletin & Review, 19, 871–878. https://doi.org/10.3758/s13423-012-0280-4
Wang, B., & Theeuwes, J. (2018). Statistical regularities modulate attentional capture. Journal of Experimental Psychology: Human Perception and Performance, 44, 13–17. https://doi.org/10.1037/xhp0000472
Wang, B., Samara, I., & Theeuwes, J. (2019). Statistical regularities bias overt attention. Attention, Perception, & Psychophysics, 81, 1813–1821. https://doi.org/10.3758/s13414-019-01708-5
Wang, L., Yu, H., & Zhou, X. (2013). Interaction between value and perceptual salience in value-driven attentional capture. Journal of Vision, 13(3), 1–13. https://doi.org/10.1167/13.3.5
Wolfe, J. M. (2021). Guided Search 6.0: An updated model of visual search. Psychonomic Bulletin & Review, 28, 1060–1092. https://doi.org/10.3758/s13423-020-01859-9
Won, B.-Y., & Geng, J. J. (2018). Learned suppression for multiple distractors in visual search. Journal of Experimental Psychology: Human Perception and Performance, 44, 1128–1141. https://doi.org/10.1037/xhp0000521
Yeh, Y.-Y., Lee, S.-M., Chen, Y.-H., & Chen, Z. (2014). Selection history modulates the effects of dual mechanisms on flanker interference. Journal of Experimental Psychology: Human Perception and Performance, 40, 2038–2055. https://doi.org/10.1037/a0037661
Zelinsky, G. J., & Bisley, J. W. (2015). The what, where, and why of priority maps and their interactions with visual working memory. Annals of the New York Academy of Sciences, 1339, 154–164. https://doi.org/10.1111/nyas.12606
Zhang, B., Weidner, R., Allenmark, F., Bertleff, S., Fink, G. R., Shi, Z., & Müller, H. J. (2022). Statistical learning of frequent distractor locations in visual search involves regional signal suppression in early visual cortex. Cerebral Cortex, 32, 2729–2744. https://doi.org/10.1093/cercor/bhab377
Funding
This work was supported by the Ministry of Science and Technology of Taiwan under Grants MOST 109–2628-H-003–004-MY2 and MOST 111–2628-H-003–007-MY3 to Chun-Yu Kuo.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Ethics approval
This study was approved by the Research Ethics Committee, National Taiwan Normal University (202005HS022).
Consent to participate
Informed consent was obtained from all individual participants included in the study.
Consent for publication
Not applicable. The participants signed informed consent regarding publishing their anonymized data, and no potentially identifiable human data are presented in this study.
Conflicts of interest
The authors have no relevant financial or nonfinancial interests to disclose.
Open practices statements
The data and materials for this study are available upon request. The experiments reported in this study were not preregistered.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported by the Ministry of Science and Technology of Taiwan under Grants MOST 109–2628-H-003–004-MY2 and MOST 111–2628-H-003–007-MY3 to Chun-Yu Kuo.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kuo, CY., Yeh, YY. & Chao, HF. The rise and fall of durable color-induced attentional bias. Atten Percept Psychophys (2024). https://doi.org/10.3758/s13414-024-02946-y
Accepted:
Published:
DOI: https://doi.org/10.3758/s13414-024-02946-y