When I’m trying to focus on reading scientific articles, my eyes keep drifting to my phone beside me—even when it’s face-down, on silent. No notifications, no vibrations, nothing calling for my attention … yet, I still look. Once I stash it in a desk drawer and it is out of sight, I can successfully focus on what I’m reading. Why doesn’t my coffee mug, also sitting right beside me, capture my attention in the same way?

For decades, visual researchers have explored the factors determining how we prioritize our attention. Some of the most well-documented influences include physically salient stimuli in our environment, like a loud phone notification, or our internal goals, such as focusing on reading a paper. However, sometimes, objects that aren’t necessarily salient or related to our current goals capture our attention – like glancing at your phone when you’re trying to stay focused on reading.

Researchers think this non-salient, goal-irrelevant attention capture may relate to our past experiences. For instance, objects with a reward history capture attention more easily, even when they are no longer associated with the reward. For example, a standard email from a journal may capture your attention simply because past emails from them brought the rewarding news of an accepted publication.

Building on this idea, recent work published in Attention, Perception & Psychophysics led by a research team, including Alenka Doyle, Kamilla Volkova, Nicholas Crotty, Nicole Massa, and Michael Grubb (pictured below), investigated another factor, besides rewards,  that may influence attention. They aimed to understand whether objects with a history of providing instrumental (useful) information may also similarly capture visual attention in the future. For example, my phone often provides me with useful information – whether it’s googling information, checking emails for important work updates, or coordinating plans with friends and family. Does this history of usefulness make my phone more likely to capture my attention in the future, even when I’m trying to focus on something else?

The authors designed a study with a training and test phase to investigate whether an object that previously provided useful/instrumental information in one task, continues to capture captures visual attention, even when it is no longer useful or relevant to the new task goal.

In each trial in the training phase, participants were presented with two colored squares serving as pre-cues. The pre-cue squares provided information about the color of the upcoming “search target,” a colored circle among other differently colored circles.  Each circle contained a line, and the search target had a line that was orientated differently (horizontal or vertical) compared to the lines within other colored circles. Participants were tasked with finding the circle with the uniquely oriented line – the “search target” and indicating its orientation.

The training phase task included two different trial types. In one trial type, both pre-cue squares were the same color, and they matched the color of the search target. In the example in the figure below labeled “reliably pre-cued trials”, both pre-cues were red, signaling to participants that the upcoming search target would also be red. This allowed participants to reliably anticipate the search target’s color, allowing them to find it quickly to report the unique line orientation.

In the other trial type, the pre-cues were different colors, and only one of those colors matched the upcoming search target, making the cue information unreliable.  For example, in the figure below labeled “unreliably pre-cued trials”, participants could not predict whether the search target would be blue or yellow because the pre-cues were both blue and yellow. In this case, the search target itself (e.g., the blue circle) became the most useful source of information for guiding attention towards the circle with the unique line orientation since the pre-cues alone were insufficient. Through repeated trials, the search target color, rather than the pre-cues, became associated with being helpful in finding the uniquely orientated line.

Given that the colors remained consistent across trial types, this creative experimental design allowed the researchers to associate different color information histories with different stimuli.  When the pre-cues were the same color as the search target, the color information the search target provided became redundant. That is, it didn’t provide any new or useful information because participants already knew what color to look for based on the pre-cue. However, when the pre-cues were different colors, they were less informative, and instead, the search target color became associated with a history of providing useful/instrumental information (e.g. participants learned the blue search target would have the unique line orientation).  Thus, after the training phase, different colored search targets were associated with either useful or redundant information.

Now that participants have different information histories associated with different colored search targets from the training phase, the researchers wanted to test how having a history of providing useful information influences attention on an unrelated task in the test phase.  In the test phase, depicted in the figure below, participants were asked to identify a colored shape (e.g., a diamond among circles) among other colored distractor targets. In half of the trials, one of the five distractor shapes was in a color that was originally associated with information that was useful (colors used on unreliable trials) or redundant (colors used on the reliable trials) in the training phase.

Using Bayesian Modeling, the authors estimated the probability of an eye movement (saccades) toward a specific distractor color (see model in figure below). They found that participants were more likely to allocate eye movements to distractors that had a history of providing useful/instrumental information, compared to those associated with redundant information, as shown in the gray distribution figure below. This increased attention allocation to the instrumental distractor items led to slower reaction times on the task, though accuracy did not differ.

 As the authors put it:

“Objects with a history of providing useful information reflexively capture our attention, even when attending to them disrupts our current goals. Further, the more that one attends to these information-associated objects, the worse one’s task-relevant performances becomes.”

These findings provide evidence that our past experiences can influence attention capture through a mechanism distinct from a history of reward associations and relies on the information history of an item.

Given that my phone has a long history of providing useful information, I’ll try to make it a habit to immediately tuck it out of my sight when I’m trying to focus my attention!

Featured Psychonomic Society article

Doyle, A., Volkova, K., Crotty, N., Massa, N., & Grubb, M. A. (2025). Information-driven attentional capture. Attention, Perception, & Psychophysics, 1-7. https://doi.org/10.3758/s13414-024-03008-z