Gaze cues are the window to incidental learning about the world

I had never given much thought to the saying that “the eyes are the window to the soul” until I watched the TV show, Fleabag. Without giving too much away (although, honestly, this paragraph will probably be a major spoiler), the show follows a woman who regularly breaks the fourth wall, where she looks away from the scene and into the camera to share her inner thoughts with the audience. Throughout most of the show, I assumed that this was just a narrative device to invite us into the protagonist’s thoughts while also protecting us from the reality that she herself was avoiding. My assumption was turned on its head when someone (a love interest, naturally) followed her gaze, and in a moment that captured the exultant terror of falling in love, there was suddenly someone else who could see the audience, and more importantly, her innermost self.

*I guess I’m going all in with spoilers. The spooky scene in question is from Fleabag series 3, episode 4. Image capture by Denise Wong from BBC iPlayer.*

What’s funny about this scene and the cliché I opened with is that this moment didn’t come from one character looking into another’s eyes, as the original “window into the soul” saying would imply. Instead, it happened because the protagonist looked away, and for the first time in her life, someone else took that as a cue to see what drew her attention. It’s a fascinating way of distorting a completely natural behavior into something close to earth-shattering in the context of the show and the character. After all, isn’t it completely normal to follow a person’s gaze when they look away?

Of course it is! Cognitive psychologists generally don’t investigate the soul per se, but there are still plenty of interesting questions about how we might use another person’s gaze to help us learn about our surroundings. If following a person’s gaze is as natural as it seems, is it possible that children can use gaze as a cue to direct their attention or remember details about their environment? After all, the ability to track another person’s gaze creates a capacity to share in an experience together, which can then allow for what cognitive psychologists call incidental learning, where knowledge is acquired without explicit instructions. In other words, it’s possible that part of a child’s experience of learning involves paying attention to social cues like gaze without obvious instructions to do so.

*Does joining a child’s attention with your own with gaze cues help them learn about their surroundings? [Public domain from Wikimedia commons]*

A recent Psychonomic Bulletin & Review report by Mitsuhiko Ishikawa and Ayumi Yoshioka investigates children’s abilities to use gaze as an attentional and memory cue more deeply. Ishikawa and Yoshioka were particularly interested in the age group between 7 and 10 years because of prior evidence suggesting that many cognitive skills rapidly expand during this period. Do gaze cues lend advantages to children’s incidental learning or would any kind of visual cue, like an arrow, suffice? Would these patterns differ between older children and younger children? I’ll let the title of their report speak for itself: “Gaze cues facilitate incidental learning in children aged 7–10 years, but arrow cues do not.”

The study consisted of two phases. In the first part, the authors used a cueing task to gauge how different types of cues influenced how children direct their attention. In this task, the children watched for the appearance of a fractal image on a screen. When the image appeared, the children would press a key to indicate whether the image appeared on the left or right side of the display. Importantly, before the image appeared, the children were shown a cue pointing to either the left or right side of the screen. In one condition, the cue was a simple arrow; in the other, the cue took the form of a person looking to the left or right. These arrow and gaze cues were not always accurate—sometimes, the fractal image target appeared on the opposite side from the direction the cue pointed. The authors then compared reaction times from “valid” trials, where the arrow or gaze cue accurately indicated where the target next appeared, to “invalid” trials. They also split the data by age to compare patterns between the younger children between 7 and 8 years of age to the older children aged 9 to 10.

*A schematic of the procedure for the cueing task, taken from Figure 1a of the featured article.*

Slower reaction times in these kinds of tasks generally suggest that directing attention was more difficult. It’s no surprise, then, that younger children generally had slower reaction times than older children (anyone who’s ever tried to wrangle a second grader to pay attention to something they’re not already interested in would probably understand). It’s also unsurprising that trials with an invalid cue also led to slower reaction times; the participants were cued to look at the wrong side, after all.

What’s interesting about these results is that there were otherwise no differences in reaction times driven by the difference between gaze versus arrow cues. In other words, gaze cues and arrow cues seem to have similar effects in directing children’s attention, regardless of their age (at least within this age group) and regardless of the accuracy of the cue.

The results of the cueing task. Taller bars indicate slower reaction times. Younger children were always slower than older children, and invalid cues always led to slower responses than valid cues. Importantly, these patterns are consistent for both gaze cues and arrow cues, indicating that both types of cues have similar effects on directing attention.

But since we’re interested in learning, the crux of the study isn’t just in how children direct their attention but also in whether social cues like gaze afford any advantages for children’s memory. The second part of the authors’ study addressed this by testing children’s memory for the fractal image targets they saw during the cueing task. In the memory portion, the fractal images were simply shown to the children one by one, and the children then indicated whether they had seen the image in the prior cueing task or if they thought it was a new image they had not seen previously. Since they were never instructed to try remembering the fractal images during the cueing task, any indication of remembering the images during the memory task can be taken as a moment of incidental learning.

A schematic of the memory task, from Figure 1b of the featured article.

While the overall memory patterns were ultimately similar regardless of age, children did seem to be better at remembering images if their attention had first been cued to the correct side by a gaze cue. Cueing their attention with arrows otherwise didn’t seem to boost children’s memory, regardless of whether the arrow cue turned out to accurately indicate which side the fractal image appeared on. This suggests that children do indeed differentiate between social cues such as gaze from other kinds of visual cues, and these gaze cues can enhance children’s memory for visual details, thereby facilitating their incidental learning about the world.

The results of the memory task. Taller bars indicate better memory. The dashed horizontal line represents hypothetical memory results equivalent to random guessing. Children’s memory was enhanced after seeing valid gaze cues but not arrow cues, and these benefits were consistent for older and younger children.

Altogether, these results suggest that gaze serves as a more powerful benefit to memory than simple arrow cues. More broadly, this could mean that children derive great learning benefits from cues that are socially engaging, and it points us toward strategies to help children learn more effectively. Instead of simply pointing out objects in a child’s surroundings, it might be better to actively gaze at these objects to help cue a child to join their attention with yours and share in an experience together. In the end, regardless of whether you think the eyes are indeed the window to the soul, maybe it’s still true that gaze could be a window to a child learning more about the world.

Featured Psychonomic Society article

Ishikawa, M., & Yoshioka, A. (2025). Gaze cues facilitate incidental learning in children aged 7–10 years, but arrow cues do not. Psychonomic Bulletin & Review. https://doi.org/10.3758/s13423-025-02657-x

Author

Melinh K. Lai

Melinh K. Lai is a graduate student in Psychology at the University of Illinois, Urbana-Champaign. Her research focuses on predictions that people make as they comprehend language, with a current focus on how both prediction dynamics and broader comprehension and memory processes change in accordance with different goals. She primarily studies these concepts through the use of event-related potentials (ERPs).
View all posts