Grandma, where’s Waldo? Comparing Goal-directed and Habit-based Attention in Older Adults

Visual search can be a very daunting task, whether it’s looking for keys among a pile of office supplies on a desk, or looking for a dime among a bunch of coins in your purse. It seems like no matter how many times you have checked and re-checked certain locations, the object of your affection is nowhere to be found. This is the classic ‘Where’s Waldo’ problem come to life.

In the picture below, try to find the small image of Waldo (not the big one waving) among the clutter.

Did you find Waldo? (Here’s a hint: he’s hiding in a pair of striped pants).

As you can see, visual search for a specific object can be quite challenging, and this search problem is only exacerbated with age. In fact, older adults have more difficulty with visual search compared with younger adults, especially when the distractors are visually similar to the target.

To compensate for issues with clutter, searchers need to allocate attention to important locations in the visual scene. Two types of attention systems that are known to influence visual search behaviors are habit-based attention and goal-directed attention.

Habit-based attention can occur when prior declarative information is used to inform search. For example, left-to-right readers scan images of text from left to right when looking for specific information. Similarly, when looking for keys or a wallet, people tend to check locations where those objects have a high prior probability of being found (e.g., in a purse or in a book bag).

Goal-directed attention, by contrast, can occur when explicit instructions are given on how to execute the search. For example, when children are looking for different objects, parents instruct them to check in their bedrooms first.

Young and older adults can use both habit-based and goal-directed attention to inform visual search. For young adults, these two systems co-exist and when they are in conflict with one another, priority is given to goal-directed attention. An interesting question is, what happens when these two attention systems are in conflict for older adults? Which system dominates?

In a recent study published in the Psychonomic Society journal Psychonomic Bulletin and Review, Emily Twedell and colleagues pitted habit-based attention and goal-directed attention against one another in a visual search task to assess which system most affects search in older adults.

Twedell and colleagues conducted a visual search experiment on both young and older adults that consisted of three phases: a training phase, a recognition test, and a testing phase. The search task for the participant was to locate the letter T among several letter Ls. The figure below explains the task and the overall methodology.

To induce habit-based attention, participants completed a training phase where the probability of the target presence was higher in one location of the display (one of four quadrants) relative to the remaining locations (50% in high probability quadrant and 16.7% in the remaining quadrants.) Importantly, the participants remained seated in a fixed orientation relative to the display.

After the training phase, participants were given a surprise recognition test where they were asked whether the target appeared more frequently in certain locations or was equally likely to have appeared anywhere in the visual display. Regardless of the participants’ answers, they were then instructed to click in the quadrant were they thought the target appeared most frequently and rate their level of confidence in their choice (1 = guessing, 4= certain).

During the subsequent final testing phase, the targets did not appear more frequently in one quadrant relative to the others, but instead were equally likely to appear in all four quadrants (25% of the time). Also, the experimenter informed the participants of the high probability quadrant and highlighted the quadrant in blue for the remainder of the experiment, before the final test commenced.

There was one more crucial twist in the procedure.

Before final testing began, participants were asked to move to the adjacent seat (oriented at 90 degrees to the first one).  This is equivalent to moving the participants from facing north to facing east or west (see the figure above) in an effort to disrupt the participants’ spatial association to the high probability quadrant. For example, in a four-by-four display, if the high probability quadrant was in the top right corner during training, when a participant shifts 90 degrees to the right during test, it becomes the bottom right corner.

To induce goal-directed attention, participants were informed that the search procedure would be the same as during training; however, they were asked to search in the identified high-probability quadrant first, on every trial. Importantly, this quadrant was in a different place relative to the participant who has changed their seat location. This design pits habit and goal attention against each other because the location of the learned, habit-based quadrant is now in a different place relative to the participants’ viewing perspective.

There were several interesting and noteworthy results. First, young adults were faster in the visual search task than older adults. Side note: older adults had higher IQ scores and were slightly more accurate in the search task (Take that young people!).

During training, both young and older adults were faster at locating the target in the high probability quadrant than in the other quadrants, demonstrating their ability to form search habits through incidental learning. Also, roughly half of the participants in both age groups correctly identified the high-probability quadrant as evidenced by the recognition test.

During the final test, both young and older adults were faster at finding the target in the former screen location of the high-probability quadrant relative to the lower probability quadrants. This suggests that both groups developed successful allocation of goal-directed attention, as they were instructed in the task to prioritize this region during search.

As for the conflict between habit-based and goal-directed attention, older adults were less able to utilize the task instructions to prioritize the high-probability quadrant compared to young adults. In other words, age affects goal-guided attention, but not habit-based attention.

This has implications for how displays should be designed in places where older adults frequent, such as in grocery stores and shopping malls. So the next time Grandma is looking for Waldo (or her favorite peanut butter in the grocery store), try to bring her in the store using the same entrance. And if all stores of a supermarket chain keep their layouts the same, then older people would presumably be particularly appreciative.

Featured Psychonomic Society article:

Twedell, E, Koutstaal, W., & Jiang, Y. (2016). Aging affects the balance between goal-guided and habitual spatial attention. Psychonomic Bulletin, & Review, DOI: 10.3758/s13423-016-12143.

You may also like