Particularly Exciting Experiments in Psychology
December 22, 2015
Do We Know When to Stop Acquiring New Information?
As the fall semester comes to an end, students all over the world are studying for exams. An inherent part of the studying process is deciding what to study, and for how long. For example, students must decide when they know something well enough to move on to the next subject, or when to stop studying altogether so they can get a good night's sleep. Because monitoring what one knows and adjusting behavior accordingly involves complex processes, it is interesting to ask whether this ability is shared with non-human primates.
To address this question, Tu, Pani, and Hampton (2015, Journal of Comparative Psychology) (PDF, 129KB) trained rhesus monkeys to classify images of birds, flowers, fish, and people. After training, monkeys were shown images that were initially concealed behind gray rectangle blockers. Pressing a "reveal" button removed one blocker. For each test image, monkeys could press the "reveal" button as often as they wanted, and were free to make a classification response at any time.
Monkeys made more "reveal" choices when blockers were smaller, and thus more blockers had to be removed to reveal sufficient information for classification, suggesting they were basing decisions about removing blockers on whether they had enough information.
In a subsequent experiment, monkeys were shown test images with the same pattern of blockers as the point at which they made a correct classification response in the previous experiment, test images with additional blockers (i.e., less of the image revealed), or test images with fewer blockers (i.e., more of the image revealed).
Monkeys never made a classification without removing some blockers, but nevertheless adjusted their use of the "reveal" button based on the initial state of the image; they made more "reveal" choices when less of the image was available, and fewer "reveal" choices when more of the image was available. These results suggest that monkeys evaluated the available information and collected more information until they were confident in their classification ability.
Tu and colleagues examined information-seeking behavior in the context of accumulating enough information about a single item to respond correctly. However, final exams cover an entire semester's worth of information about potentially unrelated topics, and "success" is defined as the overall exam grade, not performance on a single question. In this case, a student may adopt a strategy of mastering some aspects of each topic rather than trying to study everything.
Indeed, Murayama and colleagues (2016, Journal of Experimental Psychology: Learning, Memory, and Cognition) (PDF, 186KB) found that people generally believe that restricting how much is learned will lead to better memory performance than trying to learn everything. Participants were told that their goal was to remember as many words as possible, and they would be awarded 10 cents for each correctly recalled word. Participants in the control group saw 50 words one at a time. Participants in the stop group were also shown words one at a time, but were told that they could either study the whole list of 50 words, or could press a "stop" button to stop learning new words at any time.
More than half of participants in the stop group chose to stop studying new words after around 30–35 words. Importantly, however, despite the fact that stopping was meant to improve memory performance, this was not an effective strategy: participants in the stop group actually remembered significantly fewer words than participants in the control group. In fact, there was a significant correlation between when participants stopped studying new words and memory performance, such that memory performance decreased as the number of studied words decreased.
These results suggest that people make study decisions based on the false intuition that memory performance can be maximized by restricting the amount of studied information, when in fact studying more information affords more opportunities to successfully remember.
Particularly Exciting Experiments in Psychology™ (PeePs) is a free summary of ongoing research trends common to six APA journals that focus on experimental psychology.