The Hazards of Teamwork: Does Group Study Disrupt Learning?

The classic 1973 film The Paper Chase explores the challenges of first-year law students at Harvard, focusing on a handful who come together to form a study group. These groups are formed to manage the vast amount of learning that 1-L students are expected to absorb, on everything from contracts to property to the Constitution—but in this case the collaboration is a disaster. A combination of stress and competitiveness and pettiness sabotages the group effort, leaving the individual students on their own as they face the rigors of final exams.

Study groups are very popular—and not only in law school. Right now, on university campuses around the country, students of all sorts are gathering in small groups to pool their intellectual resources. Each one of these students is wagering that individual effort cannot match the mental power of the group. But is this true? And why do we believe it to begin with?

Surprisingly, the social nature of learning has been largely ignored by psychological scientists, who have for most of the past century focused on individual memory and thinking. But recently there has been a shift toward the study of collaborative learning, and this effort is already yielding some surprising insights.

In a typical collaborative memory experiment, individuals study a list of items on their own, and later try to recall them. They might recall the material individually, or they might come together as a group and work together to remember the material. In this kind of experiment, the collective recall is almost always superior to any individual’s recall. This is unsurprising in itself, though it may explain why we think collaborative learning is superior. But here’s the more important finding: The group’s collaborative learning is far inferior to the pooled learning of the individuals working alone. That is, four people trying to learn ten items might recall seven, six, eight and seven items—but with overlap. Once overlap is eliminated, the four’s pooled learning is almost always superior to that of four people gathering to see what they remember as a group.

This difference is quite dramatic—about 55 percent recall for the group, compared to almost 70 percent for the individuals pooled together. What this means is that the group is not performing at the overall potential one would expect from a collaboration. Memory expert Suparna Rajaram calls this “collaborative inhibition,” and in the most recent issue of the journal Current Directions in Psychological Science she offers some possible explanations for the phenomenon.

One is social loafing. It’s well known that individuals tend to make less effort when they don’t feel responsible for carrying the whole load. But Rajaram does not think that social dynamics can fully explain the failure of the group in these situations. Instead, she suspects a cognitive failure—specifically, a disruption of the actual retrieval strategies that individuals use in recall. Each of us has over time developed an idiosyncratic way of storing new information, and as a result we come into a group with a unique retrieval style. Simply listening to others think out loud can interfere with this individual retrieval strategy. Studies support this: For example, giving people memory cues eliminates collaborative inhibition, presumably because recall relies less on idiosyncratic memory organization.

So what’s the long-range effect of collaboration on learning—say during the exam or even beyond? There is growing evidence, Rajaram argues, that if recall for particular information is disrupted during group collaboration, that information may be unavailable even later on, when the individual is alone. In other words, once recall is disrupted by the group dynamic, it may get disorganized or stored somewhere inaccessible—so it’s in effect forgotten for good. It’s also possible that individuals incorporate others’ mistakes—picked up during collaboration—into their own memories. Such contagion can also reshape long-term memories of new learning.

It’s not all bad news, however. Working in groups appears to offer some long-term benefit by re-exposing learners to material they themselves may have forgotten, and collaboration can also “prune” individual errors of recall. The net effect of collaboration’s ups and downs is not yet tallied, but these positive effects no doubt  contribute to our powerful, shared belief that learning in groups is a good thing.

Wray Herbert’s book, On Second Thought, is about irrational decision making. Excerpts from his two blogs—“We’re Only Human” and “Full Frontal Psychology”—appear regularly in The Huffington Post and in Scientific American Mind.


Is there any evidence of “collaborative inhibition” in group activities, like meetings, that involve something other than pure recall? Brainstorming, for instance.

The task of recalling words is relatively simple compared with most learning that goes on in college or grad school. Never-the-less it is surprising that an effect was found for the group discussion interfering with individual’s memory encoding style.

In the course of a group working together, the members also learn about the other members of the group, about group norms, and other side issues in addition to the intended learning, so the net amount learned may actually be greater for the group condition, even though the intended learning was limited.

Maybe the memory limitations involved in group learning has a social facilitation function … memories of contrasting styles or opinions (that could lead to conflicts) may also be spontaneously limited.

It interesting to evaluate these findings in the context of individualized online or elearning vs classroom learning. Classroom learning is more like the group learning situation, and therefore may not be superior for learning targeted material as some have suggested. However, if the targeted material is something like ‘how to function effectively in a group’ rather than memory for words, it would be tough to believe that the individualized learning would be superior.

Bernard Schuster

Rajaram’s fascinating research seems to suggest that collaboration inhibits retrieval. Whether or not it inhibits learning seems like a different (though related) question.

Having taught for 12 years I have tried many approaches with the students and I often try to have a well balanced program that includes group study and group projects. There are certainly benefits to it, but not for everyone. The main reason in fact for offering varied instruction methods is that people are individuals and thus have individual needs. Sometimes you find the right dynamic and things go well in the group situation and sometimes they don’t. It depends on a number of factors.
For college students it is less difficult, I would imagine, to find a study group that works well together as the further up you move in the system the more students are separated and grouped according to skill set and interest (in general). Blending the right skills and stimuli to create an effective study group is a whole other thing though. As a teacher I have had professional study groups in my continuing education courses that have gone well and have also been a complete disaster. Although we have similar skill, ability and interest we can be far too different personality wise and then personal conflicts develop and they take more of our attention than the learning.
Can we better prepare our children to work together at a young age so study groups are more effective later in life or are we too different as individuals to achieve more success than we presently do?

The headline caught my eye because my most dreaded words throughout school were the teacher’s saying, “We’re going to do a group project…” Why does it seem to be assumed by so many that kids like group learning and / or that it’s beneficial? I’m not a team player; today I’m a freelance translator [] and couldn’t be happier with my career. Pupils should not be required to learn as a group.

APS regularly opens certain online articles for discussion on our website. Effective February 2021, you must be a logged-in APS member to post comments. By posting a comment, you agree to our Community Guidelines and the display of your profile information, including your name and affiliation. Any opinions, findings, conclusions, or recommendations present in article comments are those of the writers and do not necessarily reflect the views of APS or the article’s author. For more information, please see our Community Guidelines.

Please login with your APS account to comment.