Teachers are constantly seeking to outfox would-be plagiarists. But there's a wrinkle: Plagiarism isn't always deliberate.

Indeed, research has shown that inadvertent plagiarism--or cryptomnesia, as psychologists have termed the phenomenon--is a rather common memory glitch that pervades everyday cognitive functioning.

"It happens even in everyday conversation," notes University of Georgia cognitive psychologist Richard L. Marsh, PhD, who conducts research on cryptomnesia. "Somebody says, 'Mary is so effervescent,' and then the word 'effervescent' keeps coming up." Cryptomnesia stems from a failure to simultaneously engage in creative thinking and monitor where incoming ideas are coming from, according to Marsh's research.

For better or worse, he explains, "When you put people into a creative task like coming up with new ideas or solving puzzles, they're directing conscious processing at the task itself, which leaves fewer resources to monitor ideas' source."

When people are encouraged to monitor the source of incoming information, his research indicates, they are less likely to fall victim to cryptomnesia.

Ubiquitous phenomenon

Psychological research on cryptomnesia has a short history. In three experiments published in the Journal of Experimental Psychology: Learning, Memory and Cognition (Vol. 15, No. 3) in 1989, Southern Methodist University psychologist Alan S. Brown, PhD, and student Dana R. Murphy, PhD, now an assistant professor of psychology at Nipissing University in Ontario, described a series of experiments demonstrating the phenomenon.

In their first experiment, the researchers had groups of four participants generate words that belonged to particular conceptual categories--for example, sports and musical instruments. Participants were instructed not to volunteer words that had already been used. The participants took turns generating words until each participant had contributed four words. Later, they were asked to recall the four words that they had generated for each category and to generate four completely new items.

Brown and Murphy found that cryptomnesia occurred in 3 to 9 percent of cases, with participants inadvertently borrowing others' contributions more often when they were attempting to recall previously generated items or trying to generate new items than during the initial word-generation task.

In addition, the researchers found that participants rarely reused their own contributions, suggesting that people monitor their own and others' responses differently. Participants were also more likely to inadvertently plagiarize participants whose responses immediately preceded their own, suggesting that they were less adept at processing incoming information when they were anticipating their own responses.

In follow-up experiments, Brown and Murphy extended their results to show inadvertent plagiarism not only in a social, group-based situation but also when information was presented visually, on index cards. And they found that cryptomnesia occurred more frequently when the response-generation procedure was more complex, making it more difficult for participants to monitor incoming information.

"This was a seminal paper because it was the first laboratory-based, critical examination of cryptomnesia," comments Marsh. Brown and Murphy's findings, he says, spurred him to conduct a number of subsequent studies of cryptomnesia, in an effort to better understand the unconscious influence that memory has on everyday problem-solving tasks.

In a series of experiments published between 1993 and 1999, Marsh and colleagues have extended the original results. In an initial foray, he and colleague Gordon H. Bower, PhD, of Stanford University, adapted Brown and Murphy's procedure by having participants form words using the word-search game Boggle and competing against a computer partner rather than against other participants.

As in Brown and Murphy's experiments, participants inadvertently plagiarized responses that were not their own. In fact, across a series of experiments using this experimental procedure, Marsh and Bower observed considerably higher rates of cryptomnesia than Brown and Murphy had found--a finding that they attributed to the fact that the Boggle task, a more creative problem-solving task than Brown and Murphy's word-generation task, more closely resembled real-world circumstances in which cryptomnesia might occur.

In subsequent experiments, Marsh and University of Georgia colleagues Joshua D. Landau, PhD, and Jason L. Hicks, PhD, now at Louisiana State University, showed that the rate of cryptomnesia was greater under specific conditions. It increased, for example, when there were fewer perceptual and contextual cues--such as the distinctiveness of the voice associated with other-generated information--that participants could use to perform source monitoring during the task, when less time was provided during which participants could monitor the source of incoming information, and when sources of incoming information were more credible. In contrast, plagiarism rates dropped when participants were asked to focus on the origins of their ideas, heightening their awareness of the source of ideas.

Research lends insights into insights

Overall, says Marsh, "We were quite amazed at the amount of unconscious plagiarism that was expressed in these tasks." He speculates that cryptomnesia is probably "a heck of a lot more common than anybody would realize."

Despite its potential for fostering unpleasant arguments over ideas' origin, Marsh argues, the phenomenon is probably not always a bad thing. For example, he suggests, unconscious plagiarism may be an important way for children to learn social norms for behavior. And, Marsh believes, cryptomnesia may play a critical role in many other domains, ranging from how people come to believe advertising claims or adopt particular political positions to how they respond in psychotherapy.

"There's a classic phenomenon in clinical psychology, where a therapist will be trying to get a client to believe something about their behavior, and the client is often resistant at first," he explains. "Then, the client comes in one day with an 'insight.' They're often not really insights at all--they're just expressions of what the therapist has been saying."

The mechanisms that underlie cryptomnesia also have important implications for creativity, Marsh believes. Recently, he, Landau, Hicks and psychologist Thomas B. Ward, PhD, of Texas A&M University, have examined how unconscious learning affects the creative process. In one series of experiments, for example, the researchers asked participants to draw novel space creatures. They found that when participants were first shown a few examples of space creatures that all contained some features in common, such as four legs, antennae and a tail, participants reliably included those features in their own drawings--even though they were instructed not to copy any of the features used in the examples.

"If we want to understand how it is that people design skyscrapers, or write music, or write a New York Times best seller," Marsh concludes, "I think we need to acknowledge that nothing we design is ever truly novel--every creative effort contains vestiges of what we have experienced in the past."