When evidence in a criminal trial is improperly presented, judges can instruct jurors to "disregard" or intentionally forget it. But a new study suggests that even jurors who do forget a piece of evidence--not an easy thing to do--can be unconsciously influenced by it.

What's more, by forgetting, they may actually increase their vulnerability to such influences, says the study's lead author, Elizabeth L. Bjork, PhD, a psychologist at the University of California, Los Angeles.

In previous studies, Bjork and her co-author, Robert A. Bjork, PhD, have found that people are less likely to recall information they have learned after being told they will not be tested on it, although the information continues to affect them indirectly.

They hypothesized that people whose behavior is influenced by false information, but who cannot identify the source of that influence, may commit more memory errors than people who are able to both remember and dismiss it.

To explore this possibility, they asked 96 college students in two experiments to study lists of nonfamous and famous names. Half of the participants were told to remember the nonfamous names, while the other half were told to forget them. Later, both groups were asked to judge the famousness of a mixed list of famous and nonfamous names.

The critical measure was the number of false alarms--names on the nonfamous list that were misidentified as famous in the final test.

They found that the fewest false alarms came from participants in the "remember" condition, who were able to correctly attribute the familiarity of nonfamous names to the earlier presentation of those names. The most came from participants in the "forget" condition, who were more prone to misinterpreting the names' familiarity as a sign of their fame.

The finding suggests that consciously forgetting false information does not always reduce memory errors and indeed can sometimes have the opposite effect, says Bjork.

"It has this dark side," she notes. "Because the to-be-forgotten information remains in memory, it can continue to influence judgments of various types." Jurors could be influenced by a witness's inadmissible testimony, for instance, without being able to identity the source of the influence. Judges might therefore be better off instructing them to remember the testimony--and the fact that it is inadmissible--than asking them to disregard it.

The study appears in the Journal of Experimental Psychology: Learning, Memory, and Cognition (Vol. 29, No. 4).

--E. BENSON