Psychologists from USC, the University of Texas at Austin and Beijing Normal University recently have found that some of our understanding of memory retention is incorrect.

In a study published online on Sept. 9 in Science, researchers at these schools discovered that memory increases when similar patterns are repeated for strength rather than when multiple varying patterns are made.

The long-standing belief held by psychologists maintains that information is best remembered when attained or learned in varying contexts, which creates multiple patterns to remember the information.

Gui Xue, a research assistant  and professor of psychology at USC who worked on the project, said that while someone is studying an item, he or she has to process or re-activate the same pattern repeatedly in order to be able to better remember the information.

“If you create one pattern during the first learning and a different pattern during the second learning, you are going to remember worse,” Xue said.

Experiments were conducted using functional magnetic resonance imaging, or fMRI, which allowed the researchers to monitor the memory patterns used for recollection.

Subjects were given materials such as words, faces and novel texts and were asked to recall material at a later time.

“When the activation pattern is more consistent across repetitions, you are going to remember better,” Xue said.

The research is aimed at developing additional understanding of the “forgetting curve.”

This concept, developed by Hermann Ebbinghaus in 1885, holds that memory decreases exponentially with time. The strength of one’s memory is determined by how long memory traces sustain.

Although this type of research does occur often, the researchers “developed new ideas — pattern similarity — and analysis techniques,” says study co-author Zhong-Lin Lu of USC, who holds the William M. Keck chair in cognitive neuroscience and is a professor of psychology and biomedical engineering.

As the research continues, Lu said the group is working toward a larger objective.

“One aim of our research is to find neural signatures that would allow us to predict whether and when people will forget learnt materials and refresh their memory before they forget,” Lu said.

This information could then be used to determine when review is necessary in order to sustain knowledge. Having the ability to measure retention would greatly improve disciplines that rely heavily on memorization, such as language study.

Xue, who received his Ph.D. in psychology from the Beijing Normal University, said the group’s main objective is to identify more beneficial methods of learning.

“Our goal is to try to find a scientific basis for effective learning,” he said.