Etienne Benson '99 is studying the effect of cocaine addiction on information-processing and has conducted an experiment which he hopes will help clarify the relationship between craving and cognition. Photo by Jon Chase.

Cocaine's addictive properties are well documented. Lab tests have shown that monkeys will press a bar more than 10,000 times for a single injection of cocaine, choose the drug over food and water, and take it even when the behavior is punished. People addicted to cocaine behave similarly, and while studies show that cocaine use in the United States is down in recent years, dependence on the drug remains a serious problem with few effective treatments.

Cocaine dependence is characterized by compulsive drug-seeking behavior and repeated "cravings" to use the drug. However, the connection between these cravings and drug use is unclear, with some experts claiming that craving plays no role in cocaine use. What is clear is that a better understanding of the cognitive aspects of craving is needed.

Etienne Benson '99, a concentrator in psychology and biology, is studying the effect of cocaine addiction on information- processing. With support from the Harvard College Research Program (HCRP), he has conducted an experiment that he hopes will help to clarify the relationship between craving and cognition. "I've been interested in addiction since I arrived at Harvard," Benson says, "but my interest was really catalyzed in my sophomore year by a seminar on drug policy taught by William Brownsberger, a former Massachusetts assistant district attorney for narcotics. It was a small, intense class that took an interdisciplinary approach to drug policy, combining neuroscience, psychology, sociology, and public policy, among other things.

"We took a field trip to Massachusetts General Hospital so we could hear about ongoing research by [psychiatrist and instructor at the Medical School] Dr. Randy Gollub and others on the effects of cocaine on the brain," Benson continues. "When I found out that Dr. Gollub was looking for a research assistant, I jumped at the opportunity."

Benson launched his own study this fall, which he hopes will shed light on the cognitive aspects of the motivational state of craving. His study will also test the hypothesis that cocaine addiction is accompanied by the kinds of cognitive biases that are seen in a number of psychiatric disorders.

"The mixed evidence for the role of drug craving in drug use may be due to an indirect effect of craving on behavior," Benson explains. "The effect may be mediated by changes in the way cocaine addicts perceive and interpret information about the world. For instance, craving might enhance the ability of cocaine addicts to pay attention to cocaine-related cues -- and decrease their ability to pay attention to anything else -- thereby increasing the likelihood of drug use."

According to Benson, information-processing biases like the one he cites have been found in a broad range of psychiatric disorders, including panic disorder and depression. However, very few studies have been conducted on information-processing biases in addictive disorders.

Limited research suggests that pathological gamblers and cigarette smokers in short-term withdrawal have a selective bias for words related to their addiction. The same tendency has not yet been proved for cocaine addicts, but Benson feels it is likely that they will display a selective bias for words related to cocaine. The role that cocaine craving, or subjective desire for the drug, plays has not been previously explored.

To uncover information-processing biases in his subjects, Benson is employing a paradigm called the "Stroop task." It was introduced by J. Ridley Stroop in 1935, who showed that people are slower at naming the ink color of a printed word when the word conflicts with the color in which it is printed. (For example, the word "blue," when printed in red ink, might cause us to stumble.) It has since been demonstrated that emotionally charged words cause similar delays in color-naming. "The Stroop task is usually taken as an index of 'cognitive interference,' " Benson says, "and as such it's been used in studies of almost every kind of psychopathology. Patients tend to respond more slowly to words that are related to their disorder -- 'filth' for obsessive-compulsive disorder, for instance -- than to unrelated words."

Benson is using a modified Stroop task to determine whether cocaine-dependent subjects show a slowed response to cocaine- related words. He has also been measuring the levels of anxiety and craving of the subjects and attempting to correlate them with performance on the Stroop task. "My hypothesis is that cocaine-related stimuli will interfere with Stroop performance of cocaine-dependent subjects much more than it will with the performance of normal subjects," says Benson. In addition, he expects to find a relationship between the subjects' current levels of craving for cocaine and their performance on the task.

Benson will present his findings this spring in a senior thesis. His research has been supervised by Richard McNally, professor of psychology, and Gollub. "Professor McNally has provided crucial input on the design and implementation of the experiment," Benson says, "while Dr. Gollub has assisted in the day-to-day planning of the experiment, the human subjects approval process, and the recruitment of cocaine-dependent subjects."

The project has reinforced Benson's desire to attend graduate school in psychology this fall. "I haven't decided which school I'll end up at," he says. "My top two choices are Stanford and N.Y.U."

- This story originally appeared in The Harvard-Radcliffe Undergraduate Research Programs Newsletter.