Conditioning--the workhorse of psychology made famous by Pavlov and his team of dogs--has long impressed psychologists with its power to dramatically shape animal as well as human behavior. Now, a new study shows conditioning may also encourage people to start taking a drug they initially reject.

Five of six participants in the study, published in the May issue of the APA journal Experimental and Clinical Psychopharmacology (Vol. 10, No. 2), switched their preference from a sugar pill to a capsule filled with diazepam (Valium) after researchers surreptitiously paired the drug with a bigger monetary payoff than the placebo. The effect was so strong, it made up for the study's small sample size, says University of California, Los Angeles, psychologist John Roll, PhD, who conducted the study with colleagues at Wayne State University.

And the implications of the findings could be significant, he says. They indicate that people may be more likely to continue taking a drug if it's paired with a positive experience, even if the first encounter with the drug is unpleasant. And such pairings are common: For example, teens often smoke marijuana in highly social settings, and some people take amphetamines to accompany sexual experiences.

The study "nicely remind[s] us that drug ingestion by humans takes place in a rich and dynamic environment that provides many different opportunities for learning," writes University of Vermont psychopharmacologist Mark Bouton, PhD, in one of several commentaries accompanying the article.

That said, this particular study is just one step in proving that conditioning per se is at work encouraging people to take diazepam or any other drug, adds Bouton. Several other explanations are equally plausible, including the idea that people simply grow to enjoy the effects of the drug over repeated exposure.

Your choice: sedative or sugar pill?

The Experimental and Clinical Psychopharmacology study consisted of two phases of nine sessions each with each session lasting about five-and-a-half hours with at least 48 hours between sessions. At the beginning of the first four sessions of phase one, the researchers gave participants a capsule containing either a placebo or diazepam, alternating between the two so that each participant sampled the placebo twice and diazepam twice. Except for brief encounters with the researchers, participants lounged in a room filled with games, books and videos for the remainder of the session.

The next five sessions of phase one were exactly the same, except participants could freely choose between the two capsules they'd taken in the previous sessions. As expected from earlier research, most participants--five of six--chose the placebo on every trial. The ones who didn't chose diazepam every time.

Phase two was similar to phase one, but instead of being idle during the first four sessions, participants completed two computer tasks every half-hour beginning 30 minutes before they ingested the capsule and ending two hours after. Researchers told them that they would earn extra money based on their performance on these tasks, and at the end of each set of tasks the computer displayed how much they had earned.

In reality, how much money participants earned was not linked to their performance. Instead, the researchers programmed the computer to pay participants less--an average of $10.13--when they ingested the drug they preferred during the choice sessions of phase one (placebo for five of six participants and diazepam for one) and more--an average of $20.41--when they ingested the drug they rejected during those sessions. In effect, participants got surreptitiously rewarded for taking diazepam.

To help mask the fact that the amount of money people earned was linked to drug choice rather than performance, the researchers used computer tasks in which people find it difficult to gauge their own performance. The last five sessions were exactly the same as in phase one with participants allowed to choose which capsule to take and then left to their own devices for the majority of the session.

All five participants who had preferred placebo during phase one of the study preferred diazepam during phase two once it was linked with more money than taking the placebo. Indeed, while participants chose placebo on 83 percent of occasions during phase one, they chose it on only 13 percent of occasions during phase two, choosing diazepam 87 percent of the time. The only participant who didn't change his preference from phase one to phase two was the one who preferred diazepam in the first round. In phase two, he stuck with his preference.

Interestingly, pairing money with the drug not only appeared to affect which drug people chose to take, it also affected how the drug made them feel. During phase one of the study, the five participants who preferred placebo reported that it was more pleasurable than diazepam. After the researchers paired diazepam with a higher payoff in phase two, however, the same people reported feeling better after taking diazepam.

What about the one participant who defied the trend and chose diazepam consistently throughout the study? It's possible, the researchers say, that for this one participant, diazepam was a strong enough temptation from the beginning to override any reinforcement of money. "It may be that once the pharmacology of a drug is working as a reinforcer, other influences have less of an effect," says Roll. That's why he thinks this type of conditioning might be strongest when people are first experimenting with drugs.

Questioning the mechanism

As for the majority who did shift their preference from placebo to diazepam in phase two, the researchers propose conditioning is responsible. After pairing diazepam with extra money during the first four sessions of phase two, the drug acquired the properties of a conditioned reinforcer, they say. That is, people unconsciously associated the drug with more money, thereby making it more appealing than it normally would be. The study is one of the first to show that people can be conditioned to have a preference for a drug, says Roll. "We suggest the mechanism is conditioned reinforcement," he adds.

But there may be other equally plausible mechanisms at work, suggest Bouton and Concordia University's Jane Stewart, PhD, who also wrote a commentary. For example, says Stewart, it may be that diazepam had a different--and perhaps more pleasurable--physiological effect on participants during phase two when they were engaged in the computer tasks.

"If this were the case," she writes, "then the change in drug choice would not arise from associating some fixed pharmacological effect of the drug with a rewarding state of affairs; rather, the different subjective effects of the drug in the two conditions could be attributed to the differential effects of the drug on the brain."

Roll and his colleagues admit their study does not rule out alternative explanations. That's why they're continuing with this same line of research. Next, they will see if they can use the same paradigm to get people to switch from preferring a pleasurable drug, such as an amphetamine, to preferring a placebo. They also hope to use these studies as a model for designing laboratory-based studies of ways to prevent drug use.

That said, if this study's preliminary findings hold up and a drug can become a conditioned reinforcer, there are "profound" implications for drug abuse research, write Roll and his colleagues. "Consider how often a drug is introduced to someone in the context of other sources of powerful reinforcers," they write. "For example, methamphetamine is often used among certain groups of individuals to facilitate sexual activity."

Even more common, people use alcohol and tobacco in bars to "loosen" up and are rewarded with meeting people. Teens use psychoactive drugs as an inroad to forming friendships. And people who use caffeine to wake up may find their behavior becomes reinforced when they're more successful at early-morning meetings.

In some sense, says Roll, people get a "double whammy": Added to the pleasurable effects of the drug is a sexual encounter, a friendship made or a point scored at work--powerful reinforcers all on their own. Particularly for someone using the drug for the first time, such a pairing may make the drug more appealing than it would have been by itself and thereby increase the likelihood the person will use the drug again.

Beth Azar is a writer in Portland, Ore.