By James R. Ashenhurst
Every day we are faced with decisions to make, both simple and complex: should I buy the bargain brand cereal or do I really want to pay more for those Cheerios? Sometimes, we’re faced with decisions that carry a bit more risk to our health and safety: should I jump out of this plane and skydive like I planned to, or is the risk that something might go wrong too high? In the addiction world, decisions must be made about the risks of buying and using drugs and alcohol: Should I really be driving home from the bar now, risking a DUI? What if the police catch me buying crystal meth?
People naturally vary in the amount of risk they are generally willing to take. Especially when potential rewards are great, some people will take rather extreme risks, while others are more hesitant. Clearly, the world needs risk-takers to brave the waters with new business ideas, or to risk rejection to gain romance. Risk-taking is by no means a uniformly bad trait. But, when it comes to drug use, how might having a risk-taking personality affect how people choose to use? Understanding how risk-taking relates to drug and alcohol dependence (alcoholism) might help clinicians and addiction treatment centers be more effective by making patients aware of how their own risk-propensity influences their disease.
The difficult part of answering these questions is deciding how you’re going to figure out exactly how risk-taking a person is. In the past, many researchers used simple self-report questionnaires that boil down to essentially asking participants how risk-taking they think that they are. However, there is a good deal of self-report bias when using these questionnaires; in other words, the accuracy of a person’s answer depends on how self-aware they are and how well they evaluate themselves compared to others (which also requires them to evaluate others objectively). To deal with this problem, Carl Lejuez developed an elegantly simple experimental task that avoids self-report bias: the Balloon Analogue Risk Task  (named the BART in honor of The Simpsons, they also made a task called the MRBURNS).
It works like this: you see a balloon on a computer screen and you can press a button to inflate it by a small amount. Every time you inflate it, you get a small amount of money. But, there is always a chance that when you inflate it, the balloon will pop and you’ll loose all the money you’ve accumulated for that balloon. You can also decide to “cash out” at any point and add the money you’ve earned to a guaranteed bank and move on to the next balloon to pump. Participants actually receive the money they’ve banked in the task. So, how far would you go?
As it turns out, how people behave in this task relates pretty well to how they behave in the world (this is known as external validity); a person who inflates a lot (and probably pops more than a few balloons) is more likely to not wear a seatbelt, practice unsafe sex and, yes, experiment with drugs and drink problematically  [2, 3]. Also, a twin study has shown that risk-taking in the BART is heritable in males  and I have demonstrated that behavior is heritable in a rat version of the task , suggesting that at least some of it is due to nature and some due to nurture. This is good news for medical research, because it means that there is some discoverable biological pathway that determines, in part, how people behave in the BART.
Still, this preliminary research about the BART and alcohol was gathered from young undergraduates who do not have long histories of alcoholism or drug dependence. Thus, for my research, I wanted to know how older folks who are diagnosably alcoholic might behave in the BART . We invited 158 gracious volunteers from the Los Angeles community (who identified themselves as having problems with alcohol) to the lab and evaluated their dependency severity under the same guidelines used by psychiatrists in the DSM-IV. We also had them play the BART. My prediction was that participants with more severe alcoholism would also tend to be bigger risk-takers.
To my surprise, everything flipped around. People who were more risk-taking (inflated the balloons bigger) actually had fewer alcoholism symptoms. In other words, the more severe the case of alcoholism, the less risks they would take in the BART. How could this be, and what does this tell us about the role of risk-taking in alcoholism?
There are several possibilities. For one, it could be the case that while young risk-takers tend to drink problematically, as alcoholism develops, it is actually the problem-drinkers who are more risk-averse who tend to go on into more severe cases of alcoholism. This theory relies on the idea that risk-taking personality is fixed and doesn’t change much in adulthood; it might be a stable trait that influences the developmental course of alcoholism.
It could be, however, that the trait is not always stable across a lifetime, and experience with alcohol changes one’s risk-taking personality. If we assume instability, it could be either social and/or biological factors that cause the change. Maybe people with more severe alcoholism face more problems in their personal life, and this changes their temperament to be more risk averse. Or, it could be that the continued exposure to a lot of alcohol changes the parts of the brain that evaluate risks and underlie the decision-making process. It is well-known that chronic exposure to alcohol at high levels for long periods of time changes the quantity and subtypes of neurotransmitter receptors in the brain as part of an adaptive process; the brain adjusts itself to tolerate the constant signals it’s getting from alcohol. Thus, it is a reasonable idea that decision-making parts of the brain could change too.
Lastly, it could also just be an observation that is specific to this task in this population. While the task has been shown to be externally valid in the college-aged sample, we didn’t reassess that here for older alcoholics. We’re talking about people taking small risks to earn relatively small amounts of money by the end of the task. Usually, participants are rewarded with somewhere between $5 to $20, depending on the study.
What if larger sums of money were at play? Or access to alcohol was at risk? Once a person is an active alcoholic, what feels risky and what’s not might change too. Acknowledging that you have a problem and starting to try to cut down or abstain might feel more risky than continuing as normal. Nevertheless, even if this flip is specific to behavior in a laboratory task, it means that the relationship between risk-taking and alcoholism is not as straightforward as we might expect.
So, what do you think? In your experience, are the more severely alcoholic people you’ve known not big risk-takers? If you’re an alcoholic in recovery, does it seem like your risk-taking personality changed over time? Hopefully, we’ll get more clues down the line and we’ll be better positioned to say which theory is correct, and this can then help alcoholics in their own pathway to addiction recovery.
1. Lejuez, C.W., et al., Evaluation of a behavioral measure of risk taking: the Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Applied, 2002. 8(2): p. 75-84.
2. Fernie, G., et al., Risk-taking but not response inhibition or delay discounting predict alcohol consumption in social drinkers. Drug and Alcohol Dependence, 2010. 112(1-2): p. 54-61.
3. Lejuez, C.W., et al., Differences in risk-taking propensity across inner-city adolescent ever- and never-smokers. Nicotine Tob Res, 2005. 7(1): p. 71-9.
4. Anokhin, A.P., et al., Heritability of risk-taking in adolescence: a longitudinal twin study. Twin Research and Human Genetics, 2009. 12(4): p. 366-71.
5. Ashenhurst, J.R., M. Seaman, and J. David Jentsch, Responding in a Test of Decision-Making Under Risk is Under Moderate Genetic Control in the Rat. Alcoholism: Clinical and Experimental Research, in press.
6. Ashenhurst, J.R., J.D. Jentsch, and L.A. Ray, Risk-Taking and Alcohol Use Disorders Symptomatology in a Sample of Problem Drinkers. Experimental and Clinical Psychopharmacology, 2011. 19(5): p. 361-70.