The Role of Expectancy in Steroid-Related Aggression
Imagine if every day was like your first and you had to relearn everything everyday. Life would be impossible and you’d get little accomplished. Our behavior is made efficient and automatic via processes that carry our experiences forward, allowing us to anticipate situations, behavioral requirements and outcomes, although positive outcomes are never guaranteed. This discussion of such processes in AAS-related aggression expands on descriptions of the potential role of such processes in AAS-related aggression by Riem and Hersey (1995) and Sharp and Collins (1998) and also uses examples from alcohol research.
How does experience (actual or observational) guide future behavior? Current theories suggest through learned expectancies (Dragoi & Staddon, 1998; Rescorla, 1988). These expectancies reflect “if-then” associations among events, behaviors and outcomes. Experiences create associations between circumstances and their behavioral requirements (e.g., conflict – aggression) and behaviors and their outcomes (e.g., aggression – success). Such expectancies are likely to underlie physical phenomena such as anticipatory increases in endogenous T prior to competition (even in a competition as innocuous as a chess tournament; e.g., Mazur, Booth, & Dabbs, 1992). The role of these processes has been studied in hypnosis (Council, 1999), fear (Schoenberger, 1999), tobacco use (Brandon, Juliana, & Copeland, 1999), and alcohol use and subsequent behavior (Goldman, Darkes, & Del Boca, 1999), among other behaviors.
Researchers often equate these expectations with beliefs (e.g., Pope, Kouri, & Hudson, 2000), suggesting that they represent conscious knowledge that supports conscious decision-making. In fact, expectancies generally function outside of awareness. If we had to actively think about and decide upon every action we take, we would be able to do little else except think. Consider a day on which you are on your way home, but you need to go to the store first. Your attention wavers for a moment and you find yourself pulling into your driveway. If you can relate to this situation then you have experienced expectancy driven behavior – certain stimuli evoke certain behaviors automatically and conscious effort is not required to initiate this chain of events, but to change it.
Substance Use Expectancies
Why do people use AAS? To enhance performance, appearance, or mood, increase strength, aggression, interpersonal dominance, or workout intensity? These expected outcomes suggest both an existing motivational state (e.g., physique or strength concerns, desire for weight gain) and expectancies for AAS effects (e.g., that it will improve physique or mood, increase aggression). Both the motivation and the expectancy increase both the probability of drug use and the occurrence of certain outcomes. Without “wanting” these effects and “knowing” that AAS “do” these things, why would one use them? The association between circumstances (sport or physique concerns) and behavior (drug use) increases the likelihood of use and the “if-then” association – if one uses AAS, then certain outcomes result – increases the likelihood of those expected outcomes. People would probably not use AAS (or any other substance) without a perceived “need” for them and the “knowledge” (through observation, gym lore, or personal experience) of and desire for their expected effects. This also applies to training, diet and supplementation. Few people do or take anything without some information (conscious/unconscious) of the “likely” results and an existing desire or motivation to achieve those results.
Alcohol researchers have found that positive alcohol expectations predict drinking initiation, transition to problematic use, and post-use behavior (See Goldman, Darkes, & Del Boca, 1999, for a review). Individuals feel and act intoxicated when served drinks they believe may contain alcohol even if they do not (Darkes & Goldman, 1993; 1998; Marlatt, Demming, & Reid, 1973) and this placebo effect may account for approximately 75% of the effect of many widely prescribed antidepressant medications (Kirsch & Saperstein, 1998).
Expectations for Anabolic Steroids
Could anyone not have heard reports that AAS “cause” aggression? It has been reported in the popular media (Taber, 1999) and is well known in the resistance training subculture (e.g., Misc.Fitness.Weights). The Olympic games recently cast the spotlight on AAS and their alleged effects. Some activities, such as resistance training, increase the likelihood of exposure to AAS and to subculture-specific information about the “positive” effects of AAS. And, although information about potential negative effects may be encountered, it has little effect (Lindstrom, Nilsson, Katzman, Janzon, & Dymling, 1990). Negative effects are viewed by users and potential users as either improbable or a distant future possibility. Ultimately, the combination of exposure and positive expectancy makes high dose use and certain psychological outcomes more probable.
Alcohol researchers also find that sensation-seeking tendencies (an affinity for arousing experiences) predict heavier alcohol use, but act partially through expectancies for positive social effects from alcohol (Darkes, Greenbaum, & Goldman, 1996). Similarly, pre-existing factors (e.g., desire for weight gain; Wang, Yesalis, Fitzhugh, & Buckley, 1994) predict AAS use and are likely to act partially influence use through expectancies for increased aggression/dominance and physique enhancement from AAS. Believing alcohol has been consumed can activate such expectancies and result in increased social behavior, even without drinking alcohol or the recognition that social behavior is being assessed. Limited evidence suggests that AAS expectancies could lead to or enhance the expected outcomes in a manner that is partially independent of drug use.
An AAS/Expectancy Hypothesis
Alcohol expectancies have been measured explicitly (endorsement of expectations) and implicitly (via methods from cognitive psychology), evaluated using complex statistical methods and inferred from experiments that manipulate expectancies and evaluate changes in use and behavior (See Goldman, et al., 1999). Their role in alcohol use and related behavior is largely unquestioned. Few investigations of AAS information that might reflect expectancies exist (e.g., Schwerin & Corcoran, 1996). In general AAS researchers have not addressed expectancies’ role in AAS related psychological effects, much less their potential role in physiological effects.
Subjective increases in enthusiasm and aggression (Bahrke, Wright, Strauss, & Catlin, 1992) and general positive feelings (Rashid, 2000) have been associated with AAS use. Users endorsed stronger “beliefs” in positive outcomes from AAS use than did non-users (Schwerin & Corcoran, 1996), but the groups did not differ in AAS expectations of aggression, suggesting that non-users are familiar with and expect aggression to result from AAS use. Whether users and non-users evaluate AAS-related aggression differently (negative or positive) is unknown, as is the direction of causal influence, since positive beliefs may result from experiences/exposure or predate and motivate use.
To explore the interaction of expectancies, AAS use, and behavior, expectancies could be measured in AAS naïve subjects and assessed over time as use begins and outcomes occur. For instance, expectancies of non-drinking adolescents predicted the transition to alcohol use one-year later (Christiansen, Smith, Roehling, & Goldman, 1989). Alternatively, expectancies for drug administration could be manipulated and outcomes evaluated. In balanced placebo studies (fully crossing drug/placebo treatment and drug/no drug instructions), believing alcohol was consumed when it was not produced “intoxicated” behaviors (e.g., increased sexual arousal) and increased craving in alcoholics (Marlatt, Demming, & Reid, 1973; Engle & Williams, 1972). There are apparently no studies over time of the interplay between AAS use and expectancies in initially AAS naïve subjects.
Would participants administered AAS or credible (discussed later) placeboes in a balanced placebo manner exhibit the psychological effects of AAS in the absence of actual drug administration? Although the possibility remains untested, there is reason to suspect as much. This design allows the differentiation of pharmacological and psychological effects. The behavior of those who do not expect to receive AAS but do, would ostensibly be pharmacologically based. Those expecting AAS but receiving placebo would exhibit the effects of their drug expectancies.
Participants given low dose testosterone, inert placebo, or no treatment were compared and self-report measures of anger, irritability, impulsivity and frustration found that the placebo group scored more highly than the T and control groups (Bjorkqvist et al., 1994). This is consistent with an expectancy hypothesis for AAS psychological effects. Unfortunately, expectancies were not assessed and the lack of a similar expectancy effect with low dose T is perplexing. The authors suggested that low dose T may suppress endogenous T, hence no effect. Endogenous T did not increase with treatment. Although animal evidence suggests a tranquilizing effect of low dose androgens (Agren, Thiblin, Lundeberg, & Stenfors, 1999) and low dose T may be useful in treating depression in the elderly (see Seidman & Walsh, 1999 for a review), this hypothesis requires further investigation.
Although mentioned five years ago (Riem & Hersey, 1995), AAS researchers have yet to address the inadequacy of their inactive placeboes. Such “inert” placeboes solely control for the act of injecting or swallowing a drug (external context), not for side effects of active treatments that might provide internal cues regarding drug treatment and facilitate expectancy-based responses. For instance, Pope et al. (2000) injected 600 mg. of testosterone cypionate or an equal amount of sesame oil vehicle (typical of AAS administration experiments; e.g., Kouri et al., 1995; Tricker et al, 1996). Besides subtle increases in aggressive responding on the Point Subtraction Aggression Paradigm (PSAP: an analog test of aggression), group differences were found for physical factors, such as blood pressure, testicular length, and lean body mass.
Imagine participating in this study. Might effects like testicular or LBM changes and the effects associated with them suggest the nature of your drug treatment? Such effects might “break the code” and activate unconscious “if-then” associations between drug use and outcomes. Participants have detected the effects of AAS within a relatively brief period of time (Lukas, 1996). For instance, even without actual strength increases, subjects given AAS reported a subjective sense of increased strength within 3 weeks of beginning treatment (Crist et al., 1983) and, although still blinded, participants could retrospectively distinguish between drug and withdrawal phases (Su et al., 1993).
Admittedly, partitioning effects into those that mediate treatment response and those that are extraneous is difficult. But, given current understanding of the effect of expectancy, “proving” that a drug causes certain outcomes requires the effort. It may be that, in AAS research, such side effects result from the increases in blood steroid levels that are also hypothesized to cause increased aggression, hence their differentiation would be difficult. On the other hand, serum AAS levels and reported aggression rarely correlate reliably (e.g., Bhasin et al., 1996; Bjorkqvist, Nygren, Bjorklund, & Bjorkqvist, 1994; Su et al., 1993), casting doubt on such an assumption. Without such an effort, the interpretation of the results of most of the current AAS studies is a difficult and speculative endeavor.
Incidentally, the PSAP seems unique in its detection of increased aggression in experimental participants. Two studies (Kouri, Lukas, Pope, & Oliva, 1995; Pope, Kouri, & Hudson, 2000) using the PSAP (which considers punitive responses to a fictitious opponent as aggressive responding) reported increased aggressive responding after testosterone administration. However, the use of inert placebo and the absence of increased aggression reported by users’ significant others is problematic. In fact, experiments using observer ratings rarely, if ever, report changes in aggression (Bjorkqvist et al., 1994; Tricker et al., 1996).
I have engaged in what some might call rampant speculation on the role of expectancy in AAS-related aggression. If so, then these hypotheses could and should be tested by those who do AAS research. The occurrence of AAS-related aggression is not in question. This discussion should not be taken to suggest that AAS related aggression is not serious or that, in the absence of proof, it is an inevitable physiological result of AAS use, it is purely volitional. I merely suggest that the causal process is complex.
Aggression does occur in some AAS users and, regardless of the causal mechanisms involved, that aggression and its effects are real and potentially harmful to users and those around them. A finding that psychological factors contribute to such aggression does not then make it an exclusively conscious choice. The questions to be answered are not whether aggression occurs in some users, but why it does not occur in all users, regardless of variations in dose, whether AAS are the exclusive and direct cause of that aggression, and what other processes might be implicated?
What difference does it make? So, who cares? I suggest that the answer is important to correct misconceptions regarding both AAS as drugs and AAS users in general. Users have committed aggressive offenses and employed steroid defenses, engendering an indictment of AAS as a class of drugs. Rather than indict a class of drugs, a review of the data suggests an analysis of the individual risk factors for such outcomes, much as is done with other drugs. For instance, alcohol has also been related to aggression and people also have expectancies for aggressive effects from alcohol. Nonetheless, alcohol has not been widely viewed with the same disdain accorded AAS.
Beyond the practical implications of the investigation of a causal role of AAS, personality and expectancy, there is also the scientific interest. AAS researchers should be motivated by a desire to understand the role of these processes, to fully explicate the mechanisms of these effects, as opposed to accepting and espousing simple statements. The attempt to unravel such relationships, to elaborate the causal process is after all, the nature of science.
Although AAS play a role in aggression, the relationship is far from simple. AAS research has not adopted and investigated well-documented theories from the psychological literature on other substance use. There is reason to believe that expectancy factors play a role in the use of AAS and subsequent psychological effects. Until AAS researchers design and execute such studies, the explanation remains plausible, though unsubstantiated. Next month I will conclude this discussion of AAS use and psychological effects by providing a summary and proposing a model for the psychological effects of AAS.
Agren, G., Thiblin, I., Lundeberg, T., & Stenfors, C. (1999). Behavioral anxiolytic effects of low-dose anabolic steroids treatment in rats. Physiology and Behavior, 66, 503-509.
Bahrke, M.S., Wright, J.E., Strauss, R.H., & Catlin, D.H. (1992). Psychological moods and subjectively perceived behavioral and somatic changes accompanying anabolic-androgenic steroid use. American Journal of Sports Medicine, 20, 717-724.
Bhasin, S., Storer, T.W., Berman, N., Callegari, C., Clevenger, B., Phillips, J., Bunnell, T.J., Tricker, R., Shirazi, A., & Casaburi, R. (1996). The effects of supraphysiologic doses of testosterone on muscle size and strength in normal men. The New England Journal of Medicine, 335, 1-7.
Bjorkqvist, K., Nygren, T., Bjorklund, A.C., & Bjorkqvist, S.E. (1994). Testosterone intake and aggressiveness: Real effect or anticipation. Aggressive Behavior, 20, 17-26.
Brandon, T.H., Juliana, L.M., & Copeland, A.L. (1999). Expectancies for tobacco smoking. In I. Kirsch (Ed.), How expectancies shape experience (pp. 263-300).American Psychological Association; Washington, D.C.
Christiansen, B.A., Smith, G.T., Roehling, P.V., & Goldman, M.S. (1989). Using alcohol expectancies to predict adolescent drinking behavior at one year. Journal of Consulting and Clinical psychology, 57, 93-99.
Council, J.R. (1999). Hypnosis and response expectancies. In I. Kirsch (Ed.), How expectancies shape experience (pp. 383-402). American Psychological Association; Washington, D.C.
Crist, D.M., Stackpole, P.J., & Peake, G.T. (1983). Effects of androgenic-anabolic steroids on neuromuscular power and body composition. Journal of Applied Physiology, 54, 366-370.
Darkes, J., & Goldman. M.S. (1993). Expectancy challenge and drinking reduction: Experimental evidence for a mediational process. Journal of Consulting and Clinical Psychology, 61, 344-353.
Darkes, J., & Goldman. M.S. (1998). Expectancy challenge and drinking reduction: Process and structure in the alcohol expectancy network. Experimental and Clinical Psychoparmacology, 6, 64-76.
Darkes, J., Greenbaum, P.E., & Goldman, M.S. (1996). Disinhibition and alcohol use: The mediational role of alcohol expectancies. Paper presented at the 104th Annual Meeting of the American Psychology Association, Toronto, Ontario, Canada.
Dragoi, V., & Staddon, J.E.R. (1999). The dynamics of operant conditioning. Psychological Review, 106, 20-61.
Engle, K.B., & Williams, T.K. (1972). Effect of an ounce of vodka on alcoholics’ desire for alcohol. Quarterly Journal of Studies on Alcohol, 33, 1099-1105.
Goldman, M.S., Darkes, J., & Del Boca, F.K. (1999). Expectancy mediation of bioposychosocial risk for alcohol use and alcoholism. In I. Kirsch (Ed.), How expectancies shape experience (pp. 233-262). American Psychological Association; Washington, D.C.
Kirsch, I. (1999). Response expectancy: An introduction. In I. Kirsch (Ed.), How expectancies shape experience (pp. 3-14). American Psychological Association; Washington, D.C.
Kirsch, I., & Saperstein, G. (1998). Listening to prozac but hearing placebo: A meta-analysis of antidepressant medications. In I. Kirsch (Ed.), How expectancies shape experience (pp. 302-320). American Psychological Association; Washington, D.C.
Kouri, E.M., Lukas, S.E., Pope, H.G. & Oliva, P.S. (1995). Increased aggressive responding in male volunteers following the administration of gradually increasing doses of testosterone cypionate. Drug and Alcohol Dependence, 40, 73-79.
Lindstrom, M., Nilsson, A.L., Katzman, P.L., Janzon, L., & Dymling, J.F. (1990). Use of anabolic-androgenic steroids among bodybuilders – frequency and attitudes. Journal of Internal Medicine, 227, 407-411.
Lukas, S.E. (1996). CNS effects and abuse liability of anabolic-androgenic steroids. Annual Review of Pharmacology 7 Toxicology, 36, 333-357.
Marlatt, G.A., Demming, B., & Reid, J.B. (1973). Loss of control drinking in alcoholics: An experimental analogue. Journal of Abnormal Psychology, 81, 233-241.
Mazur, A., Booth, A., & Dabbs, J.M., Jr. (1999). Testosterone and chess competition. Social Psychology Quarterly, 55, 70-77.
Pope, H.G, Kouri, E.M., & Hudson, J.I. (2000). Effects of supraphysiologic doses of testosterone on mood and aggression in normal men: A randomized controlled trial. Archives of General Psychiatry, 57, 133-140.
Rescorla, R.A. (1988). Pavlovian conditioning: It’s not what you think it is. American Psychologist, 43, 151-160.
Riem, K.E., & Hersey, K.G. (1995). Using anabolic-androgenic steroids to enhance physique and performance: Effects on mood and behavior. Clinical Psychology review, 15, 235-256.
Schoenberger, N.E. (1999). Expectancy and fear. In I. Kirsch (Ed.), How expectancies shape experience (pp. 125-144). American Psychological Association; Washington, D.C.
Schwerin, M.J., & Corcoran, K.J. (1996). Beliefs about steroids: User v. nonuser comparisons. Drug & Alcohol Dependence, 40, 221-225.
Seidman, S.N., & Walsh, B.T. (1999). Testosterone and depression in aging men. American Journal of Geriatric Psychiatry, 7, 18-33.
Sharp, M., & Collins, D. (1998). Exploring the “inevitability” of the relationship between anabolic-androgenic steroid use and aggression in human males. Journal of Sport and exercise Psychology, 20, 379-394.
Su, T., Pagliaro, M., Schmidt, P..J., Pickar, D., Wolkowitz, O., & Rubinow, D.R. (1993). Neuropsychiatric effects of anabolic steroids in male normal volunteers. Journal of the American Medical Association, 269, 2760-64
Taber, A. (1999). ‘Roid rage. Salon.com: Health & Body. Available: http://www.salon.com/health/feature/1999/11/18/steroids
Tricker, R., Casaburi, R., Storer, T.W., Clevenger, B., Berman, N., Shirazi, A., & Bhasin, S. (1996). The effects of supraphysiological doses of testosterone on angry behavior in healthy eugonadal men: A clinical research center study. Journal of Clinical Endocrinology and Metabolism, 16, 3754-3758.
Wang, M.Q., Yesalis, C.E., Fitzhugh, E.C., & Buckley, W.E. (1994). Desire for weight gain and potential risks of adolescents males using anabolic steroids. Perceptual and Motor Skills, 78, 267-274