An Evolutionary Approach On Binge Drinking
lens of the costly signal (Diamond, 1992; Greitmeyer, Kastenmüller & Fisher, 2013; Kacir, 2010) and life-history theories (Hill & Chow, 2002; Hill, Ross & Low, 1997; Kaplan & Gangestad, 2004). The model proposed in this article claims that Zahavi’s handicap principle (Zahavi, 1975; Zahavi & Zahavi 1999) is the best suited at both the local and global levels of analysis integrating the three-level factors (biological, cognitive and sociological). What type of signals do the binge drinkers send? To whom is the signal directed? Are those signals courtship displays or threats to competitors in order to assure reproductive success or maintain social status? How can contextual factors influence drinking frequency and intensity? Exploring causes, correlates and predictors of binge drinking and its interdisciplinary nature should serve as a relevant starting point to then reveal the necessity of an evolutionary framework.
General health theories, like the Theories of Reasoned Action and Planned Behaviour (Johnston & White, 2003), consider that the intention to perform an action is influenced by an attitudinal and a normative component (Fishbein & Ajzen, 1975). While attitude evaluates the valence of the behavior (positive or negative), the subjective norm is conceptualized through the perception of general social pressure from important peers (later additional features included volitional control, see Johnston & White, 2003). However, such models are mainly correlational (as they observe consequences, they cannot distinguish the causes from the symptoms) and static (they are unable to describe the dynamic aspect of the maintenance or not of risky drinking). On the other hand, those explanations focus on an individual level and may not give a larger picture of the binge drinking phenomenon such as those taking place on US college campuses (Syre & McAllister, 1997).
Our model proposes to analyze binge drinking as a signal of resistance and high genetic quality. It can serve as a direct threat to potential same-sex competitors for resources or as a display of genetic quality to available mates. Binge drinking games can be viewed as a way to reproduce intrasexual competition without any form of direct aggression as it is frequent in the animal world (Smith & Harper, 2003; Smith & Price, 1973). That theory can produce precise experimental predictions as well as modeling key factors. An additional feature of such an approach is its interdisciplinary perspective reconciling the three fold-nature of binge drinking.
Defining Binge Drinking
The original definition of binge drinking simply stated that an individual needed to consume at least five alcoholic beverages during the same session (Cahalan, Cisin, & Crossley 1969). Decades later, the Harvard School of Public Health updated this definition by adding a gender distinction with five drinks for men and four for women in the same siting in a 2-hour period (Wechsler, Davenport, Dowdall, Moeykens, & Castillo, 1994) essentially due to the lower rate of gastric metabolism in women (Wechsler, Dowdall, Davenport, & Rimm, 1995). The 2-hour timeframe has since become the general definition in experimental design of binge drinking (Oei & Morawaska, 2004; Syre & McAllister, 1997; Weschler, Dowdall, Maenner, Hill-Hoyt, & Lee, 1998). A physiological characterization of binge drinking logically accompanies this definition through a blood alcohol concentration (BAC) level of 0.08 percent in a single sitting (NIAAA, 2004; Rolland et al., 2017).
Still, the World Health Organization defines binge drinking as 6 standard drinks (60 grams of alcohol) per sitting without specifying the drinking speed, in contrast to the NIAAA (WHO, 2014). To compare the sociodemographic and drinking characteristics of subjects of those two sets of criteria, Rolland et al. (2017) found that the NIAAA combined criteria included higher frequency of regular drinkers, family drinking troubles and hospitalization. NIAAA criteria also revealed that participants were more likely to be male, single and unemployed.
Pearson, Bravo, Kirouac & Witkiewitz (2017) have recently broadened the limits of the monotonic/non-dynamic validity of the 4+/5+ binge drinking criteria and urge the use of speed as the critical component to encompassing this risky phenomenon.
All in all, the use of such a definition in health and psychological studies has been fairly widespread (King, Houle, Wit, Holdstock & Schuster, 2002; Read, Beattie, Chamberlain & Merill, 2008; Wechsler et al., 1994; Wechsler, Dowdall, Davenport, & Rimm, 1995) allowing uniform reviews and comparisons across time and space (Courtney & Polich, 2009; Zeigler et al., 2005). Beyond the threshold of at least 4 or 5 drinks in a 2 hour window, the social risk to the binge drinker substantially increases (e.g., fights, drunk driving, confrontation with authorities, etc.), as do his or her health (e.g., sexual risks or dysfunction, suicide attempts, stroke, etc.) and economic hazards (lack of productivity) among others (Courtney & Polich, 2009; Miller, Naimi, Brewer & Jones, 2007; Wechsler, Davenport, Dowdall, Moeykens & Castillo, 1994).
The starting age of regular (at least twice a month) binge drinking ranges between 15 and 16 years old depending of the risk level of the children (Hill, Shen, Lowers, & Locke, 2000). While the age of onset of excessive alcohol use is relatively low, the age group with the most binge drinkers extends from 18 to 34 years old with even some 65 years and older reporting binge drinking five to six times a month on average (CDC, 2010). The gender difference in reported binge drinking episodes is another important factor with the prevalence of men being twice that of women (CDC, 2012). Approximately 50% of male students report binge drinking versus 39% of female students (BRFSS, 2016; Wechsler, Davenport, Dowdall, Moeykens, & Castillo, 1994). Social status is another important variable since binge drinking is more common among richer households in the U.S. (incomes of $75,000 or more, see CDC, 2012). Finally, racial differences were also important predictors of binge drinking, with Caucasians involved in 78% of all binge-drinking episodes in the U.S. while African and Asian Americans groups reported the lowest levels (Cranford et al., 2006; Naimi et al., 2003).
Recently, the health community has shifted its determination if alcohol abuse and alcohol dependence fall in a common substance disorders spectrum (DSM-IV, 1995; DSM-5; 2013). Originally, chronic drinking or alcoholism was defined as a prolonged disease including strong craving for alcohol, continued use despite regular interpersonal problems and inability to limit drinking (DSM-IV, 1995). Excessive alcohol use, a form of binge drinking, was conceptually diagnosed as “alcohol abuse” where “school and job performance may suffer either from the aftereffects of drinking or from actual intoxication on the job or at school; child care or household responsibilities may be neglected; and alcohol-related absences may occur from school or job” (DSM-IV, p. 196, 1995). The person may use alcohol in dangerous situations (e.g., driving) and suffer legal consequences (e.g., intoxicated behavior, driving under the influence). “Finally, individuals with Alcohol Abuse may continue to consume alcohol despite the knowledge that continued consumption poses significant social or interpersonal problems for them (e.g., violent arguments with spouse while intoxicated, child abuse)” (DSM-IV, p.196, 1995).
Surprisingly, no mention is made of data showing that most early age binge drinkers are not dependent on alcohol (Esser et al., 2014). Despite being an aggravating risk factor (Englund et al., 2013), the transition from binge drinking to chronic drinking is far from automatic and clear comorbidity between binge drinking and long-term alcohol dependence is uncertain (Hussong, Hicks, Levy & Curran, 2001). While we recognize that such approach might have clear applicable values for practitioners, we think that this conceptualization prevents us from drawing an interdisciplinary picture of the phenomenon. Evolutionary psychology and medicine can respectively serve as theoretical tools to make predictions and framework to build specific and more targeted alcohol prevention programs, particularly among youth1.
Binge and chronic drinking also display different patterns of effects. On a lower level, for instance, alcohol consumption acutely impairs functions of the pre-frontal cortex (in particular, planning, information processing, inhibitory control response flexibility, attention and set-shifting) but increases dopamine release in the ventral striatum, facilitating the onset of risky behaviors and aggressive attacks (Heinz, Beck, Meyer-Linderberg, Sterzer & Heinz, 2011). Chronic alcohol consumption acts differently by impairing serotoninergic transmissions in the prefrontal cortex and the amygdala triggering greater emotional responses to threatening stimuli eventually leading to more aggressive behaviors (see Figure 1).
Another main differences between binge drinkers and alcoholics is the dependence level. While various mechanisms (e.g. self-efficacy, self-esteem, etc.) can make the binge drinker resistant to alcohol outside a social context, the alcoholic may continuously crave alcohol, eventually being unable to resist the temptation. The social acceptance of binge drinking is widespread, especially among college students where playing drinking games can be a way to integrate to a group (Borsari & Carey, 2003) whereas isolation, shame and hiding are more characteristic of chronic drinkers (whether binge drinking alone or not, see Dearing, Stuewig & Tangney, 2005).
Figure 1. Example of the differential acute vs. chronic effects of alcohol on aggressive behaviors. a | Brain areas involved in alcohol-induced aggression. b | Neurotransmitter system variations in alcohol-associated aggression cascades are in blue while, in purple, are pathways affected by genetic and environmental influences.
Binge Drinking as a Multi-faceted Phenomenon
Empirical studies on binge drinking started more than 20 years ago and have shed light on the variety of factors that might play an important role. From cognitive genetics to sociology, the correlates and experimental results pass through different levels of organization that I summarized here, for the sake of the argument, as biological, cognitive and social. I synthesized it graphically (as seen as figure 2) and we will go through the different levels one by one. Then, we will present the different theories of binge drinking and how they integrate those different factors. Finally, we will evaluate their limitations before arguing for an evolutionary approach of binge drinking through the costly signal and the life-history theories that had been fruitful in evolutionary biology.
Fig. 2. The three-fold components of binge drinking. Solid lines indicate a positive relationship whereas dashed ones indicates a negative one (e.g. being male increases the probability of having higher sensation-seeking level compare to women but lower self-control).
In 1949, Straus and Bacon conducted a 3-year study in 27 US colleges and universities of 15,000 men and women reported in their book Drinking in college (Strauss & Bacon, 1953). Their work, confirmed by later studies (Wechsler, Davenport, Dowdall, Moeykens, & Castillo, 1994; Wechsler & McFadden, 1979), showed that a particularly high percentage of freshmen were binge drinkers. As expected, high-school binge drinking pattern was a very strong predictor of later binge-drinking in college (Wechsler & McFadden, 1979). Along with more (subtle-not the best adjective here) studies, a myriad of social factors correlated with binge drinking emerged: involvement in fraternities, party-centered life-style among men (Weschler, Dowdall, Davenport, and Castillo, 1995) but also among young women (Dowdall, Crawford & Wechsler, 1998). Fraternities, for instance, are famous for widespread participation in drinking games and 81.1% of fraternity or sorority residents in the US are binge drinkers (Wechsler et al., 1998; Wechsler, Davenport, Dowdall, Moeykens & Castillo, 1994). Another explanation could be that high school men planning to join a fraternity were already frequently heavy drinkers and that individuals at high-risk of behavior such as drinking (but also unsafe sex, risky driving, etc.) in general are more likely to be part of a fraternity (Canterbury et al., 1992).
When controlling for other demographic variables, the status of “never married” was also a strong predictor of binge drinking (see Weschler, Dowdall, Davenport, and Castillo, 1995; Schulenberg et al., 1996). Other studies are more nuanced and tend to show that the prevalence of binge drinking among single students is not remarkably higher than that among married students their freshmen year of college, but the big difference lies in the change in behavior between the freshman year and subsequent years of their college career: students self-described as “never married” tend to drink more and more in the course of their studies when compared to married students (Wechsler, Lee, Kuo & Lee, 2000). This could indicate that a relationship serves as a protective barrier against binge drinking through less exposure to drinking games, peer pressure or to single friends who engage in risky drinking. On the other hand, single students can use group gatherings and parties as a way to socialize (particularly during the freshman year when students do know not each other) and thus are exposed to drinking games and peer pressure, in particular through fraternities.
In contrast to what could be considered as a “healthy life-style”, student-athletes actually participate in binge drinking more often than their non-athletic counterparts (Weschler, Davenport, Dowdall, Grossman, & Zanakos, 1997). Taking part in athletics can trigger pressure, stress, injuries and social isolation. However, the strongest predictor among students involved in athletics was actually whether the athlete was living in a fraternity or sorority. In his book College Drinking: Reframing a Social Problem / Changing the Culture (2008), George Dowdall goes further by referencing a case of irresponsible and violent behavior among college athletes such as in the Duke Lacrosse team. This led to an ad hoc committee that recommended, in particular, a code of conduct for athletes, a need for improved communication between student affairs and athletics and an enforced alcohol policy. While still maintaining excellent scholastic or athletic achievements, sports team members in colleges might find themselves in a pro-drug, pro-binge drinking environment and project their high competitiveness into drinking games leading to disastrous consequences.
Other sociodemographic factors emerge when studying binge drinking patterns. In particular, precollege and family drinking patterns are good predictors of risky drinking. In a 2003 study, Weitzman, Nelson & Wechsler found that students who started drinking before age 16 were more likely to end up binge drinking when compared to their peers who started later in their adolescence. The early onset of drinking among teenagers might be due to their parents’ alcohol use and attitude toward drinking since two thirds of the group in Weitzman’s study reported than their parents used to drink during their childhood. Chassin, Pitts and Prost (2002) evaluated the binge drinking trajectories from adolescence to emerging adulthood in a high-risk sample. They found that family structure was a good predictor in particular for early heavy drinkers: adolescents from disrupted families (where one or more biological parent is absent) were more likely to be involved in risky drinking activities than their counterparts from unified families (where both biological parents are present). Children of alcoholics (“COA”s) were particularly vulnerable to heavy drinking from the ages of 12 to 23 (Chassin, Curran, Hussong, & Colder, 1996; Chassin, Pitts, DeLucia, & Todd, 1999). The early heavy drinkers exhibited all those characteristics ranging from parental alcoholism and antisociality to peer drinking, drug use and, for men, low levels of depression (Chassin, Pitts & Prost, 2002).
One of the cognitive aspects regularly and robustly linked to binge drinking is the seeking of sensation (Zuckerman, Bone, Neary, Mangelsdorff & Brustman, 1972). A person’s general attitude towards risk, measured through different short- or-long version questionnaires (Stephenson, Hoyle, Palmgreen & Slater, 2003; Zuckerman, 1968), can indicate how much they crave or are willing to be involved in risky activities that are not necessary related to drinking (e.g. skydiving) but can sometimes be highly correlated to it (e.g. wild parties). These personality variables have long been linked to substance use (Donohew, Hoyle & Clayton, 1999; Schwarz, Burkhart & Green, 1978; Zuckerman & Neeb, 1979). A study of an Australian adolescent male population, led by Andrew and Cronin (1997) explored the link between sensation seeking and alcohol use among 318 ninth, tenth and eleventh graders, thus permitting an evaluation before the freshman year of college. Using Arnett’s Inventory of Sensation Seeking (Arnett 1994), they found that desire for intensity was a stronger predictor of alcohol use and binge drinking episodes than desire for novelty.
Another construct known to be linked to binge drinking, although generally related to substance use, is impulsivity (Balodis, Potenza & Olmstead, 2009). Generally considered as a personality trait, it promotes rapid and unplanned responses to stimuli with a lower focus on the negative counterparts (Moeller et al., 2001; Potenza 2007). However, it is difficult to understand the causal direction since impulsivity as a personality trait may predispose young adolescents toward risky drinking but also excessive alcohol use that can impair self-control and inhibition, leading to behavior that could be characterized as impulsive. There is a debate in the literature as to whether gender differences exist regarding the link between impulsivity and binge drinking (Balodis, Potenza & Olmstead, 2009). While some studies reported notable gender differences regarding this relationship (Zuckermann & Kuhlmann, 2000), the gender difference for binge drinking itself is well established (Wechsler et al., 1994). A study of over 428 Queen’s University students between 2002 and 2007 revealed that while impulsivity did not correlate directly to binge drinking per se, it did to drinking habits in general and in particular to number of drinks per drinking occasion (Balodis, Potenza & Olmstead, 2009).
Drinking games are also an interesting platform to analyze behaviors and personality traits among players. Moreover, drinking games are one of the main channels to risky drinking among university students and their participation is a good predictor of heavy episodic drinking (Clapp et al., 2003; Clapp, Won Min, Shillington, Reed, & Ketchie Croff, 2008). Their main reasons or justification for participating in drinking games is competitiveness (i.e., “because I want to win”) and reproductive success (“in order to have sex with someone”; Borsari, 2004).
The beginning of college is often a novel period of relative leisure time without parental supervision and a sudden access to a multitude of potential mates but also competitors, making the first few semesters of college determinants for engagement in binge drinking. For instance, a study on 263 students from the University of Miami, not only showed important gender differences in the frequency of drinking game participation, but also that both social competitiveness and competitive drinking game enthusiasm were good mediators to predict the frequency of participation in drinking games in two-phase studies, confirming other previous studies (Hone & McCullough, 2015).
Alcohol expectancies (AE) and drinking-refusal self-efficacy (DRSE) are also two important predictors of risky drinking. The concept of AE states that the physiological effect of alcohol not only influences our reaction to but also the expectations we have of it (Oei & Morawska, 2004). Participants receiving a placebo while believing it to be alcohol behave in accordance with their expectations of the latter’s effects (Marlatt & Rohsenow, 1980). According to Bandura (1977, 1986) two classes of expectations can be distinguished: the efficacy expectancy – the capacity to resist alcohol in particular situations – and the outcome expectancy – that evaluates the consequences of a particular activity such as binge drinking. Alcohol expectancy seems to be a construct that develops before contact with alcohol and tends to crystalize later in life (Miller, Smith & Goldman, 1990). We are regularly exposed to alcohol through ads, family, peers, and events during childhood and progressively construct specific expectations toward the product stored in long-term memory. Those expectancies might be reinforced when one tends to selectively focus on specific positive outcomes resulting from alcohol consumption rather negatives ones (e.g. sociability vs. negative consequences). Through global positive, social and physical pleasures, AEs have been shown to explain a much larger share of adult and adolescent drinking when compared to other sociodemographic variables (10-19% of the variance observed in adults while peer influence and environment explained only 5% and 8% respectively; Oei & Morawska, 2004; Martin & Hoffman 1993).
However, while AEs might be reliable predictors for the amount of alcohol consumed on a given occasion, they are less efficient for predicting the frequency of drinking (Mooney, Fromme, Kivlahan, & Marlatt, 1987). Drinking-refusal self-efficacy (DRSE) on the other hand, which indicates the capacity to refuse to drink alcohol in particular situation, is related to the frequency of drinking. An individual with low DRSE would be more willing to drink more when given the opportunity to drink. In that sense, Vogel-Sprott (1974) described those aspects of drinking as being influenced by different factor types: the amount of drinking would be an aspect under an individual’s control while the frequency of drinking occasions will be more influenced by social factors. Thus, the dynamic might be that social pressure, or a party-centered life-style in a particular ecology would trigger the opportunity to drink and DRSE will be the switch to participate or not. AEs’ base level will then determine the quantity of alcohol drunk in that particular event. Again, alcohol dependent and binge drinkers are interesting examples of how those two particular factors (AE vs. DRSE) might interact. Chronic dependent drinkers have both positive expectations toward alcohol (high AE) and difficulty resisting it when offered (low DRSE; Cooney, Gillespie, Baker, and Kaplan, 1987). On the other hand, binge drinkers would tend to have higher DRSE and exhibit more willingness to refuse to participate when presented with an opportunity to drink (even though they could end up drinking more when they do agree to attend).
Table of contents :
Chapter 1: Why the Binge Drinkers Survive? A Costly Signal Perspective on Excessive Alcohol Use.
1.1.1. Defining Binge Drinking
1.1.3. Clinical Diagnosis
1.1.4. Distinguishable effects
1.2. Binge Drinking as a Multi-faceted Phenomenon
1.2.1. Social determinants
1.2.2. Cognitive Aspects
1.2.3. Biological components
1.3. Binge Drinking Theories
1.3.1. A cognitive model of binge drinking.
1.3.2. The theory of planned behavior
1.3.3. The social bond theory
1.3.4. The self-control theory
1.4. Is there a meaning to risky drinking?
1.4.1. The limits of social and cognitive theories
1.5. The evolutionary roots of binge drinking
1.5.1. The honest signal theory
1.5.2. Animal Literature
1.5.3. Honest Signaling as an Evolutionary Stable Strategy
1.5.4. A Competitive Theory: the Life-History Theory
1.6. Summary of predictions
1.6.1. Alcohol tolerance as a cue to potential mates.
1.6.2. Alcohol tolerance as a threat signal to same-sex rival
1.6.3. Agent-based modeling of binge drinking
1.6.4. A costly signal perspective on prevention actions
Chapter 2: An Original Method for Testing the Function of Binge Drinking: Dating Website, Eye tracking and Hormonal Data.
2.1. Costly Signal
2.1.1. Pupil dilation as a cue to attractiveness
2.1.2. The ovulatory shift
2.2. Current Study
2.2.1. Study 1
18.104.22.168. Fixation time.
22.214.171.124. Pupil dilation ratio.
126.96.36.199. Ovulatory shift.
2.2.2. Study 2
2.2.4. Study 3
2.3. General Discussion
Chapter 3: Alcohol Tolerance as a Costly Signal for Intrasexual Competition and Reproductive Success Moderated by Social Status: an Online Investigation of Impression Formation.
3.1. Binge drinking data
3.2. The Costly Signal Theory
3.2.1. Biological signal in the animal world
3.2.2. Binge drinking and costly signal theory.
3.3. Current Research
3.3.1. Study 1
188.8.131.52. Preliminary results.
3.3.2. Study 2
184.108.40.206. Preliminary results.
3.3.3. Study 3
220.127.116.11. Preliminary results.
3.3.4. Study 4
3.4. General Discussion
Chapter 4: Risks Signaling as a Prevention Tool Against Binge Drinking: a Field Experiment in a French High School and an Online Replication.
4.1. Costly Signal Theory
4.2. Study 1: Field Study in a French High School
4.2.4. Descriptive results.
4.3. Study 2: Online Replication
4.3.2. General Discussion
Chapter 5: Do Women Expose Themselves to More Health-Related Risks in Certain Phases of The Menstrual Cycle? A Meta-Analytic Review.
5.1. Hormones across the Menstrual Cycle
5.2. Key Domains of Health-related Risk-taking
5.2.1. Alcohol and Tobacco Consumption
18.104.22.168. State of the Evidence: Studies of Nonhuman Mammals Error! Bookmark not defined.
22.214.171.124. State of the Evidence: Human Studies
126.96.36.199. Insights from Evolutionary Theory
188.8.131.52. Insights from Behavioral Neuroendocrinology Error! Bookmark not defined.
5.2.2. Sexual Behavior
184.108.40.206. State of the Evidence: Studies of Nonhuman Animals defined.
220.127.116.11. State of the Evidence: Human Studies
18.104.22.168. Insights from Evolutionary Theory
22.214.171.124. Insights from Behavioral Neuroendocrinology Error! Bookmark not defined.
5.2.3. Risk Recognition and Avoidance
126.96.36.199. State of the Evidence: Nonhuman Animal Models defined.
188.8.131.52. State of the Evidence: Human Studies
184.108.40.206. Insights from Evolutionary Theory
220.127.116.11. Insights from Neuroendocrinology
5.3.1. Inclusion criteria.
5.3.2. Coding Study Characteristics.
5.4.1. Alcohol and Tobacco consumption patterns. Ovulation vs. other phases.
Bookmark not defined.
5.4.2. Sexuality pattern. Ovulation vs. other phases.
5.4.3. Risk recognition and avoidance patterns. Ovulation vs. other phases.
Bookmark not defined.
Chapter 6: Binge Drinking Frequency and Intensity across the United States (2007-2016): a Multilevel Modeling of Life-History Theory and Risky Drinking.
6.1. Life-history theory and risky drinking
6.2. Current Study
6.2.4. Descriptive results.
6.2.5. Inferential results.
18.104.22.168. Binge Drinking Frequency (2016).
22.214.171.124. Drinking Intensity (2007).
126.96.36.199. Drinking Frequency (2007).
Chapter 7. Conclusion
ANNEX CHAPTER 2
ANNEX CHAPTER 4
ANNEX CHAPTER 5
ANNEX CHAPTER 6