The Influence of Context in the Subjective Evaluation of “Negative” Alcohol-Related Consequences

Abstract

College students may subjectively evaluate the consequences of drinking in unexpected ways, rating “negative” consequences as neutral or even positive experiences. We previously gathered qualitative evidence for several contextual factors that may influence one’s subjective evaluations (e.g., social influences, intoxication level, concurrent positive consequences). The purpose of the present study was to quantitatively investigate whether student evaluations of consequences differ by various contextual factors. We administered an anonymous online survey to 214 college students (76% female, 69% drinkers, 48% heavy drinkers) in return for academic credit, on which they reported how positively or negatively they would evaluate six alcohol-related consequences (e.g., vomiting, being rude, blacking out) under different conditions. Paired samples t-tests were used to compare evaluation ratings, averaged across consequences, under different hypothetical contextual conditions (e.g., among friends vs strangers). Out of 16 contextual comparisons, all but one (first-time vs repeated consequence) revealed significant differences in subjective evaluations; the pattern held in reduced samples of just drinkers or just heavy drinkers. Examples include that consequences were viewed more negatively (a) if occurring at either a party or bar vs alone in one’s home/dorm, (b) if friends express concern vs find it funny, (c) if the consequence was expected vs unexpected, and (d) if there are lasting consequences vs none. When using recent consequences in feedback-based interventions to build motivation to change, it may be useful to discuss the context of a consequence and how this influences the perceived aversiveness of that consequence.

Introduction

Alcohol misuse among college students is both common and associated with a myriad of objective negative consequences. Over one third of college students report at least one heavy episodic drinking (HED) (4+/5+ drinks in a single sitting for females/males) episode in the past 2 weeks (Johnston, O’Malley, Bachman, & Schulenberg, 2010). Annual rates of consequences reach 646,000 for physical assaults, 97,000 for sexual assaults, 599,000 for unintentional injuries, and 1,825 for deaths (Hingson, Zha, & Weitzman, 2009). These alarming numbers highlight the need for continued understanding of contributing factors to college student drinking, in order to better design and refine preventive interventions for alcohol misuse in this population.

The consequences students experience from drinking may be either reinforcing or punishing, and therefore are a potentially important contributing factor to the maintenance or reduction of subsequent drinking. Support for such a learning theory perspective comes from research showing that young adults reporting a greater number of negative consequences demonstrate lower alcohol use and consequences later on (White & Ray, 2013), and that different negative alcohol-related consequences are associated with specific drinking trajectories (Read, Merrill, Kahler, & Strong, 2007, 2012). However, other work shows that negative consequences are not prospectively associated with drinking (Park, Kim, Gellis, Zaso, & Maisto, 2014; Park, Kim, & Sori, 2013; Zaso et al., 2016).

Beyond the objective experience of negative consequences, and consistent with Social Learning Theory (SLT) (Akers, 1985; Bandura, 1986; Maisto, Carey, & Bradizza, 1999), the way such consequences are subjectively evaluated is also demonstrated to be important. Interestingly, college students sometimes subjectively evaluate the consequences of drinking in unexpected ways, rating “negative” consequences as neutral or even positive experiences (Barnett, Merrill, Kahler, & Colby, 2015; Mallett, Bachrach, & Rob Turrisi, 2008; Patrick & Maggs, 2011; White & Ray, 2013). Cross-sectionally, more negative subjective evaluations have been linked to higher levels of readiness to change drinking (Barnett et al., 2002; 2003; 2006; Longabaugh et al., 1995) and lower levels alcohol use among college students (Gaher & Simons, 2007; Mallett, Bachrach, & Turrisi, 2008). Longitudinal work shows that when consequences are evaluated more negatively, students drink less in the near future (Merrill, Read, & Barnett, 2013). Some research has shown evaluations to predict subsequent drinking even when controlling for the number of consequences (Barnett et al., 2015), though others have not (White & Ray, 2013; Zaso et al., 2016).

Despite a potential predictive role for negative consequences and their subjective evaluations, little is known about the environmental conditions under which consequences are perceived as subjectively negative. Knowing when and why consequences are (or are not) viewed negatively has implications for improved prediction of future drinking behavior. In a recent study (citation masked for review), we gathered qualitative evidence from college drinkers for several contextual factors that may influence subjective evaluations of “negative” alcohol consequences. The purpose of the present study was to quantitatively investigate whether student evaluations of drinking consequences differ by various contextual factors, by asking students to rate how negative vs positive a series of consequences would be under a range of hypothetical conditions that were informed by our qualitative work. Those conditions associated with more negative evaluations may be those likely to result in increased motivation to change and/or self-initiated change in drinking behavior.

A few studies have examined individual-level predictors of subjective evaluations. In one study, females, older participants, and those who experienced consequences more often evaluated consequences as more bothersome, while college vs. non-college status was not associated with consequence evaluations (White & Ray, 2013). In a second study, more negative evaluations of recent consequences were associated with higher levels of past year consequences, as well as higher descriptive norms (i.e., perceptions that consequences were more common) (Merrill, Read, & Colder, 2013). However, variation in subjective evaluations of consequences is observed not only between- but also within-individuals (Merrill, Read, & Barnett, 2013). This suggests that how a consequence is subjectively evaluated may depend on contextual-level factors – things that differ within a person, across time, rather than just between persons.

SLT supports this suggestion, highlighting the importance not only of cognitive factors (e.g., evaluations of consequences), but also of environmental factors (e.g., external and internal contextual influences) in the learning of drinking behavior. Specifically, the SLT principle of differential reinforcement suggests that consequences may be evaluated differently depending on “stimulus conditions” or contextual factors. Here, contextual factors refer to both the internal and external conditions that may influence alcohol-related cognitions. Yet, we know little about what might determine whether a student evaluates a given consequence negatively on one occasion versus another.

One study examined a handful of weekly-level predictors of more negative evaluations (Merrill, Subbaraman, & Barnett, 2016). This study revealed that consequences were rated as more aversive on weeks characterized by (a) lower levels of alcohol use, (b) a greater number of negative consequences, (c) a smaller number of positive consequences, and (d) less positive evaluations of those positive consequences that were experienced. That is, variability in how consequences are rated from week-to-week in part depended on the context in which those consequences occur. However, context is comprised of much more than just the drinking behavior and its consequences.

As such, in a recent qualitative study, we sought to elucidate additional contextual-level correlates of subjective evaluations (citation masked for review). Using 13 focus groups, heavy drinking college students (age 18–20; n=62), were queried about the kinds of thoughts and feelings they have in response to alcohol consequences, as well as the influences on those thoughts and feelings. Consistent with survey studies, reactions to “negative” consequences of alcohol misuse were not uniformly labeled as negative by participants. Further, several contextual influences on reactions to consequences were revealed. Those influences with the strongest support included social factors (e.g., normative perceptions, immediate social context, discussions with friends); and additional qualitative themes highlighted a role for cognitive factors (e.g., attribution to alcohol), internal contextual factors (e.g., level of intoxication) and external contextual factors (e.g., concurrent positive consequences). Of note, these organizational categories are imposed by us, the research team, rather than representing labels provided by the participants themselves. These categories are further described below.

Social Factors

SLT posits a role of learning by observation. Perceptions of negative consequences as more common (i.e., descriptive norms; Baer, Stacy, & Larimer, 1991; Lee, Geisner, Patrick, & Neighbors, 2010; Perkins, Meilman, Leichliter, Cashin, & Presley, 1999; Read, Wood, Davidoff, McLacken, & Campbell, 2002; Wood, Read, Palfai, & Stevenson, 2001) or more acceptable (i.e., injunctive norms; Cialdini, Reno, & Kallgren, 1990) should be associated with less negative evaluation of one’s own consequences. Indeed, in our qualitative work, students suggested that both stronger descriptive norms and injunctive norms would result in less negative evaluations. Other social influences were also discussed in focus groups. For example, students acknowledged that who one is with during a consequence may influence his/her reaction; however, there were mixed reports regarding whether a consequence would be viewed less negatively if occurring among friends or among strangers. Further, we that learned students talk with their friends about their drinking experiences, which often serves to alleviate otherwise negative reactions. This alleviation seemed to occur particularly if consequences later became a funny story to tell, or if discussions resulted in more alcohol permissive perceived descriptive or injunctive norms. In the present study, we further examine all three of these - norms, immediate social context (friends vs strangers), and friend reactions - as potential social factors that can distinguish subjective evaluations of consequences.

Cognitive Influences

Students in our qualitative study reported that if they are able to attribute a consequence/behavior directly to alcohol intoxication (vs some other cause), this can lessen the negative evaluation of that consequence/behavior (citation masked for review). A second potential cognitive influence, with a key role in SLT, pertains to alcohol expectancies (beliefs about the likely outcomes of drinking behavior). If one enters a drinking event expecting negative consequences to occur, he/she may be less bothered by them, compared to consequences that are unexpected. In the present study, we examine both attributions and expectations as cognitive factors that may relate to subjective evaluations.

Internal Contextual Factors

First, (citation masked for review) found evidence that at higher levels of intoxication (notably, when consequences are more likely to occur), consequences are perceived less negatively. Second, SLT acknowledges the role of affect in its relation to cognition (Maisto et al., 1999). Positive mood during a drinking event may result in less negative evaluation of negative consequences, while negative mood may have the opposite effect. In the present study, we examine both level of intoxication and mood as internal contextual factors that may differentiate subjective evaluations.

External Contextual Factors

Both survey study findings (Merrill et al., 2016) and our prior qualitative study suggest that positive effects of drinking that occur alongside the negative consequences, or a more positive overall perception of a drinking event, may mitigate a negative reaction to those consequences. Moreover, it is plausible that the location in which a consequence takes place (e.g., public versus private location) may have implications for how it is subjectively experienced. As such, we test both the overall perception of the night (i.e., relative balance of positive and negative consequences of drinking) and location as potential external contextual factors that may be related to evaluations.

Time, Lasting Impact, and Prior Experience

Three additional potential factors are examined in the present study. First, our qualitative work indicated that the passage of time has implications for the valence of a consequence evaluation. The most negative reactions occur the morning after the event, when alone with one’s thoughts. However, these negative reactions tend to subside over time. Additionally, whether a consequence has a lasting impact, and whether the individual has prior experience with the consequence (i.e., experiencing it the first vs several times), may also be relevant.

In all then, in the present study we examined 12 potential correlates of subjective evaluations of consequences. We used quantitative methods to confirm the importance of these primarily contextual influences, by asking participants to give consequence ratings under different hypothetical conditions. We hypothesized that consequence evaluations would vary by context, mirroring the themes revealed in our qualitative work.

Method

Participants and Procedures

Participants were 214 undergraduate students (ages 18 to 25) enrolled in introductory psychology courses at a small liberal arts college in the northeast. Participants completed an anonymous online survey in return for academic credit, on which they reported how positively or negatively they would evaluate six alcohol-related consequences under different conditions. All procedures were approved by the University’s Institutional Review Board. Participant characteristics are presented in Table 1.

Table 1.

Sample descriptives

Characteristic	n/Mean	%/SD
Age	19.32	1.77
Gender
Male	51	24%
Female	162	76%
Genderqueer/Gender non-conforming	1	>1%
Racea
White	158	74%
Black or African-American	24	11%
Asian	11	5%
Native Hawaiian or other Pacific Islander	2	1%
Other	32	15%
Multiracial	45	21%
Ethnicity
Non-Hispanic	174	81%
Hispanic	40	19%
Class
Freshman	118	55%
Sophomores	47	22%
Upper classman	48	22%
Fraternity/Sorority membership
Yes	5	2%
Currently pledging	4	2%
No	205	96%
Current Residence
On-campus dormitory	57	27%
Non-dormitory university housing	3	1%
Off campus house or apartment	30	14%
With family	122	57%
Other	2	1%
Number of drinks per occasion (past 30 days)b
Average drinks	3.65	3.34
Peak drinks	7.21	12.30
Frequency of heavy episodic drinking (past 30 days)b	1.90	2.63
Number of negative alcohol-related consequencesb
Experienced in the last month	2.71	3.70
Experienced ever	8.50	5.48
Specific consequences (experienced ever)
Feeling sick/vomiting	118	55%
Saying/doing something embarrassing	124	58%
Blacking out	70	33%
Passing out	34	16%
Getting into a sexual situation that is later regretted	48	22%
Becoming very rude, obnoxious or insulting	69	32%

Note: n=214

^aParticipants checked all that apply;

^bCalculated only among drinkers (n= 147)

Measures

Demographics

Participants reported on gender, age, year in school, race/ethnicity, residential status, and Greek membership.

Contextual-level Evaluations of Consequences

Participants reported subjective evaluations for 6 hypothetical consequences (feeling sick/vomiting; saying/doing something embarrassing; blacking out; passing out; getting into a sexual situation that is later regretted; and becoming very rude, obnoxious or insulting). These six consequences were chosen from the valid and reliable measure, the Brief Young Adult Alcohol Consequences Questionnaire (BYAACQ; Kahler, Strong, & Read, 2005). We took care to select items that were not so severe that participants would be unlikely to have had experience with them to draw on in making hypothetical evaluation ratings, but that were not so mild as to result in little variability in evaluation ratings on the low end of the scale. We also chose items that seemed especially distinct from one another, that fell into both personal and social domains, and that did not represent symptoms that take time to develop (e.g., tolerance) but rather those that could occur/be assessed at a daily level.

For each of the 6 consequences, participants received the instruction set: “Imagine that you have been drinking alcohol. You [feel sick and you vomit]. For each of the conditions described below, please rate how you think you would evaluate the experience of [feeling sick and vomiting due to drinking].” They then provided a rating from 1=extremely positive to 7=extremely negative, for each of several different hypothetical contexts. Contexts fell into three social factor categories of (1) norms (if your friends told you they had done the same before, if your friends told you that had never happened to them), (2) immediate social context (in the company of friends, in the company of strangers), and (3) friends’ reactions (if later your friends thought it was a funny story, if later your friends expressed concern); two cognitive factor categories of (4) expectations (if earlier in the day you expected you might drink to the point of having that experience, if you did not expect it to happen that night), and (5) attribution (if you thought it was due to your drinking, if you thought it was due to something other than your drinking); two internal contextual factor categories of (6) mood (if you were in a good mood just prior, if you were in a bad mood just prior), and (7) level of intoxication (if you were very drunk, if you were just a little buzzed); two external contextual factor categories of (8) overall night (during an otherwise great night, during an otherwise not-so-great night), and (9) location (alone in your own room, at a party, at a bar); and three miscellaneous factors of (10) lasting impact (if this had some kind of lasting consequence, if there were no lasting consequences), (11) prior experience (for the first time, after having had this experience several times), and (12) timing (in the moment, looking back on it the next day, looking back on it a week later). For two consequences (blacking out, passing out) we did not assess “in the moment” or either level of intoxication contexts. In all, a total of 26 contexts were measured within these 12 domains. The contextual-level ratings were averaged across the 6 consequences for each person.

Alcohol use

Participants were asked to estimate the average number of standard drinks they consumed on a typical drinking day during the last month (average drinks), and to estimate the number of standard drinks consumed on the one day in the last month when they consumed the most alcohol (peak drinks). To assess heavy episodic drinking, participants responded to the item “During the last month, how many times have you consumed four (women)/five (men) or more standard drinks on one drinking occasion?” These three items were used for descriptive purposes. In addition, for subset analyses, if a non-zero was entered for either average or peak drinks, the participant was coded as a drinker, and participants who endorsed at least one heavy drinking episode were coded as heavy drinkers.

Alcohol consequences

The BYAACQ (Kahler et al., 2005) was used to assess whether each of a set of 24 alcohol consequences occurred either in the past month or ever. Items were dichotomously scored, to obtain a count of the number of negative consequences across each of the two time frames (past month, lifetime) for descriptive purposes. We also examined lifetime frequencies of the six consequences of interest (those for which we assessed contextual-level evaluations) in this study.

Data Analytic Plan

First, a repeated measures (within-subjects) ANOVA was run to provide an omnibus test of whether there were mean differences in evaluation ratings, averaged across consequences, across all 26 hypothetical contexts. Then, paired samples t-tests were used to make specific planned comparisons of evaluation ratings within each of the 12 categories (e.g., among friends vs strangers). Because this resulted in a total of 16 comparisons, we used a Bonferroni correction for Type I error, evaluating significance at p<.003. Analyses were first run in the entire sample. Sensitivity analyses were then conducted by running ANOVA and t-tests only among drinkers and then only among heavy drinkers.

Results

Complete participant descriptives are included in Table 1. Over the past 30 days, the subset of drinkers in our sample reported consuming an average of 3.65 drinks per occasion, engaging in heavy episodic drinking about twice, and experiencing close to 3 negative consequences. In the full sample, lifetime rates of the consequences of interest in this study ranged from 16% (passing out) to 58% (saying/doing something embarrassing), see Table 1. Among drinkers only, lifetime rates of consequences were as follows: 21% passing out, 29% regretted sexual situation, 42% became rude/obnoxious/insulting, 44% blacked out, 70% got sick/vomited, and 72% said/did something embarrassing.

Participants rated the consequences between 5 and 6 (on average) on a scale of 1=extremely positive to 7=extremely negative, across all contextual domains assessed, suggesting that these particular experiences are perceived negatively regardless of context. In the whole sample, the repeated measures ANOVA revealed a significant within-person effect (F [25, 5300] =50.61, p<.001), suggesting that participants reported varying mean consequence evaluation scores across the hypothetical contexts. All planned contextual level pairwise comparisons were significant (ps<.001) with the exception of whether the consequence occurred for the first time vs repeatedly (Table 2). Specifically, on average, consequences were viewed as more negative if a friend has never experienced the same consequence before vs has, if occurring among strangers vs friends, if friends express concern vs finding it funny, if the consequence was expected vs unexpected, if the person attributes the consequence to something other than alcohol vs to alcohol, if in a bad vs good mood just prior, if one was just a little buzzed vs very drunk, if it was an otherwise not-so-great vs great night, at either a party or bar vs alone in one’s home/dorm, at a bar vs party, if there are lasting consequences vs no lasting consequences, when looking back on it the next day or the next week vs in the moment, and looking back on it the next day vs the next week. Findings were the same when limiting the sample to either just drinkers or just heavy drinkers.

Table 2.

Contextual-level comparisons of subjective evaluations of consequences

	Mean	SD	Comparison	t (df=213)	p
Social Factors
Norms
Friend has	5.45	1.24	Friend has v hasn’t	−7.47	<.001
Friend hasn’t	5.91	1.05
Immediate Social Context
Friends	5.32	1.29	Friends v Strangers	−12.97	<.001
Strangers	6.25	0.92
Friend Reaction
Funny	5.51	1.30	Funny v Concerned	−4.60	<.001
Concerned	5.87	1.19
Cognitive Factors
Expectation
Expected	5.64	1.25	Expected v Unexpected	−5.64	<.001
Unexpected	5.98	1.13
Attribution
Due to alcohol	5.63	1.33	Due to alcohol v other	−3.57	<.001
Due to other	5.88	1.08
Internal Contextual Factors
Mood
Good mood	5.70	1.15	Good v Bad mood	−4.02	<.001
Bad mood	5.88	1.13
Intoxication level
Very drunk	5.31	1.44	Very drunk v buzzed	−6.42	<.001
Only buzzed	5.74	1.15
External Contextual Factors
Overall night
Great night	5.62	1.25	Great v Not-so-great night	−5.60	<.001
Not-so-great night	5.93	1.10
Location
Own room	4.94	1.39	Alone v Party	−13.27	<.001
Party	6.03	1.06	Alone v Bar	−14.90	<.001
Bar	6.25	.95	Party v Bar	−6.48	<.001
Time, Lasting Impact, and Prior Experience
In moment	5.42	1.32	In moment v Next day	−7.58	<.001
Next day	5.86	1.12	In moment v Next week	−4.01	<.001
Next week	5.71	1.13	Next day v Next week	4.16	<.001
Lasting cons
Lasting cons	6.10	1.08	Lasting v No lasting cons	6.89	<.001
No lasting cons	5.81	1.03
Prior experience
First time	5.79	1.23	First time v Repeated	−.54	.589
Repeated	5.82	1.13

Discussion

The primary goal of the present study was to examine whether, among college students, hypothetical evaluations of an alcohol-related consequence may be related to the context in which it occurs. Several themes derived from recent qualitative findings on the contextual influences on consequence evaluations were further supported using survey methodology. Here, we found more evidence that the way in which consequences are subjectively evaluated is quite nuanced, and does indeed differ as a function of social, cognitive and other contextual factors. That consequences may be evaluated differently depending on “stimulus conditions” or contextual factors findings are in line with the SLT principle of differential reinforcement. Each specific finding is described and placed in the context of prior research and theory (especially SLT) below.

Social Factors

Themes from our prior qualitative work regarding the impact of the social context on consequence evaluations were confirmed in this study. Specifically, consequences were viewed as more negative if occurring among strangers vs friends, if a friend has never experienced the same consequence before vs has, and if friends express concern vs finding it funny. The first of these speaks to the importance of the immediate social context. That is, students may reflect on who has observed the consequences of their drinking behavior in real-time when deciding whether or not those consequences are aversive. The latter two contextual differences speak to the importance of social norms. Descriptive norms refer to how common a behavior is perceived to be among one’s peers, while injunctive norms refer to how acceptable a behavior is perceived to be among one’s peers (Cialdini, Kallgren, & Reno, 1991). Our findings show that more permissive norms of either type may result in evaluating one’s own alcohol consequences less negatively, in turn perhaps reducing motivation to change behavior. That the ways in which friends react to one’s consequence is important also surfaced in focus group findings regarding the ways in which students discuss their consequences with friends, and how these discussions have strong implications for changing perceptions of consequences over time. Our findings in the social domain are in line with SLT’s emphasis on learning from the social environment – interactions with and observations of friends can in some cases reinforce drinking behavior, and this may occur via an impact on one’s subjective evaluation of consequences.

Cognitive Factors

As noted earlier, SLT also emphasizes cognitive factors on the learning of drinking behavior. While subjective evaluations themselves are a cognitive variable, they may be expected to vary along other cognitive dimensions, such as attributions and expectancies. Consequences were reported to be more negative if they could not be attributed directly to alcohol intoxication. When reflecting back on a consequence, if one is able to attribute uncharacteristic behavior to having been drunk, one may feel less badly about it than if the consequence occurred but could not be blamed on alcohol. We also found that if a student expected that he/she might experience a consequence due to drinking on that occasion, it would be perceived less negatively than if it occurred unexpectedly. This finding not only speaks to the key role that expectancies play in SLT, but can be placed in the context of other theories as well. For example, regarding the social reaction pathway of the Prototype Willingness Model (Gibbons, Gerrard, & Lane, 2003), this finding might suggest that a link between willingness to experience a consequence and later behavior could be mediated by one’s perception of that consequence (e.g., more willingness leads to less negative evaluation which in turn leads to more drinking).

Internal Contextual Factors

With respect to internal contextual factors, first, consequences were rated more negatively assuming one was already in a bad mood versus a good mood when they occurred. Not only does SLT acknowledge the role of affect in its relation to cognition (Maisto et al., 1999), that one’s mood prior to experiencing a consequence might color how positively or negatively it is perceived is also consistent with alcohol myopia theory (Steele & Josephs, 1990). Second, consequences were viewed as more negative if they occurred when one was just a little ‘buzzed’ versus very drunk. In the moment, higher levels of intoxication may interfere with one’s ability to accurately judge the magnitude of a consequence. However, given that the evaluation ratings in the present study were provided when participants were sober, it is also possible that, reflecting back, consequences that occurred at lower levels of intoxication are more unexpected and therefore troubling.

External Contextual Factors

We found evidence for both posited external contextual factors. First, if it was an otherwise not-so-great night, a consequence that occurs would be viewed more negatively than within the context of an otherwise great night. This particular finding was not only consistent with our focus group work, but also with Merrill et al. (2016), a study which demonstrated that weeks with more positive consequences were associated with less negative evaluations of the negative consequences reported. That is, the benefits of drinking may overshadow the drawbacks for some individuals on some occasions. Second, location was important; consequences were evaluated most negatively if they occurred at a bar, followed by a party, and least negatively if occurring in one’s own home/dorm room. These findings may in part reflect differences in drinking behavior and perceived norms by location. Prior research shows that drinking, social norms for drinking, and discrepancies between personal and perceived drinking may differ by location (Lewis et al., 2011).

Time, Lasting Impact, and Prior Experience

Two other contextual-level differences on evaluations were identified in this survey study. For example, if a consequence came with other, lasting consequences, it was rated as more negatively. Lasting consequences speak to the more objective severity of an event (e.g., getting a sexually transmitted infection following a regretted sexual experience is objectively worse than not getting an infection), and may make students more likely to re-evaluate their drinking behaviors. In the context of learning theories, such consequences are likely to be more punishing, in turn more likely to prompt behavior change. We also confirmed a role for the timing of a consequence evaluation on its valence. In particular, consequences are viewed least negatively in the moment (perhaps due to acute intoxication), and most negatively the morning after they occur, with those negative reactions reducing in magnitude a week later. Though this is not a contextual influence, per se, it is still of interest. Drinkers may wake up and initially regret or otherwise feel badly about the prior evening’s events, yet as these become less salient in memory with time, so too may the associated negative thoughts or feelings. Further, with time, students may discuss their consequences with peers, who oftentimes may minimize or normalize those consequences, in turn mitigating personal negative evaluations.

With respect to prior experience, the lack of difference between evaluations of a consequence that occurred the first time versus repeatedly was not entirely surprising, based on our prior qualitative work. In focus groups, we learned that consequences may be more negatively evaluated at both ends of this spectrum of experience. Students reported that they may find a consequence (e.g., blackout, hangover) aversive the first time it happens, but habituate to it upon repeated experience. However, students also indicated that experiencing a consequence several times may be a warning sign that drinking has progressed to an unhealthy or unacceptable level, resulting in aversion once again. In other words, there may be a U-shaped association between how many times a consequence has been experienced and how negatively it is perceived. Moreover, individuals appear to differ in their perceptions of which scenario is more negative. Indeed, closer examination of our data revealed that 39% of participants reported that experiencing a consequence for the first time would be relatively worse, with a near equal percentage (40%) reporting that experiencing a consequence several times would be worse. That participants were rating these two contextual comparisons in opposite ways suggests that this relationship is moderated by unmeasured variable(s), which would explain the lack of a main effect.

Limitations

This study is not without limitations. First, this study was conducted on a convenience sample of college student volunteers. The majority of these were female, and our sample may be characterized by lower levels of alcohol consequences than others (e.g., Brett, Leffingwell, & Leavens, 2017; Kahler, Hustad, Barnett, Strong, & Borsari, 2008), which limits the generalizability of findings. Second, participants were hypothetically evaluating consequences of drinking, so findings may not extend to actual evaluations that could be collected using real-time data collection methods such as ecological momentary assessment. Further, while some participants may have had experience to draw upon with certain consequences in certain contexts, others may not have. As such, ratings for experienced drinkers may have been closer to those they might actually report in real time. Of note, findings were replicated in sensitivity analyses, when limiting the sample to only drinkers as well as only heavy drinkers. Importantly, each contextual characteristic does not occur in a vacuum; several of the hypothetical contexts assessed in this study could be at play simultaneously. It is unclear how these different contextual characteristics may interact to influence one’s subjective evaluation, or whether some aspects of the broader context in which a consequence occurs are more influential than others. As such, future studies should examine actual evaluations of recently experienced consequences, along with multiple aspects of the context, to better understand the complexities in the role of context on subjective evaluations.

Conclusions

Despite these limitations, our findings support our previous qualitative analysis, extend prior literature and have some clinical implications. The subjective experience of consequences appears to be much nuanced, and it should not be assumed that any consequence is perceived negatively by a student on all occasions. When using recent consequences in feedback-based interventions to build motivation to change, it may be useful to discuss the context of a consequence and how this influences whether or not a student perceives that consequence as aversive. Understanding the full context in which a consequence occurred may also help inform clinicians as to whether negative evaluations, and therefore motivation to change, may occur.