Alcohol-related aggression: a further study of the importance of frustration

“Oh, he was just drunk” – young adults' views on alcohol as an excuse for violent behaviours

Aim

This article examines the excuse-value of alcohol in a hypothetical unprovoked male-to-male violence scenario. The excuse-value is conceptualised by two main questions. Does intoxication result in decreased blame? Does intoxication lessen the propensity to call the police?

Method

In the study, 1004 RDD quantitative telephone surveys were completed with Swedes aged 16–25. The response-rate was 73.8%. The aggressor's intoxication, the victim's intoxication, the severity of the outcome and the aggressor's state of mind were all randomly manipulated. The factors ‘should the police be called’ and the ‘attribution of blame to the aggressor’ have been analysed using factorial ANOVA in SAS.

Results

Analyses were stratified by sex. For male respondents, the aggressor's intoxication interacted with the victim's intoxication in predicting the attribution of blame. When the victim was drunk, the blame attributed to the aggressor decreased as a result of a higher level of intoxication; whilst there was no effect of the aggressor's intoxication when the victim was only slightly drunk. For female respondents, the aggressor's intoxication interacted with the severity of the outcome. Intoxication decreased the blame with the severe acts but not with the less severe acts. The aggressor's intoxication had no effect on whether the police should be called or not.

Conclusions

Alcohol intoxication provides some excuse-value for violence, but only for certain acts and under certain circumstances.

THE EFFECTS OF ALCOHOL INTAKE AND INDUCED FRUSTRATION ON THE DISPOSITION TO START FIRES

The object of the present study was to see whether there are tendencies towards reactive arson, measured as a relation between frustration, intoxication and a disposition to start fires. One hundred individuals, 50 men and 50 women, participated. All were students and they were randomly assigned to each of five experimental groups, namely a Control Group, a Placebo Group, a Frustration Group, an Alcohol Group and an Alcohol + Frustration Group. A projective test was devised which required the participant to complete a short story where the content involved a person who had suffered humiliation and persecution and who was fingering a box of matches in his/her pocket. Results indicated that frustration increases the probability of starting fires, but, in this study, alcohol did not increase the degree of relation to fire. Results also suggest that there are strong cognitive barriers in a population of university students against fantasizing about starting fires.

Alcohol and social influences: yielding among male social drinkers to social pressure

Thirty men participated as paid volunteers in an experiment exploring whether alcohol intoxication makes a person more or less likely to yield to social pressure. Subjects were randomly assigned to either a Control, a Placebo, or an Alcohol group. The alcohol dose was 0.8 ml of 100% alcohol/kg body weight. Subjects first estimated the length of a line, then received feedback from either a male or a female reference group (social pressure), and finally made a second estimation. This procedure was repeated 30 times. The analysis indicated that the only effect of intoxication was to become more resistant in terms of number of yieldings to social pressure, provided the feedback came from a male reference group. Locus of control in terms of internality significantly predicted number of yieldings for the Alcohol group, although the analysis indicated no significant differences between groups as to mean locus of control scores.

Aggressive and nonaggressive behavior as a function of alcohol intoxication and frustration in women

An experiment tested whether alcohol increases aggression in women in a situation in which both an aggressive and a nonaggressive response alternative of equal instrumentality and of equal variability are available. Subjects were assigned to one of three groups, namely, an alcohol, a placebo, or a control group. The alcohol dose was 1.0 ml of pure alcohol/kg body weight. After drinking their respective drink, subjects were instructed to supervise a bogus partner on a visual scan test over a series of trials. Each time this partner made a mistake, subjects could either give an uncomfortable electric shock (scale 1 to 10) or a comfortable vibration (scale 1 to 10) to the partner. Aggressive and nonaggressive behavior was measured as numbers, intensities, and durations of shocks or vibrations, respectively. Neither alcohol nor frustration differentiated the groups on aggressive or nonaggressive behavior. All groups were significantly more inclined to use the nonaggressive alternative irrespective of alcohol dose and level of frustration. In conclusion it was stated that women do not increase their aggression as a function of alcohol in a situation with more than one response alternative available. The need to incorporate gender differences as to aggressive effects of alcohol was stressed.

Effect of alcohol on quantity of creative production using the Purdue tests

An experiment was performed to test whether a moderate dose of alcohol (1 ml of 100% alcohol/kg body weight) facilitates the production of creative solutions as measured with a standard creativity test and whether a creative attitude interacts with alcohol in this respect. The 60 subjects were randomly assigned to either an alcohol, a placebo, or a control group, with 10 women and 10 men in each group. Two different personality tests measuring the creative attitude were administered a priori. Analysis indicated that both the placebo and alcohol groups produced significantly less creative solutions to the ten stimulus pictures. The a priori attitude scores significantly predicted the creativity scores for the placebo and alcohol subjects but not for the control subjects. Results were discussed in terms of disinhibition and cognitive changes under alcohol.

Male physical aggression as a function of alcohol intoxication and frustration: experimental results and methodological considerations

Forty-five undergraduate students were assigned to either an Alcohol, a Placebo, or a Control group. The alcohol dose was 0.80 g of 100% alcohol/kg body weight. Subjects were informed that they could win a sum of money depending on the performance of a partner. They then supervised the partner over a series of trials on a visual scan test and could influence the partner by either giving an uncomfortable electric shock (aggressive alternative) or a comfortable vibration (nonaggressive alternative) at each incorrect response from the partner. Both alternatives were said to be equally instrumental in reaching the goal of winning the money and both could be varied in intensity on a 10-point scale and without limits in terms of duration. Aggression was measured as number of aggressive responses chosen, and in terms of intensity and duration. Nonaggression was measured in terms of intensity and duration. Intoxicated subjects did not increase their aggression but all groups chose significantly more nonaggressive responses and did so with higher intensity and duration. Frustration did not significantly affect these types of responding. Results are discussed in terms of methodological considerations and the importance of using realistic experimental paradigms is stressed. Also, theoretical implications are discussed.

Male physical aggression as a function of alcohol, frustration, and subjective mood

An experiment tested the hypothesis that alcohol intoxication will increase aggression only if the subject is experiencing the intoxicated state as displeasing. Forty males drank either 0.8 mL of pure alcohol/kg body weight or a placebo drink and were then exposed to either a pleasant or an unpleasant mood manipulation. Following this, they were given the chance to aggress in a modified version of the Buss "aggression machine" paradigm. All subjects were observed under varying levels of frustration. Results indicated that alcohol-drinking subjects were more aggressive than placebo-drinking subjects, and that both intoxicated and sober subjects increased their aggression when frustrated. Intoxicated subjects in a pleasant mood were most aggressive under both provocative and nonprovocative conditions. The hypothesis was not supported and results are discussed in relation to different theoretical models and in relation to results from other similar studies.

Alcohol and the compensation hypothesis: a test with cognitive and psychomotor tasks

An experiment tested the hypothesis that moderately intoxicated subjects can compensate for the detrimental effects of alcohol. Subjects were assigned either to a Control group, a Low Incentive group, or a High Incentive group. The dose of alcohol was 1.0 ml of 100% alcohol/kg body weight. The rationale of the experiment was that the High Incentive group would be more inclined to try to compensate than would the Low Incentive group. A psychomotor and two different cognitive tasks were used. Analysis indicated that the psychomotor test was not at all affected by alcohol, so there was no possibility for compensation. On the Color Word Test, intoxicated subjects developed the strategy of taking longer time to complete the test and so keeping the number of errors low. On the Dot Test, men were not affected by alcohol and could not compensate while women were affected and successfully compensated for the detrimental effects of alcohol. The results were interpreted as a partial support of the hypothesis of compensation for alcohol effect.

Wine and Male Physical Aggression

Twenty male undergraduates were randomly assigned either to a wine-drinking group or a control group. The alcohol dose was 0.8 ml. of pure alcohol/kg. body weight. They were tested for physical aggression in a modified version of the Buss' “aggression machine.” All subjects were tested under frustrating and non-frustrating conditions. Alcohol did not increase aggression under any of these conditions although the paradigm produces increased aggression when vodka is used as an intoxicant. The results were discussed in terms of differential expectancies tied to different types of beverages.

Changes in the psychological defence system as a function of alcohol intoxication in men

Social drinking young men participated in an experiment exploring whether acute alcohol intoxication affects the characteristics of the psychological defence structure. It was found that a moderate alcohol dose (1.0 ml of 100% alcohol/kg body weight) increased the overall level of defence activity compared to a sober Control group and as measured by the Defence Mechanism Test. This increase was mainly due to a significant increase in the defence mechanism Isolation. Gender differences in this respect and the implications of this shift in the defence structure for the understanding of intoxicated behavior and motives for drinking were discussed.

The structure of psychological defense mechanisms in women as a function of alcohol intoxication

The present experiment tested whether moderate alcohol intoxication influences the relative strength of different psychoanalytic defense mechanisms and whether the overall level of defense activity is changed. Thirty-six women were randomly assigned to either an alcohol, a placebo, or a control group. The alcohol dose was 1.0 ml of 100% alcohol/kg body weight. Subjects were tested with the Defense Mechanism Test (DMT), which gives a valid measure of both elaborate or late and primitive or early defense mechanisms. Results indicated no significant effects of alcohol intoxication on any aspect of the defense system. This was taken as supportive of the psychoanalytic notion in respect of the stability of the defense system. Possible complications were discussed.

Alcohol effects on cognitive and personality style in women with special reference to primary and secondary process

Changes in primary and secondary process functioning as well as in experienced locus of control as indicators of cognitive style were examined as a function of alcohol intoxication. Women subjects drank 1.0 ml of pure alcohol/kg body weight and were compared to a placebo and a control group. Results indicated a significant shift in locus of control toward greater externality and a significant shift in cognitive style from a dominance of secondary process when sober to a dominance of primary process when intoxicated. These shifts were discussed as indicative of the pharmacological properties of alcohol and were related to the disinhibition and the arousal hypotheses. Alternative explanations were also discussed.

Human physical aggression as a function of magnitude of frustration: indirect support and a possible confounding influence

An experiment was performed to test whether thwarting close to the goal leads to higher levels of frustration and higher levels of aggression than thwarting far from the goal in a slightly modified version of the Buss' "aggression machine." Subjects first calibrated a shock scale and were then put in a situation in which they could give electric shocks to a confederate under an acceptable pretext. It was not possible to confirm the specific hypothesis directly but a positive correlation was found between over-all self-rated subjective frustration and a measure of total aggression supporting a general frustration-aggression hypothesis. However, it was also found that the initial calibration of the shock scale correlated positively with the measure of total aggression, indicating the possible confounding influences are operative in this kind of paradigm.

The blood alcohol curve as a function of time and type of beverage: methodological considerations

Six male subjects came to the laboratory on three different occasions and drank either spirits, wine, or beer. The spirits drink contained 1.0 ml of pure alcohol/kg body wt. and the other two drinks were 20% larger. The blood alcohol curve was monitored every 15 min. The mean peak level was roughly the same for the three types of beverages although the peak was reached later for wine and beer. For spirits and beer the curve dropped off smoothly after the peak was reached while for wine the peak level was maintained for about 60 min. Individual curves varied greatly for all three beverage types. The results are discussed in terms of methodological complications when doing experimental research in the laboratory.

Beer intoxication and physical aggression in males

Twenty subjects participated in an experiment testing the effects of a moderate dose of beer on physical aggression. Subjects were randomly assigned to a beer drinking group or a control group. Aggression was measured in terms of number of shocks given, shock intensity, and shock duration in a modified version of the Buss' aggression machine. There were no differences among groups and there was no interaction of beer by frustration on aggression. The results are explained in terms of expectancies held by subjects as to the effects of beer intoxication on behaviour.

Alcohol and aggression: a test of an indirect measure of aggression

28 male undergraduates were randomly divided into an Alcohol group (0.8 ml/kg) or a Control group. After drinking they were exposed to a violent movie and then they wrote TAT-stories and completed semantic differential scales. Only the indirect semantic scales differentiated the groups: intoxicated subjects accepted violence more and with less negative feeling.

Human physical aggression as a function of frustration: role of aggressive cues

An experiment was performed testing whether aggressive cues are necessary or only facilitative in increasing aggression to a frustration and whether their role is to “pull out” aggression directly or to add to the experience of displeasure. 20 subjects participated and a modified version of the Buss' “aggression machine” was used in which frustration was manipulated within subjects and aggressive cues between subjects. Frustration was of an arbitrary kind and aggression was defined to subjects to have instrumental value in overcoming the frustrative event Results indicated that (1) frustration alone is a weak antecedent of aggression, (2) at low frustration aggressive cues seem to be necessary for aggression to increase, and (3) aggressive cues apparently elicit aggression directly. Results were discussed in terms of Berkowitz' reformulation of the frustration-aggression hypothesis.

Alcohol, frustration, and aggression: an experiment using the balanced placebo design

In a balanced placebo experiment it was hypothesized that pharmacological effects of alcohol would override expectancy effects and that alcohol would increase aggression only under frustrative conditions by a process of an exaggerated subjective experience of frustration. In a modified “aggression-machine” frustration was defined as arbitrary and manipulated in a within-subjects fashion. Manipulation checks indicated difficulties in deceiving subjects as to the content of the drink, and it was not possible to evaluate pure pharmacological effects. Ratings of subjective frustration did not differentiate experimental groups. Only during frustrative conditions did intoxicated subjects increase their aggression regardless of information about drink's content. Frustration per se and expectancy factors did not lead to increased aggression. This was explained in terms of a more narrow focusing on outstanding available response alternatives and as opposed to the 1983 Taylor and Leonard explanation of alcohol-increased aggression.

Threat as a determinant of alcohol-related aggression

An experimental study was performed within a modified version of the Buss “aggression-machine” paradigm testing the hypothesis that threat will inhibit aggression in intoxicated subjects given that threat is not unavoidable. 40 males were randomly assigned to either an alcohol group, drinking .08 ml of pure alcohol per kg body weight, or a placebo group, drinking only tonic. Half of the subjects in each group were exposed to a threat and the other half to no threat. At 10 opportunities subjects could choose between a nonaggressive response or an aggressive response in the form of an electric shock given to a bogus partner. Aggression was estimated in terms of number of aggressive responses, the intensity and duration of these responses, and response latency. Results clearly confirmed the hypothesis and were discussed theoretically in a model combining pharmacological and psychological factors.

A possible confounding variable in different versions of the "aggression machine" when used in research on alcohol

When various versions of the Buss aggression paradigm are used in research concerning alcohol-related aggression, different information about the confederate's degree of intoxication is given to the subject. Sober subjects get the impression that the confederate is sober, whereas intoxicated subjects believe that the confederate is intoxicated. This constitutes a confounding variable making it impossible to attribute conclusively differences among alcohol and control groups on measures of aggression to consumed alcohol. An experiment tested whether this difference in information affects subjects' conception concerning expected level of aggression directed towards them from an opponent and their own level of aggression towards the opponent within the Taylor version of the “aggression machine”. Results clearly indicate that subjects expect intoxicated opponents to be more aggressive. Subjects are also more aggressive toward intoxicated opponents than sober ones. This confirms the importance of holding constant information about the confederate; this has not been done in many relevant studies.

Alcohol and aggression: pharmacological versus expectancy effects

An expectancy hypothesis has been put forward to explain the relationship between alcohol intoxication and aggressive behavior. An experiment tested this hypothesis using a balanced placebo design. The experiment was performed within a modified version of the Buss “shock-machine” procedure. Subjects were randomly assigned to either an alcohol-drinking group or a tonic-drinking group. Each group was further subdivided into one informed that the drink contained alcohol and one informed that the drink contained only tonic. Subjects' expectancies concerning the effects of alcohol on aggression were actively manipulated. Aggressive behavior was measured both in an absolute and a relative way. Given difficulties with the procedural deception about the drink, it was not possible to evaluate pure pharmacological effects. Results did not support an expectancy interpretation in that the two alcohol-drinking groups were more aggressive than the two tonic-drinking groups. Results were discussed in terms of methodology explaining the difference in results between the present study and a similar study by Lang, et al. in 1975.

Alcohol and aggression: a validation study of the Taylor aggression paradigm

An experimental study was performed testing the validity of the Taylor aggression paradigm when used to study the effects of alcohol on aggression. The main hypothesis stated that the Taylor paradigm does not adequately control for differences in pain/discomfort thresholds when subjects are repeatedly shocked. 36 male subjects were randomly assigned to an alcohol or a placebo condition. They were first required to adjust an electric shock to a subjective criterion (“definitely unpleasant”). This shock setting was then used to calibrate a shock-intensity scale. Subjects were given 12 shocks of varying intensities from this scale. No shock was above the subjective criterion. After the 12 shocks subjects again adjusted an electric shock to the same subjective criterion. Each shock and subjective criterion were rated by subjects for experienced pain and experienced total discomfort. Validity of the Taylor paradigm requires that pain/discomfort thresholds remain stable throughout a period of experimental observation. The results indicate that intoxicated subjects chose a higher initial subjective criterion and experienced the subsequent shocks more intensely. Intoxicated subjects in the Taylor paradigm are more provoked than sober subjects and reported increases in aggression scores for intoxicated subjects can not conclusively be attributed to alcohol alone.