Summary and Comments – A Dose of Ruthlessness: Interpersonal Moral Judgment Is Hardened by the Anti-Anxiety Drug Lorazepam

A Dose of Ruthlessness: Interpersonal Moral Judgment Is Hardened by the Anti-Anxiety Drug Lorazepam

by Perkins, A., Leonard, A., Weaver, K., Dalton, J., Mehta, M., Kumari, V., Williams, S. & Ettinger, U. (2013).

(Journal of Experimental Psychology: General, 142(3), 612-620. DOI: 10.1037/a0030256)


As unfashionable as it may be in many circles, I am a believer in free will. I have to be, for working as a psychologist or conducting psychological research to make any sense to me, personally. At the same time, I am keenly aware of how great an impact internal and external circumstances have on our decision-making capabilities and choices. As such, the effects of emotional states and responses on moral decisions are a fascinating topic, and that’s why I chose to present this article on a rather simple study on the effects of the anxiolytic drug lorazepam on ruthlessness in moral decisions.

Quite a bit of research exists on affective responses to moral dilemmas. For one, it has been demonstrated that when people are presented with moral dilemmas, areas of the brain related to emotional responses are more active when such a dilemma involves directly harming an innocent person (e.g., personally killing someone to save others, a moral-personal dilemma) than when the harm is remotely/indirectly inflicted (e.g., switching a lever that will result in one person dying, as many others are saved, a moral-impersonal dilemma). It appears that the prospect of directly harming someone, even for a good cause, arouses a stronger reflexive emotional resistance that one needs to overcome cognitively. Second, research has found that greater positive affect may in fact increase people’s tendency to endorse harmful acts in such moral-personal dilemmas, even as negative emotion may have an inhibitory role. In other words, it seems a negative emotional response, something akin to fear, concern or anxiety, tends to inhibit the endorsement of directly harmful acts in moral-personal dilemmas.

It was this possibility, namely that negative emotion, here conceptualized specifically as anxiety, may interfere with endorsing directly harmful acts in personal dilemmas, that Perkins et al. sought to explore in this study.  They did this by comparing the effects of a placebo (ascorbic acid) with 1 mg and 2 mg of lorazepam in 40 healthy participants exposed to moral-personal, moral-impersonal and nonmoral dilemmas. In each case, the participants had to respond “yes” or “no” to judge whether a suggested action was the morally correct choice in each dilemma.

Lorazepam is a benzodiazepine drug for which reasonably good evidence exists for (among other effects) a specific effect of “dampen[ing] negative emotion elicited by aversive situations”, an effect here considered as decreasing anxiety. If it were the case that anxiety keeps people from endorsing directly harmful acts, then administering such an anxiety reducing drug should increase the likelihood of participants endorsing directly harmful acts in moral-personal dilemmas, but should not affect choices in other types of dilemmas. To determine if lorazepam would be increase the likelihood of engaging in directly harmful acts (here dubbed “ruthlessness”) in general, or only in cases where there was a benefit “for the greater good”, moral-personal dilemmas were further divided into low-conflict ones, where the harmful acts were primarily selfish in motivation (e.g., done for convenience), and high-conflict, where there was a a clear utilitarian benefit to the act (e.g.saving the lives of others).


As expected, the authors found very large differences in endorsing harmful acts between moral-impersonal, moral-personal and nonmoral dilemmas. In other words, the content of such dilemmas did influence the participants’ responses. They were more ready to endorse harmful acts, if the acts were more indirect and impersonal in nature.

More interestingly, as hypothesized, lorazepam had differing effects on response choices depending on this content. Administering lorazepam 2 h prior to exposure to moral dilemmas significantly increased ruthlessness (endorsing harmful acts) in moral-personal dilemmas, but not in moral-impersonal or nonmoral dilemmas. Further, the effect had a clear dose-response effect in that 2 mg of lorazepam resulted in significantly more increase in ruthlessness than 1 mg. There was no significant difference between low-conflict and high-conflict moral-personal dilemmas – lorazepam had an equal effect regardless of the motive of the harmful act.

Due to its sedative effects, lorazepam also increased response times in all situations, but this was found not to be related to the increase in ruthlessness. Gender and personality type also had no moderating effect on lorazepam’s effects.

Discussion & Commentary

In sum, “the effect of lorazepam in moral-personal dilemmas was to cause a dose-dependent increase in participants’ preference in moral-personal dilemmas for responses that directly harm other humans.” These results concur with previous findings from neuroimaging studies that found greater engagement of emotion processing areas of the brain for moral-personal dilemmas, and extend these findings by suggesting that the specific emotion such dilemmas elicit is anxiety (or some emotion closely related to it). As the presented motivation for the harmful act made no difference to lorazepam’s effects, the authors conclude lorazepam “makes us more willing to harm people directly whether or not that harm is for the greater good.”

What do these findings mean for users or prescribers of lorazepam (or other similar anxiolytics)? Probably not that much. Though in statistical terms the effect sizes are quite significant, even large for placebo vs. 2 mg of lorazepam (η² = 0.232), in practical terms this meant an average of 2.33 ruthless decisions in 18 moral-personal dilemmas for the 2 mg group versus 1.75 ruthless decisions for the placebo group, i.e. acting ruthlessly 12.9 % of the time instead of 9.7 % of the time.

This is a noticeable effect, but for someone suffering from a great deal of anxiety, such an increase in possible ruthlessness probably doesn’t mean much in terms of whether taking lorazepam would be a good idea or not. Many other stronger reasons exist for carefully considering especially long-term use of the drug. In any case, increased ruthlessness, as measured here, can be considered a positive or a negative effect from a moral perspective depending on the situation and its baseline level. It should also be noted that this research involved acute dosing, a single-dose of lorazepam in drug-naive participants, and the effects of long-term chronic benzodiazepine use on anxiety may be wildly different.

The findings have some relevance to psychopathy research. The authors suggest that while the root of psychopathy is a cognitive style of not placing value on the fair treatment of others, such a style will be more freely expressed in interpersonal situations when there is little or no anxiety, as is the case with many (though not all) psychopaths. Interestingly, and rather controversially, the authors bring up the possibility of treating psychopaths with anxiogenic interventions, chemical or therapeutic, as a possible way of muting socially harmful behavior. Would psychopaths who ordinarily feel very little emotional response to a moral dilemma, benefit from a dose of anxiety at such times? That may be, but the ethical problems in any “treatment” that would increase anxiety are obviously enormous.

As a side note, as lorazepam has now been found to reduce the avoidance of direct physical harm to both self and other people, the authors pose an interesting question: “Why should the prospect of causing direct harm to others have evolved to be perceived in a way similar to a threat to oneself?” Several possible explanations stemming from evolutionary psychology spring to mind. Whatever the case, such empathic responses are clearly very relevant to moral decision-making.