Although Ethics Unwrapped features videos, case studies, teaching notes and much more on many ethics-related topics, behavioral ethics—the science of moral decision making—gets most of our attention. Frequently, we focus on the question: “Why do good people often do bad things?” As with everything related to the brain, there is neither a simple answer nor a single answer to this question. Many complicated factors influence human moral decision making.

Let’s say that last night Sam (who is married) had sex with a co-worker, thereby committing adultery. Why might Sam have done such a thing? Well, sex is obviously a powerful motivator. And because of the self-serving bias, most humans are wired to advance their own interests. Of course, Sam also likely wants to be a good person. And it is in his self-interest to be thought of as a good person and to think of himself as a good person. This is manageable.

How is it manageable? Well, before the encounter with his co-worker, the overconfidence bias likely caused Sam to think of himself as a good person, and this will not necessarily change because of a single case of adultery. There are too many rationalizations available to Sam that can enable him to separate his bad act from his character. He may resort to denial of responsibility, saying to himself: “Yes, I strayed, but only because I was seduced. I am still a good person.”

Or, before the indiscretion, Sam probably thought of adultery as very morally problematic, but afterward, maybe not so much. As David Luban has pointed out: “In situation after situation, literally hundreds of experiments reveal that when our conduct clashes with our prior beliefs, our beliefs swing into conformity with our conduct, without our noticing that this is going on.”

How does this work exactly? Well, one possibility is explored in an article by the McCombs School of Business’s own Dr. Adrian Ward and three co-authors in Making Molehills Out of Mountains: Removing Moral Meaning from Prior Immoral Actions—forthcoming in the Journal of Behavioral Decision Making. This research attempts to draw explanatory power from construal level theory (CLT). CLT holds that psychological distance of various types (temporal, spatial, social, or hypothetical) affects people’s mental representations of events and objects which, in turn, can affect their decision-making and judgments.

Ward and colleagues take CLT into the moral decision-making realm, noting that “[d]espite knowledge of their moral failings, individuals are often able to maintain the belief that they are moral persons,” and they suggest that one mechanism behind this capability is “the tendency to represent one’s past immoral behaviors in concrete or mechanistic terms, thus stripping the action of its moral implications.”

In other words, the authors predict that “individuals will tend to frame their potentially immoral behaviors in terms of their lower-level mechanics [e.g., “I wrote a lower number in the income field on my tax return”] rather than their higher-level meaning [e.g., “I behaved dishonestly”] and, in doing so, will effectively strip those actions of their moral meaning.”

If we understand the authors correctly, they argue that when we think of adultery by others, we are likely to construe it on a higher-level,  in a way that focuses on the implications of doing the act and the relationship between an action and the actor’s goals and values (e.g., “They cheated on their spouses”). But when we think of adultery involving ourselves, especially when we have actually committed adultery, we are more likely to construe our behavior concretely and contextually, removing it from the moral domain (e.g., “I was tired and drank too much and woke up next to this person I barely knew”). In the authors’ eyes, the latter approach strips away from our actions their moral implications, enabling us to continue to consider ourselves good folks despite our bad acts. Ward et al. performed five studies in support of their predictions.

Study #1 asked respondents to choose which of two ways best described several particular behaviors (e.g., not correcting a cashier, looking off someone’s test, telling an inappropriate joke). So, for example, respondents were asked if a low-level description (“Hooking up with someone else”) or a high-level description (“Being unfaithful”) best described “Cheating on a significant other.” They also asked if the respondents had engaged in such behavior. They learned, consistent with David Luban’s quotation above, that if respondents had themselves engaged in the particular behavior, they were more likely to choose the low-level, less morally-freighted description.

Study #2 confirmed that the findings of Study #1 are not confined to the U.S. culture by replicating the findings using both a U.S. and a UK sample of participants.

Study #3 repeated much of Study #1 with a new set of respondents, but asked those respondents to characterize both immoral actions and actions that were nonmoral (e.g., caring for houseplants, toothbrushing, climbing a tree). As in the first two studies, respondents who had engaged in immoral actions were more likely to select low-level descriptions than respondents who had not. However, when asked to characterize nonmoral actions, respondents did the opposite—they characterized the behaviors they had performed using higher-level representations. Why? The authors surmised that “adopting high-level representations of personally performed nonmoral actions imbues them with meaning and purpose, whereas adopting low-level representations of personally performed immoral actions strips them of the same.”

Study #4 again gave respondents a list of immoral behaviors, asked them whether they had participated in those behaviors, and had them choose between low-level and high-level descriptions. Unlike earlier studies where respondents could choose only between the high-level and low-level descriptions, in Study #4 they were able to respond on a sliding scale between two endpoints anchored by the low-level and high-level descriptions. They were also asked to judge the wrongness of the action on a sliding scale. Results indicated that if respondents both answered that they had engaged in the immoral behavior and chose low-level descriptions of the behavior, they were most likely to judge it as relatively harmless compared to other respondents:

We found that the main effects model that included both experience and level… fit the data better than the model that only included experience and the model that only included level….

These results indicate that level and experience both contribute to evaluations of an action’s moral wrongness. We found that participants who had completed the action in the past reported that it was less wrong and that individuals who identified actions in terms of their lower-level features also reported that the actions were less wrong. This study thus concretely ties lower levels of action identification to believing that the action is less morally wrong.

Study #5 presented respondents with apologies for immoral behavior couched in either high- or low-level descriptions, asking them to rate the moral wrongness of the behavior, the sincerity of the apology, how likely they would be to forgive the transgressor, to what extent the transgressor had good character, and to what extent the transgressor knew the right thing to do. The study produced mixed results. Participants who read high-level apologies viewed them as more sincere, were more likely to forgive the transgressor, and predicted less recidivism than readers of low-level apologies. On the other hand, the difference in descriptions (high-level vs. low-level) created no significant difference regarding respondents’ perceptions of moral wrongness of the act or the moral character of the transgressor.

This paper presents a promising explanation for the mechanism by which people manage to let themselves off the hook for their immoral actions. As the authors conclude: “experience can have a profound impact on mental representations, resulting in a tendency to automatically minimize the moral implications of one’s actions by focusing on an action’s mechanics rather than its meaning.”

However, we should remember that this is just one of several such explanations, and that it springs from an earlier paper by Eyal et al (2008) that contained several studies that have not proven replicable (though that is not the case with the Eyal et al. study that inspired this paper by Ward and colleagues).



Jens Agerstrom & Fredrik Bjorklund, Moral Concerns Are Greater for Temporally Distant Events and Are Moderated by Value Strength, Social Cognition 27(2): 261-282 (2009).

Albert Bandura, Moral Disengagement: How People Do Harm and Live with Themselves (2016).

Cara Biasucci & Robert Prentice, Behavioral Ethics in Practice: Why We Sometimes Make the Wrong Decisions (2021).

Tal Eyal et al., Judging Near and Distant Virtue and Vice. Journal of Experimental Social Psychology, 44: 1204-1209 (2008).

Tal Eyal et al., Thinking of Why a Transgression Occurred May Draw Attention to Extenuating Circumstances, Social Psychology 45(4): 327-334 (2014).

Han Gong & Douglas Medin, Construal Levels and Moral Judgment Some Complications, Judgment and Decision Making 7(5): 628-638 (2012).

Han Gong & Douglas Medin, Commentary on Zezelj and Jokic, Social Psychology 45(4): 327-334 (2014).

Chelsea Helion, Adrian Ward, Ian O’Shea & David Pizzaro, Making Molehills Out of Mountains: Removing Moral Meaning from Prior Immoral Actions, Journal of Behavioral Decision Making (2022)(forthcoming).

Joris Lammers, Abstraction Increases Hypocrisy, Journal of Experimental Social Psychology 48: 475-480 (2012).

David Luban, “Making Sense of Moral Meltdowns,” in Moral Leadership: The Theory and Practice of Power, Judgment, and Policy (D. Rohde, ed. 2006).

Yaacov Trope & Nira Liberman, Temporal Construal, Psychological Review 110(3): 403-421 (2003).

Yaacov Trope & Nira Liberman, Construal-Level Theory of Psychological Distance, Psychological Review 117(2): 440-463 (2010).

Cheryl Wakslak et al., Representations of the Self in the Near and Distant Future, Journal of Personality and Social Psychology 95(4): 757-773 (2008).

Alexander Walker et al., Controlling the Narrative: Euphemistic Language Affects Judgments of Action While Avoiding Perceptions of Dishonesty, Cognition 211 (2021).

Iris Zezelj & Biljana Jokic, Replication of Experiments Evaluating Impact of Psychological Distance on Moral Judgment, Social Psychology 45(3): 233-231 (2014).

Iris Zezelj & Biljana Jokic, A Rejoinder to Comments by Eyal, Liberman, & Trope and Gong & Medin, Social Psychology, 45(4): 327-334 (2014).



Behavioral Ethics:

Overconfidence Bias:


Self-serving Bias: