‘He was looking at me funny’: The (limited) rationality of the hostile attribution bias

Last week, an article in the Pacific Standard discussed the evolutionary origins and present-day disutility of the Hostile Attribution Bias (HAB). The HAB is exhibited when an individual automatically attributes malicious intentions to another, often in cases where that person’s behavior is ambiguous.  For example, when someone uses the colloquial phrase ‘he was looking at me funny’ as a justification for their own hostility, this is meant to imply that the utterer interpreted another person’s gaze as judgmental or even threatening; in fact, though, it may have been neither. Given that those with a propensity towards exhibiting this bias are also more likely to engage in aggressive behavior on its basis, the bias is widely seen not only to be irrational, but also detrimental. Indeed, the author of the aforementioned article says: ‘The trouble is, the more we sense hostility in others, the more aggressive we tend to be in return. And in many social contexts, hostile attribution bias is, as psychologists put it, highly “maladaptive.”’

In what way is the bias ‘highly maladaptive’? Is it wholly irrational? Arguably, reduction of this bias would have significant benefits both for individuals exhibiting reduced bias, and for wider society. A study cited in the Pacific Standard article investigated the effect of cognitive behavioral therapy (CBT) interventions on HAB, which they inferred from reductions in violent crime arrests. They took a large sample of young men from public schools in Chicago, most of whom were from significantly disadvantaged populations. Over the course of several months, the experimental group took part in about 13 one to two-hour-long sessions, parts of which were devoted to teaching CBT techniques. These techniques were designed to reduce judgment and decision-making errors, which the authors hypothesized may be an important mechanism of action mediating the effects of disadvantaged social conditions on young people.

An example of an intervention used in the study is as follows:

The very first activity for youth in the program is the “Fist Exercise.” Students are divided into pairs; one student is told he has 30 seconds to get his partner to open his fist. Then the exercise is reversed. Almost all youth attempt to use physical force to compel their partners to open their fists. During debrief, the group leader asks youth to explain what they tried and how it worked, pointedly noting that (as is usually the case) almost no one has asked their partner to open their fist. When youth are asked why, they usually provide responses such as: “he wouldn’t have done it,” or “he would have thought I was a punk.” The group leader will then follow-up by asking: “How do you know?” The exercise is an experiential way to teach youth about hostile attribution bias.

In their conclusion, the authors report that participation in the programme reduced violent-crime arrest by 44 percent and arrests for “other” offenses by 36 percent, this reduction being driven by declines in vandalism, trespassing, and weapons offenses. These impacts, they suggests, occur on the types of offenses we might expect to be most strongly affected by an intervention that places a heavy emphasis on reducing automatic, angry behavior, and common biased beliefs like those resulting from HAB.

Obviously, a reduction in violent crime is a good thing. But, the function (or dysfunction) of the HAB is not so clear-cut. The first thing to note is that a bias that results in a surplus of beliefs about the (mal)intentions of others is crucially different from a tendency towards violent behavior. Whilst the two will often come together, they can also be separated. Initial, automatic beliefs can be monitored and updated by the agent with further evidence before any action is taken, and the mere belief that someone does not wish you well does not necessarily have to result in a display of aggression. It could, instead, result in wariness or withdrawal. For example, if a stranger appears to be behaving in a threatening way, then, in the absence of knowing the stranger’s actual intentions, the prudent thing to do seems to be to err on the side of caution and, if possible, withdraw from the situation. However, this involves interpreting the behavior as a potential threat. HAB might in some circumstances have prudential value.

Admittedly, there is a difference between sustaining an intuitive belief about the malicious intentions of others and adopting a strategy of vigilance. Whilst the former will sometimes lead to false beliefs, the latter enjoins the agent to, as far as possible, affirm one’s ignorance of the other’s intentions in the absence of adequate evidence; in addition to the periodic prudential value of HAB, there are many situations in which vigilance is rationally warranted.

However, and more interestingly, we can further note that attempts to reduce HAB pose something of a prisoner’s dilemma. Of course, it is best for everyone if no one is hostile to anyone else (and therefore no one is aggressive). However, if someone else assumes you are hostile (and so is disposed to act aggressively towards you) then it is better that you have assumed that they are hostile (perhaps not in a way that makes you prone to violence, but in such a way that you are prepared withdraw or deflect). So, in contexts where HAB is widespread – leading to hostility in those who exhibit it – it is better for an agent to have the propensity to attribute hostile intentions. In some ways, paradoxically, this makes the bias somewhat rational in such cases, although its rationality is derived from the very fact that others exhibit a bias.

So, whilst HAB in many, even most, circumstances frustrates social cohesion and functioning, in circumstances where HAB is particularly rife, it may have some prudential value, and may, paradoxically, even be rational. This does not mean, however, that we should support the idea that HAB needs no attention or intervention. In particular, placing emphasis on training oneself to respond in non-aggressive ways even when suspicious of the intentions of others will be of benefit to individuals who do so modify their behavior, and to their wider communities, as the interventions in Chicago suggest.

  • Facebook
  • Twitter
  • Reddit

One Response to ‘He was looking at me funny’: The (limited) rationality of the hostile attribution bias

  • Nicholas. R says:

    This provides a good example of how humans tend to be risk averse individuals. Individuals choose to exhibit aggression because over multiple encounters this is the choice likely to result in a smaller aggregate negative effect (less bulling and harassment overall), . Its a type of biological warning saying confrontation will be costly. Is there any data that provides actual support for this type of decision making? Something along the lines of individuals with greater street cred, have fewer confrontations.
    Still there appears to be ill-logic in this decision because it would obviously be better for both individuals not to be aggressive. Could lack of perfect information play a part of this ill-logic? This would touch on how HAB is introduced to an individual. If an individual develops in an environment where they continually see violence, how likely would that individual be to try or even think of a none violent approach? Could a non-violent approach in fact go against an individual’s cultural norms resulting in an even greater distrust and violent response? (Think of a gangster bumping into you on the street and then being exceeding polite) In this case it may not be so much of an over prevalent HAB as an underrepresented non-aggressive response.


Subscribe Via Email