"When someone does something, like cut me off in traffic, I assume they did it on purpose."
The intentionality bias suggests that people have a tendency to interpret all acts as intentional by default. More specifically, it stipulates that when evaluating an agent engaged in an action, an intentional interpretation is automatically activated in our minds by default . This initial interpretation can be overridden or inhibited with additional information, but according to the intentionality bias, every action is judged to be intentional until proven otherwise. All non-intentional actions, including those due to accidents, mistakes, randomness, and luck will be evaluated as unintentional only after an initial intentional interpretation is activated, then inhibited. Because of this, we can make certain predictions about when intentional interpretations would be more likely. These would be situations where we have less ability to inhibit the initial default intentional interpretation. For instance, if we are under heavy cognitive load (e.g., multi-tasking) , or when inhibition is compromised due to alcohol , or even if we are younger and have less knowledge and processing capacities. If the research is correct in suggesting that intention is activated whenever evaluating an agent engaged in an action, this phenomenon may be quite prevalent and present in all populations to varying degrees.
Road rage provides a good illustration of the intentionality bias. When someone cuts us off in traffic, or drives too closely behind us, many people may become annoyed quite quickly. This can be understood with the lens of the intentionality bias insofar as, although the other driver may not have intended to tailgate so closely, or to cut us off, we neglect this interpretation and instead interpret the act as one of aggression. Once we interpret the act as aggressive, we may respond in kind (driving too slowly, or sharing an unfriendly gesture); the scene is then set to escalate an innocent mistake into full-on battle!
As explained previously, an accident is deemed to be an accident only by replacing the initial intentional interpretation with a non-intentional one. For instance, if asked whether an act like ‘‘setting a house on fire” is intentional or accidental, according to the intentionality bias our initial reaction is that it is intentional. This automatic inference of intention, however, may subsequently be inhibited by additional knowledge such as an understanding of human fallibility (e.g., forgetting a burning cigarette), social norms (e.g., burning houses are generally undesired), and flammability (e.g., of curtains and wood). Therefore, although an intentional interpretation is initially activated, after a beat of consideration we may decide that such an act is accidental. Importantly, however, if our access to the knowledge of fallibility and flammability is compromised, the initial interpretation of this act would be that it was intended. Similar logic applies to other non-intentional acts: access to knowledge about the event or related events can help inhibit the intentional interpretation.
This bias can have an impact in several ways. It can help explain why we blame others so easily and believe in conspiracy theories. Sometimes things just happen, and yet, we over-attribute volition. The intentionality bias can help us understand some forms of religious belief as well, insofar as we think that everything happens for a reason. Along the same lines, it can help explain why people have a hard time understanding randomness.
And of course, distinguishing levels of intent forms the crux of our moral judgments and our legal system. Perhaps the most extreme illustration is that the difference between murder and manslaughter rests on intentional inference; yet all judgments of guilt or innocence rely on how much the person intended the action in which they were involved. This bias thus highlights the importance of the reminder, so crucial to a fair trial, that people are innocent until proven guilty. Indeed, the fact that we need such a reminder is telling, as it seems we often neglect to see that many actions are caused unintentionally. Rarely, however, do we have to explicitly persuade ourselves that people act intentionally – the intentionality bias does it for us.
Thoughts on how to act in light of this bias
Give people the benefit of the doubt.
Actively seek alternate explanations for people’s behaviour, especially when negative.
Temper your interpretations of conspiracy and blame.
How is this bias measured?
This bias was first measured by having people read a series of sentences describing actions that can be done intentionally or by accident, but that are typically done by accident (such as ‘She kicked the dog.’). People were asked to decide whether the acts described something generally done ‘on purpose’ or ‘by accident.’ When they had to make their judgments quickly, they judged more actions as done on purpose . Other research has found similar effects when people are under the influence of alcohol - they over-interpret the intentionality of ambiguous acts . Both contexts prevent people from properly inhibiting the default intentional interpretation.
This bias is discussed in the scientific literature:
This bias has social or individual repercussions:
This bias is empirically demonstrated:
 Rosset, Evelyn (2008). It’s no accident: Our bias for intentional explanations. Cognition, 108, 771-780.
 Bègue, Laurent, Bushman, Brad J., Giancola, Peter R., Subra, Baptiste, & Rosset, Evelyn (2010). “There Is No Such Thing as an Accident,” Especially When People Are Drunk. Personality and Social Psychology Bulletin, 36, 1301-1304
 Rosset, Evelyn & Rottman, Joshua (2014). The Big ‘Whoops!’ in the Study of Intentional Behavior: An Appeal for a New Framework in Understanding Human Actions. Journal of Cognition and Culture, 14, 27-39.
 Kahneman, Daniel. (2011). Thinking, Fast and Slow. Anchor.
Interpersonal level, Need for cognitive consonance, Need for cognitive closure
Evelyn Rosset, PhD, introduced the intentionality bias in her doctoral dissertation at Boston University. She is currently a lecturer at Anglia Ruskin University, in Cambridge, UK, and founder of the Maac Lab (www.maac-lab.com), a non-profit organisation based in Lyon, France.
How to cite this entry
Rosset, E. (2023). Intentionality bias. In G. Béghin, E. Gagnon-St-Pierre, C. Gratton, & E. Muszynski (Eds). Shortcuts: A handy guide to cognitive biases Vol. 5. Online: www.shortcogs.com