When making key decisions like who to hire for a job or who to trust in a crisis, we all like to imagine that we are rational actors, making reliable, objective decisions. However, we are known for being quite the opposite, and bias can creep into every aspect of our decision-making, even—or especially—when we aren’t aware of it.

In a recent publication in the journal Personality and Social Psychology Review, Jordan Axt, an assistant professor in McGill’s Department of Psychology, explored how debiasing techniques designed to address judgement and decision-making errors might also be effective in addressing the bias behind intergroup discrimination.  

For researchers exploring judgement and decision-making errors, bias typically refers to how irrelevant pieces of information can inappropriately influence an individual’s beliefs and behaviours. Research in this field often concentrates on debiasing techniques, with the goal of limiting the impact of this irrelevant information on decision-making.

When researching intergroup relations, however, bias takes on a different form. In this case, it refers to our tendency to prefer a certain social group, namely our own. This is often based on personal characteristics like race, gender, or age. Researchers in this area aim to reduce intergroup discrimination—the unequal treatment of individuals based on their group membership—rather than address the bias that informs discriminatory behaviour. 

Axt reviewed four broad classes of debiasing techniques that are typically applied in judgement and decision-making research, exploring their potential to address the bias behind intergroup discrimination and ultimately reduce such discriminatory behaviour. 

The first class of debiasing techniques involves changing an individual’s ability to assess the relevance of different information through practice and training, in hopes of increasing their capacity to avoid bias. However, this kind of training can be quite difficult to implement effectively. 

“There’s some reason to believe that training could be effective in intergroup discrimination, but oftentimes I find that there’s a limited transfer effect: You can do a good job training on one thing, but it doesn’t necessarily carry over well into other contexts where you might want it to,” Axt said in an interview with The Tribune.

The second class of debiasing techniques addresses an individual’s motivation to avoid bias, rather than their ability to do so. The goal of these techniques is to get the individual to put more effort into considering decisions. For example, although financial incentives have been shown to be effective in weight loss and exercise settings, they are less effective for cognitively demanding tasks like noticing and combatting one’s own biases.  

The third class of debiasing techniques gives specific interventions and ideas to help avoid common errors in judgement and decision-making tasks, and looks at how decisions are processed. Though not yet applied in intergroup discrimination literature, this class of techniques is promising in its ability to address intergroup bias and ultimately reduce discrimination. 

“The goal of a lot of intervention research is to change the way you approach the task,” Axt said. “There’s good research showing that if you have to think not first about why you’re right, but why you might be wrong, then you include a wider range of information when you’re making your decision, and you become more accurate more of the time.”

The final class of debiasing techniques changes the context in which a decision is made. This could mean adjusting how relevant information is presented, or how the decision itself is structured.

“In cases where you change the context to give people enough time to really think through these decisions, that could be one manipulation of context that gets people to be more accurate and less likely to discriminate towards others,” Axt said.

Axt’s lab tends to focus on intergroup relations, rather than debiasing techniques. He explained that, while both areas of research are concerned with bias, there is very little overlap between the fields’ respective corpora. 

“At a very global level, both [areas of research] are embracing this idea of getting people to avoid irrelevant information in their beliefs and behaviours. I’m hoping that this [review] can paint a nice small bridge between these two literatures,” Axt said.