
WHY are we prejudiced? What happens in our brains when we make assumptions about people who look or speak differently to us? As movements such as Black Lives Matter work to expose the systemic racism in the US and Europe, such questions are taking on new and long overdue urgency. If we are to overcome our biases, we need to understand their neural and psychological roots.
Lasana Harris, a neuroscientist and experimental psychologist at University College London, is among those striving for such an understanding. His research focuses on how we think about other people’s minds, known as social cognition, and more specifically on how we perceive others. Working with Susan Fiske at Princeton University, his research on the brain mechanisms underlying dehumanisation has revealed the surprising ease with which we can stop ourselves from having empathy for the plights of others.
Advertisement
Such insights have informed his thinking on racism, too. Harris views what many people call unconscious bias as an inevitable result of the associations we learn and the way our brains react to perceived threats. Rather than something we engage in unconsciously, he argues that it is something we know we are doing but struggle to control.
Here he tells èƵ why societies condition people to be prejudiced and what the science says we can do about it.
Daniel Cossins: Dehumanisation is a horrifying word and yet your work suggests it is something we all do. Why is that?
Lasana Harris: Firstly, if I want to do something to another human being that is something I don’t typically like doing to human beings, then I’m going to need to not think about their mind while I do it. The other reason is that dehumanisation is a way to regulate emotional responses, preventing us from identifying with suffering and feeling negative ourselves. Take homeless people, for instance. If you have to feel empathic towards every single person you see living on the streets, you would be exhausted before you got home. Or you might feel worried about them and guilty that you didn’t help. That’s a feeling you want to avoid having every time you leave your house.
We have demonstrated that the brain regions that are always engaged when you think about other people aren’t engaged when you think about different types of people – in this case, what we call “extreme out-groups” like drug addicts and homeless people.
Where does unconscious bias, and particularly racism, fit in to all this? Is it also a form of dehumanisation?
When we did the first dehumanisation studies, a lot of people said it represented the worst kind of prejudice. But it turns out that when you look at the brain, prejudice is a slightly different function. They are both the result of a learning process, but whereas dehumanisation is linked to disgust, prejudice is linked to fear.
The reason why we are prejudiced – and I like to say that we are all prejudiced – is because we live in cultures where certain things get associated. For instance, if you constantly see African Americans committing violent criminal acts in the media, your brain associates violent criminal behaviour with African Americans.
Bias against LGBTQ+ people is often rooted in disgust and fear, with the perception being that they threaten traditional values around marriage and male-female gender roles. Not that every form of social bias is threat-based; elderly people, for example, are passively harmed by being ignored.

When you see an out-group member who you have tagged as threatening, regardless of whether they really are, a region of the brain called the amygdala is activated. This is what processes fear. So prejudice is essentially a threat response. This is part of the reason I have always had an issue which the term implicit bias, because you are always aware of your threat response. It’s not really about being aware and unaware; it is a basic survival response. It is going to shape how you behave, and you are going to at least be aware of your behaviour.
What is a more accurate way to describe our feelings of prejudice towards others?
I think a better term is just bias. Saying that you are biased doesn’t mean you are a morally bad person. It’s not like you decided to have these thoughts. You just happen to live in a society where that is the way things are structured. But you can still be held accountable for your bias, because you have the ability to regulate and override it.
How can people override their biases?
There are a few ways. The first and perhaps most prevalent is to encourage that awareness we just mentioned. It’s like if you know you are afraid of spiders but you also know that fear is silly because, in the UK at least, they aren’t going to hurt you. So if you are cleaning the bath and a spider pops up, you move it. Your ability to do that, even though you are terrified, is the same ability that lets you override bias, racial or otherwise. Your awareness of the threat response that you have to an out-group can allow you to regulate it.
Another way is to undo the learning. Even though the associations we learn can be hard to shake, you can do it if you have enough experiences with members of an out-group to realise they are actually not threatening. This is known to psychologists as the contact hypothesis – the idea that if different groups interact, you can reduce bias.
The third way, and the one that is least tried, is to get rid of the categories in which we put people. Prejudice begins because you have made that person a member of a category that is threatening. If you had put that person in a different category from the beginning, you wouldn’t have the threat response.
The problem is that by default we categorise people by demographic characteristics like age, gender and race. That is how we like to see the world. But you can see the world in many different ways: you can see a person as a member of a particular occupational group, for instance, or any number of categories that aren’t necessarily tied to this threat response that has been instilled over centuries and is now fundamental in our cultures.
Lots of companies are now offering implicit bias training. Does it work?
No. There has been a lot of research looking at implicit bias training, and it doesn’t work for a variety of reasons. One is that it is usually mandatory, which means that people aren’t motivated. The second reason is that it usually serves a legal checkbox function, which I think then spills over to the people who do it. So they think “I’ve done unconscious bias training, so I’m not biased”.
I think bias training can work but only if it builds that awareness, if it puts the bias in a historical and cultural context, so that people understand where these associations come from, and they begin to realise how systemic it is. That is one of the great things about the Black Lives Matter movement compared with what I have seen before. It’s really causing a lot of people to ask those questions.
“Saying that you are biased doesn’t mean you are morally bad, but you can still be held accountable”
What is the relationship between this bias and the systemic racism that Black Lives Matter has shone a spotlight on?
I think the bias results from the systemic racism. Historically, bias against black people was a way to appease the public so they would view slavery as civilised. There was a tonne of “scientific” research to demonstrate they weren’t quite human, and there were religious arguments that they were uncivilised. All of this painted people of African descent as subhuman.
So it is really important to take this long view, to understand how that history has influenced our brain processing. It also highlights how difficult it is to change this systemic bias, and why you need systemic change.
Do you think science can help achieve that sort of systemic change?
Social psychology has always been geared towards solving these problems. The field was created in response to the second world war, in response to the Holocaust. You had a bunch of researchers in Europe who were Jewish, who fled to the US and wanted to study obedience and conformity and prejudice and dehumanisation.

We have had decades of research now, to the point where we fully understand these mechanisms I have been describing. Since those early days, establishing shared goals between groups has been identified as a possible way of changing biased behaviour, a realisation that we are all in this together. It is kind of like the climate change problem: we have the knowledge to fix it – it is there – what we are missing is the political will.
It feels flippant to talk about anything else, but presumably the coronavirus pandemic will have a big effect on how we interact with one another.
The important concept here is disgust, which is an emotional response that evolved to protect our bodies from harm. Disgust has also been co-opted into the social domain, particularly with regard to the pathogens and diseases people carry. In the past, this response might have been reserved for people who look sick. We have now associated that threat response with everybody, because we have been told for months that anyone can be contagious.
Our brains have learned this response. But I don’t think it is going to last because we are such intensely social creatures. As things gradually ease up and we interact more and more, I imagine this urge to keep away from others will pass.
Social psychology has a bit of a reputation problem, largely because some of its most arresting findings couldn’t be replicated in repeat experiments. Do you think social psychology can survive this crisis?
The replication crises are something that hit all science, but social psychology is everybody’s favourite whipping boy because we study what everybody is an expert on. We all have social cognition, so we all have theories about why people do stuff. There has always been this pressure on social psychology to do better than other sciences.
What makes social psychology important is that it is the scientific study of how we think about others. Social psychology really values the scientific method, so every one of these crises actually pushes the field forward. I think we have done a lot to address this replication crisis. It really raised the standards in our field. We do preregistered studies, for example, where we design the study and it gets peer reviewed before the data is collected. If it gets accepted, then I collect the data and it gets published. That lifts the pressure for researchers to produce spectacular results.
We have done things that other fields are still not doing, and I think it will take them years to catch up.