Program: Busting bias: what works and what doesn't

Sana Qadar: By his 40s, Ben Barres had built an impressive career. He was a neurobiologist at Stanford, the head of his own lab, and he'd made a breakthrough discovery.

Jessica Nordell: He discovered a lot about the functioning of the glia, which are particular kinds of brain cells.

Sana Qadar: But that's not actually the story we're focusing on today, we're focusing on Ben's personal story. So, Ben had been born Barbara, and at this point in the mid-'90s he was still known to the world as a woman. He would soon undergo a transition, but he had a lot to worry about.

Jessica Nordell: He was extremely worried that the scientific establishment, the community and his colleagues wouldn't accept him, or he didn't know really what would happen, he didn't know if students would still want to join his lab or if he would still get invited to conferences. So he had a lot of trepidation about undergoing this transition. But he went through with it, and he found to his surprise that the scientific community did react but not in the way that he expected. What he found was that people who did not know that he had transitioned, who were just meeting him and being introduced to him as Ben for the first time, responded to him with more respect. They gave him the benefit of the doubt. He found that he was not interrupted in meetings anymore, and he was even at a scientific conference where another scientist was overheard saying, 'Well, Ben does really good work, his work is so much better than his sister's', not knowing that he was referring to actually the same person. And so this was really an eye-opener to Ben because he hadn't understood how deeply his gender was affecting the way people were treating him, until the transition.

Sana Qadar: This is All in the Mind, I'm Sana Qadar. And Ben's story is a telling portrait of how bias works, how it can fly under the radar yet still have an impact.

Jessica Nordell: Because the way he had been treated before, it wasn't like he was constantly facing over sexism or harassment, it was subtle, almost invisible everyday leading interactions that, because he didn't have anything to compare it to, he didn't realise that this was being governed by his gender.

Sana Qadar: And that's a workplace example, but we know that bias can affect the kind of healthcare a person receives, the kind of policing they are subjected to, and all kinds of other outcomes. But unconscious bias is notoriously difficult to address, and a lot of the strategies that we use, like workplace diversity training programs, for example, don't actually have all that much evidence to back them up.

So today, can bias be busted? We look at evidence-based strategies for change and the strategies that simply lack evidence.

At this point, most of us are pretty aware of what bias is, unconscious or otherwise, and some of the impact it has.

Jessica Nordell: Black and Latino patients are less likely to get pain medication from doctors than white patients who are expressing the same amount of distress. Women are more penalised for failure and less rewarded for success than men. If you are a white job applicant with a criminal record, one study found that you are more likely to get a call back than if you are a black job applicant with a criminal record and without a criminal record. You know, the list goes on.

Sana Qadar: This is Jessica Nordell, she's a science writer and journalist and has spent much of her career writing on bias.

Jessica Nordell: You know, I had been covering bias and discrimination for many years, and I became impatient with journalism's focus on describing the problem. Journalists…you know, we love exposing wrongdoing, we love talking about the problems and what's going wrong, that's an essential function in any society, but what I became increasingly concerned about was the question of what do we actually do about it, and what are the approaches, if there are any, I really didn't know at the beginning of this project, what are the approaches that have been shown to change people's behaviour in a measurable way?

Sana Qadar: The result is her book, The End of Bias: A Beginning: The Science and Practice of Overcoming Unconscious Bias. And before we dive into solutions, perhaps we should begin with first understanding how biases form in our minds. Where do we learn this stuff and why do we run with it?

Jessica Nordell: The idea of unconscious bias is really that as we are growing up in a particular culture, we learn the categories that are salient in that culture, categories of race and ethnicity or age or religion or what have you, and as we are learning those categories, we are also learning stereotypes and beliefs and cultural knowledge about those different categories, and all of that is stored in memory. It's not necessarily something that we choose but it's something that filters into our minds just as a by-product of living in a culture. And then when we encounter a person or a situation that maps onto one of those categories, all of those associations and all of that cultural knowledge is activated, and it starts to play a role in our interactions.

Sana Qadar: And so none of us are really immune from this, are we, because we all live in a culture and pretty much all cultures have weird ideas about outsiders, you know. So can we ever escape this kind of thinking?

Jessica Nordell: It's a good question. I think that we are all susceptible to expressing bias. One thing that is interesting is that the biases are different. You know, if you move from culture to culture, the categories are different, the biases are different. I had a really interesting experience when I was a teenager living in France briefly, I was a foreign exchange student, and in the process of going to high school I became friendly with a group of French students, they were really nice, they were really interested in me, and we hung out at lunchtime and chatted. And later I heard from the other French students and the French family that I was staying with that I had been hanging out with 'Arabs', and I didn't know anything about the category 'Arab' because I had grown up in north-eastern Wisconsin, that category was not salient to me at all, it wasn't relevant. And I didn't even notice anything different about these students because I didn't know what to look for. The category didn't even exist in my mind. And so after I had learned this category and I started to learn what it meant in French culture, then it was like my vision, my perception, my orientation towards the world started to get tuned to the French cosmology, and it was like I was learning a prejudice completely from scratch.

Sana Qadar: And did it change how you interacted with those students?

Jessica Nordell: It didn't change how I interacted with the students but I suddenly started to see the way that those students were different, in a way that was completely invisible to me before.

Sana Qadar: So there's the matter of culture but there is also the matter of our evolution. For a lot of human history, we've had to be suspicious of outsiders.

Jessica Nordell: Yes, I think there certainly is an evolutionary component…you know, for most of human history, people had a very limited range of people that they actually interacted with and probably had good reason to want to protect their kinship group. I also think we have an evolutionary reason to put things into categories and to make quick assumptions and judgements about those categories because we, just from a cognitive perspective, can't possibly process all of the bits of information that we get per day. So there definitely is sort of an evolutionary advantage to some of this.

The problem is when this tendency starts to interfere with our deeply held values, like egalitarianism, equality, wanting to treat people with fairness, wanting justice, wanting to form trusting relationships with others. That's when I think we start to see how these spontaneous reactions are not really serving us ultimately.

Sana Qadar: Let's talk solutions then, and I want to cover solutions that have worked of course but also solutions that haven't because they are important too. First up is the implicit association test. This is the most widely known tool for measuring unconscious bias. It burst onto the scene in 1998, and over the years it's had plenty of media coverage. It has also been the subject of a lot of debate in the field of psychology.

Jessica Nordell: So the IAT is this tool that kind of promises to ferret out implicit bias by testing how strongly certain social identities are linked with certain stereotypes in your mind, and basically the way it works is you sit at a computer and you are presented with words. So, for instance, if you take an implicit association test designed to assess anti-gay bias, you might be presented with a list of words like 'smiling' or 'good' or 'bad' or 'gay', and you are asked, word by word, to decide whether to put the word into the category 'gay' or 'bad' or the category 'straight ' or 'good'.

Sana Qadar: These words appear rapid-fire, so you have very little time to make a deliberate choice, you just have to kind of go with your gut. The test is available online for anyone to take.

Jessica Nordell: So if you see the word 'smiling', you might say, okay, that's good, so it goes into the category 'straight' or 'good'. And then you are shown another list and you're asked to sort the words again but this time the category is 'gay' or 'good', and the category 'straight' or 'bad'. So if you're faster at sorting words into 'gay' or 'bad' than you are at sorting words into 'gay' or 'good', it basically suggests that the connection between 'gay' and 'bad' is stronger in your mind. So it is really meant to ferret out how closely a particular identity is connected to a positive or negative valence in your mind.

Sana Qadar: And so why is it actually a bit problematic, this test?

Jessica Nordell: So there are a couple of really big problems with this test. One is that if you take the test at, say, 9am and then you take the test at 3pm, you're not necessarily going to get the same score. And so it is low in what psychologists call test/retest reliability. And you can imagine, if you are weighing yourself on a scale and it says one weight at one time of the day in a totally different weight at another time of the day, you might think, hmm, I'm not sure if this scale is that reliable. So that's one problem with it.

Another big problem is that your score on the implicit association test does not very strongly predict your behaviour. And so when this was developed, it first was seen as like a Holy Grail, like; oh my gosh, we've seen the foundation of unconsciously biased behaviour in the world because we have this test that measures this thing. But what became clear over time was that the test isn't a very good predictor of how you behave. So there might be someone whose score on the IAT would suggest they are very biased, but they might actually not behave in a very discriminatory way, and vice versa. And that's partially because our behaviour is the product of so many different factors, it's not just one association in our mind that's causing us to behave in a certain way, it's our alertness, our cognitive load, our motivation, our goals, the specific person we are interacting with, all of these things play a role.

Sana Qadar: So the IAT was a bit of a fizzer. It turns out workplace diversity training might also be.

Jessica Nordell: Definitely one of the hot approaches to decreasing bias and improving workplace cultures and organisational cultures is diversity training. The problem with diversity training is that it is rarely evaluated. If you could imagine a medicine that is distributed to a vast number of people who are experiencing some kind of health problem and the medicine is distributed and then no one checks to see if it has actually helped their health problem or not. Diversity trainings are a little bit like that right now. Most companies have some kind of diversity training but very few of them actually do the work of figuring out, well, what are the specific goals we're trying to achieve in terms of whether it's changing our culture or improving the retention of employees from certain backgrounds or whatever it is, and then tracking to see whether the diversity training is actually useful. It could be making things better, it could be making things worse, it could be having no effect at all. We often just don't know because they are rarely looked at in that light. And so that's one of the challenges, in addition to the fact that sometimes mandatory trainings can cause backlash. There was an analysis of many years of corporate diversity programs done by a couple of sociologists who found that after mandatory diversity trainings, women of colour in positions of management actually decreased. There were fewer women of colour in positions of management. So you have to be quite, I think, intentional, and really treat these approaches and these interventions almost on the level of a medical intervention, where you really want to be quite sure that what you're doing is meeting the goals that you have.

Sana Qadar: So if the evidence is mixed about their efficacy or not great overall, why have they become so common in workplaces?

Jessica Nordell: You know, I think they've become common because there are a lot of people offering them, they are fairly easy to implement, you know, you could bring in a consultant and have someone do a two-hour workshop and then you're able to say that you've addressed diversity issues. I don't mean to sound too cynical but I think that there can be kind of a box-checking approach or reason. And I think maybe a reason that gives companies more of the benefit of the doubt is that some of them really want to make a positive difference and they just don't know how, and this is kind of an immediately accessible approach that's really common.

Sana Qadar: For Jessica, the workplace is actually where she first became aware of bias, when it affected her for the first time. Growing up, she says, it didn't really factor in her thinking.

Jessica Nordell: I was a kid racialised as white in a majority white town. I am Jewish but I was undetectably Jewish, people didn't really know I was Jewish unless I happened to tell them, and I didn't encounter a lot of gender bias because I was a really conscientious, straight-A student. But what I found was when I started in my professional life, suddenly things looked a lot different. And in my own career as a journalist, I went through a period where I was trying to break into national magazines and newspapers and I wasn't getting anywhere. I was sending out queries and cold-pitches, as I'm sure you're familiar with, reaching across the ether hoping an editor will respond, and I wasn't getting any responses.

And I had one particular moment when I was trying to place an essay that I had written that had kind of a short window of relevance, and I knew that if I didn't place it soon it would just die and no one would ever see it. And so in kind of a moment of desperation I created a new email address and I sent out the same pitch but as JD Nordell instead of Jessica Nordell. And I thought to myself, maybe if this shows up looking like it's coming from a man, maybe the response will be different. I've tried everything else, why not try this? And the essay was accepted within a couple of hours.

Sana Qadar: Wow, that stark a difference?

Jessica Nordell: Yeah, I couldn't believe it, I honestly couldn't believe it. I thought, what, this didn't actually work, did it? But it did. That's what really kicked off my interest. I couldn't believe that this editor was intentionally rejecting pitches from women but I could believe that this editor was influenced somehow by some stereotypes that this person had absorbed that affected the way that they reacted to pitches. And so I wrote stories as JD Nordell for a few years after that.

Sana Qadar: You're back to Jessica Nordell now though?

Jessica Nordell: I'm back to Jessica Nordell now, I couldn't really manage it because I would be corresponding with an editor as JD and then they called me on the phone and then I said, 'Hi, this is Jessica,' and it was so confusing, it just didn't work. So yes, I ultimately just reverted to being Jessica.

Sana Qadar: You're listening to All in the Mind, I'm Sana Qadar. Today, busting bias. So, it took changing her name for science writer Jessica Nordell to get noticed, but that's not exactly a solution for curbing bias, it's a Band-Aid at best. We've also covered how a lot of workplace diversity training programs often lack evidence. So, what works? Well, the good news is there are some training programs that have actually been shown to change behaviour. One that has been extensively evaluated was developed by a woman named Patricia Devine.

Jessica Nordell: Patricia Devine is a pretty interesting person. She was one of the early psychologists who developed the idea that there could be stereotypes in the mind that are activated outside of conscious awareness. And some of her work more recently has had to do with developing interventions. So one of the interventions that has been shown to change people's behaviour is a training her lab has developed called the habit breaking training, and it basically looks at bias as a habit, as a habit that can be broken the way other habits can be broken. And so the approach that it uses is similar to cognitive behavioural therapy. So the training has a few different components. There is a component that's designed to build awareness, there's a component that's designed to increase your motivation, and then there's a component that provides replacement strategy. So what are the other things you could be doing instead of this bad habit?

Sana Qadar: Things like trying to think of situational reasons for a person's behaviour, instead of assuming it has to do with some inherent characteristic.

Jessica Nordell: And when they did evaluate it, they did a pretty rigorous randomised control trial. And so in one case they took a whole bunch of STEM departments at a university, departments that were pretty similar to one another and they put them in pairs, and one of the departments got this training and the other one I think got no training or got some kind of control training about something else. And then they looked to see in the following couple of years whether the training had had any kind of impact on their hiring practices.

And when they did a gender bias focused version of this training and they did a randomised controlled trial they found that the departments that had gotten the training did actually hire more women in a statistically significant way in the coming years. And so that's one finding. They also did racial biased focused version of the training and they found that a year later or more, students who had taken this training were more likely to speak up against racism when they encountered it in an online discussion.

And so I think what's really promising about these trainings is it seems like they are actually having an effect that is lasting beyond the day of the training or the next day. The research suggests that there is something more long-term that's getting shifted, which is very hopeful.

Sana Qadar: So that's a workplace training that has been evaluated. Then there is a training aimed at bias in the classroom, specifically reducing racial disparities in student discipline. This was devised by a psychologist at Berkeley named Dr Jason Okonofua.

Jessica Nordell: Because at least in the United States we see a really large disparity in the suspension and expulsion rates of African-American students compared to white students. And so what he did was he recruited a bunch of maths teachers to participate in this program. They were told that they were there to review best practices and teaching, but what they were actually there to do was absorb and read and reflect on the information about creating trusting relationships with students. So the whole intervention was really designed to increase what he refers to as empathic discipline. Like, disciplining students but with empathy and trust and understanding.

Sana Qadar: So the teachers read these short passages on trust and respect and why students need both of those. They were also given some of the same strategies that appear in Patricia Devine's training.

Jessica Nordell: Like they were told that it would be a good idea to avoid labelling students, to try to look at for situational reasons a student might be behaving in a certain way. And what the research showed was that after teachers went through this intervention, even though they didn't know that they were actually part of a research study, over the following year suspension rates dropped, and particularly suspension rates of African-American and Latino students dropped from about 12% to about 6%. This was in the last couple of years. And I think he actually just did a replication recently on a much wider scale.

But I think one thing that's really interesting about this approach is that it's not actually targeting bias, it's actually trying to amplify and elevate the values that are important, that we actually what teachers to adhere to when they are interacting with students, things like respect and trust and understanding and empathy. And so it's an interesting intervention because it's not actually telling teachers to decrease their bias, it's actually just helping them develop and really nurture these other values that can override those biases.

Sana Qadar: Yes, so it sounds like solutions that do work take effort, they are complex, but they can actually work in the end.

Jessica Nordell: Absolutely. I really undertook this project because I was trying to find out whether it's possible to actually change these deeply rooted habits and behaviours. And what I found was that, yes, there are approaches that actually change people's behaviour for the better, they do take work, they take a work ethic, if you will, they take a commitment. They have been shown to change people's behaviour so it's more in line with values.

Sana Qadar: After having done all this research for the book, are there any aspects of bias that you think remain unexamined or under-examined?

Jessica Nordell: Absolutely. I mean, if you look at the research on bias, the majority of the research focuses on race, ethnicity and gender, and there are so many additional kinds of bias that we do not have enough research about. I am thinking about particularly indigenous peoples. When I was looking for research about bias against Native Americans and the United States, almost all of the research focuses very narrowly on the issue of College University mascots and the controversy around them. There's a lot more obviously to the story, and so that's an area that I think is totally under-researched. Ageism and approaches to combating ageism, bias against people with disabilities, all of these areas I think require a lot more research and a lot more understanding.

Sana Qadar: And given that we have had a few decades of talk about bias, we are more aware of it now, do you think things have actually changed or are we just as biased as ever but aware?

Jessica Nordell: Good question. You know, if you look at…we were talking about the IAT earlier and its limitations and there are serious limitations, but one thing I think it's really useful for is looking at the bias is of a culture. Like, if we look cumulatively across an entire culture, across millions of these tests we can get a little bit of a sense of the contours of a particular culture's stereotypes. And there was recently a review of the way that implicit associations have changed over time, and some of them have stayed the same but some of them have decreased, particularly biased against people who are gay and lesbian has decreased if you look at broadly cultural trends in implicit association tests. So I think some biases have decreased, some are sticky, and, regardless, there is a lot of work for us all to do.

Sana Qadar: And did any of your ideas about bias get challenged or changed over the course of writing the book?

Jessica Nordell: I think I started this process thinking I was probably a little bit less biased than other people, which I think is actually true for most of us. I think most of us think maybe we are biased but we are probably a little bit less biased than everybody else. And I very humbly came to the realisation that that was not true and that I…how do I want to put this? I found that it was actually only through making sticks and screwing up that I was able to see my own biases and then able to actually take a hard look at them and address them. And that happened many, many times over the course of working on this project. I would say something or do something and then someone would say, hey Jessica, did you realise that you were saying this but that wasn't really right? Or they would pointed out to me and then I would go through a process I think that many of us also have gone through, which is feeling defensive, feeling a little angry, and ultimately seeing that this was an opportunity to actually grow. And so it was an ongoing process of learning and growing and changing.

What I believe, what I've come to see and what I truly believe is that while it is a lot of work and it takes a lot of humility and ability to manage one's own defensiveness, the rewards are really great because when we tackle our own biases we open the possibility of having an authentic relationship with another person, and that to me is such a huge reward. You know, being trusted and being in a trusting relationship with another person, particularly someone who has some kind of difference, you know, you exist across some kind of social difference from one another, it makes the work so worthwhile and so just life giving ultimately.

Sana Qadar: That's Jessica Nordell, science journalist and author of The End of Bias: A Beginning. And just before we go, we kicked off the show talking about neurobiologist Ben Barres and his experience of bias as a trans-man. Ben passed away in 2017 but his legacy is huge.

Jessica Nordell: Oh my gosh, he is like a towering figure. He became a huge advocate for all underrepresented groups in science and he constantly was pushing academia to evolve.

Sana Qadar: That's it for All in the Mind this week. Thanks to producer Bel Smith and sound engineer Russell Stapleton. I'm Sana Qadar, I'll catch you next time.