 On behalf of the McLean Center for Clinical Medical Ethics and the Grossman Institute for Neuroscience, I welcome you to the 15th lecture in our lecture series on neuroethics. Let me announce that this is the final talk of the winter quarter by Professor Kubota, and we will have a break until April 6th, which will be the first Wednesday in the spring quarter that we'll have a session. Bob Trug from Harvard is coming out to talk about neuroethical issues around the diagnosis of death, a very difficult contemporary problem. Today, it is my pleasure to introduce our speaker, Professor Jennifer Kubota. Professor Kubota is in the Department of Psychology and also in the Center for the Study of Race, Politics, and Culture here at the University. Professor Kubota received a joint PhD in Social Psychology and Neuroscience from the University of Colorado. Here she held a postdoctoral fellowship in Social Neuroscience at New York University, during which time she worked on projects relating to the neural foundations of prejudice and prejudice reduction. Professor Kubota's work has one primary aim, to reduce prejudice and discrimination. She seeks to identify and define what intervenes in human behavior to produce prejudice. In her work, she's identified factors that contribute to the inequitable burden and rewards as a result of certain social group memberships. Her findings reveal promising intervention techniques that may successfully diminish racial bias. To achieve this goal, Professor Kubota employs research tools of both neuroscience and psychology, including FMRIs, EEGs, reaction time, self-reports, to identify the neural and psychological mechanisms of prejudice and discrimination. Professor Kubota's research is supported by the National Institute on Aging and the National Science Foundation. To give you an idea of the kind of work that Professor Kubota publishes, here are three recent titles. One is called the Neuroscience of Race. A second one is called Insights from Functional Magnetic Resonance Imaging Research on Race. And a third one, the Neural Mechanisms of Prejudice Intervention. Today, Professor Kubota will speak to us on the title you see above me, Social Justice Neuroscience, Responsibly Communicating and Implementing Neuroscience as a Tool for Social Change. Please join me in giving a warm welcome to Professor Jennifer Kubota. Thank you for that lovely introduction and to all of you for coming. So today, I'm really going to discuss research on stereotyping and prejudice. With closing discussion about the ethical considerations of this type of, communicating this type of research. So really, when I'm talking about implementing neuroscience as a tool for social change, I'm really talking about using neuroscience to help understand the mechanisms involved in prejudice intervention. And really, this is a very new field, so I'm going to close with just a brief discussion about the ways in which we need to be very careful as we move forward in this research. So, in our daily lives, we navigate through complex social worlds. Even in these brief encounters, we are rapidly processing a variety of social information around us. Who we attend to can depend on a number of factors, including our culture, the environment we find ourselves in, and our internal states, how stressed out we are. Why might we rapidly process social information about others? Well, it's efficient. So we can gain, we can rapidly predict and gain a lot of information about people and then use that information to predict how they may behave. But that doesn't necessarily mean that that information, we glean as accurate in predicting the thoughts, feelings, and behaviors of others. So I'd like you to keep that in mind as we move forward. What's pretty fascinating about this research is that within a second, we quickly process a variety of information about others, including information about race, about gender, about age, and about emotion. Today, I'm really going to talk about social category processing with a specific focus on race because that's where the majority of my work is. But I do have work on all these other dimensions. So in the lab, we often strip away the context. So we just simply show people faces. And this is the simplest sort of experiment we do. When we do so, individuals are able to process information about race, gender, age, and emotion within 200 milliseconds of viewing these faces. So this information is getting into the system super fast. When categories are less clear in this example, when individuals are racially ambiguous, processing is a bit delayed, and interestingly can change depending on the label you assign the face. So if I label the face as black, you process the face the same way you process other black faces. So in this way, top-down information such as a label can influence really basic bottom-up perceptual processing. So you might be thinking and rightfully so that this is all about color differences in the face. But if you blur the faces, so color and contrast differences remain, but there are no features and they're indistinguishable, you find early differences on race and emotion entirely disappear. So it really is something about the feature information of the face that's leading to these early differentiation as a function of face. This work is all EEG work, which has really nice temporal resolution. So we're really able to look at early attention differences based on these social categories. Okay, so we process social category information rapidly, but how does this affect our memory for the individuals we encounter? So we're more likely to see outgroup members as all alike while we can differentiate characteristics among in-group members. This is often referred to as the outgroup homogeneity effect. Oftentimes you'll hear the really yucky phrase, they all look alike to me. That's what this effect is referencing. So in the lab we test this by simply just showing people faces and then asking them to report in a memory test if they saw the face. So in this example, the individuals saw same race and other race faces and then we test their memory. So here is just accuracy and memory on the y. These are same race faces and cross race faces. When we do so, individuals are more accurate for same race faces, the black bar, and are more incorrect for other race faces, the diagonal bar here. Okay, so it gets in quick and it affects things like memory. Well, of course, race also influences the activation of stereotypes and prejudices. So I've reviewed briefly how we process race and how the race influences memory, but in order to really talk about stereotyping and prejudice, it's important to first define what we mean by stereotypes and prejudice. So stereotypes are representations about groups and their members. So for example, you may have the stereotype that white guys can't play basketball or they can't jump. You may hold a stereotype that police officers are particularly fond of donuts. So really, in practice, stereotypes are the traits and behaviors we assign to groups and they do not necessarily have to have predicted validity, meaning they don't actually have to be accurate and on average, they actually aren't for many of the stereotypes we have. Some do predict base rates in the general population, but certainly not all. So you can think about these as the thoughts we have about groups. Prejudices, on the other hand, are the feelings we have about groups. How do we feel? How warmly or coolly do we feel? What's our general attitude about that group? Importantly, stereotypes and prejudices guide both our perceptions and our impressions of others. So once we encounter an unfamiliar individual, we rapidly, automatically, I'll show you some research on that in a second, activate these stereotypes and prejudices and they help guide how we think about the person. In many ways, they guide our expectations about how that individual should behave. So how do we measure stereotypes and prejudices in the lab, behaviorally, at least? Well, we do so in two ways. So we often give people explicit measures and these are measures of stereotypes and attitudes that individuals consciously endorse and can easily self-report. We also measure implicit associations. So implicit stereotypes and attitudes are those that are involuntary, may be uncomfortable for the perceiver to self-report and at times are described as unconscious. So often individuals don't have self-reflective access to their implicit associations. So in the lab, for measuring explicit stereotypes and prejudices, we often just give people self-report measures or questionnaires. This is a very typical one. It's really easy. We just ask people how warmly or coolly do you feel about group A versus group B? So in this way, participants have to deliberately report how they feel. It's a self-report measure and these explicit stereotypes and prejudices may reflect an individual's personal beliefs but are often influenced by things like cultural norms or the way in which individuals think they should be responding in the moment to that kind of question. They have the ability to control how they're responding on these measures. So we also measure implicit associations with measures that are more indirect. So for example, the implicit association test that I'm showing you here, the IAT, for this, individuals sort words and faces as fast and accurately as they can using these categories above. So people sort words with the labels black, bad and good and they sort the faces with the labels black and white. In this example, you can see that the labels are paired stereotypically. So black and bad share a label side or a response key and white and good share a label side or a response key. In the next part of the task, the labels are now switched. So now they're paired counter stereotypically. So black and good share a label and white and bad share a label. The IAT is one of the most widely used implicit association measures. It's not the only one, but it's certainly the most widely used. For this task, the typical finding is that individuals are faster and more accurate when responding with labels that are paired stereotypically. So white and good and black and bad compared to when labels are paired counter stereotypically. So implicit measures are thought to really reflect more cultural stereotypes or cultural learning. So these are associations that you build up over time and you have access to. So if I ask people what are stereotypes about a bunch of different groups, surprisingly there will be a lot of consensus. Even if someone says, well, I don't personally endorse that. They know that these associations exist. Those associations are in the system and they can influence a variety of things including nonverbal behaviors and fast decision-making. So these are really automatic, effective and semantic associations about individuals that are in the system. And you can have negative implicit attitudes but positive explicit attitudes about group members. Okay, so why might implicit racial associations be important? So this is a graph of IT scores in the US from a large sample, 732,000 individuals in the United States. These are participants here on the Y. These are implicit associations. You can just think about this as racial bias scores on the IT. Black bars are pro-white associations faster to pair when the labels are paired stereotypically. Here, the gray bars are pro-black individuals' scores and those are when individuals are faster to pair them when they're paired counter-stereotypically. So what we can see from this graph is that negative implicit associations are extraordinarily prevalent in the United States. This is a huge sample of individuals. And just because they're present, that doesn't mean that they have any effect so we could have these associations but they could be completely meaningless in predicting actual behavior but it turns out that's actually not the case. In fact, they predict real-world discrimination. In fact, in a recent meta-analysis which just looking across a number of studies published on the same topic, in this case, 122 studies, they found that the IIT effectively predicted racial bias in a range of social behaviors, social judgments and even physiological responses to a greater extent than explicit attitudes. So what I'm plotting for you right now is correlations here between explicit attitudes and racial discrimination. You can see that there is a correlation of 0.12. For implicit measures, we get a little bump up here to 0.24. So one thing to note, this isn't all the variance in explaining discrimination. There's clearly a lot of other things left, right? But implicit attitudes predict to a greater extent discrimination than explicit attitudes. So given that implicit associations predict some of the variance in intergroup discrimination, it's important to explore potential ways to mitigate implicit racial bias. And I'll turn more to that in a bit. But for those of you who are interested in assessing your own implicit associations for both social attitudes, but also for mental health attitudes, interestingly, you can go online to Project Implicit. That's the website that I took those data from. They've been collecting data for a very long time now and you can take a variety of IITs there. So to summarize just this initial section here, stereotypes are cognitions about groups. Prejudices are the feelings we have about group members. We can measure these both explicitly and implicitly using behavioral measures. And for many real-world behaviors, implicit measures are slightly better predictors of discrimination. So we know a lot about stereotyping and prejudices from a behavioral perspective, but we are now in the beginning of learning a bit more about the neural underpinnings of stereotyping and prejudice. So the variety of my work is outside of the area of behavioral research, but is more located in the area of neuroscience research, although I do both. It's really the hope of people of social neuroscientists who use these sorts of measures that by bridging neuroscience and psychology, we can begin to understand better ways to mitigate these sorts of biases that we have. So we can identify mechanisms, basic behavioral mechanisms. It may help us develop better interventions. So for the second part, I'm just gonna give you a brief overview of some of the work on the neuroscience of prejudice. So one of the first studies to explore the neural correlates of race was conducted by Professor Liz Phelps at NYU. This was a study of 14 white Americans who viewed unfamiliar black and white faces in a scanner, in an fMRI scanner. So the task looks something like this where they would see just faces one after another, and in the scanner, they were just asked to press a button if the face repeated. So that was just to keep them engaged in the task and paying attention. Then outside the scanner, they took an IAT, and they also had a physiological assessment of emotional arousal. In this case, it was a measure of startle reflex or blinking response. They did this as an independent assessment. Oftentimes when you use startle, you find that individuals have a larger startle response to outgroup members than ingroup members. But they use this as a measure of threat or arousal. So using functional magnetic resonance imaging that measures changes in blood flow in the brain while participants perform tasks, Phelps and colleagues found a relationship between the amygdala differences to black and white faces. Typically you see the amygdala has greater responses to black than to white faces. I'll describe why that's maybe the case in a second. So they found a relationship between these amygdala differences to black and white faces and implicit bias and race bias and emotional arousal. So correlated with the strength of both of those, the IAT and the emotional arousal or the startle reflex. So the amygdala is involved in the learning expression of fear broadly. It often responds to emotionally arousing things in the environment and it's important for social fear learning in research on emotion. So Phelps' findings coupled with the emotion literature led to the hypothesis that the amygdala is involved in the expression of implicit bias. So across a variety of studies, sometimes you see larger responses to black than white faces as a function of the amygdala, but often what you see is more a relationship between the degree of that difference and expression of implicit racial bias. Therefore under some circumstances, the amygdala is involved in processing outgroup faces, although I just want to note that that's actually not the case under all circumstances. I'd be happy to chat about why not. However, stereotyping and prejudice researchers often discuss how racial bias is also related to in-group favoritism. So this work focuses really on how do we process differently outgroup members from in-group members or how do we process them more negatively or why might we process them more negatively? What are the neural mechanisms of that? But there's a lot of work in behavioral literature that suggests that in-group favoritism and not always outgroup derogation is responsible for discrimination. So I may have a slight preference for in-group members and that preference can translate into bias decision making. So it became important as this research sort of progressed in the early 2000s to begin to understand the processing of in-group faces. One of the first attempts to do this was from professors, Golbein Eberhardt, who ran an fMRI study looking at the neural correlates of in-group memory with a focus on processes that support the improved memory for in-group members over out-group members. So in this study it was 10 white Americans and 10 black American participants. They viewed black and white faces in the scanner. And then they were taken out and given a memory test. These researchers were particularly interested in investigating how part of the brain involved in processing faces and familiar objects may support improved in-group over out-group memory. This region is known as the fusiform face area or the FFA for short. So what did they find? Well the researchers found the usual effect of greater memory for in-group members than out-group members. So what this plot here is, is memory scores. You can think about this as accuracy in memory on the Y. These are for the black participants. These are for the white participants. And in both cases they had better memory for in-group faces than out-group faces. Moreover the fusiform face area had greater activation to in-group than out-group members. So here this is just, you can think about this as the degree of activation of the FFA or how much the brain region was engaged. Here these are same race faces in gray, other race faces in white. And in both cases these are just different ways of plotting the data. You see greater activation for same race faces. And this greater FFA activation to in-groups predicted greater memory for the in-groups. So here what I'm plotting is their FFA region of activation. This is same race memory difference. So this is more advantage for same race than other race over here. This is FFA activation greater for same race than for other race. So the degree to which I had greater activation for same race faces in the FFA, the better memory I had for in-group faces. These researchers interpreted these findings as a possible neural basis for same race recognition advantage. Okay, so viewing out-group faces activates a region involved in emotion processing, the amygdala. Viewing in-group faces activates a region involved in differentiating familiar objects. Well, these are just passive viewing. So what actually happens during an interracial encounter? How is the brain engaging during that sort of event? Now this is a complicated study, so we're gonna go over it a little slower. For this study they had 15 white American participants. This research was done by Professor Richardson and was one of the first studies to look at interracial contact and its effect on neural processing. For this study, she had white participants interact with a black experimenter and then take a measure of mental depletion known as the strupe task. I'm not gonna explain the psychometric properties of the strupe task for you, but just think about it really as a measure of cognitive depletion. So they had an interracial interaction, then they took a measure of how cognitively depleted they were. Then they viewed black and white faces in the scanner. Okay, so what did they find? Well, they found that viewing black faces compared with white faces activated the DLPFC, that's the region that's circled here, the dorsolateral prefrontal cortex and the anterior cingulate cortex or the ACC. Both of these regions importantly are involved in executive functions such as self-control. So viewing black faces in the scanner, activated regions involved in self-control. Researchers interpreted this as the individuals trying to avoid prejudice and bringing online some self-regulatory control to make sure that their implicit associations didn't affect their behavior. After the interracial contact, there was more cognitive depletion on the strupe task. So again, this is after the white participants interacted with the black experimenter, they showed signs of mental depletion on the strupe task. So it was taxing. Moreover, there was a relationship between activation of the dorsolateral prefrontal cortex to black faces and the degree to which they had resource depletion after an interaction. So what I'm plotting here is the amount of DLPFC activity in the brain. Here is the depletion score and you can see greater self-regulatory engagement when viewing black faces relates to the degree to which they had mental depletion in this other independent task. So really the researchers said that this kind of supports, resource depletion is a potential mechanism through which interracial contact impairs executive function. Or put more simply, that interracial contact is taxing. It depletes resources available to avoid prejudice. So now even though we see that individuals engage sort of self-regulatory resources even spontaneously in this case, they were just passively viewing black and white faces, that does not mean that this engagement helps them avoid racial prejudice. So in fact, subsequent research since has shown by myself and many others has shown that circumstances when we see increases in DLPFC and ACC activation, that does not necessarily result in egalitarian decision making. So even though you bring online this self-regulation, it's not enough to get you all the way. So this body of research along with others has led to a neural model of stereotyping and prejudice. So we know that there's a host of brain regions involved in the processing of race. This is a review I conducted of the existing fMRI literature. These regions represent areas most often involved in the identification of race, the FFA, the evaluation of other race individuals in the amygdala, and the regulation of racial bias in the ACC and DLPFC, and are similar to areas that we see involved in the processing and regulation of emotion broadly. So this allows researchers to take a more network approach to understanding how race is being processed in the brain. So people now are moving, so I presented really the early work, but now the more current work is really looking at how these regions are activating in concert, and in fact, there's a few more regions involved to better understand when we may be egalitarian or overcome these biases. So in my research, I use these hypothesized neural mechanisms as a starting place to begin to explore the mechanisms of prejudice and stereotype intervention. So I research intergroup relations with a focus on the mechanisms that can reduce discrimination. To do so, I sort of bridge social psychology with neuroscience, effective science, and cognitive science to really understand and define the mechanisms that reduce discrimination. Now one assumption when you try to identify interventions is that implicit racial bias is in fact malleable. So today I talked a lot about already how implicit racial bias is automatic, relatively unconscious. I gave you a pretty negative view of implicit racial bias. Well, it turns out that over the past 10 years, there's been a lot of good research indicating that there's a variety of ways, in fact, that we can intervene and reduce or change implicit racial bias. Now this research is actually very new, but it's moving at a more rapid pace. So today I'm just gonna end with a discussion about two of the ways in which neuroscientists are looking at the neural mechanisms of race bias malleability. The first I'm gonna talk about is how counter-stereotypic exposure or showing someone who's counter to what we might expect a group member to be affects neural mechanisms of race. And I'm also going to discuss how childhood interracial contact influences these mechanisms. So first focusing on counter-stereotypic exposure. And what we did to begin to explore how counter-stereotypes influence the neural mechanisms involved in race bias reduction was we had people come into the lab and they were in the fMRI scanner. We had them do an IET. Then we had them read an intervention or a control story. I'll describe these more in a second. And then they took another IET. Very simple pre-post sort of intervention. So what was the intervention? Well, the intervention was exposure to counter-stereotypes where participants read a story about a white assailant and a black rescuer. It was extraordinarily vivid. The intervention was selected because of its relative effectiveness when tested in a previous study that compared 17 other implicit bias interventions. This was a huge study with a number of participants. So to really begin to understand it, we selected the most effective. Well, this intervention was more of a hammer approach. So not only did it give people counter-stereotypic information, it also gave individual strategies for response control on the IET. And asked them to take the perspective of out-group members. Half of the sample read a control story about the history of the toaster. It was selected to be of the same length and relatively devoid of social information. And I can definitely report that it was boring. Participants were not as enthused. Okay, so what did we find? So this is a community sample from New York. And for, when we looked at, this is implicit association scores here where pro-white associations are larger numbers. Here is IAT scores at pre-test. Intervention group is in red and the control group is in black. So what we found was that the intervention was successful. So when we looked first at pre-test, luckily there were no differences. They came in, had about similar IAT scores. And in fact, these are very similar to what you see in a national sample, the IAT scores. Following the intervention, we saw a decrease in implicit racial bias relative to control. So we saw a large drop here after reading this counter-stereotypic story in their IAT scores. But we didn't see such a drop for the control group. Participants on average got a little better at the test because it's a practice effect, but that was modulated by the story condition. Okay, so I'm not gonna go into too much detail about the number of different studies we're doing actually to explore this particular intervention because there's a number of them across a variety of neuroscience techniques. No, that's okay. Yeah, so there, I can describe that more in the questions, but there have been some follow-up studies and many of us are really interested in looking at the long-term effects, but you can see some of these interventions having effectiveness up to eight weeks, back to back. So it's just a five-minute time lag between the two. Okay, so we're now further investigating the neural mechanisms using both fMRI, but also patients with brain lesions. So these are a subset of the patients with prefrontal lesions. This is the data from the fMRI participants. You don't need to worry too much about the details of these, but really we're interested in looking at using patients, whether these regions are necessary for the reduction of IT scores we see after exposure to these counter stereotypes. fMRI is wonderful because it gives us a lot of information, but it's correlational in nature, so we're backing that up with our patient data. And what we're really interested in seeing is how counter-stereotypic exposure influences the network of regions involved in the processing of race, specifically to assess whether exposures to counter stereotypes decreases racial bias by altering evaluations or how we feel about the group by increasing self-regulation. So I sort of talked about those two neural mechanisms or by some mixture of both as behavioral research to date has kind of suggested. So we're currently about to actually write these data up and are collecting control participants, but we have all of our patients for this study. We have 10 patients with prefrontal lesions and 10 patients, oh sorry, 20 patients with prefrontal lesions, 20 patients with temporal lesions. So we can look at the role of the DLPFC and the amygdala specifically and these effects. However, these are lab studies, right, that I'm discussing. They're neuroscience studies, they're lab studies, they're cool, hopefully they'll give us some information about the mechanisms that are involved in counter-stereotypic exposure. But I wanna point out that it's important to explore mechanisms and computations that underlie these reductions in implicit bias to help develop more effective real-world interventions, but that doesn't mean that these interventions will translate into real-world change. So just for example, for the counter-stereotypic exposure, perhaps it's not just any counter-stereotypic exposure that matters, particularly in our complex world. So for example, this is a graph of IT scores in the US across President Obama's first campaign and when he took office. You don't need to really process too much about this graph, but one thing to note is the line stays flat, suggesting there was little movement during this time period. Now this is perhaps one of our largest examples of counter-stereotypic exposure and media coverage of counter-stereotypic exposure that we've actually had based on race in the US to date. And you can see that in the real world, it's not seeming to have an effect here. You see a little bit of reduction initially, but it's creeping back up. So it's important for researchers to keep in mind that these simple interventions we run in the lab may not translate to huge changes in the real world. In fact, they may not be effective at all in the real world. Suggesting again that researchers should be measured in interpreting their findings, but also try to translate their findings from the lab into the real world. Okay, so I said I would discuss a little bit about interracial contact. Interracial contact may also expose people to counter-stereotypes and is known to promote individualized processing and minimize prejudice. So one of the first studies that's ever looked at the effect of childhood exposure on neural activations based on race is a work done by Professor Cloutier and colleagues. Some of it was worked on here at the University of Chicago, where they looked at how childhood interracial contact influences adult neural processing. So for this study, participants were first familiarized with black and white faces, and then they viewed these faces and new novel faces in the scanner. So here they had some familiarity training. I'll describe why in a second. And then in the scanner, they looked at a variety of faces. Again, their task was just to let me know, let the experimenter know if the face repeated. And then following that, they took a childhood interracial contact questionnaire that measured both quality and quantity of contact. One thing to note, this is an extremely large fMRI study. So fMRI studies now are more modern ones are increasing their ends. We're moving away from these super small samples, really trying to get more robust assessments. So what did they find? Well, they found that there was a reduction in amygdala to familiar black faces. So here when I'm plotting is just the brain response in the amygdala. This is novel faces, these are familiar faces. Black faces are in blue, white faces are in red. What you see is that for novel faces, although this looks like it's significant, this effect is significant here, but you can see that there's an increase in amygdala activity just to novel faces in general. But there's a greater increase to black faces, for novel black faces relative to familiar black faces in the amygdala than for white faces. In fact, there's no difference for white faces in the amygdala. So it seems that for black faces, you see a decrease in what you typically see for these negative, what are assumed to be these negative effect of value of responses to black faces. So when you familiarize people to out group members, you see a reduction. More importantly, in addition, amygdala responses to familiar out group members seems to be shaped by childhood contact. So this is the brain response in the amygdala. Here on the Y, this is average childhood interracial contact. This is low, medium and high. The white line here is novel black faces, and one thing to note is, regardless of your childhood contact, you actually have elevated amygdala responses to novel black faces relative to familiar black faces. But interestingly, for familiarized black faces, childhood contact really seems to matter. So individuals who have high interracial contact seem to kind of get this increased reduction in the amygdala activity. So it seems like the biggest effects for contact on brain responses is when individuals are more familiar with out group members. So it's not that childhood contact seems to just wash away the differences in general. It actually washes away the differences in a particular way, and that's for individuals you're familiar with. Okay, so just some conclusions based on these research. So first off, social perception is extraordinarily complex. Category processing occurs extremely rapidly. It gets in quick, but not only does it get in quick, it actually influences how we behave. So I have a lot of research on social decision making and how the implicit attitudes affect the decisions we make, both in the context of the law and in economic decision making. A large network of regions and processes underlie race perception, racial stereotyping and racial prejudice. So it's not like there's one region in the brain that's coding this. That's a very simplistic approach to understanding the brain. And we're just beginning to understand how behavioral interventions shape neural representations of race. So social neuroscience is an extremely new field. So social neuroscience has been around since people argue about it, but at least in instantiation with the fMRI literature started around 2000 or so. So it's a relatively new field, right? And we're just beginning within the last three or four years to actually start to look at some of these intervention effects with the neural measures, at least with fMRI. So it's important really to think about how young the field really is. It's important to link these processes to real world discriminatory behavior. So, okay, I identify brain regions that activate or change as a function of a number of interventions. Well, again, the question is always, so what? Does it translate into anything effective? So what we typically do in the lab is we look at the basic mechanisms all the way up through translating them into real world contexts. So we develop these interventions, we test them actually in the real world. So we're starting to do some work with schools here in Chicago to look at the effectiveness of these interventions on shaping children's attitudes and that is, again, a long term study. So we're looking not only at does it change it, but does it change it in the long term? Because policy really needs to be dictated by something that's not only just short term effective, but has a long term effectiveness. And it needs to be something that can be easily implemented and actually makes sense in a real world context. So I want to end with a final note of caution. So there are now to date 20 published fMRI studies on race. There are three studies published on interventions in the fMRI literature. And if you just compare that quickly to the 428 fMRI studies on cognitive control alone, you can see really how young the field is. So I presented evidence that implicit racial prejudices can contribute to our decision making and ways to begin to identify interventions. However, it's important to keep in mind that because we're really at the beginning, we need to be measured in how we present these studies. So at least to what I consider to be three sort of important points considering the infancy of this literature. First, it's important to be very measured in your interpretations, stay very close to your data. So it's one thing to say I see a change in IAT scores or I see a change in implicit attitudes. It's another thing to say I can reduce racial prejudice in the real world with this simple study that I tested on a small sample of individuals, right? So you really need to be very measured. Need to really employ best research practices. So now we're moving towards replications of our findings and larger samples in the fMRI literature. So people are really trying to be even better than they were initially what I would say initially in the fMRI literature. Be very careful about these studies. It's very important to responsibly communicate these findings to the public. So I'll describe why this might be the case in a minute, but these measures are certainly not ready for predictive use in the real world and many researchers, including myself, are fighting very hard to keep them out of the court system, for example. So they provide us with an excellent way to begin to quantify these mechanisms, these basic mechanisms in the laboratory. We can understand the way in which these mechanisms are conserved across moderators like environment and different individual differences and participants. So they're really important for driving research forward and beginning to build an understanding of intervention. But there's still so much to learn. So although we may be exploring the neural correlates of race, it's in no way means that racism is innate, okay? So the processing of race is complex and varies by environment, context, goals of the perceiver and the culture. So what I'm showing you here are just a random sampling of a number of different headlines in the media based on this type of research. What you can see is pretty sensationalized titles. Is your baby racist? Can we develop a medical prescription to reduce racism? Like the answer is no, right? So I can tell you right now, no. Is there something in your brain that makes you a hidden racist, right? So saying that I hold these associations, right, that I know an apple is red or I have an association that an apple is red. And that is completely different than talking about the way in which that translates into behavior. So now on average, we may see these effects translating but that in no way can predict individual behavior, right? So my best guess for an individual is the average response but in no way does that mean you will respond that way. So people can hold these associations without making these crazy sensationalized claims about racism. So it certainly doesn't mean if you have them, you're racist. Racism is about behavior. And I think, you know, it's important as we begin to outline these mechanisms that we do not overstep the data and researchers responsibly communicate their science to the public and we push for more accurate representations of this research to the public. So it's really important for researchers to work really in collaboration with people in journalism in particular I would say but in other areas certainly. So I sit on panels for neural law panels and, you know, I can tell you the court system really would love to use some of this in legal defense and it's pushing forward faster than we can answer the questions and so we really need to be very measured and cautious as we proceed. So with that I'd just like to thank all my lab members and my funding agencies. We can take questions. That was a wonderful talk. I'm sure you agree that racism is just a sample of the bias we have in the world. We are full of bias every part of our life at the various degrees as Gordon Hoppe that you describe classifies it in fourth grade and racism in the grade three and we all have grade one or two and in a various ways, Republican against Democrat, Democrat, foreign and it's just full. I have been interested in this. I have, I don't think I met anyone who wasn't in certain degree bias. It is possible that you are bias against those who are bias. And so short of removing the amygdala, what kind of a, what kind of a suggestion you have because we are bound to have a bias interaction with the regular, most of bias is developed within us by people on the top, kings and presidents and those people who strongly believe in some kind of supernatural where there is no foundation to it. And so how could you do it? I saw you said you wanna change the world and if you just say a little bit about it. Yeah, so I certainly am not going to change the world through this work. Again, the idea is to understand the basic mechanisms that result in change, but implementing them on some sort of national level is not within my wheelhouse of expertise, certainly. But what I do wanna say is you're absolutely right that we have biases about all kinds of things. We have biases in decision making, just basic decision making outside of the social context. You know, we're risk averse, for example. There's all kinds of things, ways in which we have biases in how we move about the world. And some social group members, you're actually, there's actually no stigma about being negative towards them. So you mentioned political orientation. That one is certainly one where people will outright say exactly how they feel about the other members, right? And so it's one thing to say we have these differences or we have these associations. And it's another thing to say that translates into some sort of bias decision making that would affect, for example, hiring or prescriptions. What I can't prescriptions or any medical decision making. What I can say is there's an excellent study on the IIT that shows greater implicit racial prejudice in physician predicts less prescription of life-saving medication. So there is definitely something to this, but it doesn't mean it's everything. To the point about taking away the amygdala. Interestingly, if you remove the amygdala, you still see an implicit attitude. So clearly the amygdala is not the only thing here. Don't remove the amygdala. Yeah, keep the amygdala in. It's important for more than this, believe me. Thanks for an interesting talk. So have you looked at, or is there work being done in more of the cultural parts of this? Because it seems to me that some of the things like just a couple examples, reading a book by Adici, Americana, and she talks about how racism really doesn't exist in the form where in Africa or in Nigeria. And when she comes here, she sees it really clearly. And I would be really interested in knowing how those people sort of amygdala light up and what their frontal cortex does in self-regulation. And also, like for example, in Europe, there's a fair amount of experience that shows that when African-Americans go to Europe, you know, like just jazz, people who were in jazz, they met with much less racism. And those people are not exposed, they don't have early childhood exposure to people of different races, but they seem to be implicitly less racist. So could you address that? You bring up, both of these really bring up really important aspects of this research. So one thing is important to note about the work on race is that race is culturally constructed for sure. So the meaning of race changes from cultures, some cultures, in fact, skin tone is not the predominant way in which they differentiate based on race. Oftentimes you can see languages being particularly important. And actually with children, if you look, language is often the most important way in which they discriminate when you'll see who they don't wanna affiliate with. It's based on language often. So that comes in pretty early. So studying race in the US, we're studying a culturally constructed construct that changes over time. So right now Asians are often talked about as the token outgroup members. Well, during World War II, that certainly wasn't the case. So we can look at these changes as a function of history and context. And so there's a lot of excellent cross-cultural work showing if you look at some of these other countries, Europe, and you mentioned some of the African countries which are very important. You actually don't see the same black-white effects. You have to pick the group that's central there, the social groups that are meaningful there. So what does this mean? Well, it means that you can see these in-group, out-group effects, someone who's not like me or someone who's like me, really organizing how we process others. But importantly, the way in which that actually gets into our system is it's co-opted by a pretty broad neural system that's just really involved in saliency, novelty processing and processing about social others. So we didn't evolve in some way to detect skin tone differences and make judgments based on skin tone. We involved in a way that made us cautious of novelty and difference and we notice it. So cultural attitudes can then be co-opted and affect the way in which these systems are processing the information around us. So you kind of answered half of my question. So a lot of the work that you showed us was more along the idea of in-group, out-group. And I was just wondering, one of the figures you showed us early on was that the effect on memory of in-group, out-group from someone else's work on European versus African-American faces. And while they both did have an in-group effect, the African-Americans had significantly the same amount of memory for the white faces and the European-Americans had a much decreased memory for the black faces. So that obviously is along the line of some other things that we'll hear about, you know, stereotypes and implicit biases belonging to more the dominant idea than just like what group you're part of. So I was wondering if you had any sort of neural results or analysis along the idea of what the actual social stereotype is and the effect of the neural systems versus just in-group, out-group or that interaction. Yeah, so the specific stereotypes about different groups can matter a lot and it depends on the type of studies that we're doing. So if we do studies related to danger, for example, or threat, so I do a lot of work on shoot decision-making. So I didn't present a lot of my actual work, but for that work, you see the stereotypes about the group. So it's not in-group, out-group that's driving that effect. It's often groups that I have the association being dangerous or threatening that are driving the differences we see there. Importantly, one other way in which the implicit associations have been described as cultural, some of the data that has led to that description is that you see for black participants in the US, although you see more variability in their IT scores, you'd still see a large majority of those individuals having pro-white attitudes. So it certainly suggests that if you grow up in this culture, you're exposed to those associations, right? And you could be buffered from them and you can buffer yourself from them, certainly. There's a lot of really good work on that, but it doesn't protect you from having these associations in your system. They can, it depends on the task. So if it's a threat task, yes. Two comments, it was fun to see the faces of the researchers of the work that you presented. I was waiting for the test at the end regarding which ones we were familiar with. My other comment was that I've been wondering about the Obama counter-stereotypical idea. The idea that all the media coverage of Obama was counter-stereotypical. And I wonder if you've looked at whether those articles were positive or negative, because if they were negative articles, then it's maybe not actually counter-stereotypical. Yeah, so there's been some good research actually trying to do analysis, archival analysis of the coverage around Obama and relate that to these IIT effects. It does seem like there was a lot of negative coverage. So the assumption that it was all positive media coverage is certainly not true, absolutely not true. So it's not just having someone who's counter-stereotypical around or in a powerful role that seems to be important, right? It's still the way in which the cultural people with influence and who can reach out to the public are really describing that individual. And so it's very negative. It has been very negative. Any comment in by racial childrens? Can you repeat the question? By racial, to raise the parent. So there is starting, so by racial and multi-racial individuals are the fastest growing population in the United States. And there's been some really great work in behavioral research on by racial individuals. And just to sum up very quickly, because there's a lot of it, it shows that the individual's own ethnic identification can kind of predict how they will respond on these implicit association tests. So if they have a more ethnic identification with one parent versus the other, you'll see that association being similar to what you might see in that same population of individuals. So it's not clearly just also again as simplistic as the blending that seems to help. It's the way in which you internalize that information. We're actually doing a large study. We just completed it at U of C looking at the by racial and multi-racial individuals on campus and we're exploring some of these things over time. So we're tracking them for two years and doing a bunch of, taking a bunch of measures including implicit measures, racial measures. Thank you. We'll take one last question, please. Okay, this will be a quick question, but maybe not a quick answer. You mentioned, you had seen some premature attempts to use some of this research in the legal world. And I wondered if you'd like to expand on that. Yeah, so there's a big push by two companies in particular. One of the most prominent ones is called no lie MRI. And there's also work, so they try to put fMRI research as a basically a lie detector test, admissible in court. And there's also people who are working, who have private companies who are working to try to use the IET as a standard measure for use in court. So if you score some number, some positive number, then you're racist and you should get a harsher sentence for the crime you committed. The people who do this research, the inventors of the IET, they are some of the most vocal people saying, writing briefs and really communicating with people in the law to try to make sure this work doesn't get into the legal system. So there's a lot of ways in which we manipulate data, meaning we clean the data. There's a lot of standard practices in fMRI, but it's not simply some gold standard measure that is infallible. And in addition, just having these associations, again, does not translate into behavior necessarily. So to say there's an association, and then to say that made you shoot someone, for example, and these are life and death situations, is, in my mind, extraordinarily.