Neuropolarization: Brain Activity Differences Contribute to Partisan Divide


Lorenzo Norris, MD; Yuan Chang Leong, PhD


November 13, 2020

This transcript has been edited for clarity.

This interview was originally published as part of MDedge's Psychcast podcast series. In this episode, podcast host Lorenzo Norris, MD, assistant professor of psychiatry and behavioral sciences at George Washington University, Washington, DC, spoke with Yuan Chang Leong, PhD, a postdoctoral scholar in cognitive neuroscience at the University of California, Berkeley, about his research into the neural underpinnings of politically right- and left-leaning individuals.

Lorenzo Norris, MD: This is Dr Lorenzo Norris, editor-in-chief of MDedge Psychiatry. It's our pleasure today to welcome Dr Leong to talk with us on the Psychcast about a really interesting topic that people may not have been thinking about in this way: the potential neuroimaging differences involving our politics and partisanship. Dr Leong, welcome to the Psychcast.

Yuan Chang Leong, PhD: Thank you for having me.

Norris: Dr Leong and I are recording this on October 30th. If you're like many of us at this point in time, you are likely suffering from a bit of fatigue with regard to constant [political] ads. You've watched the debates, or as many of them as you could. You have likely vetted many of these issues with friends and colleagues. Your close circle may be supporting one colleague, or you may have a divided house in which people are supporting one or the other candidate. Regardless of where you fit in the political divide, people are looking for different ways in which we can start to think about the current environment.

Dr Leong and his group have conducted a really interesting study that involves the use of functional MRI, looking at how voters or potential voters respond to campaign or political messages. Dr Leong, your study showed three really interesting findings. Can you walk us through some of those findings, maybe starting with this idea of neuropolarization? Just break that down for us.

Leong: Sure. What we did in this study was to bring in participants from the Bay Area community. We brought them into the lab and scanned them using functional MRI while they watched videos related to immigration policy.

The reason for doing this is that I'm sure many of us have observed this phenomenon in our daily lives — where you show the same news footage to different people and they will state they see something different from each other; that they interpret it differently and have different responses to it. As neuroscientists, my collaborators and I were interested in understanding the neural basis of this. That's the motivation behind how and why we ran this study.

What we found was that despite watching the exact same videos, the neural responses between conservative-leaning and liberal-leaning participants diverged in their brains. They diverged specifically in an area of the brain we call the dorsal medial prefrontal cortex. They did not diverge in early sensory areas.

What this suggests is that the two groups were not literally seeing something differently, but instead they were interpreting the information differently. We don't see this divergence of responses in the areas of the brain that are associated with seeing and hearing. Instead, we only see them in this higher-order brain region that cares about interpreting and making sense of the information that you have.

What this suggests is that these divergent interpretations can be observed in the brain as well. We've coined this term "neuropolarization" to sort of parallel the political or the behavioral polarization that we see happening in our society, tying it back to a neural basis for it.

Norris: Neuropolarization. This has got my mind thinking a bit here. People are looking at the same event or the same thing, and they're having different responses. We're talking about dorsal medial prefrontal cortex, higher-order cortical structures. I always think of the dorsal medial prefrontal cortex as handling narrative content. I also think of it and the prefrontal cortex in general in terms of executive function.

Did your group find that there were any differences in subcortical functions, such as amygdala, hippocampus, or anterior cingulate gyrus, before it got to the dorsal prefrontal cortex?

Leong: We didn't see divergence in neural responses in those brain areas. But we did observe a divergence in connectivity to the dorsal medial prefrontal cortex from those brain areas, specifically from the ventral striatum.

What this means is that this brain structure that cares about reward processing responded similarly between the conservative and the liberal participants. But what was different was its connectivity or its correlation with activity in the dorsal medial prefrontal cortex.

Once again, all of this is correlational. I just want to be careful in stating the claim. What it suggests is that the information from the ventral striatum, even though it's represented similarly by the two groups, it's transmitted and conveyed differently to the dorsal medial prefrontal cortex.

Norris: I want to focus on that and dive right in, in terms of that connectivity from the ventral striatum to the dorsal medial prefrontal cortex. What I think of, just as a practical example, is large gatherings of different political groups. You'll have gatherings of large groups for a variety of different reasons. Depending on the message or the media, or even when I'm with different people, people will have very different interpretations of large group gatherings in which people aren't wearing masks. No masks, no social distancing, or anything of that nature. Our interpretations in terms of the intent, the purpose, and the threat, I find fascinating.

When you tell me that what's different is the connectivity and processing, but no difference in terms of the actual sensory information, I find that fascinating, Dr Leong. Please go ahead.

Leong: I should qualify that by saying that this could be due to the fact that many of the videos used were interviews that had two people talking. There wasn't necessarily always much action in them or physical stimuli that could be perceived differently. That could be driving the fact that we didn't find any differences in the sensory areas of the brain.

If we had news — maybe a different set of stimuli that included, for example, the protests or maybe the inauguration where two groups were professing to see two different realities, like physical sensory realities — we might see something different. This could be due to a limitation of our study.

When you don't find anything, you can't claim that it's not there. What we can say is that we did find something, and we did find something in this higher-order brain area, suggesting that we do see differences in interpretation without differences in sensory processing. We don't know whether we would never see differences in sensory processing, given a different situation.

Norris: Dr Leong, I really appreciate you saying "in terms of the limitations of the study." I believe the N was 38 people. The study had 38 people. With that being said, though, on the basis of what you've seen with the study, what type of insight or thoughts does this give us with regard to bias processing or things that we can think about in terms of further understanding neuropolarization?

Leong: I think there are two main things to mention here. One is that as a neuroscientist, I'm interested in the brain. I don't know about you, but a lot of what my brain is doing nowadays is to think about politics, and I think it's important as neuroscientists to understand the mechanisms and processes that are involved in this thinking.

The second thing, as a more societal implication piece of this work, is that I think it shows that political beliefs or prior beliefs and attitudes have a powerful influence over how you perceive, interpret, and respond to new information. As we've shown here, you can watch the same news footage and draw completely opposite conclusions depending on your political affiliation or ideology.

This highlights why it's so difficult to bridge the partisan divide. Maybe trying to persuade partisans with more information by showing them more videos might not be the most effective strategy to do this. If our goal is to reduce neuropolarization and change minds, what we might need to do is to think carefully about what information we show, how we frame it, and how we structure it to reach the most amount of people in a way that's nondivisive.

This study doesn't directly address that. It just shows that we see this divergence. I think what it suggests is that showing the same video is not going to do what you think it's going to do — or maybe people know that. I do think that it's not going to change minds just by showing people facts.

Norris: I find it interesting in terms of how we message things. I'm not sure when it really became popular, this use of red and blue states and this color coding. I'm not sure when I became so aware of it. I know when it starts to come up on the screen, it can absolutely feel like a civil war with red states and blue states.

That's such a black-and-white binary when we talk about polarization, and I think about what you said with regard to the messaging. How can we be more responsible in our messaging in terms of not causing more polarization per se, but giving people the opportunity to exercise that cognitive or executive function and interpret what's being given to them instead of their political beliefs or biases being overly activated?

Leong: I think this calls back to some classic, I would say, sociopsychological studies about minimal paradigms. The idea behind these studies is that all you need to do is to give people a fictitious group label, and suddenly they start to think more and more like each other within the group. They defend and they give more money to people in that group, even though those groups were arbitrarily defined.

These are some classic psychology studies from the 70s and 80s, and they're called "minimal group paradigms." The idea there is that all you need is an arbitrary group label, and suddenly these intergroup processes come into play. As I mentioned and emphasized, they're arbitrary groups. Can you imagine how much more powerful they are if these are groups that are tied into your actual social circle and that they are they are also sort of basically in your face in the news every day where people, as you mentioned, say things like "red states" and "blue states" and create these divisive social categories.

I do think that by highlighting the gaps and the divide, we do run the risk of driving people further apart as people become more sectarian in their beliefs by clinging to their in-group.

The corollary is that maybe one way to bridge this divide is to try to call to a more collective and broader identity that people from both sides of the political spectrum can identify with. If you can find a way to have them consider themselves as one group, for example, as Americans more broadly, that could potentially be more effective in bringing people together.

Norris: I would agree with you, and I love the way that you put that. I want to go back to some of what you described in the study. I'm curious, why did you all choose immigration as the topic in which to test this out?

Leong: There were a couple of reasons. The first reason is that science takes a while. In fact, this particular study took an entire election cycle. We actually ran this study in mid-2016, where it was the first year of the current presidency. At that time, immigration was very polarizing. It was the most polarizing issue. This was the time when there was an executive order banning immigration from seven Muslim majority countries. There was mention of building a border wall. It was a time when immigration was particularly polarizing on both sides. That was one reason.

The second reason is that I personally am an immigrant. It's something that I care about, so it made sense for me to study it in this context. I would say that those were the two main reasons why we decided to run with immigration.

Norris: I'm glad you brought that up. When I think about immigration and I think about what you said, about how can we use different language that at times is less divisive. You mentioned that you're an immigrant. I'm married. My wife is an immigrant from Trinidad and Tobago.

I'm not saying anything that we've never heard before in this country, but we're a country of immigrants. There should be at times, if we can watch our language and how we use it, some common ground. Maybe there already is common ground, but when we get polarized, we just can't see it. Because we are a country of immigrants, we ideally base our strength on that diversity, that diaspora, that melting pot, if you will.

I find it interesting that one of the principles of our country, immigration, we can also view as a threat. I know that that's been historically present whenever there's been a new group of folks coming into the country. But immigration doesn't necessarily have to be this litmus test or this flash or this flare. It could be, depending on how one processes it, something that we feel united around.

Now, with that being said, can you comment a little bit, Dr Leong, in regards to some of the findings that you may have or thoughts about language, and the use of language as it relates to your study findings?

Leong: When we ran the study, we were not only interested in showing that the neurologic responses differ between the two groups as they watch these videos. We also wanted to understand what kind of content was driving these differences.

What we did was we went with the videos, we looked at time points in the videos where we saw the greatest divergence, and we tried to infer from the audio transcript of the video what was the type of content that was particularly effective at driving the neuropolarization that we saw in the study.

What we found was that there were certain categories of words that were particularly powerful in driving these differences. One of these categories was risk- or threat-related words, so such words as "risk," "threat," "security," "safety," "community" — words that had to do with a sense of threat. Segments of the videos with a high proportion of these words tended to also be time points in which we observe great neuropolarization. That's one category of words.

The second category of words included words that were both moral and emotional in nature. They deal with questions of right and wrong, they deal with strong emotions, so such words as "violate," "harm," and also "compassion" — words that tie into both morality and emotions. They also tend to be particularly divisive in the brain driving neural responses.

Norris: That's interesting in terms of those categories of words. When I think about those categories, the first one, safety and community. I actually see those words, in my opinion, used on both sides. When I think of safety and community, I think of those who do not feel safe in their community or that their community or their law enforcement protects them.

On the other hand, I see others that feel that their community isn't safe and that their community is under siege, and the concern with regard to defunding police or attacks on law enforcement or protection that keep their community safe.

When you take a step back, in my humble opinion, these groups have many of the same concerns. They're just coming at them at times from different perspectives. I could see why that is definitely, definitely, definitely triggering.

The other thing that you said in terms of morality and emotions, like such things as right and wrong. I absolutely can see that. Your study was done in 2016-2017 when immigration, the Deferred Action for Childhood Arrivals (DACA), the wall, all of those things were really hot-button issues.

I'm in a medical center, and we were very much concerned about our international students at that point in time, whether they would actually be able to stay, all types of things.

But now, when I think about those things in terms of right or wrong, I would be remiss if I didn't bring up George Floyd and many of the elements that we have seen in the country in terms of police brutality. Again, I'm not trying to get into a political discussion here, but I would be curious — and you can't say one way or another — if you did that study now and you study folks from different political divides, what their functional MRI would look like when they witnessed the events that occurred with the murder of George Floyd.

Leong: I think you bring up a very important point, which is essentially that what our results generalize and how they change with time as the political climate, the zeitgeist, and the things that we talk about change [is a good question], and how they would be different if we were to run this study now.

I will say that, as you've mentioned already, a lot of that that the hot-button issues now are also discussed in terms of right and wrong. They are also discussed in terms of safety and threat. They can be a threat to livelihoods. They can be a threat to life. They can be a threat to even an abstract term, such as our way of life.

I think that even though the issues change, the themes around these topics have been fairly consistent. The actual issues are different, but the themes that we keep returning to — of right and wrong, of emotionality, of threat and safety — seem to be similar.

If we were to run the study again, I would expect to see these same differences. There might be other specific categories of words that may pop up that were not seen 4 years ago. I would say that the stuff that we found about threat and morality and emotions, we would find again.

Norris: I brought up images of George Floyd. I think you could also bring up images of police officers losing their lives. Many police officers have unfortunately lost their lives and have been murdered recently. I also think that you could bring up the opioid epidemic and rural America. I'd also be curious if you brought up coronavirus in terms of what people will look at across the political divide. Dr Leong, again, I think you've simulated critical thought.

Leong: Thank you.

Norris: I believe this [is about] not actually thinking right, left, liberal, conservative. I think about whether I am being mindful of the media that I'm consuming and my own biases, and actually thinking about what it is that I'm looking at. Why I'm responding this way. What makes sense and what's ground truth, what is reality, and then make decisions based on that.

I think you stimulated critical thought, at least for me. As we approach this election cycle again, I just think about, whether it seems that way or not, there are many things that we all share in common. We need to start to work on finding more of those instead of becoming more polarized. Dr Leong, the last words are yours, sir, in terms of take-home points before we approach Election Day.

Leong: I think you've brought up a very, very powerful point. It is part of the message that my collaborators and I hope people take away from this study. There is one thing that I think I would be remiss not mentioning. When people talk to me about this study, many people have focused on the differences between groups.

I think it's also important to emphasize that in this study, we also found many similarities. The neurosignal, for the most part, was very similar. It's only in one part of the brain area that it diverged, and it didn't always diverge. It only diverged in particular points in time, which were particularly divisive. For the most part, we have the same brains. I think it's important to emphasize that commonality.

Also, the things that we care about are very similar on both sides of the political aisle. People care about threats and safety. They care about what is right and what is wrong. In that sense, there is also another commonality that we share as people in this society. I think remembering that, and as you mentioned, slowing down and really thinking of how another person would not share my beliefs or might be thinking of a particular issue, can go a long way.

If we could only take a step back, be empathic, and allow ourselves to consider a perspective of how this would be seen by someone who does not share my assumptions and biases. I think at the end of the day, if people were willing to take that moment to do that, it could be very productive and promising.

Norris: Dr Leong, thank you so very much for appearing on the Psychcast.

Leong: Thank you for having me.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.