Joe Siev studies extreme political behavior and its appeal. He’s a postdoctoral fellow at the Darden School of Business at the University of Virginia. In our conversation, we talk about his research linking people’s sense of ambivalence with their willingness to take extreme action.
Things that come up in this episode:
- Ambivalence and support for extreme political action (Siev & Petty, 2024)
- The social side of political ambivalence (Siev et al., 2024)
- The feeling of certainty and support for extreme public health measures (Siev et al., 2023)
Transcript
Andy Luttrell:
You are listening to Opinion Science, the show about our opinions, where they come from, and how we talk about them. I’m Andy Luttrell and Happy New Year. 2024 was… I mean, yeah, it was. It just was. We did it. 2024 happened and now we’re doing 2025. That seems to be how time works. And here we are, another year of the podcast and there’s a lot of good stuff in store and we’re kicking things off this year with a buddy. Joe Siev is a postdoctoral fellow in consumer behavior at the Darden School of Business at the University of Virginia. But I’ve known Joe since he was a baby grad student. He started as a PhD student at Ohio State the year after I left. But I’d kept in touch with folks in the lab and got to know about this sharp kid doing interesting work, and it’s been fun to follow what he’s done since then.
Today we’ll talk about his research on how people think about extreme things in unique ways, ways that can sometimes undo assumptions that social psychologists have had for a while. This work of his is also related to the experience of ambivalence, which has come up plenty of times on this show. Ambivalence is when something seemed both good and bad at the same time, and people have a hard time knowing what to do with that feeling. For more on ambivalence, you can dig up episode 35 of this show with Iris Schneider, but I say let’s do this. Welcome to 2025. Here’s my conversation with the wonderful Joe Siev.
So the work that we’re going to talk about is sort of a combination of ambivalence and extremism. Those are like… I can’t tell which side of that coin is the more prominent. They seem like equally prominent in the work that you’ve done. But I think maybe to start, we could look at ambivalence just to sort of set that stage. And so could you give your rundown of what it means to be ambivalent and kind of what that’s meant to psychologists up until you stepped on the stage?
Joe Siev:
Absolutely. Ambivalence has multiple definitions and multiple ways of measuring it. So ambivalence in the most general sense means mixed opinions. So holding some evaluations that are positive and some that are negative of some object. So maybe you really like cheesecake, but you also wish that cheesecake was a little bit less caloric. So that could be one way to be ambivalent is having some positive evaluations and some negative. People refer to that as objective ambivalence or potential ambivalence. And what they’re getting at with that is that that has the potential to create subjective feelings of conflict, making you feel torn and internally conflicted. So that’s kind of the other version of ambivalence. There’s just the opinions one, but then that can also make you feel subjectively conflicted. So subjective conflict is what we focused on in this work, so feeling torn. And it’s also worth noting that in addition to feeling subjectively conflicted when you have mixed opinions, people can also feel subjectively conflicted for other reasons.
So for example, if you disagree about something important with someone that you care about, let’s say your partner, that can create some conflict, not because you have mixed evaluations of the thing that you’re evaluating, the cheesecake, but because that disagreement that you have with your partner can also make you feel ambivalent. And there are other ways as well. So expecting that you’re going to encounter information that counter argues your position, even if you don’t have it yet, just anticipating that there might be more to the story than you know about right now can also create a sense of internal conflict. And another version, a fourth way to get there is wanting to feel a different way than you did. So you might hate exercise, but you really wish you liked exercise. And that can create some internal conflict about exercise. So there’s the mixed evaluations and then there’s the internal subjective experience of it.
Andy Luttrell:
I think this will be relevant too. What is it about having this mixed reaction to the world that feels bad? Why would necessarily there be this road between saying, “Yeah, there’s good things and there are bad things. Cheesecake is yummy and it’s bad for me,” and going like, “Oh, but it’s such a problem?” What is that link there?
Joe Siev:
That gets at the other part of your previous question too, which is about what has ambivalence meant to the field over the years? The reason that ambivalence is important and the reason that people care about their own experiences of ambivalence is because it can impede decision making in action. So do you eat the cheesecake or not? Do you go exercise or not? So people are more able to act decisively when they feel pretty clear about one course of action, whether it’s to approach or avoid something, to give it a thumbs up or a thumbs down. But when you feel torn in both directions, it just makes it harder to move in the world. It can paralyze you. And so there’s research that shows that people will delay decisions when they feel ambivalent. They might try to outsource those decisions. It’s just tough to jump off the fence and figure out what to do.
Andy Luttrell:
Yeah, it’s about this longstanding desire for people’s opinions to say something about what they’re going to do. Generally we would say that when I like something or favor something, I’m going to do things that express that or do things that are in line with that. But it sounds like even if I generally cheesecake, if I am kind of torn about it, that liking doesn’t tell me as much about the choice I’m about to make when it comes to what I’m actually going to have before me. So generally ambivalence has been related to doing things less. Like you said, paralyzes people, leads them to delay. But you come along and you try to say, “Actually, let’s change the story a little bit.” And so under what conditions might we actually start to say, “Oh, this story we’ve been telling forever about ambivalence might not apply in every circumstance?”
Joe Siev:
Well, this was a surprise to us too when we started to see some cases where it seems like the opposite can happen and to cut right to the chase, that condition where we see that the effect of ambivalence tends to reverse. So instead of making people less likely to act and more likely to delay, it can make them more decisive and more likely to act. It’s for extreme actions. So rather than the ordinary types of behaviors that have been studied in most research in psychology and definitely in this area… That could be things like people’s willingness to vote. So if you feel ambivalent about a candidate or your choice between candidates, that makes you less likely to vote.
But in that same domain of politics… There’s a lot of interest right now in partisan animosity and support for partisan violence, for example. We see that when we get into these more extreme kind of counter normative types of actions, we see that it’s actually the opposite. It’s that people are more willing to support or even say that they’re more willing to act in extreme ways for beliefs that they feel conflicted about. So extremity seems to reverse this general tendency that we’ve known about for a long time.
Andy Luttrell:
So what is it about a behavior that makes it extreme? You mentioned political violence, but in general, what’s an extreme behavior and how is it different from a more moderate behavior?
Joe Siev:
Good question. The way that we define this is by thinking about behaviors in a particular attitudinal domain. So let’s say that we’re talking about Covid-19 mitigation. That can be our domain. And so if I ask you how do you feel about policies that are supposed to mitigate the spread of Covid-19, assuming that we’re still in the pandemic for this hypothetical, you might have an evaluation of that. You might be relatively favorable toward that and you might feel conflicted about it or you might not. But I could also ask you questions about actions that you could take that would be consistent with your position in that domain. So you could wear a mask. That would be a behavior that would be consistent with being pro-Covid mitigation, but it would be relatively ordinary in the sense that it’s common. It’s something that everyone does. And if you do it too, you’re going to be aligned with that norm.
You’ll be with the crowd. Everyone is doing this thing and you’ll be doing it too. That makes it an ordinary or moderate behavior. But what would make a behavior in that domain extreme is if it’s really quite unusual. And behaviors tend to be unusual when there are reasons for not doing them. So political violence is a very clear example of that. There are lots of reasons why most of us would not support, let alone engage in political violence. But there are some who do. Or you could take it down a notch, talk about something that’s maybe a bit extreme, but less so, getting into an argument with someone about politics in a very confrontational way. That would be more extreme because it’s really quite unusual and that unusualness has important implications.
So one important implication that we think it has here is that it sends a very strong signal. If you do something that most people are not willing to do and you express your position that way, it sends a very clear message about where you stand. And we think that that could be part of the reason that this happens, that people are more supportive of extreme actions when they feel ambivalent potentially because they want that strong signal that’s attractive to them. So the unusualness of a behavior makes it extreme, and because it’s unusual, it can send a very strong signal about who that person is on the inside.
Andy Luttrell:
So here’s where I’m going to careen into the philosophical, and this is why when I teach grad seminars, my students hate me because I get really hung up on these definitions. I remember there was one time where a student was like, “I’m interested in authenticity.” And before I could even say it, she’s like, “No, I don’t know what that means.” So that’s the thing. So I’m a little hung up still on… It’s one of those things where… When you say extreme behavior, everybody gets it. Everybody gets why violence is extreme and what was it, reading information is not extreme.
Joe Siev:
Yes, that was another one.
Andy Luttrell:
Yeah, I was looking on the list, “What was the least extreme?” And it was just reading information about the topic.
Joe Siev:
Pro-attitude. So reading information supports your view,
Andy Luttrell:
Right, right. So not a radical action, which makes intuitive sense, but I was a little surprised when you were talking about extremism that you kind of defined it within how typical it is. Does it seem like that is definitely part of the definition or just it tends to be the case that extreme actions are unusual and less extreme actions are common? Or is that baked into the definition, or is that just kind of a convenient way to describe the difference?
Joe Siev:
I think it’s baked into the definition. There’s definitely room for multiple perspectives on this. I think it’s based into the definition. There’s literature that precedes ours, which focuses on what they call frequency and intensity as the important determinants of what makes something extreme. And the argument basically is that when we call something extreme, we mean that it’s unusual, but things are unusual for a reason. And the reason that they’re unusual is because very intense in some way. So maybe they are very costly for a person to engage in. You wouldn’t donate all of your belongings, even if you really believe in a cause because that’s extraordinarily costly for you. Or it could be something that’s dangerous. Engaging in violence, in addition to whatever other qualms we might have about it, is quite risky for the person doing it.
So that would be the most formal definition I think that I would advocate for is that yes, that unusualness is integral to it and things become unusual because there are strong deterrents against engaging in them. It’s also worth adding that in common language people often use the term extreme in a evaluative sense. So if something’s extreme, it’s usually not a compliment. So there may be kind of a more value neutral and a less value neutral way of thinking about extremity. Something is extreme just descriptively, it’s kind of at the tail end of the distribution versus something is extreme, we should all object to it. So in terms of common language, I think both of those are used. In terms of a formal or theoretical definition, I would start with that unusualness piece.
Andy Luttrell:
But nevertheless, things can be unusual that are not necessarily extreme. I think that was the counterfactual that I was getting distracted by.
Joe Siev:
Yes. So I guess the intensity piece comes in there. Maybe I shouldn’t overemphasize frequency relative to intensity. In some of the work that we’ve done on this, we’ve used the language of unusualness and strength because we were trying to couch it in the attitude strength framework. So if you have an unusually strong opinion, then that’s extreme. If you have an unusually weak opinion, perhaps that’s less extreme. Or if you… Yeah, stop there.
Andy Luttrell:
Yeah, we don’t need to spiral too much. I mean I could, but for the sake of people listening, I’ll cut it off there. But one thing that I think is useful about the way that you do things is it’s not always up to you, the researcher, whether this behavior is extreme. You kind of outsource that decision to people who make these decisions for themselves. And so operationally, how do you distinguish between extreme and less extreme behaviors in the research that you’ve done?
Joe Siev:
Yeah, good question. So we asked people. We asked people, “How extreme is this behavior?” And we relied on their judgments about that. But we did it in a way that allowed us to separate those perceptions of how extreme the behavior is from the people who were expressing willingness versus not. So we had a bunch of responses from people who reported their ambivalence about some issues, so let’s say Covid-19 mitigation policies and their willingness to engage in these behaviors like wearing a mask or confronting someone who refuses to wear a mask. But then we had other participants rate those behaviors for extremity. And so wearing a mask was rated by our participants as relatively non-extreme. And confronting someone for not wearing a mask was rated as quite a bit more extreme. And then committing violence on behalf of your position on that issue was considered kind of the most extreme, near the maximum of the scale. And so then we just combined those data sets and that allowed us to use those extremity ratings as a predictor of how ambivalence and extremity were relating to each other.
Andy Luttrell:
And presumably there’s a fair amount of consensus around what’s extreme versus not. Did you get the sense that it’s like everyone goes, they go, “I don’t care what Andy is talking about. We all agree this is an extreme behavior?”
Joe Siev:
Yeah, there’s a decent amount of consensus, definitely when you talk about comparisons between behaviors. People could use the scales a little bit differently. Some people might be reluctant to give maximum ratings or there are kinds of response sets that could create some variability, but pretty much everyone is going to agree, and in our data, this was the case that violence is extremely extreme. Confrontational behavior, let’s say, or taking a risk to your own well-being, medically, let’s say, participating in a vaccine trial or something like that, also quite extreme, but maybe a little bit less so. And then everyone also agrees that compared to things like that, wearing a mask or having an ordinary conversation with someone about your opinion or voting, these things are totally ordinary. So there is a lot of consensus about that. Things could change in a different situation.
If the situation itself were extreme or highly threatening, then the extremity of behaviors could change. There’s this quote from Malcolm X that I don’t remember exactly, but the thrust of it is, “In an extreme situation, the only rational thing to do is to respond extremely.” He was talking about his agenda, he was talking about history of racism in the country and describing that as extreme and saying, “We need extreme measures to counteract that.” So similarly, if you’re in a dangerous situation, then dangerous responses are going to seem less extreme, but we’re talking about general ordinary life and that’s the context that our participants were responding in.
Andy Luttrell:
So just to refocus us back on the main thrust here, so when you get these kinds of responses, you can distinguish between extreme and less extreme behaviors, you have a sense of how ambivalent people are about these kinds of issues and behaviors. What’s the pattern of findings that keeps creeping up when you look at these data?
Joe Siev:
The pattern of findings is that the relationship from ambivalence to willingness to engage in these behaviors changes completely when we look at the ordinary behaviors versus the extreme ones. And so statistically this is moderation, this is an interaction effect, but in normal person terms, what this means is that we see a negative correlation when we’re talking about ordinary actions. So for example, the more ambivalent a person says they are, the less willing they say they are to vote or to wear a mask if that’s the topic. But simultaneously, the more willing they say they are to engage in confrontational or risky or even violent behavior. And so there a positive relationship emerges. So this same question that we ask people, “How mixed or conflicted or indecisive do you feel about your position?” It’s negatively correlated with some behaviors, the ordinary ones, and it’s positively correlated with the extreme behaviors.
Andy Luttrell:
And I can’t remember now if we already covered this part, but why should more ambivalent people be more willing to engage in extreme action?
Joe Siev:
This was the $64,000 question and what we spent a lot of time trying to pack and understand. And the reason that we think that this happens is because ambivalence is uncomfortable. Essentially it creates a motivation to compensate. We really mean this in the same sense that people mean it when they talk about somebody overcompensating in just day-to-day context. It’s behavior that’s meant to counteract some kind of tension or discomfort or psychological threat that a person is feeling. So in this case, what might be the nature of the discomfort goes back to what we talked about earlier, that ambivalence can be paralyzing. It impedes action. It makes it harder for you to decide. And that’s difficult. And not only that, it might be especially difficult when we’re talking about an issue that people feel strongly about.
So that’s true about political issues in general, but it’s especially true for people who care a lot about politics, and that’s something that we find is that people whose opinions on politics are quite strong, so maybe they are ideologically extreme or it might just be that they have a very extreme attitude on that issue if it’s a specific issue like, let’s say, gun control or Covid-19 policies, those people show this effect of ambivalence on extreme behavior especially strongly, which is consistent with this discomfort and compensation story because it seems that it’s especially disconcerting for people to feel mixed about something that’s quite important to them. So that combination seems to be especially potent and creates this kind of defensive or compensatory drive.
Andy Luttrell:
And these extreme behaviors just like are a quick way for me to establish, “This is where I stand. It’s clear to me, it’s clear to others. I don’t have to feel so conflicted about this if I could just take some decisive extreme action.” And ultimately that’s presumably why they’re doing this.
Joe Siev:
Yes, that does seem to be why they’re doing this. They have something to prove and this is an effective way to prove it. So it’s clear, it’s efficient, other people aren’t doing it, and it allows people to kind of rectify that discomfort or to try to.
Andy Luttrell:
In the paper you talk… There’s a conclusion that I didn’t quite follow, which was that this isn’t about resolving the ambivalence, it’s about something else. Could you explain a little bit what that means? Because typically what ambivalent people would say is, “I feel so mixed up about this, that what I really want to do is address that discomfort by deciding where I stand and that’s going to help me feel better.” And it sort of seemed like this would fit that, that you’d go, “What I’m ultimately trying to do is to escape this feeling by taking decisive action.” So what do you mean that that’s not the case? Or what evidence makes you think that that’s not quite what’s happening?
Joe Siev:
We thought that that would be what’s happening. That was our initial expectation, just like you and it’s functional. If you need to make a decision and you’re conflicted, if you can get over your ambivalence, then you can go ahead and make your decision. We thought that that might be what’s going on, and we wanted to gain some evidence for that possibility, but we also were aware that there could be other reasons for it. So maybe to backtrack a little bit, it’s probably worth noting that in general, outside of the ambivalence context, when people experience psychological threats or discomfort, there’s research that shows that they can respond in a variety of different ways. So one way is they can try to address the source of the problem. They can get rid of whatever’s making them uncomfortable. So when we talk about ambivalence, that’s kind of what you described.
They would be trying to reduce their ambivalence and take a position on one side, but that’s not always an option or a desirable. For any number of reasons, people might not think that they can address the source of the problem. And in that case, people do different things. They might try to distract themselves or they might try to kind of symbolize and kind of just signal that they have this characteristic that they feel insecure about, even if they don’t really, and that could also be what’s going on here. And so to get back to the ambivalence context, we wanted to manipulate that motivation to reduce the ambivalence versus not. Because if we enhance the motivation to reduce ambivalence and that’s what people are doing, then that should strengthen the effect. But if we reduce the motive to reduce ambivalence, if we make people comfortable at being ambivalent so they don’t want to change it anymore, then the effect should go away if that’s what’s happening. But if it’s not about reducing the ambivalence, then we wouldn’t expect that to happen. In fact, even the opposite could happen.
If people are not trying to take a side, they’re just kind of acting out and compensating for their discomfort, then making them more comfortable with their ambivalence is not going to change anything because maybe they feel comfortable with their ambivalence or that’s just not the problem. And so to make a long story short, we manipulated those beliefs about whether ambivalence was a good thing or not. And what we found was that it doesn’t seem like people are trying to reduce their ambivalence. When we convinced people that ambivalence is a good thing… So in one study, we just described it as reflecting, being fair, and unbiased and knowledgeable. On another study we talked about how the political domain is full of trade-offs and it’s complicated, and so it’s appropriate to have some mixedness.
What we found is that the effect was still there, and if anything, perhaps even a little bit stronger, but it was when we made people doubt their ambivalence, when we said that, “A lot of political issues are really pretty one-sided here,” or, “If you’re ambivalent, it kind of makes you wishy-washy, unreliable. Why can’t you just make up your mind?” That’s when the effect went away. And so what that suggests is that people aren’t trying to reduce their ambivalence when they endorse these extreme actions. They’re reacting to how they feel, but it’s not about trying to reduce ambivalence, it’s about symbolizing.
Andy Luttrell:
So it’s as though people can be saying like, “Oh, yeah, yeah, I know stuff is complicated and I’m conflicted and that’s fine, but God does it feel bad, so sign me up for this extreme action.” And so you don’t have a sense that it’s not as though people didn’t resolve their ambivalence. You don’t have that kind of information, but you just know that when you allow the comfort of being okay with ambivalence, they’re still or even more so having this defensive response to just feeling yucky about not having reached some conclusion.
Joe Siev:
Yep, exactly. It seems like there’s something about this combination of negative affect or bad feelings and a positive interpretation of those feelings or of the source of those feelings. So kind of feeling uncomfortable and saying, “That’s valid, that’s justified. I’m right to feel this way,” that seems to be an especially potent combination. I do wonder about the generality of that formula of negative affect and positive cognition as potentially a cause of extreme actions. But ambivalence, especially in politics, is really interesting because it’s a natural… Those two factors are naturally confluent there because we know that ambivalence is uncomfortable. And we also find, as in the study that I was just describing, that people in general seem to be relatively favorable toward ambivalence about politics. At least at this moment in history in the United States, that seems to be the case, whereas it’s very different from other domains where ambivalence has been studied.
So for example, racial ambivalence, where we’ve also seen research that’s focused on compensating for feeling uncomfortable about having ambivalent attitudes toward a racial outgroup. People don’t want to feel ambivalent about racial outgroups generally speaking. If you value egalitarianism and you have negative reactions toward a racial outgroup member, that’s not something that you want to maintain most likely. And so politics is quite different from some other domains. Relationships is another example that ambivalence has been studied quite a bit in, and generally speaking, people would rather not be ambivalent about their relationship and they think that ideally they won’t be. But I think politics can be different in that regard.
Andy Luttrell:
A slightly different take, and maybe I haven’t thought through every corner of this, but it reminds me a little bit of that old cognitive dissonance study that gave people a placebo and they’re like, “This is going to make you feel terrible.” And that kind of wiped out this conflict effect because they were like, “I feel so bad. Oh, but it’s this irrelevant thing that made me feel bad.” But the opposite was that if they say, “Oh, this is going to calm you down, this pill’s going to make you feel so good,” that’s where they jack up their reactions because people are like, “Well, I feel uncomfortable because I did something wrong and you said I’m supposed to feel good, and so I don’t know what to do with myself now.” It sort of seems like something like that could be happening too, where you’re like, “Oh, don’t worry. Everything’s fine. You’re supposed to feel that way.” And they’re like, “But I feel bad is the thing. I feel bad and it can’t be because of this. It must be something else.”
And so that’s what provokes the defensive response. Whereas if it’s the opposite… That’s why I can’t tell how well it applies to the other condition you have, but I don’t know, I’ll throw it to you. That seems at all relevant?
Joe Siev:
Yes. One example that we use in the paper and some of the other writing that we’ve done on this is the analogy to fragile self-esteem. So there’s research that shows that people who think very highly of themselves but feel conflicted about that, they have maybe some doubts about it in their heart of hearts, can behave differently in some ways. So imagine that there are two bosses. There’s one boss who thinks highly of themselves and feels quite secure and confident about that, and another boss who also thinks quite highly of themselves, but secretly maybe they have imposter syndrome or they feel insecure about their standing. You can imagine giving feedback, let’s say, some kind of criticism to the boss. You made a mistake or you mixed something up.
I think intuitively we can anticipate different kinds of reactions from those individuals, maybe a more restrained kind of action that’s just focused on responding to the issue for the secure boss, but the insecure boss is going to feel uncomfortable with that criticism and might lash out. And so imagine if you told that boss, “Hey, you’re right to have imposter syndrome. That’s good. Good read on this situation. You are an imposter.” So it’s a little bit like that, telling people that they’re right to feel off.
Andy Luttrell:
Yeah, “You’re telling me things are okay, but I still feel bad,” kind of feels like the experience that’s being captured in that condition.
Joe Siev:
Yes, I think so.
Andy Luttrell:
The other side of the coin is interesting too, which is that… So you find support for this idea that the more ambivalent people are, the more willing they are to engage in extreme action. But it’s also worth noting that even in the same sets of data, you find support for the story we thought we knew all along, which is that for other kinds of behaviors, more ambivalence is related to being less willing or interested in doing those things.
And I wondered if you think that makes us reinterpret that pattern, which is to say that is it that all along the issue has just been that we’ve been observing people’s willingness to do things that can’t address the feelings that they’re having, and it’s not so much that they have a weak attitude that provides them no guidance in these situations? That’s usually the story we tell. It’s like, “Oh, ambivalence means… I don’t really… I’m just kind of mixed. I don’t have any direction to go in.” But maybe instead it’s just that like, “I’ve you feel something that this behavior can’t help you with, and so you just kind of ignore it as an option for you.” I don’t know. That was my read. It’s like a radical retake on what we know before. I wonder if you’ve thought anything about that.
Joe Siev:
Yeah, it’s interesting. One finding that was almost an afterthought initially, but I think is pretty interesting is that when we measured those moderate behaviors after convincing people that either their ambivalence was a good thing or a bad thing, we found that typical negative effect that you’re talking about for those ordinary behavior is when people interpreted their ambivalence is a bad thing. So why does that effect happen in general? It might be that those behaviors don’t appeal to people who are experiencing ambivalence because they’re really looking for something else with more signal value, but it could also have something to do with the domain and how people are interpreting ambivalence in that domain. So a lot of research on ambivalence… I’m not sure I’m directly answering your question here, so feel free to follow up and let me know. But a lot of research on ambivalence looks at, let’s say, health behaviors or purchasing behaviors where ambivalence really is kind of an impediment.
It’s not the goal. It’s not likely to be seen as appropriate or something that you want to maintain. And we see those negative effects there usually, but in the politics domain, at least here, we needed to make the interpretation negative in order to find the effect. So I think that’s a little bit different from what you’re describing, but it speaks to the openness of the question. We’ve just kind of taken it for granted that ambivalence reduces action without exactly knowing why that’s the case. And the fact that it’s a double-sided property that can both increase and decrease behavior cracks it open even more. So I’m not sure we know the answer to that question yet.
Andy Luttrell:
Yeah, I think that’s mostly what I was responding to is it just sort of makes us go, “Oh, wait a minute. Maybe we don’t actually quite know what was going on in all of those studies that showed this one pattern and the ambivalent…” Well, this is maybe an aside we can talk about later, but ambivalence just strikes me as this idea about people’s relationship to their opinions that seems really important in the way people make decisions and do things, but researchers have loaded up the concept with a bunch of baggage, and we’re kind of at a point where we go like, “Well, wait, what is it really?” Maybe we need to peel this back a little bit again to go, “What does it mean to be conflicted? Could that mean different things to different people in different circumstances?” And we’ve been, I think, maybe a little too quick to just characterize that as one solitary experience that might actually be better understood as something a little more complicated.
Joe Siev:
Yes, it tracks at least two underlying psychological dimensions. It’s one variable and it makes sense that way, but it attracts at least two things, one of which is kind of how assured and ready you feel to act, and the other is how much tension or discomfort you feel. And those things could potentially move independently even, but most of the time they’re going to track opposite poles of ambivalence. When ambivalence is low, that tends to mean feeling ready, kind of prepared and assured about your course of action. And when it’s high, it at least can mean not only that you kind of lack preparedness or readiness, but also that you’re feeling uncomfortable. Certainty is also like that. Uncertainty is not just a lack of clarity or knowledge that you’re correct. It can also… And this is maybe more obvious point because uncertainty even outside of the attitudes and persuasion domain is so widely studied. But again, that can indicate not just the absence of something but the presence of something else. And that’s something that hasn’t really been incorporated into the literature on these topics yet.
Andy Luttrell:
I guess that’s a reasonable point to note that you’ve seen similar things with uncertainty and extreme behaviors too. So in a similar sort of way, I wonder if you see them as overlapping as I do, which is that when people are super uncertain, usually we would say that they don’t act as much because they don’t really know what to do, they don’t have a strong guiding framework for what choices to make. But you found in other contexts that when people are more uncertain, they’re more willing to engage in behavior when it’s extreme behavior and for kind of the same reason. Am I getting…? Those seem like the same kind of story?
Joe Siev:
Exactly. Yeah. We have another paper where we focus on certainty and uncertainty. We found the same reversal pattern. We also found that that was strengthened when the attitudes were strong or when the domain was perceived as especially threatening. So that also speaks to this kind of tension and compensation story. When people feel uncertain in a threatening context, that’s when it really seems to create this kind of response. If the domain is less threatening, the effect is much smaller or could potentially go away. We did also compare ambivalence and certainty to each other in this most recent paper, which focused on ambivalence. And they interact, and there are some interesting things that happen when you for each other and see how they predict the behaviors. But the interaction is quite interesting because it parallels, for one thing, some of the work that you’ve done where you’ve seen that certainty increases the impact of ambivalence.
So you’ve looked at that in terms of durability over time. So the more ambivalent a person’s attitude is, the less likely it is to stay the same over time, especially if they’re despite being ambivalent, confident about it. Am I getting this right? So the same kind of interaction emerges, but the reverse for extreme behaviors. So the more ambivalent a person is, the more willing they are to engage in extreme actions, especially the more confident they are. So confidence and ambivalence at the same time. It’s a bit like ambivalence and extreme ideological views or polarized attitudes, that combination that enhances the effect makes it even stronger.
Andy Luttrell:
So it’s sort of like, “I feel conflicted, but I also feel like I know what I’m talking about and that’s where I feel like I have the most to prove to myself and others. So I’m going to take this action to just make it obvious where I stand and what’s going on-“
Joe Siev:
Yep.
Andy Luttrell:
“Because otherwise it just feels too uncomfortable to just stew in this.”
Joe Siev:
Exactly. Yep.
Andy Luttrell:
So one other thing, you mentioned the nature of the topics, and I noticed I think that in all of the cases that you’ve looked at this effect of uncertainty and ambivalence, it’s pretty much all in political or politicized domains.
Joe Siev:
Yes.
Andy Luttrell:
Does that seem like an important part of the puzzle or it just happens to be where we care a lot about extremism?
Joe Siev:
It is an area where we care a lot about extremism, and so that’s part of the motivation for that focus. We came to understand over time that politics and politicized domains have qualities that probably lend themselves to the effect emerging. So it’s an area that a lot of people care about. It’s also an area that has two sides that many people have some sympathy toward. So it has both the conflict and the about something important ingredient. I am also really interested in seeing whether similar kinds of things could happen in other domains where maybe it would be less intuitive or less obvious.
The example that I always go to in my head, and I’m not saying this is exactly what happened or that this is accurate, but I wonder the slap, the Oscars’ slap, what was going through Will Smith’s mind in that moment? One possible story that you could tell is that he was compensating for discomfort about his relationship, which we now know was quite conflicted, but maybe he felt some commitment to respond. It seems like there was something compensatory happening there about discomfort in that situation and responding in a way that’s kind of miscalibrated, it’s off. That’s what people are getting at when they talk about overcompensating. It’s like, “Man, what is that person doing? That seems so-“
Andy Luttrell:
What is that?
Joe Siev:
So I wonder if at least in some cases, similar things could happen in other contexts, whether it’s ambivalence about a relationship or about other people. But we focused on politics because we started the project thinking about extremism. So it was like, “Which beliefs lend themselves to extreme actions?” The literature on extremism has focused lots on individual level factors, so which people are most susceptible, and some on situational factors, what kinds of social environments and networks kind of make extreme actions more likely. But the belief-based perspective, which beliefs lend themselves most to extreme actions, hadn’t really gotten as much attention. So we started with trying to understand extremism there, and so that naturally lent itself to focusing on politics.
Andy Luttrell:
Great. Well, I think the slap is a perfect place to end, so I’ll call it there and say thanks for taking the time to talk about all this stuff.
Joe Siev:
Thanks so much, Andy. This is great.
Andy Luttrell:
All righty. That’ll do it for this episode of Opinion Science. Thank you so much to Joe for taking the time to be on the show. I know he’s been a loyal listener since the beginning, so it was fun to get him on the podcast. For more about him, you’ll find a link to his website in the show notes. Be sure you’re following or subscribed to Opinion Science wherever you happen to be listening to this. You can go to OpinionSciencePodcast.com for past episodes and other fun stuff. And also thanks to folks who have pitched in financially to keep the show running. Things like web hosting and recording services carry a price tag. And I’m also hoping to do a few more in-person recordings at a local radio station. I’ve got some of that in the works right now, and I’m also getting back to having episodes transcribed, which had to be paused for a bit.
So if you’re up for kicking in a few bucks to help these things happen, you can go to OpinionSciencePodcast.com/donate. And that reminds me, we watch America’s Test Kitchen as a TV show with my daughter all the time. And the other day she told me… I asked her where we live, and she said, “We live in the United States of America’sTestKitchen.com/tv.” Anyhow, also rate and review the show if you can. The podcast has a nice collection of five star on Apple podcasts, but it’s been a minute since we got a new one. So you can make your review next. And if you write it, if you write something nice, I’ll read it. I’ll read it on the show if that… Sometimes people like when that happens. So I promise I’ll do that and there’s your incentive. Okay, I think that’s all for me. Happy New Year. May all your wishes come true, and I’ll see you next month for more Opinion science. Bye-bye.
