More than half a century of empirical research has documented a range of psychological factors underpinning ideological differences, including correlations between conservatism and authoritarianism, dogmatism, prejudice, and justification of inequality, among other things. Some people think conservatives have gotten a bad rap from the science.
One such school of thought is represented by moral foundations theory, which frames itself in reaction to another decades-long line of research into the nature of moral development, focused first on issues of justice, pioneered by Lawrence Kohlberg, and then on issues of care and harm prevention, pioneered by Carol Gilligan. Moral foundations theorists argue that these two issues represent only an “individualist” subrange of traditional moral concerns, and that consideration of the full range of moral concerns—including issues such as sanctity, ingroup loyalty, and respect for authority—would show everyone to be more or less equally moral.
I recently interviewed political psychologist John Jost, delving into some of the background of these conflicting accounts, and the problems he sees with the moral foundations approach. “I and others have pointed out that many of the worst atrocities in human history have been committed not merely in the name of group loyalty, obedience to authority, and the enforcement of purity standards, but because of a faithful application of these principles,” Jost pointed out. While he has no problem with moral foundations descriptively, as an account of what has motivated different people, he is sharply critical of making the leap to accepting such claims as truly moral—as what should motivate us going forward. And he's done studies showing precisely the same sorts of moral contradictions that he alluded to in a more mundane experimental setting.
But there's another whole perspective on what might be wrong with moral foundations theory—a perspective which perhaps in some ways cuts even deeper than Jost's does. This perspective also accepts a good deal of the descriptive findings involved, but not all of them. Or perhaps more accurately, it draws sharp attention to crucial missing findings that have been glossed over. Moral foundations theory posits a small set of distinctly different moral concerns, and it correlates those moral concerns with distinctly different emotions: harm with anger, sanctity with disgust, etc. In fact, it has been argued that moral judgements are essentially post-hoc rationalizations of distinct instinctual emotions. But the actual evidence for such distinct basic emotions is noticeably quite weak—indeed, it's virtually non-existent.
In contrast to the basic emotions interpretation, there's another long-standing approach to psychology, known as constructionism, which has been making quite a comeback recently, taking note of the gaps in the basic emotions story, and taking advantage of new technological advances in brain imaging to study what's really going on in our brains. If the constructionist account is true, then moral foundations theory is not an account of how humans are hardwired. It may provide a useful map of where we've been in the past—of social mores more than morality—but it doesn't provide us with maps for the new world we may want to shape in the future. For that, we may need a new set of maps entirely.
Such were the broad outlines of issues raised for me reading the recently-published paper “A Constructionist Review of Morality and Emotions: No Evidence for Specific Links Between Moral Content and Discrete Emotions.” I reached out to the lead author, Daryl Cameron, of the University of Iowa, beginning with an email exchange and leading to a lengthy phone interview. The two have been merged together and edited for clarity and length.
One thing I'd like to ask up front, as it may help guide my thinking: I know this approach represents a challenge to moral foundations theory--which has some other, more obvious problems, which I interviewed John Jost about recently. In particular, some of these “moral” foundations appear to motivate behavior that's hard to see as anything but immoral. Setting that to the side, however, your work strongly suggests a different sort of challenge, as part of a broader difference in how you see the psychological world. So I wanted to start with any thoughts you have that could be helpful along these lines.
Our paper is trying to ask some deep questions about “what is morality?” and “what is an emotion?” Within moral psychology, there has been a dominant way of thinking about the answer to both of these questions. Moral psychologists tend to talk about moral foundations as if they are distinct, evolved mechanisms of the mind. Jon Haidt has used the taste bud analogy—just as we have unique taste receptors for sweet, salty, savory, sour, etc., we have unique “moral receptors” for harm, fairness, authority, loyalty, and purity. Similar claims have been made about emotions. Some of the classic cross cultural work by Paul Ekman and others going back to the 1960’s suggests that we have “basic emotions” that are universal, distinct, evolved mechanisms (E.g., disgust, anger, fear, sadness, happiness). Given the similarity between these lines of thinking, it’s not surprising that researchers began investigating whether certain kinds of morality correspond to certain kinds of emotion. I.e., disgust with purity, anger with harm. In other words, the “purity taste receptor” is uniquely responsive to disgust, just as the salty taste receptor is uniquely responsive to salty foods.
We looked at the published studies and find that there isn’t support for these strong 1:1 links between specific kinds of morality and specific kinds of emotion. We provide a few methodological reasons that limit interpretation of past work, which I’m happy to go into further. We then provide a positive account of why these specific links do not emerge, by discussing “constructionist” theories of the mind. Constructionism generally suggests that discrete mental states, such as an emotion or a moral judgment, are built from more basic lower-level ingredients. A common example is how a cake is really built from ingredients of butter, sugar, and flour. In the case of emotions and morality, some of these ingredients can include how good/bad you feel (valence) and how excited/tired you feel (arousal). It may just be that anytime you’re feeling worked up from too much coffee or sluggish because you haven’t had enough coffee, that these changes in arousal shape moral decision-making.
So what’s the take away message for readers who are interested in morality? The taste receptor example breaks down as a metaphor for moral experience—there are not be specific relationships between certain emotions and certain moral foundations (and more broadly, we can question how distinct moral foundations and different emotions really are)—so instead we should look at underlying factors such as how good/bad you feel or how excited/lethargic you feel, to better understand everyday morality.
I know that there's a long history to the tradition of constructionism, even though it's been in the shadows more recently. Could you talk a little bit about that history, and what it alerts us to that we might otherwise miss?
Historically constructionism goes back to William James in the 1890s and earlier. Classically there's been this tension between constructionists and basic emotions since the inception of psychology as a science. William James, in this paper we have a quote of his that there are as many types of emotions as there are rocks in a New England farm. In other words, there's immense variety to the point where cataloging how many emotions are there across the world—Are there six universal emotions? Or seven? Or eight? etc.– I mean James said that that was kind of a fools errand. There's going to be such a large number that really, you're just kind of creating a taxonomy that doesn't really tell us much about why we have certain emotion, and how they influence us in everyday life.
Back in the 1960s there was a classic studies by Schachter and Singer, where basically it's kind of old-school, classical, confederate actors creating a real-time situation for participants. I put you in a room and make you feel a certain level of arousal, but then I shift what you attend to to try to create different kinds of emotions. The idea was that emotions are constructed from basic parts that can be moved around in interesting ways. But then, in the past say five or six years, it's really kind of taken off within modern affective science, primarily because of the work with neuroscience, this kind of higher resolution that MRI and other kinds of techniques can provide that adds a new level of sophistication to the many of these questions.
In your paper you discuss two types of feelings, affect and emotion. What is the difference between them and how are they related?
The way we define affect—we vary between calling it affect and core affect—the basic idea is the continuous barometer of how do you relate to your environment; it's always there and it involves two dimensions. It involves how good or bad are you currently feeling? And how aroused, or worked up are you currently feeling? Those are independent things. So you could be really upset, feeling negative, but also highly around. An emotional state which has that kind of profile would be something like anger, disgust, fear.
On the other hand, you can have a negative valence states that are low arousal, like sadness. When you're sad, you feel bad, but you don't necessarily feel aroused or worked up, you tend to actually feel under-aroused. Same thing with positive affect. You have a high arousal positive affect state, like excitement, joy. But you also have low arousal positive affect states, like contentment serenity and so on. So affect is essentially those two dimensions together. It's always there. It's a way for you to code your relationship to your environment.
Now, emotion is different. Affect is just one piece of a motion. It's like flour butter, affect is one of the basic ingredients that goes into an emotion, but it's not the whole thing. You need to also have conceptual knowledge about the world and what you're feeling, and that together with core affect combines in the quick unconscious categorization process to create the experience of the discrete emotion, like disgust anger or fear.
One useful metaphor for this that's been used is like the light spectrum, core affect is like a spectrum of light, it's continuous, it's always there. It doesn't contain actual colors, it doesn't contain an area that corresponds to red and green and blue and so forth. We have concepts about colors—like red, green and blue—and we use those concepts to carve out space on that spectrum, and then create the experience with discrete color. So constructionists think that emotions, they're all mental states, so you have core affect, and then you have conceptual knowledge, and then those combine, using conceptual knowledge to categorize the core affect as a specific kind of emotion. This isn't necessarily a conscious process. It's not like I'm feeling a certain way and then I think, “Okay how my feeling? I must feel anger.” Typically this is thought to be quick and reflexive. We don't notice the categorization process happening.
So, how do constructionists understand this difference, as opposed to those who aren't construction, those who believe in basic emotions, for example, and others with somewhat similar theories?
In the paper, Figure 1 shows you a continuum of the deep philosophical assumptions that separate basic emotion theory on the one hand, and on the other end constructionism. One of the easiest ways to think about this is, basic emotion theorists are committed to a very strong claim that there are between six and eight universal emotions. So you hear about disgust, anger, fear, sadness, happiness, surprise, maybe contempt, maybe some others.
What they're committed to is the idea that these are distinct biological mechanisms. These are distinct things, they're encoded over evolutionary history into human beings. This is why past work suggests that people can universally recognize emotions across cultures, and the specific claims therein are that there is a single facial expression that corresponds to a specific emotion. We call these “whole number accounts,” they posit a whole number of emotions, like six to eight. But each emotion has a distinctive facial expression, a distinctive brain signature, a distinctive physiological profile, and a distinctive judgment that goes with it, and a distinctive kind of behavior that goes with it. That's the strong form of the claim—that's what's universal across human species are these seven or eight distinct emotions. So you often hear the claim that these emotions are natural kinds in the world. There like biological kinds, built into human beings.
What constructionists differ on is, well clearly people understand what anger and disgust in fear mean, and they seem to stay, when we consciously experience emotion they seem they feel distinct, but just because they feel distinct doesn't license the inference that, on a biological level they actually are. So what constructionists would say is universal in the human species is we all have core affect, this underlying continuum of good or bad, and arousal, how aroused you feel, and also this ability to categorize things, the concepts that we have. These are things that we see in both humans, and to some degree, nonhumans. That's what's universal. But anything else is culturally- and person-specific.
So, when I categorize my core affect as disgust, I have certain concepts I've accrued from my personal experience, from people I know. If you were to do the same thing, you've had different life experiences, interacted with different people, your version of disgust might feel a little different to you, both in terms of how it feels subjectively in your conscious awareness, but also in the very physiological and neurological correlates. So, constructionists don't say that basic emotions are universal. They can vary widely across individuals. And, interestingly, too, within the same individual in different situations.
So, we have some examples of the paper,. Things like anger in different situations. So anger on the road, when someone cuts you off in traffic. You might think that would involve certain kinds of physiology, a sense of constriction, certain behaviors. But then think about anger in the workplace. It's going to feel very different; the behaviors, the judgments will be very different. So it's not like there's one emotion that's even the same across the same person in the different situations, which is what you would expect from a basic emotion perspective. Does that make sense?
Yes, in fact it leads right into something from the paper I wanted to make sure we discussed, which is the role of consistency and specificity. Could you explain what they are and the role they play in your paper and your analysis?
Consistency is the idea that, if basic emotions is correct that you have a built-in emotion program for each of the basic emotions like disgust anger fear etc. That means across different levels of the organism they're all going to fire in synchrony. So if I'm experiencing anger, I'm going to have certain thoughts, but also have certain physiology, certain brain signals, certain behavioral consequences, and all these things should kind of coral coalesce and correlate together in the same way. So all the levels the organism should kind of unify in the same direction when you have an emotional experience like that. That's the consistency claim, that across levels of analysis, brain body thought, you're going to have correlations between these different measurement levels.
Now an early inspiration for constructionism in the 90s was that, well, actually we don't see this. If you correlate emotion, experience—self-reported experience—with behaviors and physiology, often these are not correlated at all. And these null correlations were inconsistent with the classic basic emotions framework. They were kind of troubling for many people,.
Specificity is related. So, consistency is for a given emotion you since he has correlation across different measurement levels. Specificity kind of inverts that. It says, okay for any given measurement level, are there differences across different emotions? So if I wanted to look at whether certain brain areas are associated with certain emotions, that's a specificity question. At the level of MRI or other brain measurements techniques, can we say specific activations were disgust versus anger, versus fear, and so forth. And the variety of meta-analysis suggests that the answer is no, that at the level of brain experience, we don't see evidence for specific emotional signatures. The same thing can be said physiology and behavior, too.
So that's the difference: consistency is about whether across measurement levels this correspondence were given motion, specificity is within a given measurement level is there differences between different emotions.
And in both cases, you find the evidence shows that mostly that's not found?
Correct. The evidence is very weak, and certainly not consistent with the strong basic emotions perspective. That was a negative answer evidence against basic emotions framework, but then constructionism developed as a positive account of why these results might be the case, why might we not find those kinds of specificity or consistency.
So what kind of answers constructionism can provide instead?
Constructionism takes what's called the domain general view of the mind, not domain specific. It's saying that, okay a discrete emotion like a disgust, there's not a dedicated mechanism to it, rather it, like any other emotion, results from the complex interaction of domain general systems that are involved in a lot of things. Core affect can be involved in a lot of things besides emotion. It can be involved in memories, in hunger, in other things that are not specifically emotional. The same thing with concepts and categorization. They form the core of much of the human experience that's not necessarily about emotions.
The paper also has a discussion about how the realm of moral psychology also involves two broad concepts that have drawn the attention of constructionists—the realms of moral judgment and moral content—which you approach in a similar fashion. Can you summarize that?
You and I have been talking about how constructionism can apply to emotions, but we think the exact same logic can be applied to moral foundations, too. So, anger and disgust feel like distinctive experiences, but just because they feel distinct doesn't mean that they have a deep-rooted biological distinction between them. It could be these more domain general systems interact.
So, moral foundations theory talks about different kinds of morality, in the same way that basic emotion theory talks about different kinds of the emotion. It's not domain specificity, it's about moral modules, so to speak: one for harm, one for fairness, one for group loyalty, one for authority and respect, one for purity. You see this in much of the writing for the moral foundations theory advocates. It's the idea of distinct natural kinds of morality. You can essentially think about it as analogous to how basic emotions theorist talk about distinct emotion. That's what we're getting at with this notion of content. A moral judgment, is just in the moment, just making a judgment of wrong or right, a blame or praise. Now you can make a judgment in regard to any content, conceivably. So I can level moral judgment about a harm violation, about a security violation, or fairness violation.
And so the moral judgment then is general, as constructionists understand it? Correct?
Yes, we think that, much in the same way as apparently discrete emotions can be unified by domain general mechanisms—core affect and concepts—we think that moral content can be unified in the same way. So Kurt Gray [a co-author of the paper along with Kristen Lindquist] has a couple of papers on this, arguing that supposedly distinct moral foundations might be reducible to the more basic foundation of harm, but specified in situations in situationally different way. So it's just like core affect in concepts, as they relate to emotions. It's just that here, it's core affect and concepts related to harm.
So, thinking through the different foundations, you have harm—the direct obvious physical harm—but then you can think about all the other foundations that been proposed through the lens of harm. So you can think about sanctity is harm directed at the soul, or the spirit. You can think about respect for authority, or disrespect for authority, as harm directed at an ethical structure. You can think about loyalty or disloyalty is harm to the group, etc.
So, rather than positing these five distinct moral modules, we can say, “Well actually what if there's just a more basic underlying dimension of core affect and concepts about harm, but then you can be specifying it in regard to different specific situations you encounter?” Which is how constructionists talk about emotions. So with an emotion, you have core affect, and then you have some emotional concepts about—say something like disgust—but then which concepts you use to categorize core affect and disgust in that instance will depend upon the specific sort of situation you're in, other idiosyncratic thoughts in your mind, and so on.
So we're just trying to show that the logic is parallel. You can apply constructionism to undermine the idea of these natural kinds of emotion, but you can also apply constructionism to undermine these ideas of natural kinds of moral content. So when I mentioned earlier we're trying to undermine basic assumptions about emotions and morality in the field, that's what I'm getting at. Most people who have written on the relationship between emotions and morality, they talk about them in domain-specific natural kinds ways. Were trying to push back against that.
One of the things you mentioned in the paper was a constructionist model of dyadic morality, with a harm-based template of two perceived minds. I think it would be helpful to talk about that a little bit, because it adds some more specific detail.
The dyadic template is a prototype of most moral actions. It involves a moral agent, someone doing an action, causing, inflicting harm, in a suffering patient who is the recipient of the action. So when we think of moral episodes, moral transactions between people, we typically think of two people and one causing a change or harm to the other. You can also think about the inverse of that, causing an increase in welfare, so helping another person. This can capture that. There's a variety studies coming out now suggesting that harm—this dyadic template involving two people, were one person harms/helps the other—that's the unifying feature of morality.
If you ask people to think of five moral transactions transgressions, just asking for their immediate response, typically the things you'll see listed at the top list are—which is a way of assessing what's most accessible for people—are classic harm infractions: murder, rape, assault, genocide etc. So, we cite some of this work in the current review. The idea is that actually this might be one of those basic ingredients, a conceptual template of an agent helping/harming the patient, and then we specify that conceptual template about harm, depending upon the specifics of the situation. Are there aspects of the soul involved? Are there aspects of group relationships involve? But we can do that without positing these five distinct modules of morality. So that's the basic idea
In other words they arise out of a particular context, a social context in which they occur, rather than out of internal wiring.?
Exactly that's a perfect summation of what were trying to get at. It's social, it's contextual, it's going to vary across situations, across people, and we think that ultimately understanding both morality and emotions at that level will help us answer old questions and ask new ones. So, for instance, if underlying dimensions of core affect matter, for moral judgment, especially, well there's are all kinds of things that can change core affect that aren't about emotions. So you can increase arousal by going for a jog. You can decrease arousal by missing a nights sleep. These are all things that also implicate core affect.
Where does this take you in terms of looking forward to further critiquing the whole number accounts or to advancing the understanding you can get from a constructionist account? What kind of the experiments going forward are there that you think would be helpful?
In our review we have kind of two aspects. First, we have the negative aspect, we point out that, well, if you wanted to show that disgust goes with purity and anger goes with harm, etc.—which is what many of these basic emotions minded people have done in the past—here are the standards you want to meet, here are the criteria you'd need to meet to show those relationships. And so we go through them in the past work and they haven't met those standards.
So one example, a lot of work connects the emotion of disgust to the specific moral foundation of purity. That's the one you hear about the most, the disgust/purity relationship. So what they'll do is they'll induce an emotion like disgust, and they'll either compare it against a neutral control condition where nothing is induced, or an emotion induction that is something like sadness. Now, that might seem like good control, because they're both negative emotions, so if disgust influences purity and sadness doesn't, that's potentially informative.
But, as we point out in the paper, those differ on one of those critical underlying ingredients, and that's arousal, one of the dimensions core affect. So remember, disgust is negative, you're worked up, very high arousal. Sadness is negative, but you're down-beat, it's a low arousal emotion. And it could be that the difference in arousal is what matters, and not necessarily anything specific to disgust. There's more recent work showing that just simply putting you in a high arousal state, regardless of what the emotion is, makes you more blameworthy of others.
So, moving forward, a simple recommendation, if you don't want to buy into constructionism as a theory of emotions, at the very least what you'd want to do is have more well-controlled studies that try to compare disgust against other emotions that are similar, like anger and fear and so forth. That said, studies that have done that--there have been a few—haven't found such specific relationships. So there's a concern about whether those would even be found.
Okay, that sums up the critical analysis of the whole-number approach. What about the other, constructive side, what constructionism has to offer, instead?
So, that was part one. Part two is, well, constructionism is saying there's these underlying dimensions, like concepts and core affect, they're the moving parts, they're the parts of the engine that make the car run and you can try to directly target those in interesting ways. So, rather than focusing on, “Well does disgust go with this? And this anger go with that?”—those are emotion-specific questions—we say to go deeper, go into these fundamental questions of categorization and affect, and move those around in fun ways.
At the very end of our review, we do suggest one prediction that's a way of complicating things further, it takes a constructionist logic to its final extreme. Constructionism is saying if you take the category of emotions, different emotions, fundamentally they're not biologically distinct, they're made real in the minute, when you yourself use your own concepts to categorize your core affect as a specific emotional state of being. Same thing for moral content, moral foundations. There may not be these fundamentally distinct foundations, there only made real in your mind when you categorize core affect as harm in a specific situation.
But, then go to another level, you could begin to wonder, “Well, is there a real stark difference between an emotion and a moral judgment or moral content?” If they're both built from the same underlying ingredients—core affect, and conceptual knowledge about the world—then the line between them becomes a little bit blurry. So there been a variety of philosophical points about how some of the classic mental faculties are categories in mind, like emotion, cognition, memory, moral judgment, etc. We talk about them as if those are distinct things that happen in our minds, as if it's almost like a phrenologist: Here is where the emotions happen. Here's where the moral judgments happen, etc.
What if those boundaries are actually kind of fluid? And they're only heuristics we use for making sense of how we talk about mental function?, But actually, we can manipulate people so that the same basic ingredients—the flour, the butter, the sugar—can be used to make it either a muffin or a cake, and depending on what situation we put you in, those same things can manifest as an emotions or as a moral judgments?
I thought this was a particularly fascinating point, which defies some deeply-held assumptions. So, I'd like to ask if you could make it more specific, more concrete for people, to help drive it home.
Yes, we're doing a study now. One situational factor that might matter for this, for how the different ingredients combined to make different recipes, is attention: Where are you focusing your attention at the present moment? One distinction that has often been made is between focusing on your internal experiences (so you focus on your visceral sensations) or focusing outward in the world (so focusing on features like sound). You might think that by manipulating focus of attention—the self focus or the world focus—you could shape how those ingredients of core affect and concepts are used to create a mental state.
So, imagine the following example. You're driving along someone cuts you off in traffic, if you're focused on your internal experience, you're probably going to recognize that what you're experiencing is anger, or outrage. So, if you're cut off in traffic, if you're focused internally, what you'll notice is an experience of anger. But, if you're focused outward on the world as this happened, you might think that person is a jerk, that person is morally atrocious, because of how he or she is driving. So that is essentially the basic idea. It's that if you're focused internally on the self then your core affect and concepts might combine to form a emotion. If you're focused on the world, they may combined to form a moral judgment or a moral perception. Does that clarify it?
Yes, it does. And it leads to something else that occurred to me while reading the paper—the focus of attention may be involved in how ideology and morality relate to each other. One thing that's relatively well-established, in terms of the “Big Five” personality traits, is that liberals tend to be more open to experience and conservatives exhibit greater conscientiousness, more focus on order. That seems to indicate a differing focus of attention—to what's already well-organized for conservatives, versus attention to the wider world and trying to make sense out of novelty and chaos for liberals. Or to simplify further, attending to the world's orderly vs chaotic parts. That could feedback on how morality was shaped in a lot of different ways it seems to me, and it's not neatly clear-cut in the way that moral foundations very would be, but it would tend to suggest the kind of systemic effects that constructionism is better equipped to handle. Is that's just a crazy idea from a dabbling dilettante journalist, or might it be something that's got some potential?
No, I think that this is a really cool idea, it's an extension of what we wrote that as far as I know it hasn't been studied. But I think the short answer is definitely. I think that ideological differences in attention would definitely most certainly shape, or I predict they would shape how these ingredients work together. Descriptively, conservatives and liberals do tend to focus on different things, they attended different things selectively.
So if you look at ideological content, liberals focus a lot on perceived suffering of underprivileged groups, arguably conservatives are a little bit a less attentive to that, they focus more on liberty, and other kinds of things,, and you could talk about that in a way that doesn't imply like the strong moral module thing the moral foundations theory does, you can say that well descriptively sure, liberals attend to certain kinds of things, and conservatives attend to other kinds of things. But that difference in attention, and just an underlying domain general process, can explain so many things in a way that you don't have to go deep to make the stronger claim that there are these five distinct moral modules that are why conservatives and liberals differ. It could in fact be due to what they attend to, and how that shapes other things.
I'm a longtime William James fan, so your approach naturally appealed to me. But that's not to say that I dismiss all of Haidt's work out of hand—nor do I take it that you do. Liberals and conservatives do employ different sorts of language, for example, they orient around different sources of authority. But I come away from your paper with a stronger feeling than ever that most of what he's pulled together belongs in the domain of sociology, anthropology, social psychology—not individual psychology. It's all about the social surround that's shaping the environment in which these basic processes that you're writing about appear and work themselves out. That's what I came up with as my way of making sense of this, and I wanted to find out is that just me or is that how you would see it, too?
I think that's generally good way to think about it. I think I can specify the little bit. You're right, Haidt and his colleagues have been heavily influenced by anthropology and sociology, they cite Durkheim all the time, and that's great, moral psychology is interdisciplinary, and it flourishes when it is in many ways, no single field can claim to be the unique inroad [48:57] to understanding human morality.
I guess what I would say is that documenting descriptive variations, that's one thing, but understanding why we make moral judgments is a separate thing. That's more of a process focused question. So even though there is legitimate descriptive variation, and what people think morality is about, that doesn't mean that we have to add a process level to argue for these mechanisms, these moral modules.
They're not simply saying there is variation. They've also made the claim in multiple papers that there are these moral modules at the psychological level, that explain this descriptive variation. What were doing in our paper is saying that well, you we don't have to do that, that might not be the most parsimonious answer. There may still be descriptive variation that may be accounted for by something a lot simpler, which is these domain general ingredients, of core affect, conceptual knowledge, and you can add in attention to that to as a moderating factor as well.
Coming back to [William] James, he has this term, the psychologist fallacy which he talks about in the principles of psychology, it's one of my favorite bits of writing from him,, and he talks about the fallacy that when you're studying the human mind, it's tricky, right? So it's not like biology or chemistry, because it's a mind trying to study itself. So you have to be very careful about not taking how things seem to you, and then placing that back onto the world as how they actually are. So, how things seem to us, generally, we tend to see anger and disgust as different emotions. Just on the basis of our conscious experience, we might feel those are distinct things, biologically. But for James that would be a fallacy, because it's a fallacy to assume that just because something seems different in your experience that it actually is different on a more basic level.
So we're trying to suggest that we not do that, let's not commit the psychological fallacy, when we understand emotions and morality, And if we do that, if we try to dispense with any blinders we might have in our own personal biases—well bias too strong, from our personal experience—then maybe we can understand things a little better.