Borscht Belt on the Bimah: The Neuroscience of Comedy

Thank you. I feel very strange following comedy because I’m not going to be that funny. And, of course, because I’m a scientist, I have to have PowerPoint slides. We can’t speak without slides.

So my job here is to actually go back and say, “Well, why were these guys so funny?” Because they were really funny. But what is it that brings us that humor, that sense of pleasure, from what we just experienced? And there’s no way to explain this all in one fell swoop or one single theory, but there are some aspects of humor which make a lot of sense once you start to sort of dial into it. And they make a lot of sense within the concepts of modern psychology as well as neuroscience.

Why do we find things funny?

And what I want to do is start with some of that psychology part of it, and then try to pull in some of the neuroscience on what the brain is actually doing here. One of the key ideas in this field, going back now to the 1700’s, is this idea of humor being tied to incongruities, things that don’t quite fit together. And it’s been more recently articulated as “incongruity, detection and resolution.” And what that’s saying is that part of what’s going on is: first, we have to detect that there’s something that’s not actually matched up here, quite right, and then it’s from the recognition of that and the understanding of that, that some of the humor comes.

So what happens in a lot of these sorts of situations is we’re presented with something that starts us down one path, and then it shifts on us. And what we get, in fact, often are violations of conventions, violations of what we expect to come next. And that’s actually really critical here. We expected something, and something different happened.

So let me give you just an example of that, going in the theme of Jewish humor.

So imagine, Shabbat morning – Mother comes, knocks on her son’s door. “Bubby, it’s time to get up. You’ve got to go to Shul.”

“Mom. I don’t want to go to shul. I’ve got two reasons I don’t want to go to Shul. Those people don’t like me, and I don’t like them.”

The mom says back, “I don’t care. I’ve got two reasons why you’ve got to go to Shul. First, you’re 54 – and you’re the Rabbi.”

Our brains are prediction machines

Now, what happened there? Part of what’s going on is that the brain is, in fact, a machine for prediction. We are constantly predicting what’s going to happen next. Our brain is actually set up in ways, where we often think of – sensory information comes in, you think, “oh,” it goes in, hits the retina, it goes back to the visual cortex, it starts working its way forward. But each step of the way, there’s actually feedback going on, because we’re actually predicting what’s going to happen. And this is absolutely essential in language. Everything we do with language is actually – we’re predicting often what the next thing is that the person is going to say. I mean, I can start [with] easy sequences here. ABC. What comes next? If I go ABC-F, you’re like, “Wait, that wasn’t right.” If I even start just typing stuff – I mean, we see this now that we use autocorrect. How many times do you type something – it puts in a word and it’s actually right? It’s the times, though, that it’s wrong, where you’re like, “I can’t believe it did this. Why is it doing that?”

Well, it did that because statistically, that was the next thing to come. But we do this constantly. Every time you’re in a loud room having conversations, you don’t actually hear everything the person is saying, but you fill it in. Sometimes you fill it in wrong, and you get into all sorts of problems because you’re like, “Oh, yeah, what? You said that. I can’t believe you agreed with that.” But a lot of that is because we’ve got predictions going on of what’s coming next, and we’re really set up to do that.

When prediction meets incongruity

We’re also really set up to deal with incongruities – when things don’t match. Because it’s actually really hard; we have to work at it quite hard.

Famous psychology paradigms are things like the Flanker, where what you have to do is you have to say the direction of the middle arrow. And it’s really easy when that arrow fits with the other arrows beside it. But if it’s in the opposite direction, now you’ve got conflict, and you’ve got to deal with that.

Similarly, we’ve got a classic test, the Stroop task. The Stroop task is a situation where you’ve got to either read or name the color, but the written word is in conflict [with it]. So, for instance, if it’s easy, when it’s congruent – you see that red there? It’s red. Both things are feeding you in the same direction.

On the other hand, when you start to have that in conflict, you’ve got to actually stop. You’ve got to process it. You’ve got to make a decision about it and resolve that. So a lot of what happens in comedy, at least in narrative comedy, is you get a situation where we set up oppositions, and really what we’ve got is the text could be compatible. What you’re hearing could be compatible, for instance, with two stories, and each one could fit, but they don’t fit together. In fact, they’re often in some degree of opposition. So, for instance, in this sort of story here:

“Is the doctor at home?” The patient asked in his bronchial whisper. “No,” The doctor’s young and pretty wife whispered in reply, “Come right in.” So you’ve gone in a very different direction, and the story set you up in one direction, and then it takes you in a different one.

The multiple stages of humor

This often is thought of as happening in multiple stages, where there’s a first stage – where you detect that there is an incongruity, it’s been set up. You resolve it in some way, and there’s pleasure in that experience.

Now, you can say, why is there pleasure in that? But at some point, there’s an insight in there, and it’s in that insight that we actually find amusement – in the same way you get pleasure in the cases where you just solved a puzzle. “Yeah, I got that!” This happens in us. You may not even necessarily to yourself, say, “Oh, I got it.” But you do. And sometimes even it takes a while. Sometimes you get a joke and you’re processing it, and then finally you’re like, “Oh, I get it.” Some people also will say there’s a fourth stage, which is an elaboration stage, where you get it, and then the person starts to push it those steps farther, they start to embellish it to make it even more extreme, and that will just tickle us, essentially.

An important piece here also is that there is some degree of a violation, but it’s a benign one. And what am I saying when I say there’s a violation? That in some ways, this actually runs against what is standard and acceptable. It may be an ethical violation. It may be social; it may be physical. But there’s a key thing here, where it’s a little bit offensive. If it goes too far, then people will be like, “No, that’s too much.” And we can actually go and look over time. There are jokes that we would have probably found really funny 20-30 years ago, and now we’re like, “You know, that is actually incredibly insensitive, and it’s inappropriate.” And so, there’s a cultural overlay here, and there’s also a huge cultural context. There are certain themes we can talk about in Jewish humor, where we go and we identify it. And we could even say there and say, “Yeah, you take that out of the context.” And it’s like, I don’t get it. If I start making jokes about lox, if I’m out in Nebraska and someone’s never had bagels and lox, it’s not going to mean anything to them. So there’s a contextual thing that’s really important.

I think one of the fascinating things is the extent to which Jewish humor has actually exploded well beyond the boundaries of Jewish culture, particularly within the United States. It’s been really quite dramatic in that respect.

Nonsense and absurd humor

There are also other sorts of things that can be incongruous, which don’t require, necessarily, the storytelling, but there’s somehow a break with reality. But there’s always something that’s in there that’s incongruous.

So, for instance, in this first one, where the incongruity here is: okay, you’ve got a shark coming down a mountain. Well, what’s working there? Where is the connection? Well, the connection is actually quite physical, because we’ve got this line. So we can see one image here, essentially. If you didn’t see the rest of this, it’s just a fish swimming, essentially. But once you combine them, you’ve got two different stories, and then you’re having to resolve them. I think the one on the other side – just, ducks are funny. But no, I mean, ducks usually aren’t homicidal and trying to kill people. I have a friend who raises ducks, and she would actually dispute me on that.

What does neuroscience tell us about humor?

So what does neuroscience actually tell us about this? I’ve primarily, so far, just talked in terms of psychology. I run this neuroimaging center at Rutgers; what we do is we put people into a high-field magnet, the same sorts of magnets that you would use to take pictures of the structure of a brain. What we do is, we make use of a unique feature of how the brain actually, physically works. And what that is is: when an area of the brain starts working harder, its metabolic needs change. And critically, the blood flow to the area changes. And it happens to be the case that the way the MRI works is all about, actually, magnetic properties. And it turns out that there’s a change that happens between oxyhemoglobin and deoxyhemoglobin when you have that need, and that actually changes the signal that we pull out. And when we do that, we can then measure activations within the brain, and we can start to use that to study how the brain and different parts of the brain work.

And so if you apply this to something like humor, there are certain things we can do with that. We can first ask about, well, simply: where in the brain is active when someone is saying something funny and we laugh at it? We can also start to ask some questions with what we call reverse inference. If we know what a brain area normally does, we can try to infer what it’s doing in a new situation based on its normal function. So we start to sort of bootstrap our understanding of the brain, and we can also start looking at sequences of events.

So what I’m going to show you here is from a meta-analysis. What this means is they took a whole bunch of studies, and they combined the data from those studies. Most of our studies are relatively small, and they can be really idiosyncratic, but if we combine data across studies, we can start to see what the common themes are.

This was a meta-analysis of 57 different studies, using functional MRI, of different sorts of humor. And I don’t expect you all to be able to look at this and immediately go, “Oh, yeah, I can see that. I recognize that,” and it’d be better if I actually had this on a surface of the brain – it’d be easier. But the way we often look at this stuff is: we look at single slices at a time. And these are coronal slices, which means if you’re just sort of putting your head like this [in profile], I’m going, “slice, slice, slice, slice.” Luckily, not with a knife.

What it’s showing here, though, is it’s going from the very front of the brain up here, and we’re moving progressively further back in the brain as we go, “Okay.” And the green here are areas where, regardless of the type of humor stimuli they used, these were the areas that were active.

Areas of the brain involved in humor

And there are some things that are really interesting about this in terms of the areas that come up. So, one of the biggest areas you’ll see, which is along here, is the left inferior frontal gyrus. This is an area of the brain which is heavily involved in semantic integration. And if you actually go sort of up enough and back enough, you get into an area known as Broca’s area, which is heavily involved in actual language output. But the whole region is critical for semantic integration. Well, what are we talking about when we’re listening to language and listening to these stories? We’re integrating this information.

Another area that ends up being really important is this dorsal medial area, which is sort of up in there, an area that’s heavily involved in what we call theory of mind. Theory of mind is just referring. If I’m thinking about what you’re thinking, I have a theory of what’s in your mind. So [in] a lot of this, the humor comes about what we had as people’s expectations, and we’re manipulating some of those expectations. The anterior cingulate, which is this area right here, is heavily involved in incongruity. Actually, if you do that Stroop task, if you compare when it’s easy versus [when it’s] hard, that’s the area that’s going to come online, because it helps resolve that sort of situation.

And then we have some other areas that are really heavily involved in reward. So, for instance, this area down here that you see is part of what we call the ventral medial prefrontal cortex. All that means is that it’s ventral in the frontal lobe, so it’s at the bottom and it’s along the midline – heavily involved in reward. If I come out here and say, “Here’s some money,” that area is going to actually activate.

There are also some critical subcortical areas, meaning not on the cortex, but buried deep down. Here is an area called the amygdala, which is actually one of the areas that I have studied in my early years. What we would do, because we wanted to activate this area – we knew this area heavily responded to smells. I went to a guy who was the world’s authority on intestinal gases, and they would give me this, like, Teflar bag of these sulfides, and I would come up in front of the scanner, and I would release this in front of someone while they were in the scanner. And I have to admit, I almost always cracked up a little bit doing it, too. But this area is highly involved in emotional experience, in emotional arousal, and it becomes active heavily during this.

The other area that’s really critical in here is an area that I can’t quite reach here. But there’s an area in here where we’ll get into an area called the nucleus accumbens. It’s heavily involved in reward. Actually, if you take drugs like cocaine, this area is where a lot of doP, the neurotransmitter dopamine, is being released. We’ll see this better in another slide in a second.

Humor, patterns and the brain

We can also start using this, though, to try to pull apart the components of what’s going on. So, for instance, in this study by Chan et al., they actually are looking at the different phases here. And this is when they’re actually making a comparison between the detection of the incongruity itself at that phase, versus when it’s being resolved. And so here, in this one, where we’ve got these temporal lobe regions, which appear to be heavily engaged at that point where that incongruity is detected. But here we have this more prefrontal area, this medial frontal area, hitting at that resolution component. And I’ve got one more study that I just want to talk about, which is really recent.

This is a study also where they’re trying to look at the different phases, and what they do is: they’ve broken this up where every trial goes in a certain pattern, where first they have a setup, there’s always a fixation dot that comes up in between, because we try to separate this stuff out in time after the setup, they actually have, it’s sort of a first punchline where that incongruity is really just sort of laid out there.

And the second part of that punchline is where you get an actual resolution of what’s going on. And so they do this in subjects, and they start to break this apart. So, for instance, they’ve got up at the first part just what’s going on as you’re setting up the expectation.

The second part of it is where you get that first part where the incongruity gets presented. The third part is where there’s a resolution. And this is where we start to get this huge amount of frontal-lobe involvement coming in, as well as part of further part back called the parietal lobe or the inferior parietal lobe on those slices. Now, rather than up and down, these are axial slices. So they’re coming down like this. But the critical part here is this last part, which is where they start to elaborate. This is where, rather than just sort of leaving it there, they start to push it to that next level. And here what gets really interesting is, not only do we have the ventral medial prefrontal cortex coming up, we have this amygdala, the area that in my studies were the “fart area,” essentially.

But critically, here we have the nucleus accumbens. This is that area I was saying that gets pushed by drugs of abuse, like cocaine or amphetamine. This is also, actually, interestingly, the midbrain showing up there, which helps produce dopamine. It’s also, though, critical in that it fires when you get prediction errors, particularly positive prediction errors, when something is better than you expected it to be.

So what this suggests to us is, really, a lot of what’s going on here, we can easily see how it’s encoded in the brain. And it fits, in fact, with a lot of this larger theory of what’s going on. When I was asked to do this, I must say, I do not actually study the neuroscience of humor, but I do study reward, and I do study these issues, like how prediction and expectation come about. And it actually, I think, fits really beautifully. So thank you for having me.