 The big question that we're interested in is how the brain supports aesthetic experiences. For example, when you go to a museum and you find a painting to be beautiful, or when you go to a performance and that performance is touching or moves you, we want to know what brain processes and representations support those unique aesthetic moments. The more specific question we focused on in this study is a question about different aesthetic domains. In particular, what I'm referring to by a domain is something where there's a set of rules that describe how you interact with that domain and how you make judgments about that domain. So the domains we're focusing on here are artwork and architecture and natural landscapes. So the types of judgments that a person makes about a natural landscape might be quite different from the types of judgments they make about artwork. And not only that, but the way that the brain represents the relevant information is also quite different. Despite the fact that these different domains have very different properties and people engage with them in different ways and the brain represents them in different ways, it might still be possible that there's one brain system that can represent aesthetic appeal across all of these variety of differences. And we're focusing specifically on two brain networks. Because I'm interested in visual processes, we are investigating the visual system with a focus specifically on the ventral visual pathway, the part of the brain that represents what it is that you're looking at. So this is a brain network that is very extrinsically focused. It's a brain network that is active when you look at something out in the world. The second network we care about is a brain network that seems to be internally focused, and it's called the default mode network. This brain network is a brain network that is thought to be mostly involved in internally directed mentation. So for example, when a person recalls autobiographical memories, or when you are mind wandering and just thinking internally to yourself about what you are going to have for breakfast tomorrow. So typically this brain network does not become engaged when you pay attention to the external world. Therefore, we want to look in these two networks, the visual system and the default mode network, to see whether or not either of these brain networks contains a representation of aesthetic appeal that generalizes across all of these different visual domains. The primary method that we use in this study is called functional magnetic resonance imaging. This is a form of magnetic resonance imaging, or MRI, that can be used to measure a correlate of neural activity in the brain. What we do is we put people in an MRI scanner, and we use a projector and a mirror to show them images. In this experiment, we show them either images of artworks, or we show them images of architecture, or we show them images of natural landscapes. They view each image for four seconds, and then they use a trackball to respond and tell us whether or not they found that particular image to be aesthetically moving or not. So that is the basic behavioral data that we're going to be using to correlate with the brain data. Now, the complicated part of this experiment is the analysis. We take the brain signals that we measure from across the brain, and we pull out the brain signal from specific regions that we want to test. There's one set of regions from the visual parts of the brain. So in this particular case, the ventral occipitotemporal cortex. The second set of regions are from the default mode network. We take the pattern of data from these brain regions as a person is looking at, for example, an image of artwork, and we train a machine learning classifier to try to predict people's responses. If a part of the brain contains information that is relevant for aesthetic appeal, the machine learning classifier will be able to pick up on that information, and then when I try to test the classifier on a new set of trials, it will do better than chance. That's the basic method here. We train a classifier and we test the classifier on data from a specific brain region. In one type of test, we train on one domain, for example, artwork, and we test on that same domain, artwork, where we train on landscape and test on landscape. The second type of test is what I call a cross-domain test. In this case, we're training a classifier on landscapes, but then we're testing that classifier on architecture, for example, or we're training the classifier on artwork and testing it on landscapes. First, we investigated the visual regions of the brain. When we take signal from these regions and we try to train a classifier to predict people's responses, we find that classifiers are not able to do it very well based upon signal solely from these visual regions. On the other hand, when we look in the default mode regions of the brain, we find a very different picture. First of all, data from these regions is able to be used to predict people's preferences. Not only do we find a signal for aesthetic appreciation in these default mode network regions, but that signal is domain general. We can train a machine learning classifier on trials where a person was looking at artwork and we can actually predict their responses on trials where they're looking at landscapes. I think that this is really the surprising part of our finding. This brain network that is really thought to be involved in inwardly directed contemplation, a brain network that normally does not respond when a person engages with the external world. It's this brain work and not the visual parts of the brain per se that seem to really contain a domain general representation of aesthetic appeal. I think these findings are relevant in two different levels. First of all, they have some relevance within the field of cognitive neuroscience. Aesthetic experiences are important and powerful experiences. They're experiences that can really be life-changing in some cases. We think that by understanding these types of experiences that we might be able to also understand the brain dynamics of other kind of rare or peak moments of experience. Turning to the larger societal implications, we think that there could be implications both for learning and also for well-being and health. When you find something to be aesthetically appealing, these moments can really be teaching moments. They can be moments where you learn about yourself or learn about the world around you and the pleasure you get from those could potentially be harnessed to help people learn better. We're very interested in understanding whether or not the brain mechanisms of aesthetic experience can teach us anything about learning more generally. It's clear that aesthetic experiences are very important for people's well-being. The amount of money and time that we spend on aesthetic pursuits makes it clear that going to an art museum or looking at a landscape is not purely a pastime, but it's really an important key component of our lives. I think that there could also be some potential health outcomes and relevance for this work. The most immediate questions that this research raises are really mechanistic. How is it that information about your visual world then makes its way to this default mode network, to the internally directed DMN, and how is that information transformed from information about what it is that you're experiencing to then a representation of how you feel about it, of whether or not you find it to be beautiful or not? So we really need a better mechanistic understanding of how these networks talk to each other and how the information flows from one place to the next. Second, there's obviously the question of whether or not this applies to non-visual domains. So is it the case that the DMN also represents aesthetic appeal for music that you like, or when you go to a performance and you see someone dancing, or if you read a book and you find that book to be aesthetically appealing, is it really domain general in that sense, in that it generalizes across many different sensory domains? Beyond these more mechanistic questions, there are a few directions we also plan to take this research. First of all, given the importance of the DMN for internally directed thought, it's natural to ask whether or not its involvement in aesthetic experiences is related to one's sense of self, or to a sense of self-identity. And we're really interested in trying to better understand whether or not self-relevance is playing a role in these aesthetic judgments. The most moving aesthetic experiences you've had probably didn't occur in a lab, but they occurred in a museum or in a performance space. Well, we can't exactly measure brain imaging using fMRI when you are sitting in Carnegie Hall. But what we can do is try to look at bridging technologies that allow us to measure some of these same brain systems and brain processes in those naturalistic settings where people are more likely to actually have strong aesthetic experiences.