What Is the Purpose of Emotion?

EDITOR’S NOTE: This is the last of a three-part series on the Human Mind Conference, an “interdisciplinary event” held in Cambridge, England, in June 2017, “bringing together a wide range of experts from across the humanities and the cognitive sciences to discuss key aspects of mental life and experience.”

What is the purpose of emotion? More elaborately, how do the psychological and neurological mechanisms of emotion underlie a person’s point of view on the world?

These were the questions taken up by Philip Gerrans, a professor of philosophy at the University of Adelaide in Australia, in his talk at the Human Mind Conference in Cambridge, England in June 2017. Gerrans drew extensively on earlier talks by Andy Clark and Anil Seth, which examined the nature of human perception and advanced the theory of predictive processing. (Their talks were explored in another article here at Big Questions Online.)

To simplify greatly, predictive processing holds that the mind crafts perceptions based on its existing ideas or models of the world. So we tend to perceive the world in familiar terms, like seeing a face on the Moon. This happens not so much because we are self-absorbed as because all perception interprets the world in terms of ideas already known to an experiencing subject.

Something like the predictive processing model can also be applied to understanding emotions, according to Gerrans. As he argued in his talk: “emotional processes tell us not just how the world is,” i.e., the information already given to us by perception, but “how that information matters to us.” As rational organisms, we are driven by the pursuit of a myriad of goals the fulfillment of which matters to us in varying ways. “Emotions help us [to] see” and “understand how the world is mattering to the fulfillment of those goals.”

The mind, says Gerrans, is hierarchically organized to solve problems. “And what [the mind] likes to do is solve the problem at the lowest possible level.” Emotions, according to this view, serve a dual purpose. First, they alert us to the significance of things in the world. And second, by existing as feeling, they also allow us to solve many problems quickly and without much conscious deliberation.

Thus, Gerrans contends, emotions help to coordinate the mind’s hierarchical problem-solving structure and its engagement with the world. More specifically, “low-level emotional systems direct perception [and] sensation” to extracting the most relevant information from the world. High-level emotional systems manage and reflect on this information. When the emotions fail to solve a problem on their own, higher-level executive functions step in: “If it can’t be solved, up to the next level, where more integrative, more abstract systems, which are more flexible and have access to a wider range of information, try and solve the problem.”

Gerrans offers the example of meeting someone. Say you meet someone and want to know if he is disposed to you in a friendly way. You might scan him for a smile and inviting posture, automatically processing the resulting information into a warm or cold feeling. This is the “low-level,” perceptual-organizing function of emotion. An example of the “high-level” process would be expecting one response from the person and getting another, and then stewing over it afterward. This is like “when you’re sulking in your bedroom about why you didn’t get the job and why the people are all against you and how it’s not fair.”

Central to Gerrans’s talk was the theme that the cognitive and value-imbuing functions of emotion are too often conceived as set apart. If we think of them as separate functions, the question arises as to which is primary and which secondary. But Gerrans suggests that this is something of a false choice. Sulking in the bedroom, for example, is just a higher-level process of reflecting on an initial bad feeling. “But it’s not … separate” from the feeling itself.

The most fascinating “takeaway” from Gerrans’s talk was the comparison he drew between the perception of objects and the sense of self. Predictive processing tries to answer questions such as the following: How is it that when we look at an object, such as a rock, the light waves that reach our eye from the object are conceptually incoherent — they show us only one side of the rock, as an arbitrary area within a larger background image — and yet we perceive the rock as separate from that background, whole and durable?

Predictive processing tells us that it is our acquired yet deeply ingrained belief that objects are separate from their environment, whole, and durable that allows us to perceive them this way. Predictive processing does not conclude that objects are therefore constructed or somehow illusory; the point is that objects in our perceptual field are things we have to learn about. Thus our perception of objects comes about as a result of this learning, rather than the other way around.

Gerrans’s provocative claim is that the same type of process at work in our perception of objects also occurs for our emotions — and even for our very sense of self. We experience a sense of being a unified persisting self, that is, in the same way we experience the world as populated by unified persisting objects.

Gerrans’s talk raises a number of perplexing questions. The most immediate stem from old anxieties about the reality of selfhood and the emotions — about whether the world, as it matters to us, is somehow just in our heads. The idea that the self and the emotions are models constructed to suit the purposes of the brain seems to imply that they are mere “user illusions,” to use the popular phrase of neuro-philosophers.

But an alternate interpretation is possible. Few neuroscientists would really suggest that, just because our perception of objects involves a mental model of objecthood, objects are therefore illusions. Among other things, this would imply that babies who have yet to develop object permanence are more philosophically enlightened than the rest of us. The problem is that our model of objects becomes so deeply ingrained that we forget it is a model. But we forget it precisely because the model works so remarkably well in describing real features of the world.

Follow Gerrans’s analogy, then, to its logical conclusion: If selfhood and emotions are as deeply ingrained in our perception as objecthood, that may be because these models are just as powerful as objecthood in describing the way the world really is. In other words, selves and emotions gain as strong a claim as objects to actually being real features of our world. Yet, at the same time, objecthood seems to require consciousness no less than do selfhood or emotions. For none of these things — objects, selves, or emotions — does it seem true to regard them as either strictly “out there” or “just in our heads.”

Gerrans’s talk, like so many others at the Human Mind Conference, seems at first blush to reduce mind to matter, but winds up only reiterating very old philosophical conundrums about whether one can ever be brought fully under the purview of the other.