Thursday, April 5, 2012

On Emotions and Free Will: What are emotions, and why do we have them?

In this essay (broken into three parts for publishing), I will first explore ancient and contemporary theories of emotion and arrive at a practical, working concept capable of informing behavioral hypotheses. Secondly, I will reopen the debate on the existence of free will in light of recent discoveries about the role of emotions in decision-making. In the context of this discussion, I will propose a novel redefinition of the concept of free will that treats it not as a metaphysical quality of mind, but as a cultural invention serving as a notional goal of rational thought. Lastly, I will place these conclusions in the context of contemporary and future research on the phenomenon known as motivated reasoning.

What are emotions, and why do we have them?

It is unnatural for us to ask this question, for emotions are so seamlessly woven into our conscious (and unconscious) experience. So seamless are they, that they can often be mistaken for a feature of our souls. I say 'soul' because I mean to imply our sense, real or not, of our true being. For as long as we have had historical records, philosophers, poets, artists, scientists, and other delinquents prone to ask unnatural questions, have sought to disentangle emotion from the soul so that it may be understood on its own. According to psychologist George E. Marcus, the oldest presumption regarding the psychology of emotion is the separation of emotion from the practice of reason. Even today, we commonly hold the metaphorical view that, somehow, “passions arise from the 'heart' and reason from the 'mind'.” Further, we often view these “forces” as occupying the opposite ends of a linear spectrum: the more emotional someone is, the greater the likelihood that “obsession, delusion, and demagoguery will hold sway” (Marcus, p. 188). Similarly, an overabundance of reason seems fastidious, cold, and inhuman. Though essentially human in form, Dr. Spock's most alien characteristic was his lack of emotions and famously stalwart adherence to “logic”.

Historical Views

At least since Plato, the dominant perspective in psychological approaches to government has been that the role of emotion is profoundly negative. In Plato's Alegory of the Cave, he describes humanity as born in the darkness of a cave. Never knowing anything else, he confuses shadows and shapes playing on the wall before him will real things. Under the oppressive rule of corporeal desire, he spends his whole life chasing at illusions, for he could not be distracted for a moment to reflect on his depravity and find his way out to the realm of light and reason. Across the great Eurasian landmass, 12th century Buddhist luminary Shinran similarly described the passions as “an instinctual, emotive force arising from the unconscious and deeply rooted in the body” that once “entwined with egocentric human calculations ... becomes distorted and causes havoc in our lives".

For Plato, as well as virtually all western political philosophy up through the Enlightenment and the United States Constitution, emotions are considered baser instincts. Thus, a crucial goal of any governmental design is to subjugate emotion to reason. In his Second Treatise, Locke writes: “The freedom then of man and liberty of acting according to his own will, is grounded on his having reason, which is able to instruct him in that law he is to govern himself by, and make him to know how far he is left to the freedom of his own will”. In other words, the capacity for reason is precisely the mental faculty that makes intellectual freedom possible, and to ensure this it must be employed in the construction of political regimes. Some one-hundred years later, James Madison extended this idea in The Federalist Paper #10, asserting that “As long as the reason of man continues fallible, and he is at liberty to exercise it, different opinions will be formed. As long as the connection subsists between his reason and his self-love, his opinions and his passions will have a reciprocal influence on each other; and the former will be objects to which the latter will attach themselves.” Accordingly, he and his coauthors designed a form of constitutional government with institutional checks and balances capable of mitigating the most severe impacts of passion while making positive use of the motivating force of emotion.

Contemporary Views

The intuitive understanding of emotion espoused by the great philosophers has shown remarkable staying power and has played a prominent role in western history and political thought. But what is emotion really? Modern science has advanced several theories. In general, the scientific perspective on emotions is that they function as heuristics, or cognitive rules of thumb. In other words, they are a efficient shortcuts in decision-making, allowing decisive behavior in time-impacted situations, at speeds many times faster than what can be achieved through deliberate thought processes. Emotions may facilitate appraisal of a stimuli and discharge a response even before conscious awareness of the stimuli has been achieved. Evolutionary psychologists John Tooby and Leda Cosmides proffered a bare-bones definition of emotions as the basic means by which living creatures solve the problem of “approach” and “avoidance” (1990). This simplified perspective captures the idea that emotions, for humans or any creature, have the functional purpose: to rapidly establish a disposition with regard to an object in the world. If the object is deadly, such as a predator, rational cognition would be too slow. Thus, emotions evolved as a mechanism to compress the information-space to a single bit of data: approach or avoid; or alternatively, good or bad, positive or negative.

A simple approach/avoid model seems to make intuitive sense for a lizard, but what about humans? Psychologists have not strayed from the approach/avoidance assumption, but human models of emotion have generally supplied them with a greater number of “dimensions” on which to appraise an object. I will discuss briefly discuss three such theories. First, valence-arousal theory describes appraisals along two orthogonal dimensions: valence (good or bad) and level of excitation, or arousal. For example, high-levels of both arousal and valence may produce emotions such as enthusiasm, delight, or elatedness, while low-levels of both may produce gloominess or misery. Under an appraisal of high arousal and extremely low valence, we might experience fear or anxiety; under the opposite we might feel tranquil or content.

Second, the theory of discrete emotion also relies on binary, positive-negative appraisals, but applies them according to a hierarchy of considerations. For example, the precise feeling, or emotion experienced is determined by 1) valence, and 2) whether there are rewards or punishments, 3) how certain the rewards/punishments are to realize, and 4) to whom they will go (the experiencer or someone else).

A third theory, multiple systems of emotions, is more complex still. Multiple systems theory asserts that different emotional systems operate simultaneously to facilitate automatic behavior in response to various kinds of stimuli (good, bad, and everything in between), but also to control attention on any one in particular. These systems enable attention to be shifted from one thing to another but also sustain the ability to focus attention more deeply and ignore distraction. This can be crucial. For example, if a child is missing, the sense of terror communicates urgency to the mother and may block out all other concerns, focusing all of her cognitive resources on the single task of finding the child. In Marcus' view, the multiple systems view has the advantage of a more solid grounding in recent neuroscientific discoveries. Specifically, he argues that the functions of successive layers of emotional systems map on to neural subsystems in the brain, i.e. somatosensory, associative memory, and declarative memory systems. Without getting too deeply into the details of his criticism, Marcus argues that the dimensional models are overly simplistic and cannot account for the diversity of human emotional experience, nor the totality of roles they play in decision-making. I do not disagree with this, however, his theory may as well be incomplete.

While Marcus may identify a behavioral role for diverse spectrum of emotion, multiple systems theory is fundamentally a more dramatic reconstruction of a basic approach/avoidance mechanism. This view is not no much wrong as it is incomplete. In fact, it very well may suffice for most of animalia---for a given asocial creature, its relationship with respect to any object should be reflected in a one-dimensional linear distance function. On a single dimension, there are only two possible moves: approach or avoid, which may only be moderated by speed. Bases for decision are straightforward: Can I eat it? Mate with it? Will it eat me? Make me sick? Some combination thereof? Can I think about this for a while, or do I have make a decision now? For a (comparatively) small number of species, the addition of a social dimension creates opportunity for (and meaningfullness of) lateral movement. A social creature's relationship to an object is reflected in a distance function that includes not only the object, but the sum of all dyadic relationships in an n-object community. For any two social creatures, their relationship may be triangulated from their known relationship to a third. This notion is inherent in the aphorisms “The enemy of my enemy is my friend” or “Any friend of yours is a friend of mine”. Marcus appears to have attained a loose intuition this notion, but perhaps not enough to apprehend its significance. Citing a growing body of neuroscientific research, Marcus writes:

“As we gain greater insight into how the brain is organized, we may well learn more about how humans function as social and reasoning creatures (Damasio,1994; Goleman, 1995). Even in the task of moral judgment, recent research in neuroscience establishes the central role of emotions in the resolution of moral dilemmas (Greene, Sommerville, Nystrom, Darley, and Cohen, 2001) (Marcus, 2003).

Contemporary psychologist Jonathan Haidt proposes a promising alternative perspective. In his effort to understand the psychological basis for morality, Haidt draws upon lessons from biology and evolutionary psychology (the lessons of Tooby and Cosmides), and postulates five foundations of moral judgment: care for others, fairness, loyalty to one's group, respect for authority, and purity. For Haidt, these are basic criteria for moral judgment (Haidt, 2007). Emotional responses to observations of another's behavior or condition (i.e., empathy for their suffering, anger at their treachery, or indignation at their disregard for social mores) drive moral judgment, and by extension they will govern social interaction. Unlike the Marcus' multiple systems model, Haidt's model is built from the ground up as a social theory of emotion. The two are not necessarily mutually exclusive. The former may suffice to explain a basic scheme of approach/avoidance common to animalia (humans among them), while the latter sits atop in a multi-tiered theory of emotion necessary to explain the more complex geometry of a social species.

Haidt's model illuminates an adaptive, functional role human emotional diversity in a way the multiple systems model cannot. Marcus correctly points out that adaptive purposes for which emotional systems evolved are far too complex for a simple, one or two-dimensional good/bad basis of object appraisal. This is as much an oversimplification as attempting to reduce the gustatory palette, or the set of various tastes associated with different foods, to simply 'good' or 'bad'. One commonly held misconception illustrates this point. For example, if you ask a normal person (not us) why a delectable piece of fine Belgian chocolate is so delicious and they will like tell you that it is because it is so sweet and richly satisfying. If they are a bit more sophisticated, they might suggest that this is due to its abundance of fat and sugar. But there is nothing inherently satisfying about fat, nor is sugar inherently sweet. On the contrary, these 'tastes' are neurological adaptations allowing us to discern which materials from our environment are good for eating, and those that were not. The positive psychological experience accompanying the taste is merely nature's way of enticing us to choose this food over less nutritious alternatives. In the case of rich Belgian chocolate, that sweet, sensuous euphoria washing over you as it softly melts to a sultry goo in your mouth is an indirect impression of a readily digestible, energy dense material. At nine calories per gram, fats are the most energy dense food source available to us. Ergo, bacon is heavenly. Yielding almost 12 calories per gram, gasoline might taste even better if it were not, of course, extremely poisonous. Were you to try it, you would no doubt find so repulsively bitter and acrid in the extreme that you would spit it out at once and spontaneously vomit out the entire contents of your stomach just in case a small quantity slipped past your gullet. These two examples, chocolate and gasoline, may exemplify the extremes of 'bad' and 'good' experiences with food, but the palette and somatosensory system associated with it is precise instrument capable of discriminating multiple categories of nutritional properties of a material. For women, the physical demands of pregnancy can deplete their stores of various essential vitamins and minerals, resulting in sometimes (famously) bizarre cravings for exceedingly specific food types. In some cases, women may experience cravings for materials not generally considered to be foods, such as soil or chalk. In order to discern which foods can provide the nutrients she requires, she requires somatosensory spectrum capable of producing a wide-diversity of experience.

I regret belaboring this analogy with taste, but I feel it is essential to convey how important it would be for a social creature to have a similarly (or perhaps even more) precise “emotional palette” for the detection the nuanced situational dynamics of a large-n social sphere and the discharge of an appropriate behavior. As Haidt discussed with an audience at Ted Conference in 2008, “The righteous mind was designed by evolution to unite us into teams divide us against other teams.” With the righteous mind, we are able to maintain norms of social behavior that deepen cooperation and ultimately make our communities stronger and more resilient than others. With the righteous mind, opportunists may navigate the social world, steeped in culture and norms, to advance his individual interests and capture adaptive advantage from his peers. They keep us constantly apprised of fluctuations in the social fabric presenting momentary opportunities to advance our standing in the competition for status, resources, and mates. Moral judgments inform him whom to form coalitions, friendships forged with brotherly affinity, of whom to be wary, with whom to rival for power and how to charge him with impropriety. Within the social sphere, we feel camaraderie, intimidation, humility, shame, pride, indignation, and many other emotions for which theory expects no comparable experience for an asocial creature.

Summary

To summarize my views on emotions, emotions are strings and we are puppets hanging from them. The puppet master, the ultimate master for whom we dance upon the stage of the world like marionettes, is our genome. As we are enculturated within a society, we learn the consequences of our behavior in terms of emotions, feelings of guilt, shame, accomplishment or appreciation, just as we learn what foods taste like: experience interacting with innate faculties producing emotional dispositions. Culture is software, but the emotional subsystems with which we interact with culture are hardware. When children aren't too busy stuffing myriad objects into their mouths, they experiment with social behaviors like hitting, grabbing, cheating, stealing, and also sharing, cooperating, and hugging. Hopefully, there is a parent or caretaker around to apply the right amounts of discipline and praise to equip them with the correct associations of emotions with behaviors that will assist them in comporting their behavior with what society expects of them. The social emotions compress this vital information in the same way bout past social interactions, just as taste compresses information about the nutritional content of honey. They tell us who is safe to cooperate with, and who must be avoided. Further, they enable us to cultivate relationships with others that deepen cooperation, through the proliferation of norms of behavior.

Continue to Part 2: What is Free Will, and why do we think we have it?

No comments:

Post a Comment