by Yohan J. John
The osmosis of neuroscience into popular culture is neatly symbolized by a phenomenon I recently chanced upon: neurochemical-inspired jewellery. It appears there is a market for silvery pendants shaped like molecules of dopamine, serotonin, acetylcholine, norepinephrine and other celebrity neurotransmitters. Under pictures of dopamine necklaces, the neuro-jewellers have placed words like “love”, “passion”, or “pleasure”. Under serotonin they write “happiness” and “satisfaction”, and under norepinephrine, “alertness” and “energy”. These associations presumably stem from the view that the brain is a chemical soup in which each ingredient generates a distinct emotion, mood, or feeling. Subjective experience, according to this view, is the sum total of the contributions of each “mood molecule”. If we strip away the modern scientific veneer, the chemical soup idea evokes the four humors of ancient Greek medicine: black bile to make you melancholic, yellow bile to make you choleric, phlegm to make you phlegmatic, and blood to make you sanguine.
A dopamine pendant worn round the neck as a symbol for bliss is emblematic of modern society's attitude towards current scientific research. A multifaceted — and only partially understood — set of experiments is hastily distilled into an easily marketed molecule of folk wisdom. Having filtered out the messy details, we are left with an ornamental nugget of thought that appears both novel and reassuringly commonsensical. But does neuroscience really support this reductionist view of human subjectivity? Can our psychological states be understood in terms of a handful of chemicals? Does neuroscience therefore pose a problem for a more holistic view, in which humans are integrated in social and environmental networks? In other words, are the “chemical self” and the “social self” mutually exclusive concepts?
I. Dopamine: The Tastiest Ingredient in the Chemical Soup?
It's actually somewhat difficult to pinpoint when exactly the chemical soup conception of the self escaped from the laboratory and started breeding in the pop cultural sewers. For the sake of brevity let's focus on what may be the most famous ingredient in the soup: dopamine. The idea that dopamine is responsible for the feelings of pleasure that come from food, sex, love and recreational drugs is ubiquitous these days — it circulates in pop science books, newspaper articles, blog posts, and meme images, growing in strength with each repetition. A google search for dopamine and the phrase “pleasure chemical” yields tens of thousands of results, including several from mainstream newspapers and science education websites. For example, the “edutainment” website HowStuffWorks, in an article on how love works, writes that dopamine contributes to love by “producing a feeling of bliss”.
Did neuroscience ever actually establish that dopamine produces a subjective feeling of bliss? Not really. In the late 70s and early 80s a neuroscientist named Roy A. Wise developed the “anhedonia hypothesis“, which proposed that a lack of dopamine prevents pleasure. This was initially inferred from research on how rewards and punishments motivate animals to perform particular actions. If you regularly give a lab animal a reward (like food or water) when they perform an action (like pressing a lever), they will learn to perform that action more often. Certain drugs that affect the dopamine system seem to weaken the ability of animals to associate actions with rewards. It is reasonable to assume that a reward like food is accompanied by a subjective feeling of pleasure, and that animals will modify their behavior to maximize that subjective feeling. Wise therefore made the entirely reasonable and testable hypothesis that dopamine mediates the subjective pleasure of rewards, thereby facilitating the association between actions and rewards. 
Early support for this hypothesis also came from studies of the addictive properties of recreational drugs that elevate dopamine levels. When given unlimited access to cocaine, amphetamine or heroin, rats and monkeys would self-administer these drugs to the point of starvation . This kind of animal research seems to fit in with many people's prejudices regarding human drug addicts. Drugs of abuse must be so pleasurable that irresponsible hedonists can't help but gravitate towards them! With ideas like this in circulation, the anhedonia hypothesis — that low dopamine levels decrease pleasure — easily morphed into a hedonia hypothesis — that high dopamine levels enhance pleasure. The idea was popular enough that it almost eclipsed older knowledge. Long before dopamine was ever imagined to be the brain's in-house party drug, it was known to play a role in more mundane tasks such as the control of bodily movement. The involvement of dopamine in muscle movement and control is readily inferred from the observation that Parkinson's disease, which is associated with degeneration of a group of dopamine-releasing neurons in a part of the brain called the substantia nigra pars compacta, causes tremors and difficulty in initiating movements. These motor symptoms are partially alleviated by drugs such as L-DOPA, which elevate dopamine levels.
By the 1990s researchers were picking up plenty of evidence that the link between dopamine and subjective pleasure was somewhat tenuous. In the face of the new data, Wise himself later abandoned the subjective aspects of anhedonia hypothesis. Researchers discovered that dopamine is also related to motivation, to uncertainty, to errors in predicting future rewards, and even to negative experiences such as stress. In fact the earliest academic reference to the phrase “pleasure chemical” that I could find was in an article from 1996 that reviewed evidence against the idea that dopamine causes pleasure. But by this time the genie was out of the bottle, and has refused to go back ever since. The pop science filter retained the sexy bits of the research and discarded the rest.
If you're committed to the chemical soup idea, you might admit that the evidence suggests that dopamine isn't the bliss chemical, but that neuroscientists just need to keep looking until they find the right chemical. This is not an unreasonable working hypothesis. But it rests on the assumption that we only need to look inside the body for the causes of subjective feeling. What about hypotheses that include external factors? As it turns out, there are ways to study how social and environmental conditions interact with biochemical forces. Let's return to the finding that animals will self-administer drugs like cocaine and amphetamine until they starve to death. What sort of situation were the lab animals placed in?
Rats, as it turns out, are social animals. But rats in addiction experiments are often confined to small cages and denied normal social engagement — they are thus already in a state of unnatural stress prior to being given access to the drugs. They might just be self-administering drugs because there is little else for them to do. Some studies, such as Bruce Alexander's Rat Park experiments from the 1970s, suggest that when rats are free to move, socialize and eat as they please, they are less likely to become addicted to cocaine or morphine . Our understanding of addiction is far from complete, and there may be issues with the Rat Park experiments, but given how problematic the “dopamine produces bliss” research is, we have to ask the following question: Why did popular culture absorb the molecular bliss story and ignore the story of well-adjusted rats who were capable of just saying no? In other words, if there are people out there who want a pendant that represents the scientific approach to bliss or pleasure, why choose a dopamine molecule over, say, a depiction of rats having fun? To address this issue it might help to take a detour into the role of reductionism in modern society.
II. Parts and Wholes
Reductionism is the idea that to understand a phenomenal whole we just need to break it up into simpler parts. This process is known as “analysis” — which comes from the Greek for “breaking up”. The complementary process is “synthesis”, which comes from the Greek for “compounding” or “putting together”. The reductionist worldview, armed with the tools of analysis and synthesis, has been extraordinarily successful in physics and chemistry. Matter was broken up into atoms, which in turn were split up into leptons and quarks, the subatomic particles that are typically seen as the building blocks of the universe. Scientists and engineers are constantly finding new ways to assemble atoms and molecules, synthesizing substances that may never have existed before. More recently biology has also yielded some of its secrets to the reductionist's probes. Biological matter is also decomposable into atoms, but biology provides us with its own distinctive building blocks at the level of DNA, proteins and other large biological molecules. Dopamine and the other ingredients of the chemical soup show up at this level. Perhaps some people see them as the atoms of experience.
When scientists break up an object or process into smaller parts, they often say that the parts “give rise” to the whole. The analytic perspective is conducive to thinking about causality as a one-way street that extends from the very small to the very big. In the reductionist's Great Chain of Being, subatomic particles give rise to atoms, atoms give rise to molecules, and molecules give rise to the myriad forms of inanimate and animate matter we can sense directly. But what do we mean when we say that one thing “gives rise” to another? In many cases scientists haven't actually shown how the parts can be synthesized to produce the whole — in practical terms some of the links in the chain from the small to the big are missing. The most prominent examples come from biology: we can show that organisms can be decomposed into cells, and cells can be decomposed into biomolecules, but we still have a long way to go before we can start with a set of biomolecules and assemble a whole organism. Synthesis is not just analysis in reverse. As anyone who has broken a delicate object knows, it is far easier to smash something to smithereens than it is to put it back together. Causality may not be a one-way street after all.
The holistic perspective invites us to consider causal factors that reductionism often neglects. For many people the word “holism” connotes new age thinking, critical theory or social constructivism — modes of thought that are often portrayed as irrational, anti-scientific or obscurantist. I'd like to suggest that there is a form of holism that we can talk about without abandoning rationality, science, or clarity of expression. According to a holistic, non-reductionist perspective, the causal relationship between parts and wholes is not a unidirectional flow from parts to wholes. A reductionist conception of reality is a “bottom-up” notion of causality, whereas a holistic conception also includes “top-down” forces. A non-reductionist scientist does not deny that a whole derives many of its properties from its parts. But in addition, a non-reductionist will typically point out that in many situations the properties displayed by the parts are in a sense derived from the whole. The way something behaves does not depend solely on its constituent parts, but also on factors that are external to the parts. The arrangement of parts matters as much as the properties of the individual parts.
Consider the variety of substances composed solely of carbon atoms. In nature, carbon takes the forms of graphite and diamond. Scientists have also created artificial allotropes like buckminsterfullerene, graphene, and carbon nanotubes. Each form of carbon has a distinct set of properties. Graphite is brittle, flaky, and greyish-black. Diamond is the hardest naturally occurring substance, and is transparent. The wildly divergent properties of carbon allotropes depend on something more than the intrinsic properties of individual atoms. This “something more” turns out to be structure; the arrangement of the atoms crucially determines the properties of the whole. A died-in-the-wool reductionist might wish to counter that the arrangement is somehow determined by the atoms themselves, but in the case of carbon this is clearly not the case. Diamond crystals do not spontaneously form when you place a handful of carbon atoms together. The formation of diamond requires very specific conditions of pressure and temperature. If these conditions are not met, other forms of carbon may emerge. What carbon atoms do when they get together depends not only on their distinctive properties, but also on contextual or environmental factors that are external to them. To understand an object we must not only uncover what its building blocks are, but also the organizing principles that determine how the blocks are arranged.
Integrating holism and reductionism requires incorporating both bottom-up factors (the building blocks) and top-down factors (the organizing principles). To understand an object or phenomenon both as an integral whole and as a part of a larger process, you must consider at least three levels of analysis. There is the level of the object itself — a diamond crystal for instance. At this level you discover the object's properties like hardness or transparency. Below this is the level of the parts — the carbon atoms. At this level you discover how the properties of carbon atoms allow for (but do not ensure) the distinctive tetrahedral bonds that give diamonds their unique properties. Finally there is the level of the environment — the temperature, pressure and other conditions that induce a particular structure or arrangement on the constituent parts. At this level you discover the external factors that lead to diamond formation. A “pure” reductionist might seek to derive everything from parts, whereas a “pure” non-reductionist might instead point to environmental conditions as the most important causal factors. To speak of external factors is not to deny low-level causality, but to more fully acknowledge that causality in not a one-way street but a criss-crossing web of interactions across multiple levels of organization. 
III. The Chemical versus the Social — Can It Be Both?
We can now return to the chemical soup idea of human experience. Biology has in recent decades become strongly influenced by reductionist thinking. It has become more and more popular to account for a psychological phenomenon using concepts like a “gene for addiction” or a “pleasure chemical” or even an “empathy neuron” . A predictable backlash has also ensued, with journalists and academics from a variety of backgrounds pointing out that reductive explanations of complex mental phenomena leave out causally relevant environmental and social factors.
The debate between reductionists on the one side and holists, anti-reductionists and social constructivists on the other can give the impression that chemical explanations and social explanations are mutually exclusive. This is not unlike the tired old nature-versus-nurture debate. The most honest solution to this apparent dichotomy is to reject it. Nature and nurture are intricately linked. Nurture makes manifest the possibilities latent in an organism's nature. A similar relationship holds between the chemical conception of the self and the social conception. Biochemicals are the media through which social and environmental patterns become manifest inside a person as the neural correlates of experience. Society and the environment, meanwhile, are the media through which chemical changes in an individual become manifest outside a person as actions and reactions. We can describe the link between the social and chemical conceptions of the self in terms of two interlinked feedback loops. The first loop links body, brain and mind, mediating how biochemicals influence a person's behavior and how behavior (and its consequences) in turn influences the person's biochemistry. The second feedback loop links the mind and body with the environment and with society, mediating how a person's behavior influences the outside world, and how the outside world in turn influences a person's behavior.
The radical reductionist sees the self as a whole determined by biochemical parts. The radical social constructivist sees the self as a part determined by a social and environmental whole. Both of these perspectives miss the fact that the two levels are always interacting with each other. Just as temperature and pressure induce carbon atoms to arrange themselves into a diamond, the external world induces a person's biochemical processes to arrange themselves into experiential states. Just as the intrinsic properties of carbon atoms allow for the formation of diamond bonds, the biochemical properties of a person's brain and body allow for the formation of social bonds. The self, if it to be found anywhere, should be sought at the boundary, the interface, between the body-mind loop and the mind-world loop.
Of these two feedback loops, it seems as if the mind-world loop is more frequently neglected, at least in the popular domain. To describe the pleasures, pains, hopes, and fears of people as nothing more than chemical fluctuations is to undervalue the causal role of social and environmental factors. And to imply that all scientists approve of such reductionist half-truths is to ignore the diversity of academic opinion, painting science as a one-eyed monster, blind to human interaction. Echoing Margaret Thatcher, this Neuro-Cyclops seems to be saying: “There is no such thing as society. There are only individual molecules.” The parallelism between popular culture and right-wing ideology might not be coincidental. Just as a person can attribute his or her psychological problems to chemical shortcomings, a society can attribute its problems to individual shortcomings. This is convenient for anyone invested in preserving the socio-economic status quo. If one believes solely in bottom-up causality, then dealing with social problems only requires the right pills or the right loans (or the right drones) targeted at the right individuals. The top-down power relations that give societies their structure can then be treated either as nonexistent or as the cold unchangeable facts of nature, rather than as possible targets for political reform.
In psychology, reductionism is the position that neurochemicals are ontologically prior to mental states. In politics, reductionism is the position that individuals are ontologically prior to society or to ecology. These two forms of reductionism — atomism for material phenomena and individualism for social phenomena — resonate with each other. Perhaps this process of mutual reinforcement has something to do with the popularity of the Chemical Self over the Social Self. The evolutionary biologists Richard Levins and Richard Lewontin point out that our explanations of phenomena at the intersection of biology and society are frequently filtered through ideological lenses . When explaining the prevalence of tuberculosis in the 19th century and its subsequent decline, people usually bring up the discovery of treatments to deal with bacterial infections. Bacteria are clearly involved in the bottom-up aspects of the causal web. But why did tuberculosis afflict some people and not others? Levins and Lewontin remind us that the poor nutrition levels and unsanitary living conditions of industrial workers may well have contributed to the top-down component of the causal web. Similarly, brain chemicals play a bottom-up causal role in establishing particular psychological states, but social structures play an equally important top-down role in determining which groups of people are more likely to experience pain and anxiety, and which are more likely to experience pleasure and bliss.
IV. The starry heavens above me and the chemical soup within me
Lest I give the impression that critics of reductionism merely have a political axe to grind, I will now try to demonstrate how impoverished the Chemical Self picture is, even if all we care about is an accurate causal narrative. A thought experiment involving an actual moment of pleasure can highlight some of the forces left out by reductionist conceptions of subjectivity. Let us imagine the following scene: an astronomer is gazing through a telescope in her backyard. Her radio plays music softly, while her dog sniffs around on the patio. The dog bumps into her telescope, shifting the telescope slightly. When the astronomer refocuses the telescope, she recognizes a star that conjures up memories of her childhood. These memories set off a wave of emotions and behaviors: pleasure, nostalgia, tears, wistfulness, quiet laughter. The causal story we tell ourselves to explain her reactions will involve bursts of dopamine and other neurotransmitters, but only the laziest description of the actual experience will stop at the neurochemical level.
The astronomer is receiving a physical message from a star — photons born of atoms in a stellar furnace have travelled light-years to give up their energetic ghosts upon her retinas. But a part of the message they carried lives on: it is transduced into an electrochemical signal that percolates through her nervous system. This signal moves through her visual system in a matter of milliseconds, eventually becoming a coherent, conscious percept somewhere along the way. The visual percept ripples through her brain, inviting other regions to participate in a new dynamic motif woven into the unceasing pattern of neural activity. The percept finds its way to the pattern-recognition areas of the brain, which chime in with the names of the constellation and of the star. As the perceptual pattern moves and grows, it begins to trigger reactions. Brain regions involved in memory, emotion, language, and bodily control are all notified.
Perhaps the first reaction stirred is surprise. The astronomer almost forgot about that star. An expectation of what she ought to be seeing was quietly building up — perhaps in her prefrontal cortex — but it doesn't match what she sees. This mismatch between expectation and reality causes groups of neurons in various places to fire. Perhaps some of her dopamine cells are involved in sounding the alarm. But these neurons depend for their activity on other cells, such as those involved in pattern-recognition, and in the setting up of expectations. Dopamine neurons fire in response to particular configurations of neural activity arriving from other parts of the brain, and what those configurations are depends on the astronomer's experiential history, which in turn depends on her social, economic and ecological circumstances. Pleasure does not magically arrive at these neurons through the aether.
The astronomer's habitual chain of thought has been disrupted, and now memories begin to manifest themselves, perhaps emanating from the hippocampus or other parts of the temporal lobe. She recalls being introduced to star-gazing by her mother. She recalls her first telescope, given to her on her 7th birthday. She recalls the book of ancient astronomical myths that her late grandfather gave her. She recalls her college astronomy club. Her reactions reverberate in feedback loops, mingling with each other and with her initial pleasant surprise. A whole chain of bodily reactions has also been triggered. A smile, a sharp intake of breath, a change in heart rate. Meanwhile, new memories are already forming and making links with old ones. The dog has found a dead bird in the yard, and the astronomer is now trying to prevent him from eating it. Genes linked with memory-formation are being expressed in neurons, shaping new synaptic connections. Perhaps new neurons are being born, aiding the memory process. In the brain of our astronomer, the incident involving the dog and the dead bird is forming real physical connections with all the other experiences evoked by that star.
Many of these internal processes contribute to the astronomer's subjective experience. A dopamine drug alone cannot possibly generate the same pattern of neural and bodily events, for the simple reason that it would not be capable of triggering the same perceptual states. And the perceptual states are not just patterns of light dancing on the retinas. The astronomer's unique personal history has quite literally shaped the synapses and pathways in her brain that allow her to recognize perceptual patterns and assign cognitive and emotional meanings to them. The history of her family, and of her culture, and of her species also play crucial causal roles in setting the stage for this moment of pleasure. Her family inspired in her a passion for astronomy, and had the resources to support her in her pursuit of this passion. The culture she is part of allowed her to go to college and study a subject of her choice, without discrimination. In the absence of these forces, her brain and body would not have been modified in ways that allowed her to control a telescope, to recognize a star, and to derive pleasure from the experience. Forces operating on vastly different spatial and temporal scales collaborated to produce her moment of pleasure on this night.
Imagine if a dopamine pendant could conjure up this causal tapestry, rather than parodying the excesses of reductionism. Imagine if a molecule-shaped trinket could remind us that conscious experience binds together the astronomical and the microscopic, the social and the solitary, the cultural and the natural.
But then again, maybe this is too much to ask of one little molecule.
Notes and References
 A great deal of information on the various roles of dopamine in the brain can be found on Wikipedia and Scholarpedia. A recent article by Roy A. Wise. covers the historical development of the anhedonia hypothesis, and also point out that he later abandoned the idea that dopamine was involved in the subjective feeling of pleasure. Another article by Wise that is available for free on PubMed covers similar territory. An article by John D. Salamone (1996) reviews some of the evidence against the idea that dopamine causes subjective pleasure.
 For example, Aigner & Balster (1978) showed that monkeys will self-administer cocaine to the point of starvation.
 There are several excellent scientific critiques of reductionism. Perhaps most famous is physics Nobel Laureate Philip W. Anderson's classic paper “More is Different”. In it, he says “The behavior of large and complex aggregates of elementary particles, it turns out, is not to be understood in terms of a simple extrapolation of the properties of a few particles. Instead, at each level of complexity entirely new properties appear and the understanding of the new behaviors requires research which I think is as fundamental in its nature as any other.” Another physics Nobel Laureate, Robert Laughlin, criticizes reductionism in the paper “The Theory of Everything” written with David Pines. Laughlin argues persuasively for emergentism in his very readable pop science book A Different Universe: Reinventing Physics from the Bottom Down. For an interesting formal exploration of emergence check out the paper: Emergence is coupled to scope, not level. Richard Levins and Richard Lewontin mount a spirited attack on reductionism in biology in their book The Dialectical Biologist.
 “Mirror neurons” have been proposed as a possible neural basis for empathy. I explore some of the problems with this idea in this blog post: Do mirror neurons explain understanding, or is it the other way round?