4 Enlightenment
Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.
—David Hume (1711–1776)1
Our scientific method has no universal rule book; it is an uneasy coalition of divergent philosophies. On the one hand is the reductionist manifesto, expressed by Thomas Hobbes in the preface to De Cive in 1651:2
Everything is best understood by its constitutive causes. For, as in a watch or some such small engine, the matter, figure and motion of the wheels cannot well be known except it be taken asunder and viewed in parts.
Hobbes saw the mathematical certainty of geometry as that to which science should aspire. He viewed the purpose of science as being to establish causal explanations—deductions from observations to necessary conclusions. If we observe, for instance, the spiking behavior of a neuron, the reductionist invites us to ask why it behaves in that way. This question carries in two directions: what caused that behavior, and what are its consequences; and the reductionist seeks deterministic accounts of both.
On the other hand is the vision of science expressed by Karl Popper in Conjectures and Refutation (1963). This vision denied that absolute certainty can come by inductive reasoning from observations: there are always different ways of “explaining” observations, and science must find ways of choosing between them. We want explanations that are simple and elegant yet powerful—theories that imply more than we can know from the mere accumulation of facts. Such theories must forever be provisional: in predicting things that we do not yet know, they are forever open to disproof.
There is no more rational procedure than the method of trial and error—of conjecture and refutation: of boldly proposing theories; of trying our best to show that these are erroneous; and of accepting them tentatively if our critical efforts are unsuccessful.3
However, we cannot advance by always and forever questioning the foundations of our understanding. As Popper put it:
Science does not rest upon solid bedrock. The bold structure of its theories rises, as it were, above a swamp. It is like a building erected on piles. The piles are driven down from above into the swamp, but not down to any natural or “given” base; and if we stop driving the piles deeper, it is not because we have reached firm ground. We simply stop when we are satisfied that the piles are firm enough to carry the structure, at least for the time being.4
Neuroscience progresses through an uneasy combination of reductionism and the “leaping imagination” of Popperian science, and through discovery. New discoveries often arise from reductionist approaches. They often depend on technological advances: from molecular biology, in its ability to dissect, transform, and reconstruct the basic functions of cells; from physics, in new ways of imaging cells and the molecules and organelles within them; from mathematics, harnessing vast computational power in the analysis of structures and networks; or from chemistry, in the ability to manipulate the basic elements of neural signaling. Ideas are often inspired by discoveries. Discoveries about mechanisms in cells usually lead to some ideas that we can test by the same means that we employed to make those discoveries. But to understand how mechanisms in cells construct the behavior of complex systems comprising many interconnected cells is harder, and we must find ways to test our ideas, not just ways to advertise their possibility.
My home of Edinburgh was, at the close of the eighteenth century, the hub of the Scottish Enlightenment. It was home to Joseph Black, the founder of thermochemistry, who discovered carbon dioxide and magnesium; to his student Daniel Rutherford, who discovered nitrogen; to the geologist James Hutton, whose work revealed the deep time through which the world has evolved; to Adam Ferguson, the “father of modern sociology”; to Lord Monboddo, the founder of comparative linguistics; to Adam Smith, the author of The Wealth of Nations; to Colin Macfarquhar and Andrew Bell, who founded the Encyclopaedia Britannica; to Alexander Monro (primus), who founded the Medical School at Edinburgh and rescued anatomy from the dogma of classical authority; and to his son (secundus), who discovered the lymphatic system. Another native son was William Cullen, who recognized the importance of the mind in healing and the placebo effect, and who coined the term neurosis. His lectures and textbooks became famous internationally and grew the size and influence of the Medical School. One of his American students was inspired to name his son after him: William Cullen Bryant became the “first American poet.”
Friend to them all was the philosopher David Hume, whose skeptical approach led a later son of Edinburgh, the novelist Robert Louis Stevenson, to declare that Hume had “ruined philosophy and faith.”5 In his Enquiry Concerning Human Understanding (1748) Hume argued that any reasoning from mere observations is undermined both by the fallibility of our senses, on which all observations of fact depend, and by the logical inability to deduce causality from association however consistently two events might be associated in time.6 However often we have seen that day follows night, we cannot be certain that the sun will rise tomorrow, nor can we safely assume that it is darkness that causes the sun to rise.
This is not because the laws of nature might change while we sleep, but because we might have mistaken the laws by our inferences from incomplete evidence. We can never prove the truth of any theory, no matter how much evidence we might have assembled: tomorrow might yet bring a counterexample. Scientists therefore avoid talking of proof, almost as though it were unlucky to do so.
But although we can never prove a causal association we might yet disprove one, and Karl Popper’s answer to Hume’s “problem of induction”6 was that the scientific method must exploit this. In The Logic of Scientific Discovery he argued that we make progress only with the aid of imagination.4 From observations we construct hypotheses that we test by experiments, aiming not to verify them but to refute them. We develop our understanding by culling refuted theories, leaving only those that, for now, have withstood determined challenges. Popper’s vision has been embedded in the ethos of academic journals, research funders, and grant-awarding committees, all of which yearn for innovative hypotheses and “killer,” hypothesis-destroying experiments by which they can be tested.
If we observe the behavior of a neuron, the reductionist manifesto invites us to ask what caused that behavior, and what its consequences are. Here is the rub. We can speculate about these things, but as Hume put it, “causes and effects are discoverable, not by reason but by experience.”7
We gain experience through experiments, and from this we try to formulate hypotheses that can be tested. Often the only falsifiable hypotheses we can make are rather trivial: about immediate causes of a behavior of a particular neuron or about its direct consequences. We really want bold hypotheses, hypotheses about the meaning of its behavior for the whole organism—hypotheses about its physiological significance. But the behavior of any one neuron in a mouse or a man has no physiological significance. Every neuron receives thousands of inputs from hundreds of neurons and sends outputs to hundreds of others. Functions are enacted not by neurons acting alone but by populations acting on other populations.
If any neuron were fully representative of a population commonly engaged in a single function, then we might, by observing one neuron, glimpse that collective behavior. But all neurons are individuals, with individual quirks and eccentricities.
I, like you, am a human being. We share the same genes, with a few minor differences, but those did not solely determine who we are now. The environment into which we are born, our early life experiences and our interactions with close kin, define us also. Particular experiences might seem fleeting and ephemeral, yet they leave their mark on our behavior. These things that are true of you and me are true of any two neurons, however similar their genetic and developmental fates. As we will see, adjacent neurons serving a common physiological function can respond differently to the same stimulus. No neuron receives the same inputs as even its closest neighbor, or expresses exactly the same genes, or has the same morphology or the same intrinsic properties.
Our understanding of the brain is built of boxes inside boxes inside boxes inside boxes. The largest box contains about 20,000 genes and the hundreds of thousands of proteins that they encode. Some who rummage in this box are concerned with the structure of complex molecules and how the shape of a receptor enables a peptide to bind with it. Some worry about how the genes are regulated, what proteins bind to their regulatory regions, and how this affects the level of expression. Some study how proteins are cleaved by enzymes, and what chemical reactions occur. The next box contains the many different types of neurons. Here, the puzzles include understanding the properties of membranes that allow neurons to generate patterns of spike activity, the intracellular signaling pathways that are activated by signals received, the mechanisms by which activity is coupled to secretion and gene expression, and how all of these are influenced by physiological signals. The next box contains the clans of neurons, and here the questions are about how each clan is organized, how cells in a clan communicate with each other and with other clans. This box is about the networks of neurons, and how information flows through them. The smallest box of all is about systems: how the networks of clans achieve important ends, like maintaining a constant body weight through a lifetime of changing habits of eating and exercise, like delivering a baby, like recognizing a friend, like deciding when to run and when to fight. These boxes must fit together, but often it’s hard to see quite how they do. Although we think with our brains and through the stuff of our neurons, it is not clear that by studying neurons we can say much about how we think. The classical foundations of our understanding of the brain have served us well, but the question must arise as to whether these foundations are sound enough to explain the behavior of neuronal systems.
To understand the enactment of function we must explore the behavior of populations of neurons that act collectively to influence other populations. We must know what neurons in a clan have in common and what differences separate them from other clans. We must understand the interactions among neurons of a clan, and theirs with neighboring clans, and how this aggregated network responds to inputs from other networks of clans. We must also know a lot about the functions that the clans enact before we can formulate testable ideas about how those functions are enacted.
Among the advantages of neuroendocrine systems are that we can put names to the neurons we study. Vasopressin neurons do “what it says on the tin”: they secrete vasopressin, which acts on the kidneys to concentrate the urine. We can measure what is secreted and study its consequences. We can study the neurons in exquisite detail, and on a scale by which we can infer the behavior and properties of the population. Neuroendocrine systems are tractable, amenable to reductionist interrogation, and this enables us to build an understanding that can generate the bold hypotheses we seek. They are also important, if we accept that our health and survival are important. Understanding them must underpin our assault on obesity and diabetes, stress and depression, hypertension and infertility, and disorders of social and sexual behavior. I wouldn’t argue that we should study only systems that seem important in this narrow sense. When I lose my keys, as I often do, it might seem sensible to look in obvious places, but if they were in obvious places they wouldn’t be lost. It is often better not to look for your keys at all, but to do something else that leads you to different places, where you might find your keys or something much better.
Each neuroendocrine system is different from every other. Will anything we learn be relevant to the rest of the brain? I think so, for two reasons, both rooted in Hume’s Enquiry.
The utmost effort of human reason is to reduce the principles, productive of natural phenomena, to a greater simplicity, and to resolve the many particular effects into a few general causes, by means of reasonings from analogy, experience, and observation.7
Karl Popper, in The Logic of Scientific Discovery, emphasized the importance of imagination: “Bold ideas, unjustified anticipations, and speculative thought, are our only means for interpreting nature: our only organon, our only instrument, for grasping her.”8 Where do these ideas come from? They come, said Hume, not from our reason but from our experience, and they come as analogies. We see patterns in clouds and stars, emotion in abstract images, messages in parables, and metaphor in poetry. The oxytocin and vasopressin systems are important “model systems” in neuroscience not because they have revealed universal truths, but by inspiring hypotheses that might or might not hold elsewhere in the brain.
My second reason is that the hypothalamus affects our emotions and behavior. Some populations of neurons determine not only how much we eat and drink but also what and when we eat and drink. Others drive daily rhythms of drowsiness and wakefulness, others control aggression, others sexual behavior, others love and friendship, others our moods and our responses to stress and threat.
If we ask someone why they did some particular thing, as likely as not they will give a reason that is a “rationale” for their actions. But how much of our behavior is really governed by reason? How many reasons are mere rationalizations of behavior driven by passions, biases, and instincts unacknowledged? I’ll give Hume the last word:
This operation of the mind, by which we infer like effects from like causes, and vice versa, is so essential to the subsistence of all human creatures, it is not probable, that it could be trusted to the fallacious deductions of our reason, which is slow in its operations; appears not, in any degree, during the first years of infancy; and at best is, in every age and period of human life, extremely liable to error and mistake. It is more conformable to the ordinary wisdom of nature to secure so necessary an act of the mind, by some instinct or mechanical tendency, which may be infallible in its operations, may discover itself at the first appearance of life and thought, and may be independent of all the laboured deductions of the understanding.9
Notes
8. Popper, Karl (2002) The Logic of Scientific Discovery. Routledge Classics, 280.