The Neural Basis of Free Will: Criterial Causation

by Peter Tse

The issues of mental causation, consciousness, and free will have vexed philosophers since Plato. In this book, Peter Tse examines these unresolved issues from a neuroscientific perspective. In contrast with philosophers who use logic rather than data to argue whether mental causation or consciousness can exist given unproven first assumptions, Tse proposes that we instead listen to what neurons have to say. Because the brain must already embody a solution to the mind--body problem, why not focus on how the brain actually realizes mental causation?

Tse draws on exciting recent neuroscientific data concerning how informational causation is realized in physical causation at the level of NMDA receptors, synapses, dendrites, neurons, and neuronal circuits. He argues that a particular kind of strong free will and "downward" mental causation are realized in rapid synaptic plasticity. Recent neurophysiological breakthroughs reveal that neurons function as criterial assessors of their inputs, which then change the criteria that will make other neurons fire in the future. Such informational causation cannot change the physical basis of information realized in the present, but it can change the physical basis of information that may be realized in the immediate future. This gets around the standard argument against free will centered on the impossibility of self-causation. Tse explores the ways that mental causation and qualia might be realized in this kind of neuronal and associated information-processing architecture, and considers the psychological and philosophical implications of having such an architecture realized in our brains.

Genres: Philosophy, Neuroscience, Science, Psychology, Nonfiction, Artificial Intelligence

456 pages, Hardcover

First published February 15, 2013

Reviews

[Pa · Rating 5 out of 5]

Finally we hear from a neuroscientist arguing that a form of free will is real. The book is extremely clearly written, and is actually quite brief (230 pages) because it is about half endnotes and indexes. It seems that most neuroscientists and philosophers are coming down on the side that free will is illusory, mainly on the basis of a handful of experiments by Libet and Wegner, all of which involve repetitive hand movements. Tse argues that free will is operative in the domain where we play scenarios out internally, weigh options and deliberate using our capacities of working memory, attention, and consciousness, all of which are closely related. Repetitive and automatized hand movements are not where the action is in free will he says. There are some good reviews of the book on amazon, but in a nutshell he argues that the neural code is not just one about how neural firings trigger neural firings, but is first and foremost a code in which neurons rapidly rewire each other by changing synaptic weights within milliseconds. This makes neurons respond to different information in the immediate future than they otherwise would have. He makes a compelling case that emerging evidence (which he documents meticulously) of rapid synaptic reweighting is a game changer for our understanding of mental causation in general,and volitional mental causation in particular.

[Mark Deardorff · Rating 5 out of 5]

I am a philosophical libertarian and believe, despite my acceptance of a divinity, in Man's absolute free will. That it is such without divine permission but irrespective of divine will. That being said, since John Archibald Wheeler first proposed that information was the ultimate reality, I was quite certain that freedom of action was unrestrained.

This book just gave me a little support for my intuitive beliefs. It is also useful in some pseudonymous work in preparation.

It is gratifying to some evidential reasoning that supplements the work of Dennet although the Wikipedia editors ignore this book.

[Arthur · Rating 4 out of 5]

This book would be five-stars if it was called anything other than: "the neural basis of free will." Tse writes a very compelling explanation of just how complex neural computation actually is. I learned a number of things about neurons that are glossed-over even in graduate level neuroscience courses, such as the Boolean computational possibilities of dendrites. Furthermore his theory of criterial causation is a very helpful way of thinking about neural computation. Indeed, even his explanation of the influence of quantum randomness of the functioning of neurons rides the fine line between exaggerated claims of "magical" influence and the overly reductionist perspective of ignoring their influence. Unfortunately, at least in my reading of the work, he was unable to show that all of this necessitates strong free will. Certainty the work shows how it is perhaps more possible, but the actual proof of it wasn't very convincing. For a work whose title suggests what it does, I was a little disappointing.

[Craig Martin · Rating 4 out of 5]

Thirty years ago, Benjamin Libet, an American neuroscientist, reported experimental findings that if you asked people to make voluntary movements, their brains initiate the movements before they become consciously aware of the intention to move, thus eliminating the possibility of free will. Everything is determined. Scientists and writers, including Robert Sapolsky Determined: A Science of Life without Free Will, have fashioned strawmen out of the results and built a hard determinism that feels cold. Those who can't go that far have constructed a compatibility framework - Compatibilism is the view that free will and determinism are not mutually exclusive and can coexist.

Peter Tse is a Professor of Psychological and Brain Sciences at Dartmouth College. He is an entertaining writer and presenter and takes great pains to demolish the significance of the Libet experiments and to rally against the hard determinists and the dead compatibilists (including David Hume and, more recently, Daniel Dennett). Tse is also a very competent neuroscientist, and he provides details down to the neuronal level for his belief in ‘Libertarian’ free will. He is passionate about the subject, and in addition to this book written in 2015, he has a series of excellent and highly recommended MOOCs on Coursera about the broader topic of free will (before reading his book, I completed the three modules of Libertarian Free Will: The Basics of Libertarian Free Will, Free Will and Neuroscience, and Neural Basis of Imagination, Free Will, and Morality).

It is dangerous to distil a detailed 240-page book (plus another 220 odd pages of Appendices, Notes and References) into a few critical statements, but here goes:

There is randomness at the quantum level, amplified at the macro level. The release of neurotransmitters and diffusion across a synaptic cleft (a 20-40 nanometer gap ‘connecting’ one neuron to another) exhibits a random walk. This creates randomness in spike timings. Actions in the brain regions combine many neurons, sending bursts of signals in a yet-to-be-discovered neural code. The code is non-algorithmic, and the future state of the neural network is, in part, determined by the response of the input signals to a collection of synaptic weights. Collections of inter-connected neural regions, each with many distinct types of neurons, and each served by mysterious glial cells, combine to bring the ‘being you’ into play.

Tse argues powerfully that introducing randomness from the microscopic domain to neural networks and, therefore, to the domain of mental causation is central to having free will in the 'strong' sense.

Humans are motivated by reward: incentive drives behaviour. At the neuronal level, this is through the release of dopamine, serotonin, and oxytocin, amongst others. We get adrenaline (norepinephrine) and dopamine-induced hits to consume calories to keep the system alive; we desire warmth when cold, triggered by signals (shivers, goosebumps) and our 'feeling of cold'. We also seek to reduce error signals in our patterned world. We like patterns - even seeing faces in objects, Pareidolia - and we resolve those patterns against our expected worldview (possibly through predictive coding). Beyond the unconscious, our conscious choice drives us to make good decisions. We consciously choose to follow a moral, cultural, or religious code, and that code sets a framework that adjusts our behaviours. Our goals may be at odds with self-preservation - sharing food or carrying out acts of heroism or self-sacrifice. So, our free will is a volitional chain reaction where our changed network of synaptic-weighted decision points conditions our motor outputs. We even learn new behaviours and controls, which can update our weights in milliseconds and over the years.

Tse also argues that the dominant view of neuronal computing (from McCulloch and Pitts in 1943, until today) is incomplete. These early models put all the ‘decision-making' at the heart of the neuron. In contrast, more recent work (including that by Christof Koch) indicates that the dendrites - the input connections to a cell - are active and can themselves carry out ‘computing’. The greater complexity of the human brain pushes against the enthusiasm of in-silico artificial neural networks as a proxy for what happens in the human brain. We are ‘non-computable’. (See Non-Computable You: What You Do That Artificial Intelligence Never Will).

Tse has written an important and detailed book that appeals to computational neurobiology, computational psychiatry, neuroscience enthusiasts, and those interested in philosophy and the theory of mind (there are even several references to Ludwig Wittgenstein). Tse is well-read and well-researched.

Tse has articulated a good argument against Hard and Compatibilist determinism. He is a physicalist and a strong advocate for Free Will from a purely libertarian perspective. Although his German mother was a woman of faith, he sees no place for religion in his writing - like Hume. He does, however, talk in his courses about the neural basis for meditation and mindfulness, perhaps influenced by his paternal Chinese heritage.

Through my own volition, I read the book, liked it, and gave it four stars.

[Cashy John · Rating 5 out of 5]

This book is not only about free will. It is more about how mental events can be causal in the brain and about the role of attention and consciousness in that causal chain. A very interesting view of how the brain works. Highly recommended for those who love science and the deep questions.