Online research in philosophy

It is not widely realised that Turing was probably the first person to consider building computing machines out of simple, neuron-like elements connected together into networks in a largely random manner. Turing called his networks unorganised machines. By the application of what he described as appropriate interference, mimicking education an unorganised machine can be trained to perform any task that a Turing machine can carry out, provided the number of neurons is sufficient. Turing proposed simulating both the behaviour of the network and the training process by means of a computer program. We outline Turing's connectionist project of 1948.

Consciousness is a pretty sexy topic right now, as the plethora of recent books on the subject demonstrate. Everyone is having a go at it: philosophers, psychologists, neuroscientists and physicists, to mention just a few. And for every discipline or sub-discipline that pretends to some insight on the matter we find not only a different explanatory strategy, but a different take on the explanandum – there is widespread disagreement about _what_ a theory of consciousness should actually explain. However, one thing seems to be agreed by all concerned: consciousness, whatever it is, is deeply mysterious.

Since there isn't a computer that seems conscious at this time, the idea of machine consciousness is supported by thought experiments. Here's one old chestnut: "What if you replaced your neurons one by one with neuron sized and shaped substitutes made of silicon chips that perfectly mimicked the chemical and electric functions of the originals? If you just replaced one single neuron, surely you'd feel the same. As you proceed, as more and more neurons are replaced, you'd stay conscious. Why wouldn't you still be conscious at the end of the process, when you'd reside in a brain shaped glob of silicon? And why couldn't the resulting replacement brain have been manufactured by some other means?".

We don’t need science in order to know that we are conscious or aware. But science suggests that we are conscious largely due to our brains. And science perhaps further suggests that our conscious states are identical to certain states of our brains. This later idea—that conscious states just are certain brain states (hereafter, the Identity Thesis)—is the main aim of this book to defend.

An easily-accessible introduction is provided for theauthor''s book Enchanted Looms , which is reviewedelsewhere in this volume by Jesse Prinz and by MarcelKinsbourne, and also for the article Didconsciousness evolve from self-paced probing of theenvironment, and not from reflexes? , which alsoappears in this volume and which summarises theauthor''s more recent thoughts on consciousness.

I dispute that consciousness is generated by core circuitry in the forebrain, with predominance of motor areas, as Cotterillproposes in Enchanted Looms and other theorists do also. Ipropose instead that conscious contents are the momentary modeof action of the integrated cortical field, expressed as a point vector ( dominant focus ), to which, in varying degree, allsectors of the network contribute. Consciousness is the brain''saccess to its own activity space, and is identical with the moment''sdominant mode of activity. The dominant focus is generally weightedtoward enactively encoded percepts. Anticipation and preparation,perception and action, inextricably interdigitate. I also dispute the view of Cotterill and others that consciousnesshas unique agency, which bestowed adaptive advantage when the brain evolved. Being identical with the activity of the network,consciousness can have no additional agency, and it can offerno adaptive advantages beyond those that characterize the network.

In Enchanted Looms , Rodney Cotterill defends the hypothesisthat conscious sensory experience depends on motor response. Thepositive evidence for this hypothesis is inconclusive, andnegative evidence can be marshaled against it. I present analternative hypothesis according to which consciousness involvesintermediate level sensory processing, attention, and workingmemory. The circuitry of consciousness can be dissociated fromaction systems and may mark an evolutionary advance from a priorphylogenetic stage in which motor outputs and sensory inputswere more intimately bound.

I address whether neural networks perform computations in the sense of computability theory and computer science. I explicate and defend
the following theses. (1) Many neural networks compute—they perform computations. (2) Some neural networks compute in a classical way.
Ordinary digital computers, which are very large networks of logic gates, belong in this class of neural networks. (3) Other neural networks
compute in a non-classical way. (4) Yet other neural networks do not perform computations. Brains may well fall into this last class.