Computers on the Brain: Why We Need Philosophers


Everyone says that brains are like computers. Well, maybe not everyone, but neuroscientists and philosophers of mind use this analogy in their attempts to understand how brains work. On the face of it, the comparison is clear. We know what computers are and we know what brains are, after all. Just like we know that a rock is a rock and an apple is an apple and a democracy is a democracy, right?


Actually, it’s not that simple, as Hebrew University philosopher Oron Shagrir explained yesterday in a talk he gave at Bar-Ilan University as part of its Science, Technology, and Society Colloquium.

There are different kinds of computers, Shagrir explained—and he didn’t mean PCs and Macs. Rather, there are different models by which computers can be built and can work.

Specifically, some computers—like the one you are presumably reading this on—are built on an algorithmic principle. Some sort of input is converted into a code (like a binary code of zeroes and ones), which the computer manipulates and combined with other data according to a given algorithm, or rule, to produce another set of code that is then recoverted into an output.

But there are other models of computers. Shagrir called one such model an “analogic” computer. In an analogic computer, the input is converted into a state of the computer which in some way “maps” the input. It is then manipulated and combined with other data to produce a new computer state that in turn is an image or map of the output.

Sound confusing? The essential point is that in an analogic computer the relationship between the data before and after manipulation is analogous—in other words, it bears some sort of resemblance or relationship to—the state of the world outside the computer. In an algorithmic model, the input is turned in the computer into an abstraction that, like all those zeroes and ones, bears no resemblance at all to the world outside.

You might say that in an analogic computer, if you get garbage out after putting garbage in, you’d also find something inside the computer that looks a lot like garbage. But if it were an algorithmic computer, all you’d see would be zeroes and ones.

Now, neuroscience research shows that the brain also takes input and manipulates it. Shagrir used the example of how the brain collects and manipulates data collected by the eye so that organisms can orient themselves in space. I won’t go into the details here, but basically the eyes provide two kinds of data, each of which is recorded by a different kind of cell in a region of the brain called the PPC, or posterior parietal cortex. One kind of cell registers the eye’s orientation in relation to the head and another the stimulation received by light hitting the retina. A third kind of cell combines this information into a map of where the head is located in space.

To make a long story short, what experiments show is that the brain combines the information recorded by the first two types of cells into a position in space using a mathematical relationship. The same mathematical relationship can also be used to describe the relationship between the head and the object in the outside world.

In other words, the brain computes analogically. The relationship between the states of the cells in the brain is analogous to the relationship between the things in the outside world that they are recording. It’s not just a bunch of zeroes and ones.

So what’s the problem? The problem, Shagrir said, is that when most philosophers of the mind talk about the brain as a computer, they base their theories on an algorithmic model of a computer. That is, when philosophers use the word “computer” in reference to the brain, they actually mean something different than what neuroscientists mean when they use the word “computer” in reference to the brain.

Which obviously makes it difficult for the philosophers and scientists to agree about what the brain is and how it works.

This is why we need philosophers. Philosophers are people who are trained to think carefully and clearly about what we mean when we make assertions of various types. Language can seem precise when it is actually muddled. If “computer” can mean two different things, consider how many conflicting and contradictory meanings words like “democracy,” “justice,” “victory,” not to mention “Jewish,” can have. And the muddle over words like these can cause a lot more agony and destruction than the confusion over what kind of computer we mean when we say the word “computer.”

2 thoughts on “Computers on the Brain: Why We Need Philosophers

  1. This strikes me as a little confused. In the first place it makes some highly materialist assumptions about the mind. No amount of speculative hand-waving and vagaries about the organisation inside the black box will solve the conceptual problems if the basic premises behind the confusion are left in place. I suggest Philosophical Foundations of Neuroscience and Bennet and Hacker’s discussion of the ‘mereological fallacy’ in neuroscience (p. 73).

    Sheldrake’s The Presence of the Past explains a useful analogy where the brain is likened to a television receiver. As Sheldrake points out the complete inability for the materialists to explain memory (and it isn’t for want of trying) should suggest there is something wrong. In Sheldrake’s metaphor it would be like hacking a TV receiver apart looking for the programme.

    you could also check out an article I have written, Consciousness Really Explained? from another angle, and also Henry Stapp’s Mindful Universe provides yet another angle in trying to drag neuroscience, kicking and screaming out of the nineteenth century.

    In summary, Physicalism is a busted flush. We should have realised this in the first half of the 20th century with the failure of classical physics to account for the very small, but the denial (surely) can’t be kept up for much longer.

Leave a Comment