How Do We Make Decisions?
The most common question I get in response to the idea that experience is a property of individual cells is how we can make decisions if there are lots of experiences at once in our heads.
In response to this question I would ask what mechanisms we would expect to have evolved in our brains under pressure to make the optimum response to any particular sensory input. I think the answer would be that we would expect there to be mechanisms that allow us to pass copies of the sensory input to a very large number of modules at once so that we can obtain all of the vast range of possible relevant responses as quickly as possible. If one response is obviously salient it may simply beat the others to it, as when we grab at a vase about to fall over. If several responses compete for appropriateness we will want to allow some sort of weighted play off, but what we do not want is to have to go through all the possible responses one by one.
Thus if we see a face we will want our brains to send copies of the face to at least a few thousand modules and for those that most strongly signal a match to a known individual to produce a response on a single pass. If we see some pears on a fruit stall while hurrying to the bus stop we will want the sight of the pears to be sent to modules generating all sorts of responses. Do they look ripe? Do we remember if fruit from that stall is good quality? Would we miss the bus if we stopped to buy? Was the fruit bowl at home full or depleted this morning? Was there change in my pocket or would I need to get my wallet out? And so on and on. What we find is that if any of a hundred and one questions about the presence of the pears yields an obvious ‘No, forget it.’ it usually pops up at once without any need to go through a list of what the questions might be.
It seems fairly obvious, therefore, that the whole point of having massively parallel pathways in our brains is to allow us to send vast numbers of copies of sensory data to different places to answer all the relevant questions about it at once. What we think of as automatic decisions must depend on all the possible responses.
We might think that ‘considered’ decisions work in a rather different way that involves some ‘controlling agent’ weighing things up. Yet there is no particular reason why we should not have much the same system operating, if at a smaller scale. If we start to puzzle over whether the face in the film is that of Laurence Olivier or somebody rather like him, or maybe someone not like him that we associate or confuse with him I suspect that signals encoding the sense of puzzlement and the concept of the problem are sent to lots of places as before, maybe a bit further forward in the frontal lobes, and that the final decision draws on lots of responses that are now filtered by additional weightings triggered by inputs related to the sense of puzzlement and so on.
We have an intuition of being conscious of making our decisions, but as indicated in the section on consciousness as a causal relay race, we can only expect to be conscious of decisions having been made, along with a sense of being generated internally, just as for the sense of raising one's arm.
The basic fallacy that the question about how ‘we’ make decisions reveals is our intuitive sense that there has to be some single ‘controlling’ causal influence in a mind, often called an ‘agent’ by philosophers. ‘Control’ in this sense has no meaning in a coherent causal account. It is a pseudoconcept that we grow up with. There are no causes that are somehow ‘stronger’ than others, or that inject causation into an otherwise automatic world. The idea that there has to be a single copy of experience that is responded to by a controlling agent is pure supernaturalism. What is so odd is that neuroscience has rejected the idea of such a central controlling agent but in general refuses to abandon the idea, that would naturally go with it, that there is only one copy of experience. Once we are released from both ideas I suspect we can do some biology and make sense of what and where experience is.