Let us look at various possible stages in the emergence of consciousness.
As a possible first stage there may evolve something that acts like a centralized warning, that is, like irritation or discomfort or pain, inducing the organism to stop an inadequate movement and to adopt some alternative behavior in its stead before it is too late, before too much damage has been done. The absence of a warning like pain will lead in many cases to destruction. Thus natural selection will favor those individuals that shrink back when they receive a signal indicating an inadequate movement; which means, anticipating the inherent danger of the movement. I suggest that pain may evolve as such a signal; and perhaps also fear.
As a second stage, we may consider that natural selection will favor those organisms that try out, by some method or other, the possible movements that might be adopted before they are executed. In this way, real trial-and-error behavior may be replaced, or preceded, by imagined or vicarious trial-and-error behavior. The imagining may perhaps initially consist of incipient efferent nervous signals, serving as a kind of model, or symbolic representation of the actual behavior, and of its possible results.
Richard Dawkins has brilliantly developed some such speculations about the beginnings of mind in considerable detail.18 The main points about them are two. One is that these beginnings of mind or consciousness should be favored by natural selection, simply because they mean the substitution of imagined or symbolic or vicarious behavior for real trials which, if erroneous, may have fatal consequences. The other is that we can here apply the ideas of selection and of downward causation to what is clearly a choice situation: the open program allows for possibilities to be played through tentatively — on a screen, as it were — in order that a selection can be made from among these possibilities.