I don't think this isn't fully known. Neural science has long been a fascination of mine and the conclusion I've come to is that it can happen in two ways. The

I don't think this isn't fully known. Neural science has long been a fascination of mine and the conclusion I've come to is that it can happen in two ways. The first is direct. In order for this to happen there must be a stable representation of the input and output. In this case, bright colors must have a stable representation, and fear must have a stable representation. This gives the opportunity for neurons to know how to connect the two so that one leads to the other. How does the individual neuron do that? The body is full of chemical gradients and they function kind of similar to landmarks that can be used for navigation by the growing neuron. The second is indirect. In this case there is nothing directly encoding the connection between brightness and fear, instead due to factors of the brain structure and the environment the animal exists in it is highly likely to learn the association. For instance the low-level visual processing and fear processing areas of the brain may be almost but not quite wired up. Now there may be a common insect that is brightly colored and just a little bit poisonous. Due to the innate curiousity the animal will try to eat the insect and get an upset stomach. The dormant connection is then molded to associate these two things.