Models of the Mind

two functions in the two different types of connections on to Purkinje cells. The granule cells pass along sensory signals; specifically, different granule cells fire depending on which sound is being played. The climbing fibres tell the cerebellum about the air puff; they fire when this annoyance is felt. Importantly, this means the climbing fibre

based coding.

Without a plan to say what action is taken, how do we know what the next state will be? This

The brain has hubs. In humans, they’re found sprinkled throughout the lobes. The cingulate, for example, which curves around the centre of the brain, serves as a hub; as does the precuneus, which sits atop the back of the cingulate.4 In studies of sleep, anaesthesia and people in comas, activity in these areas correlates with consciousness.

The rule says that if you want to know the probability that the die is weighted, you have to combine two different terms. The first is the probability that the rolls you’ve seen could come from a weighted die and the second is the probability of the die being weighted in the first place.

Attractors in a ring network, on the other hand, are continuous. With continuous attractors, transitioning between similar memories is easy. Rather than being thought of as a trampoline with people standing at different points, models with continuous attractor states are more like the gutter of a bowling lane: once the ball gets into the gutter it

If a neuron was very active in response to a particular image, for example, the connections from its very active inputs would be strengthened. As a result, that neuron would respond strongly to that and similar images in the future. This makes neurons responsive to specific shapes and different neurons diverge to have different responses. The netwo

balance must be struck between the cost of wiring and the benefit of information sharing. Small worlds do just this. In a small world, the more common connections are the relatively cheap ones between cells in a local cluster. The pricey connections between faraway neurons are rare, but there are enough to keep information flowing.

the visual system rapidly adapts its code to the motion information it’s currently getting. Specifically, to be an efficient encoder, a neuron should always fire at its peak firing rate for the fastest motion it sees and at its lowest for the slowest.

a balance between excitation and inhibition creates havoc in a neuron and helps explain the variability of the brain.