The basic mechanism of decision making is [[evaluative neural network]] (ENN). ENN has two or more paths for reaching a goal. For every [[node]] in a path, there are several [[implications]]. Every implication has a value which is represented by the [[reward system]]. the value can very from positive to negative. The strength of the reward will be composed by the intensity of the reward that the reward system cell produce, and the closeness of the reward to the action, where immediate reward will increase the strength of the reward to the implication, while delayed reward will result reduced connection between implication and reward. The strength will be produced according to the learning rules of [[LTP]] and [[STP]].

'''[[FFFF]]''' and '''[[exploitation]]''' will heighten the threshold for using paths, therefore only strongest path will be used. If there aren't very strong paths available, no path will be selected, and the decision making system will freeze. The strongest path will be produced by conducting more repetitions in this path relative to other paths. '''[[PFC]]''' and '''[[exploration]]''' on the other side of the scale will reduce the threshold, and therefore will enable a larger variety of options. the option that will be selected will be the one with the most rewards in non-self control mode. In self-control mode, a grater rewards in the future will be preferred to immediate high rewards.