A principal function of neuronal populations is to enhance profit, maximizing the organism's receipt of reward over time or their chance of survival under changing and unpredictable conditions. Yet, individually, most neurons contact only an exceedingly small fraction of the population and have no immediate access to information about changes in their external environment. We have set an experimental framework in order to study the interactions among neurons in monkey's brains. Using inferential statistics and the so called Generalized Linear Models (GLM), we statistically reconstruct the interactions between neurons during two different tasks, a learning task and a recalling task. We have reconstruct the neural networks for several monkeys performing these two tasks and by measuring the graph density and degree of the networks we have found that the recalling activity is usually more demanding than a learning process. This is an ongoing project so these results are only preliminary.