A method for extracting as knowledge causal relationships between input variables and an output variable of a neural circuit model, said neural circuit model being of a hierarchical structure constructed of an input layer, at least one hidden layer and an output layer and having performed learning a limited number of times by determining weight factors between mutually-connected neuron element models in different layers of the input layer, hidden layer and output layer, wherein with respect to plural routes extending from a neuron element model, corresponding to a particular input variable, of the input layer to a neuron element model, corresponding to a particular output variable, of the output layer by way of the individual neuron element models of the hidden layer, a product of the weight factors for each of the routes is determined, and the products for the plural routes are summed, whereby the sum is employed as a measure for the determination of the causal relationship between the particular input variable and the particular output variable.