1. a model used in functional magnetic resonance imaging to describe the shape of responses. A neuron receives a number of inputs, each of which comes via a connection that has a strength or weight; these weights correspond to the synaptic efficacy of the neuron. Each neuron also has a single threshold value. Activation of the neuron is determined by the weighted sum of the inputs minus the threshold. The activation signal is passed through a transfer function to produce the output of the neuron.

2. in time-series analysis, a type of model used to forecast a time series that is influenced by present and past values of other time series.