Neural Network Models in Phonology

Wikimedia Commons

When we talk about neural network models, we are talking about artificial neural networks.






Artificial neural networks are made up of artificial neurons.





An artificial neuron is a mathematical model that (crudely) replicates the behavior of a real neuron by summing its inputs to produce an output.






The artificial neurons in a network have connections to each other, and/or to themselves.

Wikimedia Commons

Connections are either excitatory or inhibitory.


2 major categories of neural network models:

Cognitive Science-style                    Neuroscience-style

              (connectionist/connectivist)                                                                                                       

Norris 1994;                                                                                                   Rybak, Paton, & Schwaber 1997

Central pattern generators and the control of rhythmic movements

P & P applications for neural networks :
perceptual/auditory grouping, phonological learning, OT modeling, category formation, gestural coordination, . . . .