A connectionist model of English past tense and plural morphology

The acquisition of English noun and verb morphology is modeled using a sing le-system connectionist network. The network is trained to produce the plur als and past tense forms of a large corpus of monosyllabic English nouns an d verbs. The developmental trajectory of network performance is analyzed in detail and is shown to mimic a number of important features of the acquisi tion of English noun and verb morphology in young children. These include a n initial error-free period of performance on both nouns and verbs followed by a period of intermittent over-regularization of irregular nouns and ver bs. Errors in the model show evidence of phonological conditioning and freq uency effects. Furthermore, the network demonstrates a strong tendency to r egularize denominal verbs and deverbal nouns and masters the principles of voicing assimilation. Despite their incorporation into a single-system netw ork, nouns and verbs exhibit some important differences in their profiles o f acquisition. Most importantly, noun inflections are acquired earlier than verb inflections. The simulations generate several empirical predictions t hat can be used to evaluate further the suitability of this type of cogniti ve architecture in the domain of inflectional morphology.