A Connectionist Study on the Interplay of Nouns and Pronouns in Personal Pronoun Acquisition

Cascade-correlation learning is used to model pronoun acquisition in children. The cascade-correlation algorithm is a feed-forward neural network that builds its own topology from input and output units. Personal pronoun acquisition is an interesting non-linear problem in psychology. A mother will refer to her son as you and herself as me, but the son must infer for himself that when he speaks to his mother, she becomes you and he becomes me. Learning the shifting reference of these pronouns is a difficult task that most children master. We show that learning of two different noun-and-pronoun addressee patterns is consistent with naturalistic studies. We observe a surprising factor in pronoun reversal: increasing the amount of exposure to noun patterns can decrease or eliminate reversal errors in children.

[1]  T. Shultz,et al.  The learning of first and second person pronouns in English: network models and analysis , 1999, Journal of Child Language.

[2]  K. McRae,et al.  A Computational Developmental Model of the Implicit False Belief Task , 2008 .

[3]  Dirk van Rijn,et al.  Proceedings of the 31st annual conference of the Cognitive Science Society , 2003 .

[4]  Johanna D. Moore,et al.  Proceedings of the 28th Annual Conference of the Cognitive Science Society , 2005 .

[5]  K. McRae,et al.  Proceedings of the 30th Annual Conference of the Cognitive Science Society. , 2008 .

[6]  R. Charney,et al.  Speech roles and the development of personal pronouns , 1980, Journal of Child Language.

[7]  Thomas R. Shultz,et al.  Modeling Cognitive Development on Balance Scale Phenomena , 2004, Machine Learning.

[8]  S. Chiat,et al.  Context-specificity and generalization in the acquisition of pronominal distinctions , 1981, Journal of Child Language.

[9]  N. B. Schiff-Myers From pronoun reversals to correct pronoun usage: a case study of a normally developing child. , 1983, The Journal of speech and hearing disorders.

[10]  Bella M. DePaulo,et al.  The effect on language development of the special characteristics of speech addressed to children , 1978 .

[11]  William C. Schmidt,et al.  A Cascade-Correlation Model of Balance Scale Phenomena , 2000 .

[12]  Christian Lebiere,et al.  The Cascade-Correlation Learning Architecture , 1989, NIPS.

[13]  Thomas R. Schultz,et al.  A Connectionist Model of the Development of Transitivity , 2004 .

[14]  Thomas R. Shultz,et al.  A Compositional Neural-network Solution to Prime-number Testing , 2006 .

[15]  Thomas R. Shultz,et al.  Modeling Acquisition of a Torque Rule on the Balance-scale Task , 2009 .

[16]  T.R. Shultz,et al.  Network analyses: the case of first and second person pronouns , 1995, 1995 IEEE International Conference on Systems, Man and Cybernetics. Intelligent Systems for the 21st Century.

[17]  Y. Oshima-Takane,et al.  Children learn from speech not addressed to them: the case of personal pronouns , 1988, Journal of Child Language.

[18]  Thomas R. Shultz,et al.  A connectionist model of the learning of personal pronouns in English , 1994, COLT 1994.

[19]  Dirk Schlimm,et al.  Learning the structure of abstract groups , 2009 .

[20]  J. Oller,et al.  The diagnosis of language disorders in bilingual children: surface-oriented and pragmatic criteria. , 1983, The Journal of speech and hearing disorders.

[21]  Y. Oshima-Takane Analysis of pronominal errors: a case-study , 1992, Journal of Child Language.

[22]  T. Shultz,et al.  Learning the Structure of a Mathematical Group , 2007 .