The emergence of words: Modelling early language acquisition with a dynamic systems perspective

This paper introduces a computational model of early language acquisition that is able to build word-like units from cross-modal stimuli (acoustic and pseudo-visual). The architecture, data processing and internal representations of the model strives for ecological plausibility, and is therefore inspired by current cognitive views of preverbal infant language learning behaviour. In this paper, we attempt to visualise the emergence and development of the models internal representations as an epigenetic landscape, which is a popular method for depicting the evolution of behaviour through the dynamic systems theory. We show that our computational model, through a general statistical learning mechanism, displays similar properties to the dynamic systems theory and supports the empiricist view of human development.

[1]  Kj Friston,et al.  Dynamic patterns: The self-organization of brain and behavior , 1997 .

[2]  Dylan M. Jones,et al.  Perceptual organization masquerading as phonological storage: Further support for a perceptual-gestural view of short-term memory , 2006 .

[3]  John P. Pinto,et al.  Continuous processing in word recognition at 24 months , 1999, Cognition.

[4]  P. Kuhl.  Early language acquisition: cracking the speech code , 2004, Nature Reviews Neuroscience.

[5]  Roger K. Moore,et al.  The application of dynamic programming techniques to non-word based topic spotting , 1995, EUROSPEECH.

[6]  Guillaume Aimetti,et al.  Modelling Early Language Acquisition Skills: Towards a General Statistical Learning Mechanism , 2009, EACL.

[7]  J. Gerard Wolff,et al.  Language acquisition, data compression and generalization , 1982 .

[8]  Michael Don Palmer,et al.  Reflections on language , 1977 .

[9]  Linda B. Smith,et al.  Development as a dynamic system , 1992, Trends in Cognitive Sciences.

[10]  Unto K. Laine,et al.  A noise robust method for pattern discovery in quantized time series: the concept matrix approach , 2009, INTERSPEECH.

[11]  Bert Cranen,et al.  A computational model for unsupervised word discovery , 2007, INTERSPEECH.

[12]  D. Lewkowicz,et al.  A dynamic systems approach to the development of cognition and action. , 2007, Journal of cognitive neuroscience.

[13]  Linda B. Smith,et al.  Infants rapidly learn word-referent mappings via cross-situational statistics , 2008, Cognition.

[14]  Hugo Van hamme,et al.  Automatically learning the units of speech by non-negative matrix factorisation , 2007, INTERSPEECH.

[15]  Geert J. P. Savelsbergh,et al.  THE DEVELOPMENT OF COORDINATION IN INFANCY , 1993 .

[16]  B. Morrongiello,et al.  Crossmodal learning in newborn infants: Inferences about properties of auditory-visual events , 1998 .

[17]  Robert F. Port,et al.  The Dynamical Systems Hypothesis in Cognitive Science , 2006 .

[18]  Julia L. Evans.  The Emergence of Language: A Dynamical Systems Account , 2008 .

[19]  S. Counce.  The Strategy of the Genes , 1958, The Yale Journal of Biology and Medicine.

[20]  Linda B. Smith,et al.  A Dynamic Systems Approach to the Development of Cognition and Action , 2007, Journal of Cognitive Neuroscience.

[21]  James R. Glass,et al.  Unsupervised Pattern Discovery in Speech , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[22]  E. Thelen,et al.  Newborn stepping: An explanation for a "disappearing" reflex. , 1982 .

[23]  K. Newell,et al.  A dynamical systems interpretation of epigenetic landscapes for infant motor development , 2003 .

[24]  Joseph B. Kruskal,et al.  Time Warps, String Edits, and Macromolecules , 1999 .

[25]  S. Pinker.  The Language Instinct , 1994 .