Simulated Evolution of Discourse with Coupled Recurrent Networks

The origin and evolution of language have been the subjects of numerous debates and hypotheses. Nevertheless, they remain difficult to study in a scientific manner. In this paper, we focus on the string-context mutual segmentation hypothesis proposed by Merker and Okanoya, which is based on experimental findings related to animal songs. As a first step in formally exploring this hypothesis, we model the evolution of agent discourse using coupled recurrent networks (RNNs). This model is a simplified representation of this hypothesis; that is, agents are situated in a single context (e.g., behavioral, social, or environmental) and they mutually learn their utterance strings from the prediction dynamics of their RNNs. Our simulation demonstrates the emergence of shared utterance patterns, which are culturally transmitted from one generation to the next. Furthermore, the distribution of the shared patterns changes over the course of this evolution. These findings demonstrate an important aspect of language evolution: "language shaped by society."

[1]  T. Ikegami,et al.  Emergence of net-grammar in communicating agents. , 1996, Bio Systems.

[2]  George Kingsley Zipf,et al.  Human behavior and the principle of least effort , 1949 .

[3]  Makoto Taiji,et al.  Dynamics of internal models in game players , 1998, adap-org/9807004.

[4]  Jeffrey L. Elman,et al.  Language as a dynamical system , 1996 .

[5]  Inman Harvey,et al.  Fourth European Conference on Artificial Life , 1997 .

[6]  Charles E. Taylor,et al.  Artificial Life II , 1991 .

[7]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[8]  Angelo Cangelosi,et al.  Emergence of Communication and Language , 2006 .

[9]  Alison Wray.  Dual processing in protolanguage: performance without competence , 2002 .

[10]  Alison Wray,et al.  The transition to language , 2002 .

[11]  Takashi Ikegami,et al.  Evolution of Birdsong Syntax by Interjection Communication , 2007, Artificial Life.

[12]  Morten H. Christiansen,et al.  Learning to Segment Speech Using Multiple Cues: A Connectionist Model , 1998 .

[13]  Kazuo Okanoya,et al.  Neural Substrates for String-Context Mutual Segmentation: A Path to Human Language , 2007 .

[14]  Simon Kirby,et al.  Natural Language From Artificial Life , 2002, Artificial Life.

[15]  M A Nowak,et al.  Evolution of universal grammar. , 2001, Science.

[16]  B. Merker,et al.  The Natural History of Human Language: Bridging the Gaps without Magic , 2007 .

[17]  Angelo Cangelosi,et al.  Learning and the Evolution of Language: The Role of Cultural Variation and Learning Costs in the Baldwin Effect , 2002, Artificial Life.

[18]  Noam Chomsky,et al.  The faculty of language: what is it, who has it, and how did it evolve? , 2002 .

[19]  Jordan B. Pollack,et al.  The induction of dynamical recognizers , 1991, Machine Learning.

[20]  Janet Wiles,et al.  Methodological issues in simulating the emergence of language , 2002 .

[21]  S. Mithen,et al.  The Singing Neanderthals: the Origins of Music, Language, Mind and Body, by Steven Mithen. London: Weidenfeld & Nicholson, 2005. ISBN 0-297-64317-7 hardback £20 & US$25.2; ix+374 pp. , 2006, Cambridge Archaeological Journal.