Language Evolution and Computation Bibliography

Computer models of evolutionary phenomena often assume that the fitness of an individual can be evaluated in isolation, but effective communication requires that individuals interact. Existing models directly reward speakers for improved behavior on the part of the listeners so that, essentially, effective communication is fitness. We present new models in which, even though 'speaking truthfully' provides no tangible benefit to the speaker, effective communication nonetheless evolves. A large population is spatially distributed so that 'communication range' approximately correlates with 'breeding range,' so that most of the time 'you'll be talking to family,' allowing kin selection to encourage the emergence of communication. However, the emergence of altruistic communication also creates niches that can be exploited by 'information parasites.' The new models display complex and subtle long-term dynamics as the global implications of such social dilemmas are played out.

Recurrent neural networks can be trained to recognize strings generated by context-free grammars, but the ability of the networks to do so depends on their having an appropriate set of initial connection weights. Simulations of evolution were performed on populations of simple recurrent networks where the selection criterion was the ability of the networks to recognize strings generated by grammars. The networks evolved sets of initial weights from which they could reliably learn to recognize the strings. In order to recognize if a string was generated by a given context-free grammar, it is necessary to use a stack or counter to keep track of the depth of embedding in the string. The networks that evolved in our simulations are able to use the values passed along their recurrent connections for this purpose. Furthermore, populations of networks can evolve a bias towards learning the underlying regularities in a class of related languages. These results suggest a new explanation for the ``critical period'' effects observed in the acquisition of language and other cognitive faculties. Instead of being the result of an exogenous maturational process, the degraded acquisition ability may be the result of the values of innately specified initial weights diverging in response to training on spurious input.