From language-ready brain to languages

This talk will address the bridge from emergence of the language-ready brain (as described by the mirror system hypothesis, MSH) to the actual development of full-blown human languages, playing variations on the theme of “construction”: the evolutionary concept of niche construction, the practical skill of constructing more complex objects from simpler components, and the notion that languages can be seen as combining a lexicon with a construction grammar that combines form and meaning. I will assess the relevance of two approaches to Construction Grammar: Luc Steels and his colleagues see Fluid Construction Grammar (FCG) as part of a larger system embedded in each of two or more interacting robots. More specifically, FCG is coupled to a visual system and a motor system in each robot. FCG has been developed to support modeling of cultural evolution of human language. I will explore the question: “If we assess the capabilities of an FCG-endowed robot, to what extent do they model underlying properties akin to the innate perceptual, motor and learning capacities of the human brain, and to what extent do they correspond to properties that the human child only gains as a result of development within a human society richly structured by prior cultural evolution?” A (somewhat) related issue: Peter Dominey's approach to a neural implementation of a (limited portion) of CG is based on a prior model of cortico-striatal interactions underlying oculomotor control and learning. In terms of MSH, one wants to know (i) What is the analogous model for general primate manual control (i.e., shared by macaque, chimpanzee and human)?; and (ii) CanMSH be refined to support the transition from such a system to one that can support language? In the process, I will offer a preliminary assessment of how well Dominey's approach to embedding constructions in the brain stands up to the demands of such an account. Background Reading Arbib (2011a) combines an exposition of MSH with some ideas on the way complex imitation may have fostered the emergence of language and sophisticated tool use, with the latter building on the work of Dietrich Stout (2011). An initial linkage of diverse notions of construction (Arbib, 2012) appeared as a commentary to a target article in BBS (Vaesen, 2012). Arbib (2011b) develops the theme of niche construction by building on and critiquing the approach taken by Bickerton (2009) in his book Adam’s Tongue. My group’s perspective on construction grammar can be seen in several papers (e.g., Arbib, Gasser, & Barrès, 2014; Barrès & Lee, 2014). Many papers explore the diverse work of Luc Steels and his colleagues on FCG (e.g., Beuls & Steels, 2013; Steels, 2013; Steels & De Bele, 2006; Steels & Spranger, 2008). Peter Dominey’s expertise in modeling the role of the basal ganglia in the macaque neurophysiology of saccades (Dominey, Arbib, & Joseph, 1995) provided the basis for his