Open challenges in understanding development and evolution of speech forms: The roles of embodied self-organization, motivation and active explora

📝 Abstract

This article discusses open scientific challenges for understanding development and evolution of speech forms, as a commentary to Moulin-Frier et al. (Moulin-Frier et al., 2015). Based on the analysis of mathematical models of the origins of speech forms, with a focus on their assumptions , we study the fundamental question of how speech can be formed out of non–speech, at both developmental and evolutionary scales. In particular, we emphasize the importance of embodied self-organization , as well as the role of mechanisms of motivation and active curiosity-driven exploration in speech formation. Finally , we discuss an evolutionary-developmental perspective of the origins of speech.

💡 Analysis

This article discusses open scientific challenges for understanding development and evolution of speech forms, as a commentary to Moulin-Frier et al. (Moulin-Frier et al., 2015). Based on the analysis of mathematical models of the origins of speech forms, with a focus on their assumptions , we study the fundamental question of how speech can be formed out of non–speech, at both developmental and evolutionary scales. In particular, we emphasize the importance of embodied self-organization , as well as the role of mechanisms of motivation and active curiosity-driven exploration in speech formation. Finally , we discuss an evolutionary-developmental perspective of the origins of speech.

📄 Content

Open  challenges  in  understanding  development  and  evolution  of  speech  forms:   the  roles  of  embodied  self-­‐organization,  motivation  and  active  exploration1  

  Pierre-­‐Yves  Oudeyer2   Inria,  France   Ensta  ParisTech,  France  

  Abstract:   This   article   discusses   open   scientific   challenges   for   understanding   development  and  evolution  of  speech  forms,  as  a  commentary  to  Moulin-­‐Frier  et   al.  (Moulin-­‐Frier  et  al.,  in  press).  Based  on  the  analysis  of  mathematical  models  of   the   origins   of   speech   forms,   with   a   focus   on   their   assumptions,   we   study   the   fundamental  question  of  how  speech  can  be  formed  out  of  non-­‐speech,  at  both   developmental   and   evolutionary   scales.   In   particular,   we   emphasize   the   importance  of  embodied  self-­‐organization,  as  well  as  the  role  of  mechanisms  of   motivation  and  active  curiosity-­‐driven  exploration  in  speech  formation.    Finally,   we  discuss  an  evolutionary-­‐developmental  perspective  of  the  origins  of  speech.    

  Keywords:   Origins   of   speech   forms,   self-­‐organization,   curiosity,   social   reinforcement,  active  exploration,  development,  evolution,  evo-­‐devo  

  1.  Comparing  theories  of  speech  formation  in  a  unified  Bayesian  framework  

  Studying   the   forms   and   formation   of   speech   has   long   been   a   topic   of   tremendous   interest  for  cognitive  science  in  general.  It  has  been  repeatedly  used  in  the  last  century   as  the  cradle  in  which  alternative  theories  of  language  as  well  as  sensorimotor  control   have  been  expressed  and  debated.  Jakobson  (Jakobson,  1941)  used  it  as  a  strong  ground   for  the  early  elaboration  of  structuralist  theories  of  cognition.  Later  on,  it  has  been  the   pivot   of   theories   of   perception,   and   their   potential   links   to   action   (Galantucci   et   al.,   2006),  as  well  as  theories  of  language  development  in  the  child  (Oller  et  al.,  2013).  It  has   also   gathered   efforts   in   the   quest   for   understanding   the   origins   of   language,   where   a   mystery   is   how   linguistic   forms   can   arise,   be   shared   and   evolve   in   a   population   of   individuals  (Steels,  2011;  Oudeyer,  2006;  Kirby  et  al.,  2014;  Moulin-­‐Frier  et  al.,  in  press).    

  Across  these  scientific  enterprises,  mathematical  and  computational  modeling  has  been   prominently  used  in  the  latest  decades,  grounded  in  the  physics  of  the  speech  system   and   in   the   dynamics   of   neural   and   learning   architectures.   Such   models   constitute   a   formal   language   allowing   us   to   formulate   and   analyze   precisely   hypotheses   about   complex   mechanisms.   Yet,   an   obstacle   to   scientific   progress   has   been   that   alternative   theories   have   often   been   expressed   through   different   formal   languages,   making   it   challenging   to   articulate   and   compare   them   in   a   single   framework.   This   challenge   applies  both  to  models  of  speech  evolution  (e.g.  Liljencrants  and  Lindblom,  1972;  Berrah   et   al.,   1996;   Browman   and   Goldstein,   2000;   de   Boer,   2000;   Oudeyer,   2005;   Pierrehumbert,   2006;   Wedel,   2011)   and   models   of   speech   acquisition   (e.g.   Guenther,  

                                                                                                                1  Oudeyer  P-­‐Y.  (2015)  Open  challenges  in  understanding  development  and  evolution  of  speech   forms:  The  roles  of  embodied  self-­‐organization,  motivation  and  active  exploration,     Journal  of  Phonetics,  http://dx.doi.org/10.1016/j.wocn.2015.09.001.     2  Email:  pierre-­‐yves.oudeyer@inria.fr  ;  Web  :  http://www

This content is AI-processed based on ArXiv data.