Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Origin of language: (at least) 4 ways of thinking Punctuationist Naîve Evolutionist Pre-adaptationist; Exaptationist Exaptationist AND Cultural evolutionist Language Sub-faculties to be considered Language Faculties of vocal learning, syntax, and semantics (at least) merged together to enable language Okanoya, 2006 Sub-faculties and exaptations Vocal Learning Syntax From intentional respiratory control From sexual displays that are sequentially organized, such as dance and song Semantics From hard-wired, situation specific communication signals (i.e, alarm calls) To meaningful action coordination Species that show vocal learning 81 species of whales ~5000 species out of ~9000 species of birds (songbirds, parrots, and humming birds) Only one species of primates out of ~220 Okanoya, 2010 Species that learn vocalizations Some whales Some birds One primate (!) Common anatomical feature: Direct pathway from motor cortex to medullar respiratory/vocal nuclei Okanoya, 2006 Reasons for having the direct path Birds and whales: accurate respiratory control when flying or submerging Humans: ???? Baby cry hypothesis (Okanoya et al 2002) No other primates cry as much as human babies No risk of predation by crying Manipulation of caregivers via crying was effective Okanoya, 2006 Cry development in human infants 1d 18d 1M Baby cries begin as repetitions of simple elements. Baby cries become complex both acoustically and syntactically after 1 month. This maybe related with myelinization of the direct pathway. 4M Okanoya, 2006 Mutual segmentation of baby cry and mother semantics Okanoya, 2006 Example 1: Evolution of Language via Gesture From gestures to language Not a new idea, really (Condillac, 1746) “Sign languages” (W. Stokoe) “From Hand to Mouth” (M. Corballis) Etc. Example 2: Evolution of Language via Syntax (and vocalization) Birdsong syntax Song is a sexual behavior: sung in a mating context Okanoya, 2006 Song syntax in two strains of birds Munia (left) and its domesticated form, Bengalese Bengalese Finch sings complex songs Munia sings simple songs Okanoya, 2006 Handicap Principle (Amotz Zahavi) It takes some effort to maintain excessive ornaments. Surviving with such excessive ornaments implies that the individual must have higher survival value and such males are selected by females. Excessive ornaments can then be indicators of the fitness. Bengalese finch mating song Bengalese finch song Only males sing to seduce females. Learning takes place when birds are at juvenile stage. Learning proceed in two phases: sensory phase and sensory-motor phase. Cortex- basal ganglia pathway for song learning. It can be expressed by a Finite-state syntax Finite-state song syntax in Bengalese finches In this example, ab, cde, and fg are organized into “chunks”. These three chunks are sung in various orders. Song learning in a multi-tutor environment 1st generation Male A 1st generation Male B 2nd generation 1st generation Male C Takahasi 2010 Ethology Brain of the Bengalese finch LMAN NIf HVc Lateral NIf AreaX RA Medial OKANOYA Lab. Bilateral NIf Lesion simplified Syntactical Song Before After Hosino & Okanoya, 2000 Birdsong studies: Summary Juvenile Bengalese finches segment parts of songs from multiple tutors and recombine these to invent their original songs. To sing normal songs, avian forebrain areas equivalent to human inferior frontal cortex and basal ganglia are necessary. Cost of complex song syntax Predation Time to find food Neural substrates (more brain space) Higher testosterone level (harms immune system) Need to maintain fine muscle tune First two points could be masked by domestication Functional un-grounding: Spotted Munia, a sympatric species of white-backed munias in Taiwan Field work: Taiwan study site T: Taipei H:Huben (Suburban) (Woods) Mountain running South to North M:Mataian (Rice field) More sympatric species, more linear (less complex) songs Song linearity H < M, T Sympatric ratio H < M, T Effect of sympatric species Higher sympatric ratio is correlated with lower song complexity. Interpretation: songs cannot be complex to secure species identification. Domestication is a special case in which no sympatric species exist. Degeneration of species identification led to evolution of syntactical complexity? 1. 2. Domestication degenerated constraints By domestication, several constraints (species recognition, social learning) against developing complex, elaborated songs are masked. Domestication promoted Sexual Selection Female choice for complex song syntax is strongly promoted by domestication. Degeneration by self-domestication as a driving force for Human Language evolution? wild wild domestic domestic Some points Possible biological exaptations for language Vocal plasticity > vocal learning String segmentation > syntax Context segmentation > semantics A mechanism for signal complexity Functional un-grounding increases complexity (e.g., “domestication”) Evolution of Language: Some conclusions Most efforts (overemphasis?) have been dedicated to investigate grammar/syntax and phonology/vocalization Crucial ingredients that are under-represented … (or ignored). The inherent role of: 1. 2. 3. Semantics (i.e., meaning emergence, not just lexical, sentential, propositional) Co-speech motor action (e.g., gesture) Coordination of top-down mechanisms (e.g., cultural evolution)