Lecture 3 - Uniqueness of Language Flashcards
Uniqueness of language - according to Lenneberg
“species specific”
“species-uniform”
don’t find cultures where kids don’t acquire language till they’re ten
kids learn to speak without a whole lot of encouragement
not learnable by other species
all languages have “universals”:
language universals
not totally clear
shared traits of sound/grammar/meaning
how many languages have to have them for it to be a “universal”?
highly predominate word orders: subject - verb-object & subject-object-verb
Psycholinguistics
psychologists/cognitive scientists that study language
Noam Chomsky
somewhat fomented modern study of psycholinguistics
how could we learn all this stuff if it wasn’t built in?
acquisition of grammar, syntax
child-directed speech
why is it that we speak differently to infants and kids?
does it help them learn language more easily?
some data says they learn better with child-directed speech or kids reinforce you for doing it because they pay more attention and if they pay more attention they learn better
cultural learning (everyone does it, so you do it)
statistical language learning
maybe rather than having certain sound patterns built in maybe you can detect statistical recurrences of stuff in the speech stream you are getting
you can get all this info from the data you’re getting so you don’t have to have it built into your language module because
cross-cultural studies
most of these studies are done in english
international adoptions
longitudinal study
taking one to three kids and follow them to see when they develop certain properties (when they start producing “wh” questions)
cross-sectional study
less time
take one set of one year olds, one set of 2 year olds, and one set of three year olds and look at the abilities of each age group
observational study
track kid’s natural productions
experimental study
try to get kid to produce certain words at certain age time points
experimenter controlling the situation
standardized study
standardized assessment measures of typical infant development, typical early responsiveness to language, vocabulary
“relative to the norming sample here is the size of this kid’s vocab”
CHILDES
(system for online data sharing)
– So researchers can analyze transcripts (now,
video data too) rather than reinvent the wheel
Big Data
idea is: maybe we can have computers to studies for us on 1000 kids
LENA system
records every vocalization kids and people around them make
software categorizes things
get a bunch of recordings and have TONS of data
