Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... ·...

18
Language Definition: shared symbolic system for communication Human speech vs Sign language vs Animal communication Linguistic universals Semanticity: meaningful Arbitrariness: random connection between sound and meaning Flexibility of symbols: can change Naming: everything can be named Displacement: future or past tense Productivity or generativity: can create new

Transcript of Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... ·...

Page 1: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Language

Definition: shared symbolic system for communication

Human speech vs Sign language vs Animal

communication

Linguistic universals

Semanticity: meaningful

Arbitrariness: random connection between sound and

meaning

Flexibility of symbols: can change

Naming: everything can be named

Displacement: future or past tense

Productivity or generativity: can create new

Page 2: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Language research

Psycholinguistics

Chomsky’s “Syntactic structures”

General structure and rules of language

Acquisition

Comprehension

Spoken or written

Words to sentences

Production

Physical and mental

Representation of info

Neuroscience

Broca’s and Wernicke’s areas

Page 3: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Structure of language

Structure RULES Grammar: complete set of rules to produce sentence

Phonology: sounds into words

Syntax: word order into sentences

Semantics: communicate meaning (word or sentence)

Word level Phonemes: smallest units of sound ( /b/ )

Morphemes: smallest unit with meaning

Lexicon: knowledge of word meaning, sounds, etc.

Sentence level Hierarchy: words - phrases - sentences

Pragmatics: social rules

Page 4: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Perceiving words

Phonemic restoration effect Warren (1970): cough replaces phoneme

Speech segmentation Pronunciation of words in sentence different than alone

• Coarticulation – sounds mush together

Pollack & Pickett (1964): only 50% acc for recognizing individual words Ss previously said within a conversation

Page 5: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Understanding words

Question: What factors influence ability to understand words?

Word frequency effect

Lexical decision method

Faster RT for high frequency vs low frequency words

Context effects

Identify words alone or in conversational speech

Faster RT if words fit expected schema for sentence

Lexical ambiguity

Lexical priming method

Words can have multiple meanings – need context of sentence to clarify

Semantic ambiguity

Garden path sentences

Page 6: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Lexical decision

Meyer & Schvaneveldt (1971)

Priming task: word activates lexicon

RT task for word pairs: Word or non-word?

Experiment 1: y/n both words?

Pairs: Associated words, unassociated,

word-nonword, nonword-word, nonwords

Simultaneous presentation of stimuli

Experiment 2: same/different judgment

Page 7: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Meyer & Schvaneveldt (1971) Exp 1: Y/N both words?

RT of correct responses Y-N task

400

600

800

1000

1200

Associated Unassoc W-non Non-W Non-Non

Associated vs. unassociated: 85ms difference

Serial decision model: 2-stage decision process

1st non-word: quick N decision

Word-nonword pair: more errors b/c fast Y to word

Page 8: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Lexical ambiguity

Page 9: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Lexical ambiguity Swinney (1979)

Examine access to meaning of ambiguous words

Method: Listen to sentence & lexical decision

Results: RT same to “ant” and “spy” when hear “bug” 200ms later: “ant” faster

Conclusion: Context takes time

“Rumor had it for years, the govmt

building had been plagued with problems.

The man was not surprised when he

found several spiders, roaches, and other

bugs in the corner of the room.”

Page 10: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Syntactic ambiguity “The spy saw the man with the binoculars”

How do we separate a sentence into smaller phrases

Mechanism: Parser

Syntax-first approach

Use syntax to infer meaning

Late closure: “B/c he always jogs a mile seems like a short distance to him.”

Method: garden-path sentence

“The horse raced past the barn fell.”

“The horse (that) raced past the barn fell.”

Temporal ambiguity

Interactionist approach

Simultaneous use of semantics and syntax

Page 11: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Tanenhaus et

al. (1995)

Eye-tracker

“Put the apple

on the towel in

the box.”

Interactionist

approach

support

Page 12: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Semantics

“Colorless green ideas sleep furiously”

Semantics can overpower syntax

Fillenbaum (1974): paraphrase sentences

Change expected meaning (“don’t print that or I won’t sue you”)

Change order of sentences (“John went into the store and got off the bus”)

Findings: Reordered sentences

Conclusions: Effect of top-down processing and schemas

Semantic grammar approach

Begin to analyze the sentence immediately

Process: role of words contributes to overall comprehension

Garden path sentences

“After the musician had played the piano was quickly taken off stage.”

Eye-tracker: eyes fixate longer when encounter error

Page 13: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Is language innate or learned?

Nativist (Chomsky)

Innate language acquisition device (LAD)

Localization of language in brain

Behaviorist (Skinner)

Reinforcement and imitation

Gradual mastery

Interactionist

Combine innate capacity and social environment

Page 14: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Syntax

Chomsky’s theory

Proposed innate language

skills

Phase structure grammar

• Allows for generativity

“Visiting relatives can be a

nuisance.”

Transformational rules:

• Surface: actual words

• Deep: meaning of sentence

• Transformational rules:

Conversion of deep to

surface structure

Page 15: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Deep vs surface structure

“Change the towels in the green bathroom”

“Dust the furniture”

“Draw the drapes when the sun comes in”

“Trim the fat before you put the steak in the icebox”

“And please dress the chicken”

“Lexical ambiguity”

Page 16: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Do these speakers think differently

b/c of how they talk?

English: “the elephant ate the peanuts”

Mandarin: when event occurred is optional separate word

Russian: need tense and gender on verb, and if ate all or portion of the peanuts

Turkish: must specify if event was witnessed or hearsay

US English: “making money”

English: front/back terms for time

Mandarin: also up/down metaphors for time

Page 17: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Sapir-Worf hypothesis

or Linguistic relativity hypothesis

Language shapes how we think and perceive

Strong version: controls thoughts

Weak version: influences thoughts

Another alternative: Changes what we pay attention to

Cross-cultural tests

Roberson et al (2000):

Color identification for British and Berinmo (New Guinea)

Result: British 8 color names; Berinmo 5

Categorical perception: harder to discriminate stimuli in same category

British Ss

Berinmo Ss

Page 18: Language - Wofford Collegewebs.wofford.edu/boppkl/courseFiles/Cognition/CogPPT10... · 2010-11-18 · Syntax-first approach Use syntax to infer meaning Late closure: “B/c he always

Boroditsky’s et al. (2002)

Spanish and German speakers tested in English

Examine objects with opposite grammatical gender assignments in 2 languages

Experiment 1: Proper names assigned to objects Apple = Patrick or Patricia

Results: Consistent object-name was remembered better

Experiment 2: Name 3 adjectives for objects Results: adjectives consistent with grammatical gender

• German “key”: hard, heavy, jagged, metal, useful

• Spanish “key”: golden, intricate, little, lovely, shiny

Experiment 3: Rated similarity of pictures of objects and people

Results: Similarity ratings consistent with grammatical gender