exam 3 Flashcards
what are the properties of language?
- creative: language has infinite potential meaning an infinite number of sentences that can be created
- rule governed: sentence structure, word combination, syntax
what are the structure of sentences ?
- surface structure: organization of words
- deep structure: the underlying meaning of a sentence
Sentences can have different surface structure but the same deep structure and vice versa
what are the classes of words?
- content words: describe objects, acts, properties
2. function words: show relationship among other words (the, a)
what is a morpheme?
smallest unit of meaning in a language
ex: dogs the s means its plural
sound structure: what is a phoneme?
smallest unit of sound that can change the meaning of a word
ex: mad vs bad
what is a phone?
the sound that is actually produced
/pin/ vs /nip/ different sounds of /p/
Functions of language
speech acts: the goal of an utterance that relies on what the receiver of the speech knows
includes given information: speaker assumed lister knows info and new information: speaker assumes lister doesn’t know information
language comprehension
syntactic analysis: cue word strategy
relative pronouns signal a new clause
In a study, people had a difficult time paraphrasing and understanding the sentences they heard when the sentences didn’t contain relative pronouns
*relative pronouns lead to a better understanding of information
*in English relative pronouns aren’t always necessary for sentences to be grammatically correct but they aid in understanding
language comprehension SVO strategy
SVO strategy: subject, verb, object
In English, we assume that in a sentence the first noun is the subject the first verb is the main verb and the second noun is the object
When a sentence doesn’t follow this order it is more difficult to parse
What kinds of knowledge is necessary for comprehension?
- semantic knowledge (vocabulary)
- syntactic knowledge (grammar)
- world knowledge (understanding words in the context of the world)
- conversation postulates (turn-taking, telling truth)
after we understand what do we remember?
study: took complex sentences and broke them down into 1s,2s, and 3s. participants when asked test phrases some they had heard some they hadn’t The 4 was the full sentence which everyone thinks they have heard but they hadn’t.
The study showed that we don’t “tape record” what we hear but we gather a general understanding and summary.
stages of language production
stage 1 - select a meaning to be expressed
stage 2 - select a syntactic structure for clauses in a sentence
stage 3 - insert stems of content words in syntactic structure, forms the bare bones of a sentence
stage 4 - specify morphological forms of the content words and select function words, adding things like: -ing, -s,-ed, the, a
stage 5 - specify phonemic representation for the clause, each word is assigned phonemes, at this point haven’t said anything but you have the sounds you will need
stage 6 - select the motor commands needed
stage 7 - utter the clause, then proceed to the next clause in the sentence
evidence that stage 2 is before stage 5 language production
study - analyzed when people made misordered phonemes
found that people in their sample 98/100 times people make phonemic errors within one clause
This showed evidence that syntactic structure was in place first
evidence that stage 3 is before stage 4 language production
study- participants were given a verb and told to create the noun associated with it
reaction time was longer for words that in order to change it to a noun you had to change the stem
This study showed evidence that stem words (stage 3) are added before adding morphological elements (stage 4)
selecting motor commands in language production: 2 hypotheses
motor command hypothesis: each phoneme is concerted into a particular set of motor commands and it is always the same regardless of context *no evidence for this
coarticulation effect: differences in motor commands of a phoneme depend on its context, we anticipate the vowel sounds that follow (ex: tea and too) *lots of evidence to support this
speech perception :listening 2 hypotheses
- invariance hypothesis: states there is an invariant set of acoustic cues for each phoneme, says the pattern is always the same for phonemes *a spectrogram disproves this hypothesis, also no evidence
- parallel transmission: information about phonemes is transmitted simultaneously to the ear, cues for phonemes intersect and cannot be isolated within a word
analysis by synthesis theory (words): listening
3 processes that occur
1. bottom-up processing: collecting phonemes we think we heard and performing a feature analysis, taking in what we heard from environment
2. top-down processing: person’s expectation of what word will be said: “the dog with big teeth was chasing me and…..” you would expect bit to be the next word
if you cannot understand what was said by using those two processes you resort to:
3. third track: synthesize what other sounds might be correct, take into account accents or how you might make the sound
listening: understanding sentences
study by picket and pollard
participants were asked to perceive a string of words, listened to recorded conversations and then removed and isolated a word from the sentence, the isolated word was played back
less than 50% of participants could understand the word when it was isolated
*shows the importance and reliance on context to make sense of what we hear
Lexical analysis How do we determine the meaning of an ambiguous word?
Swinney study
Participants were given a sentence with an ambiguous word that would lead them to conclude one meaning of the word. The question was if we would look up both definitions or not.
The findings were that immediately after all possible meanings of the words were activated (bottom-up processing)
After 3 seconds and the letter string being showed, only 1 meaning is activated (top-down processing and context kick in)
2 models of reading
- recording hypothesis: words are recoded from visual to speech codes, either you say it out loud or to yourself
visual->auditory-> meaning (used for harder texts or unfamiliar words), strategy early readers use - direct access hypothesis: reader goes directly from the visual to meaning (used for familiar or easy to understand texts)
visual-> meaning
evidence for recoding hypothesis
Rubinstein study
given a letter string and decide if it is a word or not
researchers were interested in nonwords and presented participants with nonwords that sounded like real words
Found that reaction time was slower for identifying nonwords that sounded like real words
evidence fir direct access hypothesis
profoundly deaf individuals can’t code things auditorially
role of context at the word level
study by Jenkins and Strange
gave participants 2 passages containing the same four words then assessed them on what words they think they heard
researchers were trying to show how words take on different meanings in context
results were that participants reported seeing words that were consistent with the story they read even when they had not seen the words
role of context at the passage level
once you have context of a passage you can understand what it means (ex being read the story about flying a kite in class)
what explains context effect?
schemas
- can be observed from the environment or from memory
- can be hierarchically stored in other schemas
- can vary in abstraction
- are active, we see the world through schemas (ex in class: burglar vs home buyer reading about a house)
psycholinguistics
an interdisciplinary field that examines how people use language to communicate ideas
Chompsky’s theory
- research failed to support his prediction that people would take longer to process sentences that required failed to support
- his theory argues that all languages share universal grammar
impact of affirmative and negative messages on cognitive process
our cognitive process handles positive information better (faster and less errors) than negative information
incremental interpretation
- language comprehension occurs incrementally
- we don’t wait until an entire sentence is spoken or read before making judgements about what it means
online language processing measures
self-paced reading tasks - shows the amount of effort it take the participant to read various words
good enough approach to language comprehension
we frequently only process part of a sentence, people do not work hard to creat the most accurate understanding of every sentence they read/hear
neurolinguistics
discipline that examines the underlyingneurological structures and systems that support language
event related potential technique
record brief fluctuations in the brain’s electrical activity, doesnt provide info on where activity is
broca’s aphasia
effects left hemisphere, diffculty with expressive language
wernicke’s aphasia
- difficulty with language comprehension and production
broca’s area
not only related to grammar or language but more of a general attention center that engages when a representational conflict occurs
lateralization
each hemisphere of the brain has different functions
right hemisphere
responds different;y to connected language than to disconnected language
assists left hemisphere in language comprehension often more abstract language
mirror system
mirror neurons are activated when you watch someone perform an action and you perform the action yourself, can comprehend messages from the actions of other people
inter-speaker variability
speakers of the same language produce the same sound differently
phonemic restoration
can fill in a phoneme using contextual information
the McGurk effect
influence of visual information on speech perception when individuals mustinegrate bothvisual and auditory information, area of the brain that does this is the superior temporal sulcus
phonetic module
script
simple well structure sequence of events
boundary extension
refers to our tendency to remember having viewed a greater portion of a scene than was actually shown
how do we comprehend photographs?
perceive them by activating a perceptual schema
constructive model of memory
people integrate information from individual sentences in order to construct larger ideas
pragmatic view of memory
proposes that people pay attention to the aspect of a message that is most relevant to their current goals
memory integration
taking in new information in a schema consistent fashion
in a study people read a story and were asked to recall it 15 min later and then three days later. As more time passed they relied more and more heavily on their prior knowledge
*showed that scheme can miss lead us to make errors
implicit association test
participants had a difficult time associating math and science to women and a difficult time associating poetry and dance to males
countries with the highest scores on IAT were likely to have the highest “male advantage” scores on math and science tests
slips of the tongue 3 types
errors in which sounds or entire words are rearranged between two or more different words
- sound errors
- morpheme errors
- word errors
