S2 L2 - Language Comprehension intro Flashcards
(22 cards)
What is language?
An exchange of information
What is the mental representation for comprehension?
Input (of speech) > Activate (existing mental representations of sound) > link to meaning > resulting in output (comprehension (understanding)
What are the challenges of comprehension?
Ambiguity in the speech stream -
Impossible to know where one word ends and another one start
(gaps in word boundaries do not necessarily correspond to word boundaries eg ice cream or i scream)
Ambiguity at the word level - Words that sound the same but have different meanings
Ambiguity at the phoneme level - Words that change the way they sound depending on their environment
What are some word level ambiguities?
Homonyms - Words that sound the same and are spelt the same
BANK - BANK
Homophones - Words that sound the same
MUSSEL MUSCLE
Homographs - Words that are spelt the same
BOW (TIE) BOW (AND ARROW) BOW (BENDING OVER)
Where is the place of articulation?
Air is pushed up the vocal tract and hits points in the mouth that form an obstruction resulting in different types of consonants
What is Categorical Perception of sounds?
Being able to work out where one sound ends and another sound starts
Able to distinguish between sounds based on VOICE ONSET TIME (VOT)
b and p are bilabial consonants articulated in the same “place” but they sound different because they have different VOT
VOT is defined by the point at which vocal chord vibrations start relative to the release of a closure
How does ambiguity in speech stream get filtered out?
Categorical perception of sounds that are on a continuum filters out some of the ambiguity
What is the invariance problem?
The inability to define the acoustic properties that facilitate categorisation of phonemes
not easy to define a range of sounds
What are the properties that can be used to define phonemes?
Categorical perception - use the voice onset time to help distinguish phonemes
Top - down processing - activation of existing lexical representations may help us understand words, dialects, foreign accents and manage incomplete bottom-up information
How can we work out what we’re hearing?
Top down processing of existing mental representations of words stored in our mental lexicon can be used
What affects how quickly you can retrieve a word in Lexical Access?
as you are being spoken to youll look up the words in the meaning
More frequently used words will be faster accessed than words not frequently used
eg
gall vs ball
the strength of the representation in our lexicon is stronger and more accessible
Are shorter words quicker to access or slower to access from the Mental Lexicon than longer words?
Quicker to access
Words that are neighbours are slower to be accessed by the Mental Lexicon. why?
Because the words are similar - competition to be selected is increased.
A word w lots of neighbours ^ competition to be activated.
Finding the word will take longer.
How can we test for Lexical Access?
Lexical Decision Task (LDT)
Participants press one key for “word” another key for “non -words”
Reaction times are measured as the time between stimulus presentation and button press
Slow reaction times are found for words w lots of neighbours compare to words w fewer neighbours
How does context relate to Lexical Access?
Sentence context helps us activate potential word candidates, but does not facilitate the selection of the word
What are Priming Paradigms?
give a prime by showing word on screen
showing this word spreads activation for representations of words
target word is shown (a word / non-word) relalted word
Doctor and nurse are semantically related.
Spreading activation allows nurse to become active when doctor is presented
What are the links between lexical items?
Priming Paradigms
What are Cross Modal Priming tasks?
primed w a word thru headphones (auditory prime)
and have a visual target word
related prime - target pair
If you present a word fragment what can happen?
Through spreading activation you can access words (not just word fragments)
What does word monitoring suggest?
That ps can use context to activate lexical candidates that fit the context of a sentence
What does Cross Modal Priming evidence suggest?
Ps do not use context to only activate items that fit with the context
What does the system access lexical items based on?
Acoustic input
Top down processing
Lexical characteristics(eg frequency, neighbourhood density, length of words)
Activation of related items (via spreading activation)
context?
The language system activates a huge amount of information to predict what might come up next and process that information rapidly
