What is the two stage model of lexicalisation?
- Finding words: Retrieval of abstract form of a word – it’s lemma’- from the mental lexicon based on concepts that the speaker wants to convey
- Building words: Specification of the form of the word, the lexeme
What does the book say the two pieces of data are relating to lexical selection?
-Pausing and predictability
-Speech errors and lexical selection
What influences the ease at which words can be accessed from the mental lexicon? What does this mean for fluency?
Words that are used more frequently or are predictable under certain contexts are accessed from the mental lexicon easier = greater fluency/ less pauses in speech (opposite also true)
How does the number of pauses for content words compare to function words? What are two possible explanations for this?
There is more pauses before content words compared to function words
This could be because there is more options when choosing a content word (‘open class’)
Alternatively, function words could become available at a later stage of production, when the grammatical sentence frame is constructed explaining the delay
What is corpus linguistics?
The study of large collections of language data. These corpora provide information on word frequency, as well as on the patterns of phenomena such as pauses.
What is one famous test that measures transitional probability? What do you expect to find?
The guessing game: have a series of sentences with words missing and the participant has to guess what fills the blank
When this word is very predictable e.g. a saying, there is a high transitional probability, and less likely to be a long pause. Opposite true also.
what are the three ‘sources’ of speech errors discussed in lecture ?
- Grammatical similarity = swapping of words between ‘slots’ in the sentence frame. Usually occurs within the same grammatical category e.g. Seymour slices the salami with the knife -> seymour sliced the knife with the salami. Noun for a noun.
- Phonological similarity = Swap or confuse words because they sound similar e.g. Put it in the car park -> put it the par cark
- Semantic similarity = Words that have related meaning are often confused i.e. substitute antonyms, near-synonyms, blend, associated meanings via collocation
Freudian Slips
errors that are claimed to reveal repressed thoughts or feelings
e.g. I wish you were ‘her’ instead of ‘here’
According to the book what are the three broad types of speech errors?
Mis-selection (substitution, blend)
Mis-ordering (anticipation, preservation, exchange)
Other (omission, addition)
Why can speech errors be difficult to interpret?
Sometimes it can be hard to tell when analyzing language data what is actually an error. We can often be inferring the intention of the speaker, we could also make errors in hearing what they have said. Often we know if there is a self-correction, pause/ stumble or if the word doesn’t make sense in the context.
What is a substitution error? How are they proposed to occur?
A type of mis-selection error
One word replaces another
Substitutions more likely to involve words that are antonyms
Lemma level relationships that are associative- some associations may involve colocations of words that have opposite meanings
The word with the opposite meaning to the target lemma can then be accidently activated via associative links and inserted in the sentence
It may be that the antonym is a more frequently used word so takes preference or that the target is temporarily unavailable but the exact reason for its selection as the lexeme is unknown
What is a blend error? How are they proposed to occur?
A type of mis-selection error
Two words are merged
Blends frequently involve words that are synonyms near synonyms
Concept-level relationships involving pre-linguistic abstract ideas, both words are competing for selection (alternative plans)- this does not resolve and so they are combined
Out of substitution and blend errors which is more likely to involve antonyms ?
Substitution
Out of substitution and blend errors which is more likely to involve synonyms?
Blends
What are malapropisms? How are they said to occur?
They are a type of substitution error
The word produced is similar to the intended word in its sound shape, but not necessarily in its meaning
e.g. He is the very pinnacle of politeness → He is the very pineapple of politeness
This could be because there are links between the sounds contained in the target word to other words that contain the same sounds, there may therefore become activated and if strong enough replace the target word in utterance
What is the difference between serial models or lexicalisation and interactive models of lexicalisation?
Serial model (figure 3.3) = speaker has access to one word at a time following a discrete, unidirectional flow of information between levels.
Interactive model (figure 3.4) = Information between levels flow in more than one direction. From one level down to multiple units at the next level, but then also back up to higher level units.
What is the role of inhibition in lexicalisation?
The stronger the activation of the target word the more inhibition to other competition or linked words. This is sometimes referred to as the ‘rich get richer’.
What 3 types of errors fall under mis-ordering?
Anticipation = Word inserted too early in the sentence frame
Preservation = A word that has already been used remains active and available for re-insertion
Exchanges = Two words swap places
In general what is occuring in mis-ordering errors?
The speaker gets the required lexical items from their mental dictionary but something goes wrong in assigning each term to its correct position (i.e. grammar is the issue)
Sometimes self-corrected, sometimes not
Tends not to be a strong meaning relationship
What is the influence of grammatical category in mis-ordering errors? What does it suggest about language production?
Often words invovled in the error are from the same grammatical category I.e. noun for noun, verb for verb -> this suggests that when words are selected for production information about their grammatical category is available (meaning and grammatical information not entirely separate?)
Association norms - what are they? How are they studied?
Words that are evoked in the minds of native speakers when a target word is presented to them. These lists are created by asking a large number of participations to write down or say the first word that comes to them when they are presented with a target word.
What are common ‘categories/ types’ of association norms?
Some are opposites (antonyms) e.g. high + low, king + queen
Some are words you would expect to hear in the same sentence (collocational links) e.g. hammer +nail, butter + bread
In what direction do co-locational association norms tend to be strongest?
Collocational links tend to be stronger in the left-to-right direction I.e. bread produces a greater proportion of butter responses than vice versa probably because of the phrase “bread and butter”
What do errors suggest about how our brain is organised in terms of language?
Our brain can’t be like a filing cabinet that a single file at a time is pulled from. It must be more like a network whereby when one part of the network is activated other parts it is linked to also become activated.
Frequency, recency effects = It is easier for the mind to find frequently used/ recent words
-
Background Terms/ Revision52
-
Week 1- What is Psycholinguistics?14
-
Week 1/ Week 2 - Planning utterances37
-
Week 3 - Finding Words28
-
Week 3 - Building Words25
-
Week 4 - Monitoring and Repair26
-
Week 4 - Speech Perception22
-
Week 5 - Spoken Word Recognition38
-
Week 5 - Visual Word Recognition41
-
Week 6: Syntactic Sentence Processing24
-
Week 6: Interpreting Sentences15
