language
= verbal part of communication –> communication also consists of non-verbal parts (body language)
pragmatics
= difference between literal and intended meaning of a sentence
pragmatics
- figurative language
= language not intended to be taken literally
- metaphors
- why? combine (1) content + (2) relationship type
1. content/message/info you want to convey
2. relationship types: - dominance (boss-employee): commanding and following
- communality (family, friends): sharing
- reciprocity (buyer-seller): wanting something in return
figurative language
- standard pragmatic model
UNSUPPORTED
= processing figurative statements (metaphors)
- 3 states
1. assess literal meaning
2. decide wether it makes sense in current context
3. if inadequate: search or additional meaning
–> unsupported
x metaphorical and non-metaphorical meaning are understood at the same speed
figurative language
- predication model
= metaphor understanding consists of 2 components
1. latent-semantic-analysis component: assessing meaning of words based on relation with other words
- non-directional process of finding common meanings
2. construction-integration: use info from 1 for construction of an interpretation
- directional process from argument being projected to predicate
√changing the argument, changes the features selected
x does not refer to correlation metaphors
pragmatics
- common ground
= shared knowledge and beliefs
- can be effortful and attention demanding
- achieved rapidly by forming simple associations in long-term memory
- cultural factors
- egocentric heuristic: listeners interpret what they hear based on their own knowledge rather than common ground
- -> can cause listener to misunderstand what the speaker is trying to communicate
pragmatics
- differences in WM
- individual differences in working memory capacity influence language comprehension
hypotheses: high WM…
1. = high attention & attentional control = reduced mind-wandering = better comprehension (supported)
2. = form more effective situation models when reading (supported)
3. = better at discriminating between relevant & irrelevant info
–> support: low WM = greater seductive details effect (= tendency to reduce comprehension of a text with irrelevant info)
x individual differences also correlate strongly with IQ & vocabulary
x processing differs a lot across individuals
pragmatics
- cultural differences
LOW CONTEXT CULTURES (US): more direct/explicit speech
- little involvement –> less shared information –> stronger need to communicate explicitly
HIGH CONTEXT CULTURES (ASIA): more indirect/implicit speech
- deeply involved with each other –> much shared information –> no need for explicitness
direct vs. indirect speech
DIRECT
- relies on mutual knowledge
- if not given, risks relationship type
INDIRECT
- relies on individual knowledge
- you can still pretend to mean it directly –> avoid risk
- good for testing/exploring relationships
discourse
= language that is a minimum of several sentences in length (written text and connected speech)
discourse processing
- inferences
- logical inferences = depend only on meanings of words
- bridging inferences = backward inferences = establish coherence between current part of text and preceding text
- stages of forming bridging inferences
(1) bonding: automatic activation of words from preceding sentences (simple)
(2) resolution: involves ensuring that the overall interpretation is consistent with context
- anaphor resolution: pronoun/noun has to be linked to previous one (e.g. Tom sold Dave his CD because he hated it)
- causal inferences: work out causal relationships between sentences - elaborative inferences = add details to text by using world knowledge to expand on the information it contains
- -> forward/predictive inferences: anticipating future (type of elaborative)
inferences
- constructivist approach
= readers construct mental models of situation + events described
- numerous (automatic) elaborative inferences are drawn during reading even when not required
- correct inferences: enjoying when reading in a leisurely pace
inferences
- minimalist approach
= inferences are either automatic or strategic (= goal directed)
- some automatic inferences: establish local coherence (= make sense of a few sentences)
- some automatic inferences: rely only on information readily available (explicitly stated in the text)
- strategic inferences: formed in pursuit of readers goals, sometimes produce local coherence
- most elaborative inferences are made at recall rather than during reading
- correct inferences: quickly reading text, when text lacks global coherence, when reader has little background knowledge
discourse comprehension
- schema theory
= schemas contain much of information needed to understand what we read; allow us to form expectations
- comprehension/memory for discourse depends on top-down processing triggered by schemas
- 3 types of error
1. rationalisation: error in story recall that conform to one’s cultural expectations
2. levelling: omitting unfamiliar details from recall
3. sharpening: selecting certain details for embellishment
x only re-description of data: no evidence of actual activation of schemas
x not sure when/how given schema is activated
x exaggerate how error prone we are
discourse comprehension
- construction-integration model
- turn sentences into propositions representing its meaning
- propositions are stored briefly together with inferences
- many unimportant storages - integration process: spreading activation process selects propositions for text representation
- highly interconnected proposition attract most activation –> greatest probability of inclusion - 3 levels of representations are constructed
- surface representation (= text itself)
- propositional representation/ text base (= propositions formed from text)
- situation representation (= mental model describing referred to in text) that as only one mostly depends on integration process
- Kintsch
construction-integration model
- limitations
x unlikely that only involves bottom-up processing (finding that goals can influence readers attention very early)
x disregards emotions, goals & imagery
x ignores text genre level, communication level
x doesn’t say which inferences play a role
x not specific about how situation models are constructed
discourse comprehension
- event- indexing model
= comprehension process when someone reads a narrative text
- monitor five aspects, see whether their situation-model needs updating
1. protagonist: central character in present event compared to previous one
2. temporality: relationship between times at which present and previous event occurred
3. causality: causal relationship of current event to previous one
4. spatiality: relationship between spatial settings
5. intentionality: relationship between characters goals and present event - -> discontinuity in any of those aspects leads to more processing effort
- -> monitored independently = processing effort greater when more than one is outdated
- update outdated info:
- -> here-and-now view (EIM): most current info is more available than outdated one
- -> resonance view: new info in a text echoes with all text-related info stored in memory (= outdated info can influence comprehension process)
discourse comprehension
- event segmentation theory
= continuous actions are segmented into events
- 2 main forms of updating situation model:
1. incremental updating: “brick by brick” (EIM)
2. global updating: whole model is replaced; “from scratch” - -> both supported
thought and language
- whorfian hypothesis
= language affects thinking
= linguistic relativity: ways in which speakers think are influenced by language they speak
- strong position: differences in language determine/cause differences in thought (= unable to think about a topic if we don’t have the relevant words available to us)
- weak position: language influences certain aspects of cognition (= language causes preferences but these can be easily eliminated if not useful)
- effects of language very task-specific (e.g. temporal estimation)
whorfian hypothesis
- evidence (LECTURE)
support for strong:
- numerical concepts: without words for numbers, no numerical concepts; different spatial mapping of numbers
support for weak:
- metaphors
- perception/categorisation
–> colour: different categorisation, influences perception of colours
–> spatial perception: absolute (cardinal directions), relative descriptions (left/ right)
–> articles (der, die, das)
thought and language
- universalism
= thought as basis for language
universalism
- evidence (LECTURE)
against linguistic relativity:
- thoughts are not presented as words in our memory (pictures, sensations…)
- not having the definition/word, still experiencing it
- object segregation in infants
bilingualism
- compound bilingualism: e.g. parents speaking different languages
- coordinate bilingualism: native language vs. learned language in class
- language switching activates DLPFC –> managing attention to language
advantages of bilingualism
- enhancement of control mechanisms, cognitive control
- mentally more flexible (= able to adapt to ongoing changes, process info efficiently)
- better executive control
- -> better inhibition + conflict monitoring
- -> better selection of goal
- -> better sustained attention
- buffer protecting against dementia
