Language Flashcards
1
Q
Why study language?
A
- Language is a defining feature of humans – there is not a single culture without a language. It is unique only to us and is heavily entangled in our culture and cognition. By understanding language and language disorders we can learn more about ourselves.
2
Q
What are the functions of human language?
A
- Communication / Expression / Interaction / Play / Control / Identity / Thinking
3
Q
What is the distinction between a language and a communication system?
A
Language = Symbolic / Discrete / Combinatorial / Productive
4
Q
What are the consequences of human language being symbolic?
A
- The link between FORM and MEANING is often arbitrary. This allows for greater abstraction and flexibility but requires social interaction in order to learn and understand.
5
Q
What are the consequences of human language being discrete?
A
- There is a finite number of individual elements as well as rules on how to combine them
6
Q
What are the consequences of human language being combinatorial?
A
- Language creates meaning by combining smaller elements to build larger structures (words, phrases, sentences) with properties that are distinct from its sub-parts (‘more than the sum of the parts’).
7
Q
What are the consequences of human language being productive?
A
- We can combine existing elements in novel ways with almost no limitation. E.g. “I’m gonna have to science the s***t out of this”. Especially when using recursion
8
Q
What are the basic rules of sentence structure?
A
- Word categories > phrases > sentences.
- Rule for a Noun phrase = (Determiner + Adjective) + Noun
- Rule for a Verb Phrase = Verb (+ Noun phrase)
- Rule for a Sentence = Noun Phrase + Verb Phrase
9
Q
What is the difference between Language Competence and Language Performance?
A
- It is what you know about a language vs. how we produce and comprehend the language. I.e. What we can do vs. what we actually do.
10
Q
How do we deal with ambiguity in language?
A
We use parsing. There are two types: Syntax-first Parsing and Interactionist Parsing. It helps us with Garden-Path sentences.
11
Q
Psycho-linguistics
A
Psycholinguistics or psychology of language is the study of the psychological and neurobiological factors that enable humans to acquire, use, and understand language. Initial forays into psycholinguistics were largely philosophical ventures, due mainly to a lack of cohesive data on how the human brain functioned
12
Q
Phonemes
A
A phoneme is a sound or a group of different sounds perceived to have the same function by speakers of the language or dialect in question. An example is the English phoneme /k/, which occurs in words such as cat, kit, scat, skit.
13
Q
Morpheme
A
A morpheme is a meaningful unit of language that cannot be further divided. Morphemes can be words and affixes-prefixes and suffixes. Examples of Morpheme: -ed = turns a verb into the past tense. un- = prefix that means not
14
Q
Linguistic Competence VS. Linguistic Performance
A
Linguistic competence is the system of linguistic knowledge possessed by native speakers of a language. It is distinguished from linguistic performance, which is the way a language system is used in communication
15
Q
Parsing
A
There are two different theories about how humans parse sentences. The syntax-first approach claims that syntax plays the main part whereas semantics has only a supporting role, whereas the interactionist approach states that both syntax and semantics work together to determine the meaning of a sentence.
16
Q
Garden Path Sentences
A
A garden-path sentence is a grammatically correct sentence that starts in such a way that a reader’s most likely interpretation will be incorrect; the reader is lured into a parse that turns out to be a dead end or yields a clearly unintended meaning.
17
Q
At what point of a speech do we start to comprehend it?
A
- Comprehension occurs incrementally and is often anticipatory. As shown through Garden-Path Sentences and various studies. We often comprehend word by word
18
Q
How do we measure people’s comprehension of speech?
A
- Self-Paced Reading = Simple & Cheap / Unnatural
- Eye-Tracking during Reading = Natural & High resolution / Complex & No distinction between types of processing problems
- Eye-Tracking of an Image = Interactive & Spoken Language / Complex & Can induce strategy
- Event-Related Potentials = Multi-Dimensional & Information about types of processing / Resource Intensive & Unnatural
19
Q
Do we anticipate language?
A
Yes and no. According to Pinker there are >10 different words that could be inserted at any point during a sentence, so it is very unlikely we’d be able to predict language. However our brains still try to do it automatically.
20
Q
N400
A
The N400 is part of the normal brain response to words and other meaningful (or potentially meaningful) stimuli, including visual and auditory words, sign language signs, pictures, faces, environmental sounds, and smells.
21
Q
Bottom Up VS. Top Down Processing
A
Bottom-up vs. Top-down Processing. There are two general processes involved in sensation and perception. … Bottom-up refers to the way it is built up from the smallest pieces of sensory information. Top-down processing, on the other hand, refers to perception that is driven by cognition.
22
Q
Word Superiority Effect (Reicher, 1969)
A
The word superiority effect (WSE) refers to the phenomenon that people have better recognition of letters presented within words as compared to isolated letters and to letters presented within nonword (orthographically illegal, unpronounceable letter array) strings
23
Q
Rayner et al. (1983)
A
We don’t wait until we know which phrase the word belongs to before starting to comprehend
24
Q
Yates et al. (1974)
A
Implicit Causality - We don’t wait until we know who or what the word belongs to before starting to comprehend
25
Kutas et al. (1980’s)
N400 in ERPs - We don’t wait until we know what the word if before starting to comprehend
26
Altmann et al. (1999)
Anticipatory Eye-Movements – “The boy will eat the cake.” – We anticipate the kind of word
27
Delong & Kutas (2005)
Anticipatory ERPs – “…The boy went outside to fly…A kite.” – We anticipate specific words
28
Nieuwland $ Van Berkum (2006)
“When peanuts fall in love”
29
Van Berkum (2007)
Speaker Identity and Context
30
Bransford & Johnson (1972)
Washing Clothes
31
Erickson & Mattson (1981)
80% of participants experienced the ‘Moses illusion’
32
Sanford et al (2011)
Mini-stories rather than questions
• Explicit detection task (‘OK’ vs ‘Nonsense’)
• 90 different stories
• 63% detection (= 37% failure to detect!)
33
Why do semantic illusions occur?
‘Goodness of fit’ of the imposter word affects how likely people are to be misled
Likelihood of illusion increases with:
the semantic feature overlap between impostor word and correct word
the number of words in the context that are associated with the correct word
34
Hannon & Daneman (2004)
“What should the authorities do?”
35
Hannon & Daneman, 2001
Context ∼ ability to process and store in working memory
Imposter ∼ ability to retrieve and process from long term memory
Imposter effects are stronger
36
Cooperative speaker hypothesis
Communication has evolved as a cooperative effort
So we assume by default that a speaker is intending to communicate a coherent message
And cooperate in trying to recover that meaning
Cooperative principle: why would anyone try to trick you?
This explanation can explain why people sometimes ignore the actual input
But doesn’t fit all the evidence
37
Shallow processing
Illusions arise from shallow processing of the input
– partial analysis of the input (e.g., not all meaning features of a word)
People process words less deeply when they are a good fit to the described situation
Or when they are related to the correct word
Superficial coherence leads to heuristic processing
– The partial match between the actual input and the ‘correct’ version means people don’t detect the anomaly
This explains why illusions occur even in explicit error detection tasks
And also fits with ERP evidence that people do not detect a semantic violation when they encounter an imposter word that is consistent with the overall (global) context
38
Sanford et al. 2011
Semantic illusion sentences (whether detected or undetected) do not induce an N400
– As if people do not detect that the word is semantically anomalous
*Shallow Processing*
This account is also consistent with evidence that individual differences in reading skill affect susceptibility to semantic illusions
39
Individual differences Hannon & Daneman
Same pattern of effects but skilled readers show higher detection rates overall
*Shallow Processing*
40
Negated sentences
“No head injury is too trivial to be ignored”
41
Redundancy errors (Frazier & Clifton, 2011)
“Many people often thought that she uses whipped cream”
People interpret these sentences as if they only contain ‘many’ or ‘often’ instead of both
42
Passive sentences Ferreira & Stacey (2000)
Spoken sentences:
1. The man bit the dog
2. The man was bitten by the dog
3. The dog bit the man
4. The dog was bitten by the man
Question: Who is the do-er? Only 75% correct for sentence 4
• People may use heuristics like word order rather than computing a full syntactic structure
• First entity = do-er; second entity = done-to
43
Lingering misinterpretations
Christianson et al. (2001)
a. “While Anna dressed the baby played in the crib” (gardenpath)
b. “While Anna dressed, the baby played in the crib”
Question
1. Did the baby play in the crib? (100% correct in both cases)
2. Did Anna dress the baby? (50% wrong after (a))
We don’t fully correct our errors: they linger
44
Who is ‘bilingual’?
Bilingualism is the use of two (or more) languages in one's everyday life
(Grosjean, 1982)
45
A monolingual view of bilingualism
(Grosjean, 1989)
A bilingual is two monolinguals in one
–Someone who is equally and fully fluent in both languages
–Has two separate and isolable language systems
Consequences of taking this monolingual view:
Other types of bilingual are ‘special’
–(e.g., unbalanced bilinguals)
Bilinguals’ language abilities are judged by monolingual standards
Bilinguals feel inadequate or insufficient about their language abilities
46
A bilingual view of bilingualism
(Grosjean, 1989)
‘Sum is more than the parts’
–Coexistence and constant interaction of languages produces a new linguistic entity
–Bilingual Monolingual + Monolingual
–Bilingual ≈ Monolingual × Monolingual
–Rarely equally and completely fluent
47
–Monolingual speech mode
Situations where only one language is being used
May show interference effects from other language
48
–Bilingual (mixed) speech mode
Situations where both languages are being used
Code switching – swapping between languages
Word borrowing
49
L1
First language learnt
50
L2
Second language learnt
51
Simultaneous bilingualism
L1 & L2 from early childhood (before c. 8-12 years)
52
(Early) sequential bilingualism
L1 first, then L2, from early childhood
53
Late bilingualism
L1 first, L2 later in life (after c. 8-12 years)
54
Factors in bilingualism research
Age of acquisition
Modality of acquisition
Frequency of exposure / immersion
Frequency of use
Socioecononomic statusSocial context
Linguistic distance
Language testing modality
Current age
‘Talent’
55
Language acquisition differences in bilingual infants
Quantitative difference from monolinguals:
–reduced exposure for each language
Qualitative differences from monolinguals:
–Must learn to detect the different languages
–Must learn to separate and parse the information relevant to two different language systems
56
Bosch & Sebastian-Galles, 1997
Infant language discrimination
(4-5 months)
Monolingual (Spanish) infants:
faster to turn towards familiar language (Spanish)
Bilingual (Spanish/Catalan) infants:
faster to turn towards unfamiliar language (English)
Not yet clear how to interpret this
57
Sebastián-Gallés, AlbaredaCastellot, Weikum and Werker (2012)
Infant language discrimination
(8 months)
Habituation: videos in language A; Test: either A or B
Discrimination based on articulatory gestures (mouth movements)
Bilingual Spanish-Catalan 8-month-olds can discriminate two unfamiliar languages (English/French) when watching silent videos – but monolingual infants can’t
58
Monolinguals and bilinguals don’t show major time difference in establishing the ____________? Despite the fact that bilinguals learn two languages
Monolinguals and bilinguals don’t show major time difference in establishing the phoneme repertoire
59
From 6-9 months onwards, monolingual/bilingual infants are sensitive to what is or isn’t ______ in one’s language
From 6-9 months onwards, monolingual infants are sensitive to what is or isn’t allowed in one’s language
60
Monolingual children seem to use a __________ to learning new words in new contexts
Monolingual children seem to use a mutual exclusivity
61
Word learning: linking words to objects
| Houston-Price et al., 2010
Monolinguals look more at the novel object when they hear the novel word
Bilinguals don’t look more at the novel object when they hear the novel word - possibly because one concept has two words for them?
62
Bilinguals know _____ words per language and have ____ verbal fluency
Bilinguals know fewer words per language and have lower verbal fluency
63
Joint Activation
Fluent bilinguals show some activation of both languages and some interaction between them at all times
They must continually select between activated alternatives (e.g., “car” vs “voiture”)
Faster when L1 and L2 words are similar
64
Partial activation of an irrelevant lexicon during comprehension
Spivey & Marian (1999)
Russian-English bilinguals carry out task instructed in Russian
When they hear ‘Poloji marku…’ (‘pick up…’), they look at the marker
marker shares phonetic features with Russian word marku
– Suggests they access English lexicon even in Russianspeaking context
65
Benefits and Drawback of Bilingualism
cognitive benefits that last throughout life
- brain plasticity
- Better grammatical detection (bialystok, 1988)
- Less intereference on Simon effect tasks
- Later onset of Alzheimers (increased cognitive reserve)
cognitive costs:
- Size of Vocabulary
- Access Speed
- More frequent tip-of-the-tongue states
However, still a debate in research
66
Aphasia
Aphasia is an impairment of language, affecting the production or comprehension of speech and the ability to read or write. Aphasia is always due to injury to the brain-most commonly from a stroke, particularly in older individuals
Broca's Aphasia
- Simplified speech – key content words included, but word forms & grammatical constructions particularly simplified
Wernicke's Aphasia
- Paraphasia: wrong (combination of) words and morphemes
- Neologisms: Making up new words
- Impaired comprehension & repetition
67
Wernicke-Geschwind model
The Wernicke-Geschwind model concerns itself largely with the connections between specific areas of the brain. It holds that certain areas communicate with others in order to extract information about the meanings of words, before that information is passed to the speech production and motor areas of the brain.
According to the model, information from the written word first arrives from the eyes into the occipital lobe in the primary visual cortex. From here, the information is passed to the angular gyrus, and then onto Wernicke’s area. The arcuate fasciculus is a white matter tract that then sends information from here into Broca’s area. Finally, information from Broca’s area is passed to the motor cortex, which can send an outward motor command.
The Wernicke-Geschwind model is largely outdated due to inconsistencies in human case studies where an injury doesn't cause the expected deficiency based on the model.
- Language is processed not just in the left hemisphere, but the right too
- Other cortical regions are also involved in language function
- People can’t even agree where Broca’s and Wernicke’s areas are! (Tremblay & Dick, 2016)
68
Caramazza & Zurif (1976)
Caramazza and Zurif showed that Broca's aphasics failed to comprehend sentences that required syntactic analysis. (sentence-picture mapping)
Broca's area is the seat of syntax.
However, subsequent work (e.g., Linebarger, Schwartz, & Saffran, 1983) showed that Broca's aphasics had not entirely lost their syntax -- they could still make grammaticality judgments pretty darn well.
On the other hand = Fadiga et al. (2009) Broca's area is the seat of syntax via its more general role [?!] in hierarchical processing of any kind.
69
Voxel-based Lesion-Symptom Mapping
Past approach: lesion overlap
- identify groups of people with common area of injury, and compare their behaviour with control group
- But this loses information when multiple subregions each contribute to behaviour, and ignores regions outside area of interest
Voxel-based Lesion-Symptom Mapping (VLSM) avoids losing detailed spatial or linguistic information
Spatial information: continuous lesion information
– Resolution: voxel =1 mm by 1mm MRI scan of the brain
Linguistic information: continuous behaviour information
– Western Aphasia Battery subtests: Fluency, Auditory Comprehension
Per voxel, divide patients into groups that have or do not have a lesion affecting that voxel, and test whether the groups differ in behavioural scores (t-score).
Found that:
Broca’s area not especially important for fluency after insula lesions factored out
Wernicke’s area not important for auditory comprehension after middle temporal gyrus factored out
70
Left frontal lobe damage is characteristically associated with _____ deficits
Language fluency deficits
71
Left temporal lobe damage is characteristically associated with _______ deficits
Language comprehension deficits
72
Irregular forms of words are ____ in _____ but regular forms are generated by ______
Irregular forms are stored in memory but regular forms are generated by rules (+ed)
73
Pinker & Ullman (2002)
Words are stored as part of declarative memory
–Temporo-parietal cortex
Rules are stored as part of procedural memory
–Basal ganglia and frontal cortex (e.g., Broca’s area)
74
Anomia
Anomia: impairment in word findin
75
Agrammatism
Agrammatism: impairment in producing fluent grammatical sequences
76
Words vs Rules and degenerative diseases
Alzheimer’s disease
```
Alzheimer’s disease
–Especially atrophy in the temporal lobe
= declarative memory
–Loss of lexical/conceptual knowledge
–Grammatical processing relatively intact
```
More problems with irregular than regular
Unimpaired with suffixing (adding endings) to novel words
Over-regularisation errors
77
Words vs Rules and degenerative diseases
Parkinson’s disease
Parkinson’s disease
–Basal ganglia degeneration
= procedural memory
–Loss of motor skills (and grammatical skills)
–Relatively intact use of words and facts
More problems with regular than irregular
Impaired suffixing for novel words
Do not produce over-regularisation errors
78
Words vs Rules and degenerative diseases
Huntington’s disease
Huntington’s disease
–Degeneration of various basal ganglia structures
= procedural memory
–Disinhibition of frontal areas > unsuppressible movements
Produce extra suffixes: walkeded & dugged
–As if can’t suppress applying rule
But no analogous errors with irregular inflection (duggug)
‘Unsuppressed regular suffixation’
79
Sign aphasia Hickok et al. (1998)
Hand position can be phonemic (i.e., make minimal lexical distinction cog vs log)
Space can be used for inflectional morphology (e.g., past tense)
Error where one element that distinguishes one meaning from another is wrong (e.g., cog instead of log)
Same hemispheric asymmetries as in spoken aphasia
Similar symptoms at various levels of representations (phonological errors versus morpheme errors)
Also, LHD/RHD hemisphere difference is similar in sign aphasia and hearing aphasia
80
Phonemic paraphasia
wrong sounds, - ‘brogger’ replacing ‘brother’
81
Semantic paraphasia
wrong word, - ‘uncle’ replacing ‘brother’
82
Paragrammatic paraphasia
Neologism or wrong morphemes (‘brotherist’)
83
Broca’s area is involved in _______ ASL signs
Broca’s area is involved in producing ASL signs
84
Bilingual aphasia Fabbro (2001)
Sometimes asymmetric symptoms, either L1 or L2 is affected
Cannot predict recovery on basis of:
lesion type/site
context or frequency of use
type of aphasia
```
Most to Least Frequent:
Parallel recovery (both recovered together)
```
Differential recovery (better recovery in one)
Blended recovery (pathological lang. mixing)
Selective recovery (one lang only)
Successive recovery (one then other)
85
Pragmatics
the branch of linguistics dealing with language in use and the contexts in which it is used, including such matters as deixis, the taking of turns in conversation, text organization, presupposition, and implicature
86
Speech acts
A speech act in linguistics and the philosophy of language is something expressed by an individual that not only presents information, but perform an action as well.[1] For example, the phrase "I would like the mashed potatoes, could you please pass them to me?" is considered a speech act as it expresses the speaker's desire to acquire the mashed potatoes, as well as presenting a request that someone pass the potatoes to them.
87
Speech Acts Can Be...?
Representative
–asserting fact and speaker’s belief in it
Declarative
–bringing about new state of affairs
Directive
– Speaker tries to get addressee to do something
Expressive
– Speaker wishes to reveal psychological state
Commissive
–Speaker commits to some future course of action
88
Direct speech acts
Direct speech acts: the speaker’s intention is revealed directly by their words
–Please bring me a chair [Directive] –I hereby sentence you to five years [Declarative]
89
Indirect speech acts
the listener has to infer the speech act
There are some empty chairs at the back
90
Brown & Levinson, 1987
Increasing indirectness of speech acts conveys increasing politeness
```
–Make me a cup of tea
–Will you make me a cup of tea
–Can you make me a cup of tea
–Could you make me a cup of tea
–Is there any chance of a cup of tea
–Would there be any chance of a cup of tea
```
91
Grice’s maxims of conversation
Quality – Make your contribution true – do not say what you believe to be false. Do not say that for which you lack adequate evidence
Quantity – make your contribution as informative as required, but no more
Manner – Be clear: avoid obscurity, ambiguity, wordiness and disorder
Relevance (or relation) – Make your contribution relevant to the aims of the conversation
Conversations quickly break down if conversational partners deviate from these maxims for no reason
92
Conversational implicature
A particular type of inference that allows us to work out what the speaker intends to convey that they don’t directly express in their utterance
We can use a conversational implicature to interpret an utterance if its literal meaning doesn’t appear to respect one of Grice’s maxims
–as long as we assume the cooperative principle
Conversational implicatures play a fundamental role in everyday language use e.g. Metaphors
93
Audience design
Speakers tailor their utterances to suit their addressees’ needs
–What they know, their language abilities, what they are interested in
–Child-directed speech (‘motherese’)
–‘Foreigner talk’
94
Common ground
How do we come to have common ground?
Physical co-presence: being in the same physical environment
– e.g., we are both standing by the same building
Community membership: we are members of particular communities that share specific knowledge – Edinburgh students
Linguistic co-presence: what we have talked about together before – and how we have talked about it (the words etc we have used)
Conversational partners can also use common ground to conceal their meaning from other people
95
Common ground
Brennan & Clark (1996)
Round 1 - no linguistic common ground:
Dog 75%
Round 2 –
Labrador
Round 3 – using linguistic common ground from Round 2:
Labrador 52%
96
Clark & Schaefer (1987)
One student described pictures of campus locations to a friend
Task:
to describe each location so their friend could identify the correct picture
but while making sure that another student (unacquainted to the partners) could not identify the landmark
Conversational partners are adept (though not perfect) at using common ground based on co-presence to communicate and conceal:
– Addressee accuracy: 92%
– Overhearer accuracy: 47%
97
Clark, Schreuder, & Buttrick (1983)
Addressees also use common ground, when interpreting a speaker’s utterances
Q; What do you think of this clock? (same size clocks)
A; Which one do you mean?
Q; What do you think of this clock? (one clock is bigger))
Here, addressees do not query the speaker’s intended referent
– Common ground based on physical co-presence
– What is visually salient to speaker and hearer
98
Pragmatic impairments
People with these impairments have special challenges with the semantic aspect of language (the meaning of what is being said) and the pragmatics of language (using language appropriately in social situations)
99
Brownell et al. (1983)
The quack was selling a potion which he claimed would make men live to a great age. He claimed he himself was hale and hearty and over 300 years old. ‘Is he really as old as that?’ asked a listener of the youthful assistant. ‘I really can’t say’, said the assistant…
- I don’t know how old he is (Coherent non-humorous)
- I’ve only known him 100 years (Coherent humorous)
- There are over 300 days in a year (Incoherent)
RHD patients were worse at selecting correct punchline
• 60% correct vs 81% in controls
RHD patients often picked incoherent endings
• 50% of their errors vs 18% in controls
• RHD patients knew the ending should be surprising, but couldn’t maintain coherence
100
Developmental pragmatic impairments
– children who can process literal meaning of language but not its pragmatics.
– Autistic Spectrum Disorder, Pragmatic language impairment (social [pragmatic] communication disorder)
These impairments may be linked to their difficulties with Theory of Mind
