Word Recognition And Sentence Processing (Chapter 4, Part One)

Front 1

what is parsing?

Back 1

the process of unravelling the grammatical structure of a sentence

Front 2

what are the two stages of word recognition?

Back 2

1) a match between a written word and word stored in lexicon is made
2) lexical access- all the info stored in this representation, including this representation (syntactical and morphological and meaning) becomes available for future processing
Note- sometimes lexical access and word recognition are used interchangeably

Front 3

language-nonselective lexical access vs language selective lexical access

Back 3

coactivation of information in the other subsystem (when a bilingual encounters a word it activates both language systems) vs exclusive activation of information in the contextually appropriate language system- data supports parallel access to both systems

Front 4

what are word naming tasks? lexical decision task?

Back 4

-used to study bilingual word recognition
- word naming: printed words follow one by one and you have to read them
- Lexical decision: presented with letters and have to decide if it is a word or not (non words are usually pseudo words - ones whig follow orthography of the test lang)
- neither are a true representation of word recognition- in lexical the persons criterion varies with each trial, and in naming the words are in script so you can bypass recognition and just read, and also performance requires both recognizing and properly pronouncing word

Front 5

perceptual identification (2) what is progressive demasking?

Back 5

-task where the stimulus word to be identified is presented in a "data-limited " way - presented too briefly or vaguely to be clearly seen, and the participant must guess
- sometimes a word priming technique is presented first, or cross modal priming (word in aural target visual)
- progressive demasking is when the visual presentation of target alternates with a mask, and gradually the presentations of the target increase and mask decrease

Front 6

describe the generalized lexical decision task and the language specific lexical decision task

Back 6

-developed to study bilingual word recognition
- generalized: a yes response is given if the word id a word in either language
- language specific- a yes is given to only strings that are words in the specified lang

Front 7

What is gating?

Back 7

a perceptual identification technique where increasingly larger fragments of spoken words are presented and they have to guess from which words the fragments are derived

Front 8

language go no go task; language decision

Back 8

bilingual participants have to produce an overt response whenever stimulus is a word in their one language but refrain from responding when it is a word in the other lang; bilinguals presented with words of both languages mixed and have to decide form which language it comes from

Front 9

explain and give examples of online vs offline techniques

Back 9

Online= what is happening in the moment-more sensitive, cog processing (ERP, eye movement)
Offline= what aspect of a sentence's surface form impact how semantic relations between words are determined, assesses ultimate knowledge (self paced reading with moving window, grammaticality judgement or rapid serial visual presentation)

Front 10

eye-mind hypothesis

Back 10

at every moment in time the mind is processing the words of that which the eyes are looking at

Front 11

how is 'gaze duration' defined in eye movement recording techniques? (2)

Back 11

-the time between the moment the eyes first lang in a particular region of interest and then move elsewhere
- we can look at total fixation time, first pass time, and regression path time (includes when we regress back to the word after moving on)

Front 12

What is one advantage of studying using fMRI, PET, ERP?

Back 12

no behavioural response is required! it is just a straightforward interpretation of mental processes

Front 13

Modularity of the mind theory (2)

Back 13

-language is a module and isn’t affected by other processes going on in the brain so you would have an L1 module and L2 module
- these mental modules are characterized by: they operate in a domain specific way (only respond to a particular the of input), they operate fast and they are informationally encapsulated (they are impenetrable by higher level info)

Front 14

lexical ambiguity, often used to study what?; homograph effect

Back 14

words with two or more meanings i.e. cognates- often used to measure non-selective access; a difference in time and number of errors (shorter or longer depending on circumstances) between IL homographs and controls

Front 15

homophone effects (3)

Back 15

-orthographic overlap between the two languages produced facilitating effects (bilinguals responded faster to inter lexical homophones)
- where as phonological overlap produced inhibiting effects
-this supports language nonselective processing

Front 16

Describe the BIA bilingual interactive activation model (4)

Back 16

- has four levels of representational units: visual letter features, letters, orthographic forms of whole words, and language information
- it is interactive- the units at one level can inhibit or activate representations at other levels
- assumes inhibitory connections between orthographic nodes- lateral inhibition!
-in BIA if you think the word is in one language node, there is feedback inhibition to inhibit ** different from other models the other language

Front 17

what have studies on different written script language bilinguals concluded?

Back 17

reading by different alphabet bilinguals (or bi-dialect bilinguals) involves the same automatic phonological activation across languages as observes in same alphabet bilingualism
- in studies,
- monolinguals and bilinguals affected by semantic relatedness of prime -target, but only bilinguals affected by orthographic form relatedness! this is huge because the whole study was done in english, there was no filler or prime, the person just automatically conjures up the symbol in chinese from seeing the english word

Front 18

what does the BIA model adduce about frequent words? what is the implication? (3)

Back 18

- their nodes will have a higher baseline rate of activation
- this might explain why we misread infrequent words as frequent ones
- this also explains the fact that the size of a homographs effect depends on the relative frequency of its two readings

Front 19

How does the BIA model explain homographic effect?

Back 19

the difference in the baseline activation state of the recognition system’s nodes upon the presentation of a homograph vs a control word
there are dual nodes for ILH

Front 20

What is SOPHIA model (BIA+)? What does it say about lexical decision tasks?

Back 20

-a revision of the BIA which includes phonology and also removes the idea of feedback inhibitory connections

- in a lexical decision task if you are getting more info from the semantics level from both lang, you will be faster- you don’t care about the semantic particulars because you are just pressing a button if its a word or not
- this feedback would interfere with priming

Front 21

What is a word's “neighbourhood”?

Back 21

a set of words that share a substantial part of their othrographic and/or phonological form with the target word
they found that the time to recognize a word is influenced by the number and frequency of its neighbours
this mens that word recognition doesn’t take place independently and that a word doesn’t just activate its own representation

Front 22

What have cross-language neighbourhood studies shown?

Back 22

-lexical decision times were longest when the word presented had more neighbours in the non-target language than the target.
-They were shortest when the word had lots of neighbours in the target language
-thus orthographic overlap with L1 had inhibitory influence on L2 words recognition

Front 23

translaton priming effect; when do we see an enhanced effect?

Back 23

difference in response times for translation stimuli on one hand and unmatched, unrelated primes on the other
we see an enhanced effect for cognates which suggests that under certain circumstances bilinguals may automatically activate spelling-sound correspondences in non-target language

Front 24

what is a "cohort"?

Back 24

a cohort is the set of activated representations by a spoken word in spoken word recognition studies

Front 25

what does it mean to be in monolingual mode?

Back 25

in experiments when the participants have no idea they are participating in a bilingual experiment- used to study between language and within language effects. opposite is bilingual mode

Front 26

talk about speech perception in bilinguals in L1 as opposed to L2 (2)

Back 26

- speech perception in L1 may be immune to influence of L2 but not other way around, and not for cognates, which are open to competition cross language
- spoken L2 words give rise to language nonselective phonological activation in a bilinguals lexical memory

Front 27

What is the BIMOLA model (bilingual interactive model of lexical access ? (4)

Back 27

- developed to account for spoke words recognition
-represents auditory features (shared by L1 and L2), phonemes (separate), and word nodes (separate)
- like BIA (and SOPHIA) it assumes inhibition and activation of nodes however BIMOLA restricts lateral inhibition to units (words or phonemes) of the OTHER language
- doesn't believe in language nodes

Front 28

what is base language? base language homophone effect? (BIMOLA)

Back 28

-except in translation situations, bilinguals usually select one language for communication. the other language may be called the guest language
- phenomenon that words from the other language that have close homophones in the base language are more difficult to process than other guest language worse

Front 29

unit similarity effect (BIMOLA)

Back 29

- a unit, like a phoneme or word, presented in one language and sharing properties with a unit in the other language will co-activate the other

Front 30

what things (4) modulate the homograph (phone) effect?

Back 30

-task demands (generalized vs lang specific lexical decision task- effect is larger in the latter)
- composition of stimulus set (prescence vs absence of words from non-target language)
- frequency of ILH in taget and non target lang- effect is larger if iLH more frequent in non-target lang. “manger in french way more frequent than in english"
- processing lang (are you doing task in l2 or L1?- effect may be dampened when processing in L1, resulting in selective lexical access)