Study Guides (248,490)
Canada (121,589)
Psychology (1,715)
Midterm

Midterm 3

40 Pages
126 Views
Unlock Document

Department
Psychology
Course
Psychology 2134A/B
Professor
Marc Joanisse
Semester
Fall

Description
3.1 – Reading 11/3/2012 8:54:00 PM How Do We Read Anyway? - what happens in your head when you read a word  what kind of information are you using?  what kind of information are you accessing?  How do you do it? How Does The Brain Read? - no single ‘reading center’ Overview  visual input  letter features  letters  orthography (spelling)  phonology (sound)  semantics (meaning) Where Does Reading Come From? - humans evolved to talk - did we evolve to read?  First evidence of written words: 3500 years ago Speaking is Different From Reading - reading a sentence  each letter is discrete  spaces between words  minimal variability o fonts are all basically the same o spelling is (usually) correct  self-paced - hearing speech  coarticulation o phonemes are discrete o no spaces between words  lots of variability o between speakers o within speakers  always done online Reading Is A Hard Task - speech recognition is fast, online, and efficient - written word recognition is hard  learned later in life  slow to progress (only fluent after many years)  slower than listening (speed reading doesn’t count) - some people never learn to read  dyslexia; illiteracy; cultures without orthography Writing Systems of the World - logographic system  each word or morpheme has a symbol - syllabic system  each distinct syllable has a symbol - alphabetic system  each phoneme has a symbol Except.. - no language fits perfectly into one category - English is usually alphabetic, but what about  save, shoot, through - English has logograms  $, &, % Orthographic Depth - the transparency of mapping spelling to sounds - deep  Chinese, Hebrew - shallow  Italian, Finnish - English  somewhere in between Transparency in Chinese? - mostly no - but some examples of phonetic regularity - example:  words containing 大 symbol have similar pronunciations  大调 dàdiào (major)  大蒜 dàsuàn (garlic) Theories of Lexical Knowledge - we know lots about words  orthography, phonology, semantics, syntax - how do we represent this knowledge?  Locally: as a list of words  Distributed: as bundles of information - how do we access this knowledge during reading? Measuring Lexical Processes - lexical decision tasks  deciding whether a letter string is a real word o HAVE, MAVE, ASRMP, XXXXX - naming tasks - priming - dyslexia  acquired, developmental Lexicality - words are easier to process than nonwords  slower reaction time for MAVE than SAVE - not all nonwords are created equal  MAVE vs. AMVE Frequency, Length - slower recognition for lower frequency words  GREAT (high freq.) vs. PEAT (low freq.) - length effects  longer words are slower to recognize  WOO < WEATHERPROOF o Both have same frequency! Reading Is Automatic - you can’t help but read stuff - Stroop task  naming the color of a word o RED o RED  Task is harder when it conflicts with the word itself - note the reverse isn’t true – word reading isn’t harder when color conflicts Orthographic Effects on Auditory Word Recognition - Seidenberg & Tanenhaus (1981) auditory rhyme monitoring task  pie-tie  rye-tie - orthographically dissimilar words: slower recognition - spelling code is automatically activated, even when you’re hearing a word Other Things About Reading - homophony: words can have same sound but different spellings/meanings  bear-bare  carrot-carat - polysemy: same word can have multiple meanings  bank-bank  bass-bass Spelling-Sound Regularity - how do you pronounce words ending in ‘-ave’?  RAVE, SAVE, KNAVE, GAVE, SHAVE  HAVE ???? - lots of words follow the rules  regulars - some don't  exceptions (words with irregular spelling) Exceptions - irregular spelling  inconsistent with other similar words  pint (mint, stint)  great (beat, seat, treat) - strange words  not really inconsistent, but seem wrong  colonel, yacht Regularity Effects - regular words are usually faster to name than exceptions  only for naming, not LD - frequency by regularity interaction  naming low frequency exception is much harder Why Regularity Matters - how do we recognize words? - problem with memorization: we want to be able to generalize to novel words  NUST  BLOG, PODCAST Decoding - using what we know about spelling-sound correspondences - ‘phonics’ - big question:  how much of reading is decoding?  How much is recalling a word from memory? Phonological Priming - do we ‘decode’ when we read a simple word? - masked priming study  BREAK – take  Faster when they prime the word  (not faster for BROKE – take) - what does this tell us? Models of Reading - describing the mental processes involved in reading - how can we explain effects of  frequency  regularity/neighborhood  lexical influences (eg., priming) - what does reading tell us about how the mind works  modularity, local vs. distributed information The Autonomous Search Model - the lexicon is organized like a list  lexical access: searching through that list - each lexical entry is like a library book  each item is ‘looked up’ using orthography or phonology  searching serially through the lexicon for the right word - a bottom-up model of lexical access  is this problematic? Problem with Autonomous Search - can’t account for phonological effects  assumes we don't decode familiar words - how do we generalize to new words?  MAVE 3.2 – Reading 2 11/3/2012 8:54:00 PM The Dual-Route Model - Coltheart (1978): an extension of Forster’s (Autonomous Search) model - we can access the meaning of a word in several ways  orthographically  phonologically - lexical route: whole word recognition - rule-based route: grapheme-phoneme correspondence rules Evidence for 2 Routes - nonword reading is fast and easy  so we have a spelling-sound route - exception word reading is fast and easy  so we need a lexical route - one way to test it: reading in different languages  Hebrew: v. deep  English: in between  Serbo-Croatian: v. shallow Priming Experiment (Frost et al.) - semantic priming effects across languages - prediction from dual-route model:  lexical route: semantic priming effects  GPC route: no semantic priming effects - results confirmed this  the need to use the GPC route varies  some languages rely on it very heavily: shallow orthographies o less semantic priming  others hardly at all: deep orthographies o more semantic priming Seidenberg & McClelland Model - totally different approach  connectionist model - words are made up of  letters (orthography)  sounds (phonology)  meaning (semantics) - reading involves learning how each of these is connected to the other Lexical Processing in this Model - takes part of the pattern as an input  eg., orthography of a word - activation is passed from one layer to another  via neural connections  amount of activation passed depends on the connection strengths between units Explaining Lexical Effects - priming: preactivation of a word’s node due to previous inputs  HAT primes CAT  Because seeing ‘HAT’ activates phonological features of ‘CAT’ - frequency effects:  learns via experience with words  frequent words = stronger connections  so it’s more efficient at computing frequent words (like ‘break’) than less frequent one (‘freak’) What It Can Do - codes word similarity as overlap in units - so it can generalize to new words  reads ‘NUST’ but learning MUST, NUTS, NUMB, CUSP, etc. Learning to Read - different teaching techniques have tried to emphasize a specific ‘way’ of reading - phonics: old fashioned, ‘sounding it out’ - whole-language: newer approach (1980s) – focuses on learning words, not spelling rules The Whole Language Debate - phonics: translating spelling to sound, then getting meaning from sound  C..A..T..  [kaet]  (meow) - whole language: cuts the middleman  teach children to read like adults - whole language experiments have been major disasters (California; Ontario)  kids who do learn to read end up learning phonics to compensate Developmental Dyslexia - general definition: a deficit in reading not caused by more basic impairment  slow, laborious reading  many errors  normal IQ - has nothing to do with reversing letters and seeing words upside-down Varieties of Dyslexia - all people with dyslexia are poor at reading  some are markedly worse at certain word types - phonological dyslexia  poor at reading nonwords - surface dyslexia  poor at reading irregular words Phonological Dyslexia - phonological deficit  eg., poor phonological awareness  saying split without the ‘p’  judging whether sandal and candle rhyme - phonological problem leads to difficulty learning spelling-sound regularities  problems with nonwords - compensate by reading via ‘direct’ route from spelling to meaning Surface Dyslexia - difficulty with ‘direct route’  spelling to meaning - okay with phonological awareness and nonwords - rarer form of dyslexia  not as well studied 3.3 – Language and the Mind 11/3/2012 8:54:00 PM ‘Language’ vs. ‘Cognition’ - can we study language independently of other mental abilities? - same as  using logic?  Doing math?  Recalling information from memory? - do these processes use a language-based code? Language and Memory (G. Miller) - Paired Associate Learning  PLM – DAT, GRD – FLO  Subjects were ‘cheating’ by using other strategies - STM tasks  remembering lists of things  people increase their memory using strategies like chunking - these showed that we need to be studying more than stimulus-response Multistore Model of Memory (Atkinson & Shiffrin)  input  sensory register  short term store  response generator  output  long term store  response generator  output Short Term Memory - Miller  ‘the magical number seven, plus or minus two’ - tested immediate recall of lists  much better for lists of 1-7 items  very few can do better than 9 - when recall is delayed, or there is interference  worst performance - suggests that STM is limited, volatile Testing the Multistore Model - tested how people store information in memory  ‘farmers milk cows once a day, early in the morning’ - hard to remember the correct form:  cows are milked by farmers once a day, early in the morning  farmers milk cows once a day, early in the morning - but easy to remember the correct meaning  cows are milked by farmers once a day, early in the evening  cows are milked by farmers once a day, early in the morning - verbatim effect  people are better at remembering the meaning of sentences, worse at remembering the form of a sentence What This Says About LTM - people have trouble remembering the exact form of a sentence - but they are good at remembering the exact meaning of sentences - structure of sentences is not stored in long-term memory - comprehending sentences involves integrating general knowledge Beyond the Multistore Model - the multistore model  a framework for understanding memory, language and reasoning  divided memory into and LTS and STS - newer research has further subdivided LTS and STS into subcomponents Testing the Size of the STS - Baddeley: studied how STS works, its capacity - interference: how much harder it is to do two language tasks at the same time - doing a verbal logic task while  remembering two digits  remembering six digits  repeating a word, over and over Effect of Concurrent Tasks - verbal reasoning gets harder while performing other tasks - remembering more digits increases errors - repeating words also increases error rates What This Tells Us - verbal reasoning (‘A comes after B’) uses STM memory resources - digit load: more digits increases interference - articulatory suppression task  repeating words also takes up memory resources - hypothesis: there must be some language-based code in STM Visuospatial Task - working with a mental image in memory  eg., tracing the letter F in your ‘mind’s eye’  judging whether two shapes are the same Effect of Digit Load - what happens when you have to hold digits in memory while doing a visuospatial task? - much less interference than when doing a logic task Different Types of Interference - modality-specific interference  doing a language intensive task and a verbal STM task at the same time  big interference effects - modality-general interference  doing a visuospatial task and a verbal STM task at the same time  small interference effects Model of Working Memory Attentional resources  central executive  phonological loop  visuospatial sketchpad What This Has to do With Language - the use of language is affected by short term memory - but: some memory codes use linguistic information, while others do not - which WM tasks interfere with language?  Ones that engage the phonological loop  But not visuospatial ones (or at least not as much) Does Language Influence How You Think? - Watson (1925): thinking is silent speech  Smith ‘experiment’ with anesthetic curare  Ability to speak isn’t necessary for thought - Vygotsky (1934): famous developmental psychologist  language helps children’s cognitive development thanks to a ‘voice inside their head’ - Chomsky (1958): language and thought are completely different The Sapir-Whorf Hypothesis - Whorf: engineer at MIT, worked at a fire insurance company  investigating fires, noticed that people take the meaning of ‘empty gasoline drums’ too literally - came to the conclusion that people’s thoughts are heavily influenced by language - developed (with Sapir) a theory of the relationship between language and thought Linguistic Relativity - lexical and grammatical differences between to languages are reflected in the people who speak them - cultures that have two words for related objects tend to think of them as different - cultures with only one term for two objects will think of them more similarly Evidence? - Eskimo reported to have 100 words for snow  their culture sees ‘snow’ differently based on whether it’s falling, slushy, etc. - Hopi has one word for ‘flying things’  so they don’t think of birds differently from how they think of airplanes - Hopi speakers have a ‘different’ sense of time Except.. - ‘Eskimo’ (actually a group of Inuit languages) has closer to 10 words for snow  and so does English (slush, powder, avalanche) - Hopi speakers can describe other flying thing, coin new phrases to describe them - and Hopi does have a present, past and future tense  (in spite of what Whorf thought) Experiments to Test Linguistic Relativity - Heider (1972): studied color terms and color processing in different languages - Dani (spoken in New Guinea)  two color terms, black and white - showed them two color chips - then tested whether they could remember which one they saw Is Geometry Universal? - Dehaene: does language influence how a culture processes spatial relationships? - Mundurukú: Amazonian language  has fewer words for shapes, spatial relationships  no words for directions, shapes, north, south… - map reading task  look at a map that shows a hidden object  then find the object in real life Results - Mundurukú adults perform at about 71% accuracy  better than chance, but only about as well as American school children  but American adults are closer to 90% - so what’s the deal  hypothesis: language ability could be influencing the ability to code geometric relationships - but maybe: nothing to do with language, just experience  they find map reading ‘weird’ Codability - describes how easily you can describe a concept in a language - example: flavor words: salty, sweet, bitter, sour..  Japanese uses a word ‘Ajinomoto’ to describe flavor of MSG  English has no traditional word for this – even though we can taste this flavor - test case: does codability influence how people think about concepts like color?  Eg., Tarahumara has no word distinguishing green vs. blue Does Language Help Learning? - number terms: English and French have odd terms for 10-20 (twelve, thirteen..) - Japanese and Chinese don't:  ichi, ni, san (1,2,3)  juu-ichi, juu-ni, juu-san (11,12,13) - does this change how children ‘think’ about the number twelve?  Maybe…they do better when tested  But could also be due to cultural differences Language and Memory for Visual Scenes - subjects watching a video of car crash - the questions they’re asked afterwards affects their responses  ‘how fast were the cars going when they were hit?’  ‘how fast were the cars going when they smashed into each other?’ - people tend to estimate higher when you say ‘smashed’ Social Inference - does using non-sexist language matter? - people tend to pay attention to the gender in pronouns  hearing ‘he’ affects people’s attitudes toward a passage, even when it’s said in a general sense - should we change the words to ‘O Canada’ to be gender neutral? 3.4 – Language Acquisition Under Special Circumstances 11/3/2012 8:54:00 PM Creolization Pidgins - rudimentary languages created by speakers of different languages - usually occur as a result of slavery, trade - not ‘real languages’ in the usual sense  have small lexicon  few if any polysyllabic words  no productive morphology  fixed word order, short sentences  no function words (a, the, in, on) Creolization - Bickerton (1980s): many pidgin speakers in Hawaii due to influence of English - studied pidgin speakers who learned it as an adult vs. their children who learned it from birth  children showed more complex language  had grammatical structure parents didn't - children seem to be going beyond the input available to them - parent tutor is simpler than what children learned Pidgin vs. Creole Grammar - typical pidgin sentence:  aena tu macha curen, samawl churen, haus mani ei  (and too much children, small children, house money pay) - typical creole sentence  im fi kom op ja (he ought to come up here)  we de a London (We are in London) - creoles have  rich inflectional morphology, complex word order, bigger vocabulary, more phonological complexity - creoles share grammar properties regardless of base language Language Bioprogram Hypothesis (Bickerton, 1990) - complex language can develop in spite of impoverished inputs - creoles around the world bear remarkable structural similarity to each other  different vocabularies but similar grammars  all are SVO languages regardless of base language grammars - hypothesis: creoles are the result of a biological endowment for language acquisition Role of Biology - hypothesis: we don't learn language in the traditional sense - we acquire it  using innate principles - Lenneberg (1960s): there is a critical period for language acquisition  idea from biology: animals are able to learn certain behaviors early in life, but not later  imprinting in chicks  song learning in songbirds - language has lots of characteristics consistent with critical periods Critical Periods - hard to design an ethical study to test this  but there is evidence from ‘natural experiments’ - ‘wild’ children  no language till late in children - acquisition of sign language  early vs. late signers - second language acquisition Feral Children Victor of Aveyron - found wandering the woods near Toulouse  appeared to be living in the wild for all/most in life - enlightenment period: much interest in what distinguished humans from animals  ability to learn language? - Jean Marc Itard – medical student  attempted to ‘civilize’ him - showed significant progress in most respects - but could never learn more than a few words of language  ‘lait’, ‘mon Dieu’ Genie - discovered in Los Angeles suburb in 1970 at age 13  mother was blind, father was just plain crazy  father locked her in a room, tied to potty chair, bound in a sleeping bag at night.. - science to the rescue?  The ‘forbidden experiment’: attempt to teach her language - Genie learned vocabulary quite quickly - but did not progress past 1-2 word sentences  ‘applesauce buy store’ Not Terribly Happy Ending - during the study, major custody battles over Genie  lived with researchers involved in study for a while - disagreement among scientists and clinicians about treatment - in 1974, NIH cut funding - custody returned to mother, who sued the scientists - today, Genie still lives in a foster home Genie – What Have We Learned? - is there a critical period?  What aspects of language developed?  Did some problems persist? - how are language and IQ related?  Evidence that Genie had an intellectual disability? Language in the Deaf - L1 signers vs. late signers - prelingually deaf individuals  normal late acquisition - how about:  hearing loss later in life?  Limited input early in life? Role of Input - deaf children of hearing parents:  parents are not L1 signers  so input to children is poor - children quickly become better signers than parents Deaf Children With Limited Input - not all parents of deaf children start with sign - many try to teach child spoken language  strategies such as lipreading 
More Less

Related notes for Psychology 2134A/B

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit