38 Pages
Unlock Document

Carolyn Ensley

Psycholinguistics Language: identifies generations, binds groups, exclude those who aren’t peers, gender differences Psycholoinguistics: study of psychological factors that influence language use -similarities and differences in processing and using language, whether word order rules are innate or learned -psychological factors in development of language Acquisition: how language develops over lifespan Production: how we construct utterances Comprehension: how we perceive written language and speech Most knowledge of language is tacit Tacit knowledge: implicit, performing something we’re not aware of how we do it Explicit knowledge: knowledge of processes or mechanisms underlying acts, consciously aware of this Cognitive processes: perception, memory, thinking Garden path sentences: begin leading you it is about one thing but at the end it is about another, shows us cognitive factors influence language, the relationship between words causes us to choose which meaning we should assume Indirect requests: “can you open the door?” rearranged so it is less threatening, pragmatic knowledge shows us this request is more likely to be granted Aphasia: loss of language, shows what parts of language are independent of one another Wernicke’s aphasia: has syntax but no meaning Wundt: Theory of language production, sentence as the smallest unit of language, we take a concept or thought and create segments sequentially into a sentence. When we hear something we break it down in chunks and turn it into abstract thought. (not accurate) Behaviorism: Verbal behavior theory of language, language the result of being raised in an environment where language was reinforced speech (the role of adult speech in acquisition is not that direct) made associative chain theory (turned out to be wrong). Associative chain theory states that a sentence consists of a chain of associations between words in the sentence, word by word, one after another. Chomsky: proved this to be untrue, there are sentences that have no association but are syntactically acceptable, chain only works one way when association should work both ways. We don’t utter the same sentence in the same way twice. Discontinuous constituents suggest there are long range relationships between words. Argues that language is innate (nativist). Children pick up on rules without having them explicitly explained (poverty of stimulus argument). “prelinguistic communication:” -infants can’t shut off airways to talk, this is so they can breathe while nursing -exists until early part of second year -pointing, waving are basic but they tell us a lot -they show that the child understands that actions can be used to achieve goals -if there is an absence of this communication it does not bode well for the future ability to speak -communication occurs after children develop a theory of mind, they learn that other peopleare separate from themselves, this often occurs at the same time as pointing -used by anthropologists to imagine how language may have evolved -babbling, communication precedes phonology, syntax, and semantics -pointing as a universal human action unique to our species Phoneme: smallest unit of language that conveys meaning Morpheme: smallest grammatical unit in language (ex. S) Syntax: grammatical rules that tell how to put words together Pragmatics: social conventions that tell us what words to use when Prenatal language recognition: -evidence they recognize it later -newborns prefer their mother’s voices, suggests we are prepared to perceive language at birth Parentlanguage: tacit knowledge of how to teach children language, it is universal -higher in pitch, more variable, more exaggerated -probably designed to maintain attention -evidence infants prefer baby talk -caregivers use a different intonation to introduce a new word -assertions (use of an object for attention) or requests (use of an adult as a means to an object) Shatz 1978- younger children respond better when they use the object they talk about, but the assumption was made that a child’s response to a complex sentence meant they understood it Early Phonology: Phonology develops alongside other communications skills When they learn early speech sound they use them for the same thing as gestures First attempts at sound are for practice not communication (they babble more when they are alone than with people) May know more about phonology than they can produce, limited capacity to form different sounds They can identify distinctions even if they can’t produce they “fis” phenomenon (fis instead of fish), we cant use production to understand phonological development in children Catergorical perception in infancy: -Perception of a phoneme as belonging to one group or another even if the aspects of it change (a continuous sound from b to g sounds to us as b then g) -these boundaries are learned based on what language we are exposed to -children are born with the ability to distinguish category boundaries and they later lose it suggests this is an innate ability -6-9 months children can recognize probable syllables in comparison to not improbable ones -infants can recognize their own name by 4.5 months -the probable/less probable distinction is important for word segmentation -evidense 8 month olds can parse long streams of syllables -Marcus (1999) shows infants notice a new syllable added to nonsense syllable they have been exposed to at length -newborns can distinguish between languages that have different rhythm patterns (prosody) -prosodic factors may precede categorical perception -many evolutionary researchers claim prosody was first evolutionarily Babbling: 2 months- coos, more varied than cries, start at back of the mouth (ga, ka) 7 months- reduplicated babbling repeat babababa 11/12 months- variated babbling, intonation, first word, categorical perception for other languages declining at this age, shows increase for phonetic knowledge for native language. Babbling is more for play than communication. End of first year- gestures and phonology come together, ready to communicate meaning, greater motor control, cognitive maturation (things have names, referencing), sometimes make up own names for things (idiomorphs) that they may use consistently, sometimes reference to adult speech or object Idiomorphs indicate language is creative, children have learned it is important to be consistent when referring to objects Age 1- start to use words, sometimes sound correct, sometimes not Reduction- taking away a phoneme Coalescence- put a phoneme in the wrong order Assimilation- make up a word that sounds similar Reduplication- use a phoneme already found in the sentence Must use central exeutive to do it all as it is new, system can’t handle all the details so simplifications are used to deal with limited capacity working memory system, consistent with the fact that children can imitate a word but not use it spontaneously Early utterances -most are one word long -by about 19 months they may put two together -during one word phase they master the names of common objects, interactions become tangible and they can begin to learn from the parent -begin commenting using just one word, practice for longer utterances later (these are underpinnings of conversation, speaking not to get something but just to speak) Lexical development -age 6 a lexicon of 14 000 words -starting at 8 months this is about 8 words a day -children “know” words in different ways, several variations of a verb mean this estimate is higher than the entries, may not know every word on lexical semantic syntactic levels Early words: -begin on words with here and now significance, food, toys, clothes -bias to moving objects, learn moving objects names faster -not limited to nominal (nouns) -specific nominal: Mommy -action words: go, up -modifiers: pretty, dirty -social words: please, want -function words: what, for -non-manipulated items like diapers, clothing and trees tend not to be on the list The role of adult speech: -stimulation in environment that provides them with information -children cross culturally ask “what’s that?”, or adult may ask child to name an object -caregivers usually choose the basic level words, can move up or down hierarchy from there Holophrases: single word that means more than one word, show continually with prelinguistic gestures that use one hand to get an idea across. Children tend to use holophrases as a midpoint between one and two word sentences, use the same words in different context due to fixed vocabulary Early Grammar: -niether imitation of adults or full adult grammar, similar in signed and spoken languages -Age 2 begin to use word combinations: what needs to be learned varies from language to language (in English must learn word order, in Turkish must learn inflection) these impact how grammar develops -age is not a good predictor of syntactic growth, have constructed an index to facilitate the comparison across children at the same stage of language development -MLU mean length of utterances in morphemes -birth/day counts as two to an adult but only one to a child unless they know both words Stages of MLU divided language 1. 1-2- single word or putting words together 2. 2.25- learning to modulate using grammatical morphemes 3. 2.75 –learning more complex abilities like questions and negatives 4. 3.5 5. 4.0 or more- grammar should be fully formed Most children are at stage one at 24 months but the age for other stages is more variable. MLU most valuable during stage 1 and 2 Evidence that the utterances are grammatical 1. Children tend to start with content words and leave out function words, shows child understands grammatical distinction of what is critical to the meaning and what isn’t 2. Words are put together in a systematic order Word order by syntax? Doesn’t seem to fit with early utterances, evidence that children don’t treat complicated sentences differently. Not a lot of support. Word order by semantics? Meaning determining word order, By position? Preference for putting certain words in certain positions, it is a more simplistic way of order -evidence for understanding of syntax before it appears in speech Different styles and strategies in how children use languages: 1. Referential: most common, nouns and verbs in environment, begin with words and combine into sentences. Mothers of these children are more descriptive 2. Expressive strategy: social interaction determines language acquisition. They have more diverso vocabularies and say things like “stop it” and “I want it”, use intonation, use “dummy terms” when they don’t know word to make sentence sound correct, more social, language used more to communicate as a social tool , learn whole sentences and break them into words. Mothers of these children are more interactive. 3. By age six they both seem to assimilate and learn the other aspects of language We look at language in terms of -environmental factors -cognitive processes -linguistic mechanisms Studies of feral children show severe neglect may retard language acquisition, but we are unsure as to why this is Victor 1785 found, unable to learn after 5 years of training, only grunted, could name things but not request by name, couldn’t generalize nouns Why couldn’t victor learn? Maybe he was abandoned due to learning deficit, perhaps his teaching was bad, he missed critical period for language exposure Isolated children: who have almost no human contact Genie- had no contact from 20months to 13.5 years found in 1970, father didn’t want children, mother did, father murdered first two children, a third child sent to live with grandparents, genie was the 4 .h Had average weight, hip required a splint, pediatrician said perhaps retarded and father used this as justification for neglect. She lived strapped to a chair, put in a straight jacket, caged at night, father never spoke to her, the rest of the family kept from seeing her, not allowed tv or radio so was completely unexposed to language. She was beaten if she made any sound. Was found by accident when her mother took her along to get services for herself. Charges were filed against the family, Genie’s father committed suicide. -she made slow progress, rapid development of vocabulary (faster than a baby’s first year of language, had difficulties with prosody, eventually learned to put words together. She strung together content words without normal syntax but with clear meaning. Cognitive abilities exceeded linguistic intelligence. Critical period hypothesis: neurological changes that occur leaves the brain less accessible to language later, must be used earlier in life, few empirical studies 1989 Johnson and Newport Looked at families of Korean and Chinese immigrants, found early arrivals had a better idea of grammar rules, ages between 3-7 were as good as those who had been speaking English since birth Motherese: may be necessary to develop language properly, but we speak to dogs this way and they don’t learn to speak, but perhaps it is important in assisting in development Correlation study: limited relationship between the way adults speak and child language development, but odd that simpler speech leads to more complex development Experimental approach: -sentence recast (repeat sentence) -given new info -new sentence group, parents gave unrelated sentence to child’s sentence -control group, no special treatment (parents generally used a combination of all three) Recast group showed slight advantage over other groups in verb measures Sown that adult language is linked to children’s linguistic development. But not shown that specific adult speech is necessary for language development. Chomsky says- poverty of stimulus argument, must be innate forces as well Some say motherese is an argument for importance of environment as key for acquisition of language Best conclusion: environment is necessary but not sufficient for acquisition of language Cognitive processes: Obviously need some cognitive processes in place to acquire language (storing, retrieving, panning, analyzing incoming sounds) Necessary but not sufficient- feral children who didn’t acquire language may have had normal cognition Slobin -have to pay attention to ends of words -phonology of forms of words systematically modified -pay attention to order of words and morphemes -avoid interruption or rearrangement of linguistic units Children in all languages use fixed word order regardess of whether required in their language, over regualization, cognitive prerequisites for learning language. Problem: slobin’s argument is circular, came up with principles that form language by looking at language, no other evidence of these cognitive processes occurring outside of language. Sensorimotor schema: About age 2, banging, sucking, throwing, taking in sensory information by acting on it. When they develop object permanence (8mo) this stage tapers off. Shouldn’t language change as well? A change in cognition should go along with a change in language. 1. Children who do not have object permanence will only use language related to immediate environment (lots of evidence supports this) 2. When object permanence is mastered children change language to refer to things not immediately available (lots of evidence for this) ex all gone, more 3. Cognition and language are linked but the links may be very close to eachother. -links between language and cognition have been found when there are cognitive delays ex down syndrome, language delays proportional to other cognitive delays -some cases there is a big difference between cognitive and linguistic functioning ex. Genie, some children unable to learn toilet training but can learn syntax William syndrome: syntax can exist without semantic knowledge, many case studies, very happy, sound like other people but major cognitive problems, unable to take care of themselves. If normal cognitive development is necessary for normal language then this shouldn’t happen. Need to look beyond cognitive factors to explain how language is acquired Biological- Innate mechanisms: Most controversial of the process, some think it is necessary, others don’t believe it at all Language bioprogram hypothesis: Bickerton, -every child born with an innate grammar that’s available if our environment is insufficient -Pidgin and creole languages provide evidence -pidgin: members of mutually exclusive languages interact, a version of the dominant language, just enough to get by -creole: when children of these immigrants acquire pidgin without exposure to the dominant language (a full language) -creoles abounded during colonization area, specifically studied in Hawaii, pidgin speaker’s speech was rudimentary, low complexity sentences, no grammar, no syntax, little word order, sometimes no verbs, no anaphor system, very relevant and specific to the speakers themselves. The pidgin speaker’s children developed a creole, a language that uses consistent word order, use of complete sentences, structural rules from other languages unlike pidgin’s lack of rules. Proof that an innate language bioprogram creates a language when there is insufficient environmental stimuli Criticisms of Bickerton: sophistication of creole may be based on exposure to English on plantations, but it differed from English in many ways despite little contact. Parents emigrated from Portugal which is similar to English in some ways. -further support from congenitally deaf children, (not taught sign language, taught to lip read) developed homesign that was similar to normal language, developed homesign in same sequence and had morphemes for temporal and spatial -sign language in Nicaragua- when no formal education for deaf children in Nicaragua, deaf children brought together created sign pidgin, children exposed to this language eventually developed a full creole language (ISN) Parameter setting: -Bickerton’s program is instance of parameter setting -grammar is a set of finite switches -different languages have different parameters on or off, all languages based on these parameters in different combinations -Chomsky: all children are born able to use any one of these, in language acquisition they learn whether to have parameters on or off, only focusing on those important to their native language Ex. Null subject parameter- can drop subjects in some language but not in English -Hyams claims that we have a default setting for null subject parameter, children born with this in null until they learn otherwise, explains why children ignore subjects -Valian argues children are born with potential for both forms of null subject parameter and learn which to use This could explain how children learn grammar rapidly Pinker: difficult to learn language with positive evidence alone (parents speaking properly), need negative evidence to rule out competing grammars. Pinker claims we don’t provide enough negative evidence therefore there must be an innate linguistic mechanism like subset principle are needed. Brown and hanlon: looked at how parents corrected their children, made few corrections of syntactic errors but more of semantic errors, yet children learn language Brocas aphasia- problems in speech production (near motor cortex) syntax intact but semantically wrong Wernickes-problems in syntax Conduction aphasia- cant communicate between the two areas Caramazza and Zuriph -given sentence and had to match it to a picture, nonreversible (semantically important), reversible sentences (syntax is important) -brocas aphasia and conductive: didn’t do well on reversible sentences, shows that it is not completely segregated -wernicke’s showed priming and broca’s did not, broca’s can not activate words quickly enough, comprehension is slowed because of a general slowing of processing Broca’s patient comprehension strategies: Can comprehend: the girl chased the boy Can’t comprehend: it was the boy whom the girl chased, when you have a memory problem it becomes more difficult when first noun is treated as an agent Category specific dissociation: lose ability to name colours, vegetables or other categories, suggests that neurological damage can selectively affect the lexicon and that the lexicon is organized neurologically Left hemisphere: seat of language in most people Right hemisphere: more involved in spatial skills, split brain patients could use left hand to identify shapes by feel but not right hand, some limited language ability Darwin: Right Ear advantage for most in dichotic listening tasks for determining pitch contour, but left ear advantage for words Laterality- idea of LH or RH being responsible for different things LEA for most individuals, prosidy Birle et Al -humor and comprehension in RH and LH damaged individuals -shown 2 cartoon frames and had to find the funniest -RH damage had difficulty with humor, picked frame unrelated to context -LH chose related to context but not as funny -RH better at surprise -LH better at continuity -humor depends on both hemispheres -LH seems to suppress inappropriate meanings than the right Burgess and Simpson -lexical task with priming -primed by dominant or subordinate meaning of the word RVF- immediate facilitation that decreased over time for subordinate LVF-immediate facilitation for both and even increased -automatic spreading activation occurs in both hemispheres but only LH suppresses it LH dominant for lexicon Lenneberg: claims no laterality at birth -looked at results of hemispherectomies, the children develop grammar -BUT they showed difficulty in complex grammar despite language being mostly intact -brain is much more plastic when we are younger -some laterality likely at birth (due to L hand R hand preference) -laterality could still exist but increased plasticity could allow for greater recovery -ties in with “sensitive period” for language- perhaps during this time we are using LH for language it is otherwise used for things like dexterity REA for speech LEA for non-speech Evolution of language Many theories- arose from animal calls, imitation of physical sounds, grunts -known as “bow-wow”, “ding-dong” and “heave ho” -they were told don’t bother studying origin of language -we are now pretty sure these are wrong, too simplistic -Humans didn’t evolve from chimps we had a common ancestor (we aren’t sure what it was) -Darwinian logic would assume it was for survival -at some point some of our ancestors had language, some didn’t but the ones who did survived We do behavioral studies of other animals’ language capacities, lateralization of the brain -our closest relatives have no language -primates use gestures, noises, alarm calls, mating displays, submissive gestures Language must be Arbitrary- not attached to environment or instinct Duality- can rearrange small set of phonemes for infinite communication Studies in chimps: -can’t teach them to speak, physically unable -have tried using signing, tokens, language boards -chimp acquires small vocabulary, uses it to get what it wants, put symbols together but no grammar, can teach to offspring Kanzi: -may have grammar Pinker says he doesn’t, there are no function words, inflections, tenses or distinguish between statements, questions or commands -chimps learn 47 symbols by 5 yrs, human child learns 14 000 at 5yrs Survival value: 1. Can learn about food/ predators indirectly 2. Share food more easily 3. Learn from other members of the group 4. Better maintain group knowing each other’s state of mind -spoken language developed recently -developed larger pharyngeal area needed for [a], [u], [i] -must have been important because adaptation of speaking makes us more vulnerable to choking -first we developed bigger brains then we started speaking Donald’s Hypothesis -drought lead to deforestation 7mya -quadrepedal pre-hominids starting to walk on 2 legs -faster, reach more food, sun protection -6mya Lucy first bipedal hominid -pelvis for upright walking -found in Ethiopia, Australopithecines, had free hands, increase in manual dexterity, small increase in brain size, due to pelvis babyhood lasted longer -there was a need for family groups and food sharing which likely lead to rudimentary communication (likely prosodic) -prosody are emotional sounds like cries, yelps, distinct from language, this also lead to bigger brains Homo habilus -stone tools, somewhat larger brain, looked more human -bigger brain for tool dexterity and planning also provide skills needed in language -only needed one dexterous hand so LH became highly dexterous, pressure for bigger brain Homo erectus: -more complex tools -complex gestures due to better dexterity -lived in cold environments, probably in family groups -increased brain size may have allowed for gesturing with sounds Homo sapiens: -earliest was 125 000 ya -have structure for speech -probably used gestures and spoken language at first and then reduced the gestures t -essay/research paper, find/ create a theory and use relevant articles and sources to prove or disprove it Nov 21 First midterm Linguistics finds commonalities across languages 1. Duality of patterning: all languages have a limited number of phonemes that are combined into a potentially unlimited number of meaningful words, one major difference between humans and animals, animal communication does not show this duality. 2. Morphology: phonemes are combined to make different words, use different forms of the same word to convey slightly different meaning, the rules for this is morphology, smallest meaningful unit of language is a morpheme. Free morpheme is a word on its own ex bed, bound morpheme must be with another morpheme to express any meaning ex. Er. Major grammatical; morphemes express number, person (first person, second person etc.), tense or aspect. All languages have morphological systems but they differ in the impact morphology has on grammar ex. In English the number of pronouns and verb must agree and must decide on when an action took place for tense, this is not in all languages 3. Phrase structure: sentences can be divided into groups of words or constituents. Constituents: there are two constituents in a declarative sentence, a noun phrase and a verb phrase 4. Linguistic productivity: no limit to the number of sentences in a language, vast number of sentences are novel but grammatically acceptable, the exception being clichés or proverbs that are repeated. Our ability to understand these novel structures is linguistic productivity or linguistic creativity. -any theories of language and psychology needs to explain these 4 factors Phone: different instances of the same phoneme with different features the “p” in “pill” is different than the “p” in “spill”, phones don’t affect meaning, we don’t perceive these and the brain hears all the p sounds as the same Phonemes: differences in sound that make a difference in meaning ex /b/ and /d/ distinguish between big and dig, phonemes determine comprehension of the word, English has about 47, some languages have 17, the most is in the 70s, its not any harder or easier with different numbers of phonemes. Distinctive features in English are manner of articulation (voiced, voiceless) and place of articulation (ex bilabial- uses both lips) our categorical perception is learned at a young age, we learn how to segregate these phonemes properly Phone is shown with brackets [] and phonemes with // Miller and Nicely 1955:  These features of phonemes have psychological validity  Used consonants followed by letter a ex /b/a  Syllables presented under different listening conditions ex. White noise, volumes, and they had to determine whether or not it was a match.  The more similar the speech sounds were on manner of articulation or place of articulation scale were more likely to be confused with one another Hockett, Chomsky -the human brain is finite, infinite sentences? -instead of storing all sentences possible we store rules for creating sentences, a finite number of rules can create an infinite number of sentences Lasnik -recursive rule, we can embed a sentence within another “the child thinks the man left”, “the woman knows the child thinks the man left”, this appears to be a resilient part of human language Animal communication: seems to be a small number of discrete symbols, a sound is used to mean only one thing, they don’t invent words as they are needed Some parts : morphologically irregular, most in English are past tense ex ate, went, fell, children have trouble with these Grammar: in psycholinguistics grammar refers to a person’s linguistic knowledge Definition of language: infinite set of well formed sentences Definition of grammar: finite set of rules that allows us to generate infinite set of sentences Grammars are thus theories of languages How do we evaluate a theory of grammar? 1. Observational adequacy- grammar must specify what is and isn’t acceptable in the language 2. Descriptive adequacy: grammar must specify relationship between sequences, if two sentences are similar in meaning but different in structure grammar should explain this 3. Explanatory adequacy: must be easy for a person to learn, everyone uses language, must explain this Chomsky’s evaluation -these criteria led Chomsky to evaluate various theories of grammar at the time, eliminated theory of phrase structures because they do not have descriptive adequacy, does not explain relationship between sequences Chomsky’s solution: transformational grammar Three arguments made about the usefulness of distinguishing surface structure and deep structure 1. Deep structure ambiguity- 2. Sentences can have similar surface structure but different deep structure 3. Can have different surface structure but can have same deep structure Therefore a grammar that includes only one level of structure cannot be adequate Transformational rules: Transformational grammar- derivation of a sentence is a two part process 1. Phrase structure rules used to generate tree structure 2. Series of transformations is applied to the deep structure and the intermediate structure ultimately generating the surface structure -transformational grammar based on the assumption that the rules were psychologically real -derivational theory of complexity: if we had to do all this it would take more time, distance between surface and deep structure measured by number of transformations required -early studies confirmed transformational grammar, but no experiment has confirmed it, no psychological difference between sentences that have been transformed and those that haven’t.  Lexical theories of grammar o In most grammars a lexical entry includes a words, meaning and its spelling, pronunciation, syntactic properties… o Brennan and his lexical functional grammar lexical entries also include variations on a words (ex all tenses) and the different kinds of sentences it could go in o For verbs this includes the agent and the patient as well as surface structure designation such as subject or subject o The explanatory burden is on the lexicon not on transformational rules o Makes good psychological sense  Retrieval of items from memory is easy compared to transforming items in short term memory o May explain why we find sentence comprehension so effortless in everyday life o Also known as ‘psychologically realistic grammar’ Jackendorf 2003 A parent, apparent, pronounced the same so we need some sort of semantic processor to sort out word boundaries and understand this sentence, suggests phonological processor and a semantic one -may be easier to understand evolution of language, primates seem to understand meaning before syntax, evolutionarily we may have had semantics first then added syntax Cognitive information processing memory system consists of: Working memory Temporary storage of information being processed in any range of cognitive tasks, we need it for holding conversations, remembering a phone number, storing information we work with Long term memory Semantic memory Episodic memory Baddley and hitch: 1. Central executive (attention) limited capacity pool of general processing resources, determines the tasks the “slave systems” (other systems) should be doing at any given time. 2. Visuospatial (map, imagine images and manipulate them) 3. Phonological loop (short time we can rehearse materials or keep them in phonological store), includes articulatory rehearsal system (rehearsing in our heads our out loud to solidify memory) Prediction: when people make errors in working memory they tend to be for similar sounds, regardless of being presented visually or phonologically, we rehearse this with articulatory rehearsal system, ex an error with F is more likely to be with S than with E, we turn language we are thinking abut into sound. Prediction: if a person has to remember letter while speaking sounds over and over the memory will be impaired, it occupies the phonological side of working memory, supported in experiments Prediction: word length affects memory, longer words are harder to remember, suggests working memory relies on speech like mode of mental representation -evidence working memory is separate from long-term memory, shown in people with dissociation disorders, people who can’t use speech muscles still have inner speech and can still use working memory -people who have trouble with central rehearsal attention) show poor memory span, this makes sense if rehearsal loop is controlled by central executive Are tasks with divided attention easier with people for longer memory span? Daneman and Carpenter -studied working memory span and attention, did digit span, complex span (remember info while understanding sentences), looked at correlations between the two, looked at correlation between each span test and reading comprehension -complex span predicted SAT scores for reading comprehension, digit span did not -consistent with idea that working memory stores and processes information and is of limited capacity, found complex span also related to arithmetic, counting and spatial cognition task Prediction: if executive allocates attention then divided attention tasks should be more difficult for people with smaller working memories Kane and Engle (2003) working memory predicted performance on Stroop task Prediction: people with strong working memories perform better on anti-saccade tasks: fixate and wait for stimulus, just before stimulus is presented there is an attention grabbing stimulus on opposite side, a test of being able to focus attention Prediction: people with high math anxiety have shorter attention span, anxiety may produce intrusive thoughts that distract articulatory loop, and experiments have proven this Long term memory: Semantic organized knowledge of words, concepts, symbols and objects, includes motor skills such as swimming, general skills such as grammar, spatial knowledge like the layout of your house, social skills like how to begin a conversation Episodic memory -people who have high skills have more semantic memory for that area -ability for long term memory related to ability for language -expert chess players better able to remember specific lay outs in a chess board in a way that was realistic than novices, not just chunking, they lay out the game in long term memory -suggests as linguistic knowledge increases, we should have more semantic knowledge of language in semantic memory Episodic memory: memory for specific events a person has experienced, lots of evidence that it us separate from semantic memory, idea that animals have episodic memory Tulving, schacter etc (1988) -studied KC who was in a motorcycle accident -able to remember playing chess (semantic), complete amnesia of past and inability to imagine himself in the future (semantic) -tulving thinks that episodic is more recent and that’s why it is separate, perhaps semantic came once we needed to know symbols -working memory can only hold 7 (+-2) bits of information at a time, likely we can only hold 7 constituents at a time -processing function of working memory serves to organize phrases into constituents -long term memory serves several roles: Semantic memory contains information about speech and sounds of words Episodic memories built from conversations Serial (one step at a time) and parallel processing (many steps at once) -many recent models are serial because this is how computers work Serial model: -starting point would be the idea -end point would be articulation of the sentence -what happens in between? Divide task into stages, phase structure, retrieving lexical items, pronunciation -would assume that these stages occur with no overlap Parallel model: -same start and end point, same tasks but all occur at once with some interaction possible -Rumelhart Mclelland PDP parallel distributed processing model, views the mind as massively parallel, everything always interacting with one another, evidence for this in reading -we know that activity of a neuron has effects on neighboring neurons, it is hard to stop the spread of activation on the brain Bottom up processing: would assume that all lower levels operate without influence from higher levels (phonemes separate from syntax), not likely true for language, usually seen as serial Top down: if expectations are different from outcome, responding is slowed, lots of proof of context having effects on perception (ec garden path sentences), they are usually parallel processes Automatic and controlled processes: -extent to which speech and language are automatic or effortful -we may have fixed processing capacity which is probably true for language -when we have a complex language task we may use up the resource and not leave any for other tasks -controlled tasks are complex and take up more resources -automatic tasks are more simple and take up less resources -automatic tasks: unintentional, uncontrollable, efficient, fast, some may even be biological, tying shoelaces, not disrupted by a controlled task, hard to describe -developing phrase structure is not automatic -recognizing frequent words is automatic Bresnan: phrase structure model was controlled whereas lexical model was more automatic -model of lexical processing that was more automatic would be easier to use and would be more in line with the fact that we seem to use language easily  Manner of articulation o Stop consonants  Obstruct airflow at some point  B d g p t k o Fricatives  Obstruct airflow but don’t completely stop it  F s o Affricatives  Stop like closing that is slowly opened  C j o Voicing  Difference between b d g and p t k  B d g are voiced since the vocal chords are used vs. p t k where the vocal chords are not used  Vowels o Distinguished by where in the mouth they are made (front, center, back) and by tongue position (high middle low)  Acoustic phonetics o Reading machine for the blind  This has proved difficult  Taking the phonemes and making words never sounds real, we all know that from AI type machines o Why?  There is no direct link between the letters that make up the word and the exact sound of the word  Suggests that we may process speech differently than other sounds  Evidence for this  Evidence that speech sounds are different o We can process far more speech sounds per second than other sounds (25-30 per second) o Experiments have compared experiences and report of participants listening to sounds like a buzzer tone bell and a siren close together to speech sounds close together o The buzzer combination sounds like indistinct blur o Speech sounds are perceivable  Spectrograms o Sound spectrograms o Filters analyze sound and translates them onto phosphor o Frequency on the y axis and time on the x axis  Formant transitions o Rises and falls occur at the beginning or end of a syllable in the middle is called the steady state o Transitions tend to correspond to the consonants, the steady states, the vowels  Parallel transmissions o Different phonemes in the same syllable are processed simultaneously o No distinction between the t and the u and the l in tool o We hear three distinct phones but the spectrogram reveals that they aren’t distinct in the speech signal  Context-conditioned variation o Exact spectrographic appearance of a phone is context dependent o Best example is with varying vowels  The motor theory of speech perception o Liberman et al 1967 o Based on the notion that perception proceeds by reference to production o Listeners use implicit articulatory knowledge as an aid to perception o Motivated by economy: use similar processing for perception and production o Deals effectively with the lack of invariance o We perceive sounds similarly even when they are produced differently  Confirmation o Does seem to be a link between perception and production o Articulating new sounds seems to help people learn to understand language better o Experimental evidence  Mc Gurk and MacDonal  The McGurk effect  When visual and auditory information are in conflict we use both to arrive at a stable perception  Basically if lips indicate velar consonant like ga while the sound hear concurrently is bilabial ba perceivers hear da  Alveolar stop that has some components of both ga and da  Place of articulation is cued by the eye, manner of articulation by the ear  Criticisms of the theory o Studies of infants show they can perceive phonemes not in their own language  Can’t speak yet o Hard to reconcile with the model that states that perception and production are linked o Also, the way in which we produce sounds is context dependent o If production is not invariant than how can it account for invariant perception?  Libberman and Mattingly (1985) o Updated theory o Objects of speech are the intended phonetic gestures of the speaker o Phonetic gestures include  Rounding of the lips  Raising the jaw o Argued that acoustic signals are converted to intended phonetic gestures automatically o Hasn’t really been tested  Testable hypotheses of motor theory o Predicts that the perception and processing of phonetics would be linked in the brain o Evidence from people with bran damage and from brain imaging o Infants can distinguish phonemes before they can produce them o Predicts that is perception and production are
More Less

Related notes for PS366

Log In


Join OneClass

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

Sign up

Join to view


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.