Study Guides (248,583)
Canada (121,629)
Psychology (952)
PSYC 2650 (71)
Harvey (2)
Final

Final chapters review.docx

50 Pages
199 Views
Unlock Document

Department
Psychology
Course
PSYC 2650
Professor
Harvey
Semester
Winter

Description
Chapter 9: Language  Chapter 9 (page 323-355) • Virtually every human knows language • To find a human who cannot use language, we must seek out people in extraordinary circumstances, such as those who have suffered serious brain damage or have grown up completely isolated from other humans • No other species has a language comparable to ours in complexity or communicative power • Language us at the heart of, and essential for, a huge range of human activities and achievements The Organization of Language • Language at its heart involves a special type of translation • How does the translation from ideas to sounds and then back to ideas, take place? – Language use relies on consistent and well-defined patterns – patterns in how individual words are used, patterns in how words are put together into phrases – Follow those patterns when I express my ideas, and the same patterns guide you when you’re figuring out what I just said • What are the patterns of language that we all know and use? – Language has a structure – At the highest level of the structure are the ideas intended by the speaker, or the ideas that the listener derives from the input – These ideas are typically expressed in sentences: a sequence of words that conforms to the rules of syntax (and so has the right constituents in the right sequence) – Words are composed of morphemes: the smallest language unit that carries meaning. Psycholinguists distinguish content morphemes (the primary carriers of meaning) from function morphemes (which specify the relations among words) – Some morphemes are units that can stand alone, and they typically refer to particular objects or ideas or actions (ie. “talk” – Other morphemes are “bound” onto these “free” morphemes and add information crucial for interpretation (ie. “ed” or “s”) – In spoken language, morphemes are conveyed by sounds called phonemes: the basic categories of sound used to convey language. For example, the words “peg” and “beg” differ in their initial phoneme – [p] in one case, [b] in the other – Some phonemes are easily represented by letters of the alphabet, but others are not – Within each of these levels, language is also organized in another way; in each level, people can combine and recombine the units to produce novel utterances – assembling the phonemes into brand-new morphemes or assembling words into brand-new phrases – Crucially, though, not all combinations are possible Phonology The Production of Speech • In ordinary breathing, air flows quietly out of the lungs, through the larynx, and up through the nose and the mouth • Noise is produced, however, if this airflow is interrupted or altered, and this allows humans to produce a wide range of different sounds • Within the larynx there are two flaps of muscular tissue called the “vocal folds” (or “vocal cords”) – The vocal folds can be rapidly opened and closed, producing a buzzing sort of vibration known as voicing: one of the properties that distinguishes different categories of speech sounds. Asound is considered “voiced” if the vocal folds are vibrating while the sound is produced. If the vocal folds start vibrating sometime after the sound begins (ie. with a long voice-onset time), the sound is considered “unvoiced” • You can also produce sound by narrowing the air passageway within the mouth itself • The various aspects of speech production provide a basis for categorizing speech sounds • We can distinguish sounds, first, according to how the airflow is restricted; this is referred to as manner of production: the way in which a speaker momentarily obstructs the flow of air out of the lungs to produce a speech sound. For example, the airflow can be fully stopped for a moment, as it is in the [t] or [b] sound; or the air can continue to flow, as it does in the pronunciation of [f] or [v] • Second, we can distinguish between sounds that are voiced – produced with the vocal folds vibrating – and those that are not – The sounds of [v],[z], and [n] (to name a few) are voiced; [f],[s],[t], and [k] are unvoiced – Finally, sounds can be categorized according to where the airflow is restricted; this is referred to as place of articulation: the position at which a speaker momentarily obstructs the flow of air out of the lungs to produce a speech sound. For example, the place of articulation for the [b] sounds is the lips; the place of articulation for the [d] sound is created by the tongue briefly touching the roof of the mouth – Thus, you close your lips to produce “bilabial” sounds like [p] and [b]; you place your top teeth close to your bottom lip to produce “labiodental” sounds like [f] and [v]; and you place your tongue just behind your upper teeth to produce “alveolar” sounds like [t] and [d] • This categorization scheme allows us to describe any speech sound in terms of a few simple features – First, we specify the manner of production; second, voicing; third, place of articulation – These features are all we need to identify the sound and if any of these features chances, so does the sound’s identity – These features, in varying combinations allow us to describe all the sounds our language needs – In English, these features are combined and recombined to produce 40 or so different phonemes – The phonemes are created by simple combinations of the features just described The Complexity of Speech Perception • The features of speech production also correspond to what listeners hear when they’re listening to speech • Phonemes that differ only in one production feature sound similar to each other; phonemes that differ in multiple features sound more distinct – This is reflected in the errors people make when they try to understand speech in a noisy environment: Their misperceptions are usually off by just one feature, so that [p] is confused with [b] (a difference only in voicing), [p] with [t] (a difference only in place of articulation), and so on • This makes it seem like the perception of speech may be a straightforward matter: a small number of features is sufficient to characterize any particular speech sound – All the perceiver needs to do, therefore, is detect these features, and, with this done, the speech sounds are identified • It turns out though, speech perception is more complicated that this – Speech is fast; normal speaking rate is around 180 words per minute – about 15 phonemes per second – but people can follow speech that’s as fast as 250 words per minute – Amplitudes, in the form of air-pressure changes, that reach the ear, and so, in an important sense, the figure shows the pattern of input with which “real” speech perception begins • Within the stream of speech there are no markers to indicate where one phoneme ends and the next begins • For the most part, there are no gaps to indicate the boundaries between successive syllables or successive words – First step, prior to phoneme identification, need to “slice” this stream into appropriate segments – speech segmentation: the process through which a stream of speech is “sliced” into its constituent words and, within words, into the constituent phonemes • Most of us are convinced that there are pauses between words in the speech that we hear, and its these pauses, we assume, that mark the word boundaries; however, this turns out to be an illusion, and we are “hearing” pauses that, in truth, aren’t there – This is evident when we “hear” the pauses in the “wrong places” and thus segment the speech stream in a way that the speaker didn’t intend – Illusion is also revealed when we physically measure the speech stream, or when we listen to speech we cannot understand (foreign); we lack the skill to segment the stream, and are unable to supply the word boundaries and so we hear what is really there: a continuous, uninterrupted flow of sound • Speech perception is further complicated by a phenomenon known as coarticulation: a trait of speech production in which the way a sound is produced is altered slightly by the immediately previous and immediately following sounds. Because of this “overlap” in speech production, the acoustic properties of each speech sound vary according to the context in which that sound appears – In producing speech, you do not utter one phoneme at a time – This overlap allows speech production to be faster and considerably more fluent; but the overlap also has consequences for the sounds produced, and so the [s] before one vowel sounds different from the [s] getting ready for another vowel, so we cant point to a specific acoustical pattern – The acoustical pattern is different in different contexts; speech perception has to read past these context differences to identify the phonemes produced Aids to Speech Perception • The speech your encounter, day by day, is surprisingly limited in its range • We know tens of thousands of words, but most of these words are rarely used • The perception of speech also shares a crucial attribute with all other types of perception: you don’t rely only on the stimuli you receive; instead, you supplement this input with a wealth of other knowledge – On one proposal, the moment you hear the first phoneme in a word, you activate all the words in your vocabulary that have this starting sound; the moment you hear the second phoneme, you narrow this cohort of words so that you’re thinking only about words that start with this pair of phonemes – In this way, speech perception ends up as not just a matters of receiving and identifying sounds, but instead, it is a process in which you actively seek a match between the sounds arriving at your ears and the words actually in your vocabulary • In other cases, speech perception is guided by knowledge of a broader sort, knowledge that relies on the context in which a word appears – This is evident in the phonemic restoration effect: a pattern in which people “hear” phonemes that actually are not presented but that are highly likely in that context. Thus, if one is presented with the word “legislature” but with the [s] sound replaced by a cough, one is likely to hear the [s] sound anyhow – to demonstrate, researchers modify tape-recorded sounds; this now degraded stimulus is then presented to participants, embedded in a sentence – when asked what they had just head, participants reported hearing the complete word, accompanied by a burst of noise – apparently they used the context to figure out what the word must have been, and then supplied the missing sound on their own – the participants didn’t just infer what the missing sound was; they literally seemed to “hear” the sound – the participants cannot tell when the noise occurred within the sentence; cannot tell which sounds within this word were truly part of the stimulus participation and which were missing but supplied by the participants themselves – from these recordings, they spliced out individual words and presented them, now in isolation; with no context to guide them, participants were able to identify only half the words – benefits of context are considerable Categorical Perception • speech perception also benefits from a pattern called categorical perception: the tendency to hear speech sounds “merely” as members of a category – the category of “z” sounds, the category of “p” sounds, and so on. As a consequence, one tends to hear sounds within the category as being rather similar to each other; sounds from different categories, however, are perceived as quite different – meaning you are much better at hearing differences between categories of sounds than you are at hearing the variations within a category of sounds; you’re surprisingly insensitive to differences within each of these categories, so you have a hard time distinguishing say one [p] sound from another – you want this pattern because it allows you to hear the differences that matter without hearing (and being distracted by) inconsequential variations within the category – demonstrations of this generally rely on a series of stimuli, created by computer (ie. a [ba] sound that is distorted to sound like [pa]) – this creates a series of stimuli, each slightly different from the one before, ranging from a clear [ba] sound at one extreme, through a series of “compromise” sounds, until we reach at the other extreme a clear [pa] sound • how do we perceive these sounds? – As we move through the series, we might expect people to be less and less likely to identify each stimulus as a [ba], and correspondingly more and more likely to identify each as a [pa] – This would be a “graded-membership” pattern: as we move away from the this prototype, cases should be harder and harder to categorize – However, the actual data don’t fit with this prediction – Even though the stimuli are gradually changing from one extreme to another, participants “hear” an abrupt shift, so that roughly half stimuli are reliably categorized as [ba], and half are reliably categorized as [pa] – Participants seem indifferent to the differences within each category • It seems that your perceptual apparatus is “tuned” to provide you just the information you need • Perception serves you well by largely ignoring these “subphonemic” variation Combining Phonomes • Each language has a limited stock of phonemes, but they can be combined and recombined to produce thousands of different morphemes, which can themselves be combined to create word after word after word – Language users need to know both how to identify the individual phonemes, and – crucially – how to put them together into larger packages • There are rules governing these combinations of phonemes, and users of language reliably respect these rules – Ie. in English, certain sounds can occur at the end of words but not at the beginning such as “ing” • There are also rules that govern the adjustments that must occur when certain phonemes are uttered one after another – Ie. the [s] sound pronunciation or the [z] sound depending on how the base noun ends – If it ends with a voiced sound, the [z] ending is used to make the plural; if the base noun ends with an unvoiced sound, the plural is created with an [s] Morphemes and Words • Avg high school grad knows 45,000 different words; college grad: 75,000 – 100,000 • For each word, the speaker knows several bits of information – First, the speaker knows the word’s sound – the sequence of phonemes that make up the word – Second, in a literate culture, the speaker generally knows the words orthography – the sequence of letters that spell the word – Third, the speaker also knows how to use the word within various phrases, governed by the rules of syntax – Finally, the speaker needs to know the meaning of a word; he must have a semantic representation for the word to go with the phonological representation, basically connecting the meaning to the sound Word Meaning • What a word refers to is called the word’s referent: the actual object, action, or event in the world that a word or phrase refers to • Within this context, one might propose that the meaning of a word or phrase is linked to the word’s (or phrase’s) referent • There are key differences between a word’s reference and its meaning – Some phrases have no referent because they refer to things that don’t exist, but even so, the phrases seem meaningful – Sometimes a word’s reference is temporary or a matter of coincidence (ie. referent of president of US changes at regular intervals, but meaning of phrase seems more stable) – Word meanings must involve more than reference • Many words do express single concepts, and more generally, you can understand a word’s meaning only if you understand the relevant concepts – Large part of knowing a word is knowing the relevant concept – Conceptual knowledge turns out to be complicated, even for simple concepts like “bird” – Same complications apply to semantic knowledge Building New Words • Size of someone’s vocabulary is quite fluid – One reason is that new words are created all the time; this happens, for example, whenever a new style of music or clothing demands correspondingly, a new vocab – These new words don’t arrive in the language as isolated entries, however, because language users immediately know how to create variations on each word by adding the appropriate morphemes – The added morphemes allow you to use these words in new ways – indeed, the morphemes allow you to create entirely new words • This all highlights the generativity of language; generativity: the idea that you can combine and recombine basic units to create (or “generate”) new and more-complex entities. Linguistic rules, for example, are generative, because they govern how a limited number of words can be combined and recombined to produce a vast number of sentences – Thus, someone who knows English (or any language) has not just memorized the vocab of the language and some set of phrases; instead, they know how to create new forms from within the language: knows how to combine morphemes to create new words, how to adjust phonemes when they’re put together into novel combinations, etc – This knowledge isn’t conscious; nonetheless, speakers honor these principles with remarkable consistency in their day-to-day use of the language and in their day- to-day creation of novel words Syntax • The generativity of language is even more salient when we consider the upper levels of phrases and sentences • You can combine production features to create a few dozen phonemes, and you can combine production features to create a few dozen phonemes, and you can combine these to produce thousands of morphemes and words • Sentences can range from brief to absurdly long; most sentences contain 20 words or fewer • With this length limit, it has been estimated that there are 100,000,000,000,000,000,000 possible sentences in English • Once again, there are limits on which combinations are acceptable and which are not – Speakers respect the rules of syntax: rules governing the sequences and combinations of words in the formation of phrases and sentences – One might think the rules of syntax depend on meaning, so that meaningful sentences are accepted as “sentences” while meaningless sequences are rejected as nonsentences – This suggestions is wrong, as one concern, many nonsentences do seem meaningful – It seems that we need principles of syntax that are separate from considerations of semantics or sensibility Phrase Structure • What is syntax? – Answer is complicated, but one part seems to involve phrase structure rules: a constraint that governs the pattern of branching in a phrase structure. Equivalently, phrase structure rules govern what the constituents must be for any syntactic element of sentence – Also specify the overall organization of the sentence – One phrase structure rule, for example, stipulates that a sentence (S) must consist of a noun phrase (NP): one of the constituents of a phrase structure that defines a sentence and a verb phrase (VP): one of the constituents of a phrase structure that defines a sentence – Adifferent rule stipulates that noun phrases can include a “determiner,” some number of adjectives, and then the noun itself – Verb phrases can take several different forms but often consist of a verb followed by a noun phrase – One way to depict these rules is with a tree structure: a style of depiction often used to indicate hierarchical relationships, such as the relationships (specified by phrase structure rules) among the words in a phrase or sentence – you can read the structure from top to bottom, and as you move from one level to the next, you can see that each element has been “expanded” in a fashion that’s strictly governed by the phrase structure rules Prescriptive Rules, Descriptive Rules • injunctions are the result of prescriptive rules: rules describing how things are supposed to be instead of how they are. Often called normative rules and contrasted with descriptive rules – describing how language is “supposed to be”; language that doesn’t follow these rules, its claimed as “improper” or maybe just “wrong” • we should be skeptical about these prescriptive rules • language changes with the passage of time, and what is right in one period is often different from what seems right at other times; this pattern of change makes it difficult to justify prescriptive rules • the selection of prescriptive rules; therefore, may simply reflect the preferences of a particular group – and, in most settings, the group that defines these rules will of course be the group with the most prestige or social cachet • thus, people will often strive to follow these rules with the simple aim of joining these elite groups • phrase structure rules, in contrast, are not at all prescriptive; they are instead descriptive rules: rules that simply describe the regularities in a pattern of observations, with no commentary on whether the pattern is “proper”, or “correct” or “desirable” The Function of Phrase Structure • in some fashion, we all have internalized these rules; this is evident in the fact that many aspects of language use are reliably in line with the rules • the groupings provided by the phrase-structure therefore, organize a sentence, and this shapes our intuitions about a sentence’s parts • the organization can also influence memory • once organized into a phrase structure, these sequences were much easier to recall • phrase structure rules help us understand the sentences we hear or read, because syntax in general specifies the relationships among words in each sentence • this role of phrase structure, in guiding understanding, can be confirmed in a fashion that informative and often funny: sometimes, two different phrase structures can lead to the same sequence of words, and if you can encounter those words you may not know which phrase structure was intended • with multiple phrase-structures available, there should be more than one way to interpret the sentence Linguistic Universals • small number of elements can be used to create a vast number of combinations; the combinations at each level seem to be rule-governed – the rules determine which units can be combined, and in what order; they also specify a structure within the larger units – remarkably the rules are quite similar as we move from one language to the next • there are a number of regularities across language – some experts have argued that these regularities constitute linguistic universals: a rule/rules that appear to apply to every human language – these are of various sorts; some identify the inventory of linguistic constituents, and so all language, ie, include both nouns and verbs; all language include pronouns – some universals are framed in terms of probabilities – some sequences of words are common, while other sequences are quite rare; subject of a sentence tends to precede the object in roughly 98% of the world’s languages; sequence of subject before verb is preferred in 80% – often universals concern linguistic features that seem to come and go together – existence of these linguistic universals opens a possibility  how is all this complexity mastered so quickly by children? • Some researchers suggest that language learning occurs so rapidly because each child begins the process with an enormous head start: a biological heritage that somehow stipulates the broad outline of human language – The child begins language learning already knowing the universal rules; the task is for the child to figure out exactly how the rules are realized within the language community in which they are raised Sentence Parsing • how do we figure out the phrase structure in the first place? – Sentences are not uniform in their structure and are more variable, which makes the identification of a sentence’s phrase structure appreciably more difficult • how do you parse a sentence – figure out each word’s syntactic role? – Parsing: the process through which an input is divided into its appropriate elements – for example, dividing the stream of incoming speech into its constituent words – or in which a sequence of words is divided into constituent phrases – One possibility is that you wait until the sentence’s end and then go to work on figuring out the structure – With this strategy, comprehension might be slowed a little (due to waiting for the sentence’s termination), but you would avoid errors, because your interpretation could be guided by full information about the sentence’s content – People don’t use this strategy in perceiving words; instead they begin the identification process as soon as they hear the word’s first phoneme – Evidence suggests the same pattern in interpreting sentences; people seek to parse sentences as they hear them, trying to figure out the role of each word the moment it arrives – This is more efficient, but can lead to error Garden Paths • Even simple sentences can be ambiguous • Temporary ambiguity is also common inside a sentence; the early part of a sentence is often open to multiple interpretations, but then the later part of the sentence clears things up • Garden-path sentences: a sentence that initially leads the reader to one understanding of how the sentence’s words are related, but then requires a change in this understanding in order to comprehend the sentence. Example, “The old man ships” – Highlight the risk attached to the strategy of interpreting a sentence as it arrives: the information you need in order to understand these sentences arrives only late in the sequence, and so, to avoid an interpretative dead end, you’d be advised to remain neutral about the sentence’s meaning until you have gathered enough information – This is not what you do, instead you commit yourself fairly early on to one interpretation and then try to “fit” subsequent words as they arrive, into that interpretation – This strategy is often effective, but it does lead to the “double-take” reaction when late-arriving information forces you to abandon your interpretive efforts so far Syntax as a Guide to Parsing • Why do we initially choose one interpretation of a sentence, one parsing, rather than another? – Many cues are relevant, and this tells us that many types of information influence parsing – Parsing is guided by an assumption of so-called minimal attachment: a heuristic used in sentence parsing. The listener or reader proceeds through the sentence seeking the simplest possible phrase structure that will accommodate the words heard so far – People tend to assume that they’ll be hearing (or reading) active sentences rather than passive, so they generally interpret a sentence’s initial noun phrase as the “doer” of the action, and not the recipient – Most of the sentences we encounter are active, not passive, so this assumption is usually correct; however, this assumption works against us whenever we do encounter a passive sentence, and that’s why active sentences are usually easier to understand than passive sentences • Parsing is also influenced by the function words that appear in a sentence and by the various morphemes that signal syntactic role • Garden-path sentences can sometimes be enormously difficult to comprehend Background Knowledge as a Guide to Parsing • Parsing is also guided by background knowledge – People parse sentences in a way that makes sense to them – Your background knowledge tells you about aspects of the sentence • How can we document these knowledge effects? – Several studies have tracked how people move their eyes while reading, and these movements reveal the confusion readers go through in reading garden-path sentences – The moment readers realize they have misinterpreted the sentence so far, they backtrack and reread the sentence’s start, and with appropriate instruments, we can easily detect these backwards eye movements – Using this technique, investigators have examined the effects of plausibility on readers’expectations for the words they’re seeing – People are also sensitive to statistical properties in the language, and so, if a word has several meanings, they assume its most frequent meaning whenever they encounter the word – They tend to assume that “train” means the thing on tracks rather than the activity to teach – Also tend to assume that adjectives will be followed by nouns; this isn’t an obligatory pattern but it is certainly a frequent one, so the assumption seems safe – These assumptions prime us to read sentences a certain way – We rely on frequent meanings The Extralinguistic Context • Make use of another factor in parsing: the context in which you encounter sentences, including the conversational context • As important is the extralinguistic context: the social and physical setting in which an utterance is encountered; usually, cues within this setting guide the interpretation of the utterance The Use of Language: What is Left Unsaid • What does it mean to “know a language”? – Each language user seems somehow to know (and obey) a rich set of rules – with these rules determining which sound combinations and which sequences of words seem acceptable, and which do not • Your understanding of the sentences you hear is guided by the principles of  syntax, semantics, statistical, other pragmatic • These points still understate the complexity of language use and with that, the complexity of the knowledge someone must have in order to “know a language” • Another information useful in parsing: the rise and fall of speech intonation and the pattern of pauses • These rhythm and pitch cues, together called prosody: the pattern of pauses and pitch changes that characterize speech production. Prosody can be used (among other functions) to emphasize elements of a spoken sentence, to highlight the sentence’s intended structure, or to signal the difference between a question and an assertion – These play an important role in speech perception – Prosody can reveal the mood of the speaker, also direct the listener’s attention by, specifying the focus or theme of a sentence – Prosody can also render unambiguous a sentence that would otherwise be entirely confusing • pragmatics: a term referring to knowledge of how language is ordinarily used, knowledge (for example) that tells most English speakers that “Can you pass me the salt?” is actually a request for the salt, not an inquiry about someone’s arm strength • these topics – prosody, production, and pragmatics – are central concerns within the study of language • key emphasis of the chapter: each of us uses language all the time, it requires little effort, but language is a complicated tool, and we are skilled in its use The Biological Roots of Language • many people believe that humans are equipped with sophisticated neural machinery specialized for learning, and then using, language Aphasias • brain damage can cause a variety of effect depending on where and how widespread the damage is • for a number of brain sites, damage causes the disruption of language, known as aphasia: a disruption to language capacities – they take many forms and are often quite specialized, with the particular symptoms largely dependent on the locus of the brain damage • 2 broad classes of aphasia: – damage to the left frontal lobe of the brain, and especially the Broca’s area: (an area in the left frontal lobe of the brain; damage here typically causes nonfluent aphasia.), usually produces a pattern of symptoms known as nonfluent aphasia: a disruption of language, caused by brain damage, in which someone loses the ability to speak or write with any fluency. – In severe cases, the patient is unable to write or utter a word; less severe cases – only part of the patient’s vocab is lost, but the speech becomes labored and fragmented, and articulating each word requires special effort – Different symptoms comes from damage to the Wernicke’s area: an area in the left frontal love of the brain; damage here typically causes fluent aphasia – Fluent aphasia: a disruption of language, caused by brain damage, in which afflicted individuals are able to produce speech but the speech is not meaningful, and the individuals are not able to understand what is said to them. – This broad distinction between fluent and nonfluent aphasia captures the data only in the broadest sense – One reason lies in the fact that language use involves the coordination of many different steps, many different processes – Loss due to aphasia could be quite specific depending on where the damage is • Brain lesions – produced by injury or stroke – are often large enough to disrupt multiple brain areas, and so can produce a mixture of symptoms – This makes aphasias often difficult to classify precisely, and its sometimes hard to see a correspondence between symptoms and the site of the brain damage • Humans do have a considerable amount of neural tissue that is specialized for language • Damage to this tissue can disrupt language understanding, production, or both • Our skill in using language rests in part on the fact that we have a lot of neural apparatus devoted to precisely this task The Biology of Language Learning • Biological roots of language show up in the fact that language is learned • This learning is remarkably fast • This occurs in a wide range of environments • Children learn language even if their communication with adults is entirely nonlinguistic • Sophisticated learning capacities that all humans share, capacities that contribute to many aspects of the young child’s development, and not just language • Other psychologists offer a different claim: that the human brain contains several brain mechanisms specifically evolved for language learning  language learning is wired into our brains from the start • What might these mechanisms be? – Specific language impairment (SLI): a syndrome in which individuals seem to have normal intelligence but problems in learning the rules of language The Processes of Language Learning  • Learning does play a crucial role in the acquisition of language • Language learning depends on the child’s picking up information from her  environment  • Overregularization errors: an error in which someone perceived or  remembers a word of event as being closer to the “norm” than it really is.  For example, misspelled words are read as though they were spelled  correctly; atypical events are misremembered in a fashion that brings them  closer to more­typical events; words with an irregular past tense (such as  “ran”) are replaced with a regular past tense (“runned”) ­ these errors are interesting… notice that they rule out any direct contribution  from imitation, for the simple reason that adults almost never produce these  over regularizations. The children therefore, are producing forms they’ve  “invented” for themselves, and not a form they’re imitating from others.  • Studies of child­adult interaction make it plain that adults rarely correct their  children’s grammar, nor do they reward children for speaking “correctly”  • What do the learning mechanisms involve?  ­ children are exquisitely sensitive to patterns and regularities in what they hear ­ language has many elements (syntax, semantics, phonology, prosody, etc.) and  these elements interact in ordinary language use  ­ semantic bootstrapping: an important process in language learning in  which someone (usually a child) uses knowledge of semantic relationships  as a basis for figuring out the syntax of the language.  Language and Thought  • language can influence how you reason and how you make decisions  Linguistic Relativity  • linguistic relativity: the proposal that the language that we speak shapes our  thought, because the structure and vocabulary of our language create certain  ways of thinking about the world  Chapter 11 Judgement: Attribute Substitution • Frequency estimates are often crucial for our judgements  o Assessments of how often various events have happened in the past  influence judgement  • People likely rely on attribute substitution, a strategy of using easily available  information that is a plausible substitute for the information one seeks o Specifically, you are likely to do a quick scan through your memory to  look for relevant cases, allowing you to make a judgement  o If you can easily think of four friends who did well in Organic chemistry  you can conclude that this is a frequent occurrence and that this course  isn’t challenging   With this strategy you are basing your judgement on availability  • That is, how easily and quickly you can come up with  relevant examples  • Called the availability heuristic strategy  o This is still a case of attribute substitution  • Others use probability to make judgements or even resemblance  o i.e. in a job interview, the employer might look to see if you resemble past  successful employees  o This substitution is referred to as the representativeness heuristic  • Therefore the two heuristics in judgement are availability and representativeness  The Availability Heuristic     • Heuristics are efficient strategies that lead you to the right answers (usually)  o Efficiency is gained but there is room for errors  The above two heuristics are easy to access (efficient)   Increased availability acts as an index for frequency   Also, many categories are homogenous enough so that members of  the category do resemble each other • This is why you can often rely on resemblance as a way of  judging probability of category membership  • For example, people are asked if there are more words that start with R or more  words with the letter R in the third position of words o Why do people get this wrong?  The answer lies in availability  • There are more words in your memory that start with R  then with the R in the third position that are available  • Thus, the organization of memory creates a bias in what is  available which leads to an error in frequency judgement  The Wide Range of Availability Effects: • People often overestimate the frequency of events that are quite rare  o i.e. thinking you may win the lottery  • What causes this pattern? o There is little reason to spend time thinking about familiar events but you  are likely to notice and think about rare events   Especially rare emotional events   This will make these memories more available to you  o As a consequence, if you rely on the availability heuristic you will  overestimate the likelihood of similar events happening in the future  • Study: participants were asked to think about episodes in their lives in which they  had acted in an assertive fashion o Half of the participants were asked to recall six of these episodes o Half were asked to recall 12  o Then all  were asked some general questions, including how assertive  overall they thought they were  • Results: participants had an easy time coming up with 6 episodes, and using the  availability heuristic, concluded that they must be assertive  o Participants asked to think of 12 episodes had some difficulty and thus  concluded that they are not assertive  • What seems to matter for judgements of assertiveness, then, is the ease of coming  up with the episodes, not the amount  The Representativeness Heuristic: • How does this heuristic work? o Many of the categories you encounter are relatively homogenous, and this  is what this heuristic capitalizes on  o Therefore, judgements of category membership can be made by examining  resemblance to a category  Reasoning From a Single Case to the Entire Population: • In the gamblers fallacy (seeing heads nine times in a row makes a person believe  that tails will come up next) people seem to believe that each subset of a category  should have the same properties of the category overall o This is a consequence of people assuming that categories are homogenous  o Study: participant were shown a videotaped interview in which a prison  guard discussed his job  In one condition, the guard was compassionate and kind  In another, the guard expressed contempt for the prison inmates  and scoffed at the idea of rehab   Before seeing either video some participants were told that this  guard was typical and others were told he was atypical   Participants were later questioned about their own views of the  criminal justice system o Results: those who had seen the humane guard believed that prison guards  were decent people  Those that saw the inhumane guard had a more negative view  towards prison guards   Both groups ignored whether the guard was typical or not though  o These participants made conclusions about an entire category  Detecting Covariation: • Errors caused by heuristics can trigger other sorts of error in judgements about  covariation o i.e. X and Y covary if X tends to be on the scene whenever Y is and if X  tends to be absent when Y is   Exercise and stamina covary  • Covariation is important in cases when you are checking a belief about cause and  effect  Illusions of Covariation: • Do specific responses on the Rorschach test covary with certain personality traits? o A group of researchers created a fictitious transcript of peoples responses  to the test  o They also made up fictitious descriptions of the people who offered these  responses  o They then randomly paired the transcripts and personality descriptions and  were shown to students   They were asked to see if they covaried  o The students did see a pattern in the data   But this pattern was an illusion  What Causes Illusions in Covariation? • One proposal focuses on the evidence that people seem to consider only a subset  of evidence when judging covariation  o A biased input would lead to a biased output  o Specifically, when judging covariation your selection evidence is likely to  be guided by confirmation bias   A tendency to be more responsive to evidence that confirms your  beliefs rather than evidence that will challenge them  Base Rates: • Assessment of covariation can also be pulled off track by another problem,  neglect of base­rate information o Information about how frequently something occurs in general  o i.e. if you are testing a drug for HIV and find that 70% of patients taking  the drug recover it is important to know how many people recover without  taking the drug  • Study: participants are asked “If someone is chosen at random from a group of 70  lawyers and 30 engineers, what is his profession likely to be?” o Most say being a lawyer is .70, here they are using their base­rate  information properly o Other participants were given a similar task, but instead of base­rates they  were given brief descriptions of certain individuals and asked whether  each was likely to be a lawyer or engineer   Some were crafted to suggest one or the other  o Participants responses were guided by these suggestions  o But what happens if we provide both sorts of information?  Here, participants rely only on the descriptive information about  the person  • Reversal of base­rates for lawyers and engineers has no  effect on judgements (this confirms that they are ignoring  the base rates) • So what produces this neglect of base­rates? o Part of the answer is attribute substitution   People seem to turn the above question about category  membership into a question about resemblance (they rely on the  representative heuristic) • Thus they ask themselves how much this one person  resembles a lawyer  Dual­Process Models: More Sophisticated Judgement Strategies • One advantage that we have is we know it is wrong to draw conclusions from  small sample sizes  o We know that a pattern in a larger set is less likely to be the result of an  accident or luck  System 1, System 2: • There seems to be a mixed pattern of evidence here, so how can we make sense of  it? o The suggestion is that people must have two ways of thinking  One type of thinking is fast and easy (heuristics)  One type is slower, more effortful and more accurate • This is why people don’t always make heuristic­based  errors  • What is proposed is a dual­process model  o System 1 will be the label for the fast way of thinking o System 2 will be the label for the slower way of thinking  • When do people use system 1 or 2? o One idea is that people choose when to rely on each system   Thus, people switch to system 2 when making a meaningful  judgement   But we see that people rely on system 1 heuristics even when an  incentive (money) is offered to choose system 2  o Evidence suggests that system 2 comes into play only if triggered by  certain cues and only if the circumstances are right  The Importance of Data Format: • What factors other than time pressure and focus govern the use of both systems? o Some evidence suggests that features of the judgement being made can  either lead to activation of system 1 or 2   i.e. maybe the base­rate neglect example is the result of system 1  thinking   it turns out that people are more likely to pay attention to base rates  if they are represented in frequency format and not probability or  percentage  • Therefore some presentations are more “user friendly” than  others  Codable Data: • The use of system 2 is more likely if the role of chance is more conspicuous in a  problem  • If this role is salient, people are more likely to realize that the “evidence” they are  considering may be a fluke with no pattern  o With this, people are more likely to pay attention to the quantity of  evidence, with the idea that larger sets of observations are less vulnerable  to chance fluctuations  o Study: participants were asked about someone who assessed a restaurant  based on just one meal  The participants were more alert to considerations of sample size if  the diner chose his meal by blindly dropping a pencil on the menu  o Study: participants were asked about a high school student who chose  what college to go to based on her campus visit   They too were more alert to issues of sampling  • People are also more accurate in their judgements and less prone to heuristic use  when confronting evidence that is easily understood in statistical terms  o i.e. sample size  • When an employer interviews a candidate, evidence as to his job suitability is not  easily coded as, say, determining how good an athlete is by looking at their career  stats o Thus this person is not likely to use his system 2 thinking  o Instead the employer is likely to rely on simple heuristics  Background Knowledge and Skills: • It’s clear, then, that the use of system 2 thinking depends on both factors in the  situation (time pressure) and in the evidence considered (where the evidence is  presented in terms of frequencies)  • The use of system 2 thinking also depends on the knowledge and skills each  person has  o i.e. people are more sensitive to base­rates if their background knowledge  leads them to see a meaningful linkage between the base rate and the  dimension being judged  o Study: participants were asked to predict whether a specific student would  pass an upcoming exam  The participants were told some facts about the  student but they  were also told the base­rate (only 30% of students who took the  exam passed)  In this situation the participants perceived the base­rate as  meaningful o Results: base­rate neglect was not seen as it was meaningful to the  judgement  • You are also more likely to use system 2 if you are educated in the right way o Study: participants were provided with 30 min of training, focusing on the  importance of sample size   They were reminded that accidents do happen but they do not keep  happening over and over again  o Results: participants were more likely to incorporate this new knowledge  to novel cases  Confirmation and Disconfirmation: • So far we have been looking at judgements that fall within the domain of  induction o The process through which you make predictions about new cases, based  on the cases you have observed so far  • Deduction is just as important though  o Cases in which you start with claim
More Less

Related notes for PSYC 2650

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