Psychology 258 Midterm 2 PDF

PSYCHOLOGY 258 MIDTERM 2 Long-Term Memory Learning Outcomes 1. What is the serial position effect, and what is the evidence for it and against it? 2. What is the duration, capacity, and coding of the long-term store (LTS)? 3. Explain the differences between implicit and explicit memories. 4. H...

PSYCHOLOGY 258 MIDTERM 2 Long-Term Memory Learning Outcomes 1. What is the serial position effect, and what is the evidence for it and against it? 2. What is the duration, capacity, and coding of the long-term store (LTS)? 3. Explain the differences between implicit and explicit memories. 4. How does Tulving categorize different kinds of memory? What does amnesia reveal about memory? 5. What is levels of processing theory, and what is the evidence for it and against it? 6. Describe how the encoding effects of self-reference and generation influence memory. 7. What is the testing effect and how does it work? 8. How do external context and internal states affect memory? 9. Describe how repetition, spacing of practice, interleaved practice, and sleep affect consolidation of memories. 10. How are memories stored in the brain? 11. Contrast the standard model of systems consolidation with the multiple trace theory of consolidation. The Serial Position Effect Glanzer & Cunitz (1966): gave participants lists of words to remember results:  primacy effect: better recall for items at beginning of list; greater rehearsal causes transfer to LTS recency effect: better recall of items at end of list, which were still in STS; mental math (recall delay) affected recency only Problems: similarity of STS & LTS - coding: both use acoustic, visual, and semantic - forgetting: due to interference argument from parsimony - memory varies along a continuum (short--long) - why have different memory systems? - simpler to assume one memory system Craik & Watkins (1973): STS to LTS transfer - participants heard a list of items; had to remember last word starting with “g” e.g., daughter, oil, rifle, garden, grain, table, football, anchor, giraffe - words differed in opportunities for rehearsal e.g., giraffe value = 0; grain value = 4; garden value = 1 - participants did math task, then free recall of all words - memory for “g” words: - no advantage for greater rehearsal Bekerian & Baddeley (1980): role of repetition - BBC advertised new radio frequencies 10×/hour - casual listener exposed to message 1,000× - most listeners didn’t remember the message - repetition ≠ retention Long-Term Store LTS Duration (see next topic) LTS Capacity Von Neumann (1958): - calculation based on neural firing rate, estimated number of neurons, and a lifetime of 60 years - 2.8 × 1020 bits, or about 1 exabyte (1 billion gigabytes) Landauer (1986): - estimated “functional information content”: amount of information adults need to do normal tasks - compared rate of information processing vs. forgetting on a number of tasks e.g., reading, pictures, nonsense syllables - people retain 2 bits/sec - conclusion: adult LTS contains 109 bits, or 125 megabytes - not the same as storage capacity  Reber (2010): - the brain has ~100 billion neurons--but not all are involved in forming long-term memories - ~1 billion neurons likely participate in memory × ~1,000 synapses each = 1012 (1 trillion) total synapses - but each synapse does not store one memory - rather, memories are distributed across our brain’s neural networks, exponentially increasing capacity - memory capacity estimated to be 2.5 petabytes (2.5 million gigabytes) bottom line: LTS capacity unknown; practically unlimited LTS Coding Shulman (1972): - found that falsely recalled words had a similar meaning e.g., “baby” for “infant”, or “ship” for “boat” - depended on position of list item: semantic confusions at start (words in LTM), acoustic confusions at end (words in STM) - words that were held in LTS were coded by meaning Types of Memories explicit memory (a.k.a. declarative memory): conscious, intentional remembering of knowledge or an event - recall: reproducing previously encountered information from memory - recognition: identifying previously learned information implicit memory (a.k.a. nondeclarative memory): unconscious retention due to previous experience - repetition priming effect: previous experience with a stimulus facilitates later response to the same (or similar) stimulus e.g., word-fragment completion: fill in missing letters from a word: e.g., word-stem completion: given the first few letters in a word, complete the word as fast as possible: “per...” Theory of Memory Systems Endel Tulving (1972, 1985): episodic memory - memory for events that occurred in your life - tied to specific learning episode or experience semantic memory - stores words, concepts, rules, abstract ideas - general knowledge not tied to any experience procedural memory - underlies motor & cognitive skills - e.g., doing math, playing chess, riding a bike Evidence: - certain patients with amnesia may have damaged one kind of memory system but have other ones intact (this is a dissociation effect: one specific ability is affected, but another is not) retrograde amnesia: inability to remember events that occurred before a traumatic event anterograde amnesia: inability to form new memories of events that occurred after a traumatic event The case of patient H.M. (Scoville & Milner, 1957; Corkin, 2013): - had progressive uncontrolled epilepsy since age 10 - to control his seizures, had bilateral medial temporal lobe (MTL) resection at age 27 (in 1953) - hippocampus, parahippocampal gyrus, and amygdala removed - surgery was successful (fewer seizures), but resulted in profound amnesia - normal IQ, STS (e.g., digit span), language (speech, writing, and reading) - could remember events/facts in his distant past, but had temporal graded retrograde amnesia for events 11 years before the surgery - could not learn new facts or remember information about events since his surgery: could not find new home, even after 10 months language frozen in the 19502 forgot who he was talking to if he turned away - however, he could form new procedural memories (e.g., mirror tracing task), but had no conscious recollection of previous training episodes - see also Memory’s Ghost: The Strange Tale of Mr. M. and the Nature of Memory (Hilts, 1995) and Permanent Present Tense: The Unforgettable Life of the Amnesic Patient, H.M.(Corkin, 2013) The case of patient K.C. (Moscovitch, Schacter, Tulving & colleagues, 2005): - suffered a closed-head injury in a motorcycle collision in Toronto at age 30 - severe damage to MTLs and almost complete bilateral hippocampal loss - left-hemisphere lesions to posterior occipital-temporal and anterior frontal-parietal cortices (not going to ask us about the parts of the brain for a test) - personality changed from outgoing to more tranquil - had retrograde amnesia for episodic memory, although semantic knowledge was intact - also had some anterograde amnesia: no episodic memories formed after the accident (episodic amnesia) - but could form new semantic and procedural memories: played chess but did not remember playing a game knew where family cottage was but did not remember ever going there learned the Dewey decimal system for his job at the library, but did not know when he learned it The case of patient D… Not an accurate example of “short term memory loss”. Conclusions: unimpaired STS implies it is biologically different from LTS hippocampus (and MTL?) not a storage site, but important for consolidation of explicit memories (STS to LTS transfer) H.M. provides evidence that implicit memory is dissociated from explicit memory K.C. provides evidence that, within explicit memory, semantic memory is dissociated from episodic memory Encoding Levels of Processing Theory (Fergus I.M. Craik & Lockhart, 1972): Encoding processes: shallow (e.g., structural: based on appearance of a word) intermediate (e.g., phonemic: based on sound of a word) deep (e.g., semantic: based on meaning of a word) “Deeper” encoding enhances memory. Why? maintenance rehearsal: repetitive processing e.g., repeating a phone number over and over elaborative rehearsal: more meaningful processing, relating item to other knowledge e.g., grouping digits in a phone number into meaningful patterns - may work by organizing, connecting, or integrating memories (Bellezza et al., 1977; Mandler, 1979) Evidence: Craik & Tulving (1975): - participants given a list of words; asked to do one of three tasks: case: Is the word in capital letters? rhyme: Does the word rhyme with weight? sentence: Does the word fit in this sentence: “He met a _____ in the street”? - measured recognition performance and latency on incidental (surprise) memory test - is performance due to processing time? structural task: Does word match this pattern of consonants and vowels? e.g. CCVVC semantic task: Does the word fit in this sentence? e.g., “The man threw a ball to the ______.” - results:  Task Time Recognition Structural 1.70 s 57% Semantic 0.83 s 82% - depth of processing more important than processing time Problems: What does “deeper” mean? - circular definition: deeper better memory Morris, Bransford, & Franks (1977): marching encoding and retrieval - encoding task: given the word table phonetic/rhyming: Does it rhyme with cable? semantic: Does it fit in this sentence? “Put the dish on the _____.” - incidental recognition tests: phonetic/rhyming: Circle what rhymes with word presented earlier: Fable Desk Window semantic: Circle word presented earlier: Table Desk Window - results: - memory performance not dependent on depth alone - also depends on match between encoding processes and type of test Kapur, Craik, Tulving, & colleagues (1994): - shallow task: Does the word contain the letter a ? - deep task: Does word represent a living thing? - better recognition for deep task - greater activation in left inferior prefrontal cortex during a deep task - shows more activity, not deeper activity Other Encoding Factors self-reference effect (Rogers, Kuiper, & Kirker, 1977): memory is enhanced by relating items to one’s self - presented 40 adjectives to participants e.g., kind, rigid, brave, etc. - words evaluated one of four ways: structural, phonemic, or semantic tasks  self-reference group asked if each adjective described them - results of incidental recall test: - self-reference group surpassed semantic encoding group Generation effect (Mäntylä, 1986): memory is enhanced for items that are actively produced - presented 600 nouns (e.g., banana, freedom, tree) - task: generate 3 cue words associated with each noun (e.g., yellow, bunches, edible for banana) Gp1: generated cue words Gp2: saw the nouns, and the 3 cue words generated by other participants Gp3: just saw the 3 cue words generated by other participants - results of incidental recall test of nouns: - self-generated cues lead to better remembering - cues were most helpful when they were both compatible with the target word and distinctive e.g., given the noun coat, the cue jacket is compatible and distinctive, but wool is not (it may cue other nouns like fabric and sheep) Retrieval Testing effect (Roediger & Karpicke, 2006): - stimuli: two prose passages from TOEFL prep book (~260 words each) - learning conditions: study, study: 7 minutes of studying, retention interval, another 7 minutes of studying study, test: 7 minutes of studying, retention interval, 7 minutes of testing - retention interval: 5 minutes, 2 days, or 1 week - results: - conclusion: testing (via retrieval practice) improves long-term retention How does the testing effect work? additional (focused) presentation of material retrieval itself is a memory modifier, with increasing retrieval demand/effort enhancing later retention processes applied during an initial test are also used during the final test, resulting in positive transfer (i.e., a practice effect) Matching Encoding & Retrieval Encoding specificity principle: if conditions at retrieval match those at encoding, memory is enhanced (Tulving & Thomson, 1973) External Cues context-dependent memory: memory enhancement that occurs when retrieval context matches encoding context Godden & Baddeley (1975): - scuba divers given a list of 36 unrelated words via diver underwater communication sets: a) underwater, or b) on land - recalled words in same (matched) or different (mismatched) context - results: different contexts: 24% correct same context: 35% correct - evidence for context-dependent memory - strong implications for scuba divers & remembering! Context effects obtained for:  music: Mozart vs. jazz vs. quiet (Smith, 1985) background colours (Dulsky, 1935) smell of chocolate (Schab, 1990), but not unpleasant odours (Rotton, 1983) Bjork & colleagues (1978): - participants given a list of 40 four-letter nouns, studied in two sessions - sessions were held in the same room twice, or in two different rooms: context 1: small, cluttered, windowless room context 2: room in a modern building with windows overlooking a courtyard - students were tested in a third, “neutral” room - results: students who studied in the same room twice: 40% correct students who studied in two different rooms: 61% correct - implication: studying in a variety of contexts leads to better remembering, greater stability and “generalization” of knowledge Internal States state-dependent memory: if internal state during retrieval matches that during encoding, memory may be enhanced e.g., mood or emotional state, physiological arousal, sober/drunk states (Eich, 1980) mood dependence: when mood at retrieval matches the mood at encoding, memory is facilitated--however, the memories themselves are typically not emotional in nature (Lewis & Critchley, 2003) mood congruence: mood at retrieval influences the kind of memories retrieved e.g., in a sad mood, you’re more likely to remember sad events - implicated in vicious cycle of depression (Baddeley, 1993) Transfer of Information Consolidation: process of integrating new information into stored knowledge Practice Effects savings curve (Ebbinghaus, 1885): - more initial practice less time to relearn list of nonsense syllables the next day (savings) spacing effect (Bahrick, 1979): - participants learned English translations for 50 Spanish words by completing six training sessions - kinds of training: ▸ massed practice: information studied repeatedly at one time ▸ distributed practice (or spaced repetition): information studied repeatedly in separate sessions over a longer period of time - independent variable: spacing gap ▸ 0 days (massed practice) ▸ 1 day (distributed practice) ▸ 30 days (distributed practice) - participants were given a final test 1 month after the last training session - results: ▸ greater spacing between training sessions produced better remembering ▸ follow-up study 8 years later showed 30-day spacing group outperformed the other two (Bahrick & Phelps, 1987) - spacing effect: distributed practice produces better remembering than massed practice - why? ▸ multiple encoding contexts theory: multiple study sessions lead to multiple types of encoding, thus greater chance of matching during test conditions ▸ REM theory: the more rapid eye movement sessions following study sessions, the more consolidation that occurs Kornell & Bjork (2008): interleaved practice - goal: identify painting styles of different artists by inductive learning (observing exemplars) - stimuli were multiple landscape paintings by relatively unfamiliar artists, including Henri- Edmond Cross, Marilyn Mylrea, and YeiMei - independent variable was the nature of practice ▸ blocked practice: participants studied artists one at a time (e.g., 6 paintings by Cross presented consecutively, followed by 6 by Mylrea, etc.) ▸ interleaved practice: participants studied artists mixed up (e.g., 1 painting by Cross, then 1 by YeiMei, then 1 by Mylrea, etc.) - both groups studied 6 paintings by each of 12 artists, then were tested with 4 previously unseen paintings each of the 12 artists - results: ▸ blocked: 50% correct ▸ interleaved: 65 correct - in one experiment, participants experienced both blocked and interleaved practice: 78% of participants did better with interleaved practice - but when asked to rate practice effectiveness, 78% rated blocking as good as (or better than) interleaving - conclusion: interleaving enhances inductive learning of concepts and categories - implications: even though it may seem less effective, to maximize learning of different concepts, multiple examples should be presented and practiced interleaved - interleaving has been shown to work in basketball, baseball, identifying birds, electrocardiogram interpretation, learning microsurgery, and assessing complex legal scenarios - may work by requiring the learner to differentiate among related concepts, or to retrieve the correct strategy for each different type of problem spaced vs. interleaved practice: - interleaving can be seen as a subtype of spacing REM/Sleep Effects David Hartley (1791) was the first to suggest dreaming might alter the strength of associative memories REM sleep was discovered by Aserinski & Kleitman (1953); linked it to dreaming REM sleep has been found to improve learning complex logic games, foreign language acquisition, visual discrimination tasks, and intensive synchronous may be due to synchronous brain activity, neuronal replay, changes in neuromodulators, or regional brain activation (Stickgold, Hobson, Fosse, & Fosse, 2001) Contradictory evidence (Siegel, 2001; Vertes & Eastman, 2000): results of animal studies are equivocal: REM deprivation does not necessarily impair learning/memory REM deprivation confounded with stress some antidepressants suppress REM sleep, but learning/memory is not disrupted bilateral lesions of pons abolish REM sleep, but patients led normal lives conclusion: REM sleep has no role in consolidation What about the effects of sleep overall? sleep deprivation reduces working memory by 38% sleep increases performance of procedural memories, like motor skill learning  daytime naps enhance declarative memories, maze learning, complex visual stimuli, paired words); performance correlated with the amount of NREM (non-REM) sleep Conclusions: sleep has an important role in memory, attention, and executive function (Lowe et al., 2017) sleep actively promotes processing of information learned during wakefulness sleep is important, not only for memory consolidation, but also for memory enhancement Neurons, the Brain, and Memory Dual trace hypothesis (Donald Hebb, 1949): - memory formation (synaptic consolidation) requires two processes: 1. short term: experience causes activity in certain neural circuits 2. longer term: continuous or repeated activity creates stable change in nervous system - “neurons that fire together, wire together” - found in long-term potentiation (LTP): electrically induced changes in neural response (Hawkins & Kandel, 1984) - kinds of neural change found to support learning: increased neurotransmitter released increased size of postsynaptic region more synapses formed Standard Model of Systems Consolidation (Squire & Alvarez, 1995): - initially, incoming information activates cortical areas (e.g., for vision, audition, etc.), and is integrated and coordinated by the hippocampus into a memory trace - consolidation occurs via reactivation: hippocampus reactivates connections to the cortex associated with a memory occurs during sleep or by conscious rehearsal of a memory - after repeated reactivation, long-lasting cortical interconnections are strengthened and/or new connections established new memories are integrated with existing ones activity in the hippocampus is no longer required - predicts that retrograde amnesia (due to damaged hippocampus) should be temporally graded: memories formed closest in time to the traumatic event are most likely to be lost Multiple Trace Theory of Consolidation (Nadel & Moscovitch, 1997) - proposes that semantic memories stored in the cortex gradually become independent of the hippocampus (as in the Standard Model) - in contrast, all episodic memories always involve the hippocampus, which provides spatial and temporal context (where and when) - predicts that complete lesions of the hippocampus should impair all episodic memories-- whether recent or remote Autobiographical Memory Learning Outcomes 1. Describe the structure of autobiographical memory. 2. What are infantile amnesia and permastore? 3. What are three kinds of distinctiveness that can affect memories? What are flashbulb memories? 4. How does Elizabeth Loftus use false memories to show the constructive nature of memory? 5. How does the source monitoring framework explain memory errors? 6. What is misinformation, why do people believe it, and what can be done about it? Autobiographical Memory a form of long-term memory that comprises the self-knowledge, episodes, and experiences of a person’s life is crucial to our self-conception, helping us to understand who we are and how we relate to the world utilizes the two primary types of declarative memory: - personal semantic memories: - episodic memory: memory of events we have experienced organized by tiers of specificity (Conway & Pleydell-Pearce, 2000): - lifetime periods: distinguishable time periods in one’s life e.g., “When I was in university...” - general events: represent repeated events or a sequence of related events e.g., “...I was a member of the debate club...” - event-specific knowledge: details about a single event e.g., “...and I won the annual debate championship.” - this knowledge is used by the working self: current personal goals and self-images e.g., “This experience makes me an excellent candidate for this job.” Lifespan Autobiographical Memory childhood (or infantile) amnesia: adults’ inability to remember events before ages 3-5 - average age of earliest memory is ˜3.5 years of age - females report earlier memories than males - recall of memories over the lifespan falls off before age 8 compared to other periods - Freud (1916): “the remarkable amnesia of childhood” is due to repression of traumatic experiences of early psychosexual development - contemporary theories: ▸ hippocampus (involved in storage of long-term memories) not sufficiently mature before age 2 ▸ incomplete language development is a barrier to encoding autobiographical memories (which include episodic and semantic components) ▸ episodic memories are tied to one’s sense of self, which develops later in life Bahrick, Bahrick, & Wittinger (1975): - studied retention of names and faces of people’s high school classmates (vs. strangers) - permastore: very long-term storage that can last your entire life Distinctiveness Distinctiveness can help to make the relationship between cue and item unique Unique item activates fewer memories in LTS and reduces interference primary distinctiveness: incongruity defined with respect to the immediate context Rabinowitz & Andrews (1973): - isolation effect (von Restorff effect, 1933): a distinctive stimulus is more likely to be remembered - control list: words printed in black capital letters - isolated list: one item printed in red capital letters  secondary distinctiveness: incongruity is defined with respect to past experience e.g., first day of university, first time in a big city Hunt & Elliott (1980): - orthographic distinctiveness: word having unusual letter and spelling patterns - better remembered than more typical words e.g., llama, khaki, afghan (vs. leaky, kennel, airway) - due to unusual lower-case word shape (no effect when in CAPITALS) emotional distinctiveness: emotional events are remembered in greater detail flashbulb memories: detailed, vivid recollection of circumstances surrounding hearing about a surprising/emotional event Brown & Kulik (1977): - asked people about their personal circumstances when they heard about the assassination of John F. Kennedy - people gave many details, and were confident in their memories - this effect is linked to the release of adrenaline, causing greater amygdala activity (McGaugh, 2004; 2013) - seems intuitive; much anecdotal evidence - problem: were the reported memories accurate? Solution is repeated recall: record person’s memories and experiences immediately following an emotional event and compare to later surprise reassessment. Bohannon (1988)/Bohannon & Schmidt (1989): - tested students’ memory for events surrounding explosion of space shuttle Challenger - 3 groups: tested 2 weeks, 8 months, or 15 months later - measure of emotion: How upset were you (1-7)? - shuttle facts: How many were on board? How long into the flight did it explode? - personal discovery: Where were you when you heard of the explosion? What were you doing? - emotion appeared to enhance memory, but memory was not perfect The Constructive Nature of Memory visual illusions: mismatches between percept and objective stimulus - sensory data available - persist even when we know our interpretation is incorrect memory illusions: erroneous judgments based on memories - sensory data no longer available - difficult to become aware of; we often cannot compare our memory to the actual event False Memories Roediger & McDermott (1995, based on Deese, 1959): - participants studied 15-word list e.g., water, stream, lake, Mississippi, boat, tide, swim, flow, run, barge, creek, brook, fish, bridge, winding - recall of studied words: 65% - recall of strongly semantically related critical lure (river): 40% - confidence measured in recognition task (4-point scale) - results: Studied Items Unrelated Lure Critical Lure (river) confidence: 3.6 1.2 3.3 - shows memory is associative Freud (1901): - repression: active submerging of a painful memory without conscious awareness - considered this to be the most powerful defense mechanism used by the ego to reduce anxiety - not confirmed experimentally False memory syndrome: memory of traumatic experience which is objectively false, but in which the person strongly believes it to be true - “recovered” memories may be false memories, based on events that never occurred. How? - FMS can be elicited by “recovered memory therapy,” in which a therapist encourages a client to identify repressed memories (e.g., of abuse), despite the lack of evidence or memory of any past abuse - psychotherapists ask leading questions that may elicit compliance with generation of false memory - this approach (often using hypnosis) can cause false memories to be implanted e.g., one woman believed she had endured ritualistic abuse, was forced into a pregnancy, and the baby was cut from her uterus--but there were no physical scars or any other physical evidence - led to the “memory wars” debate over the existence of repressed memories of abuse Loftus & Pickrell (1995): - can people be made to (incorrectly) remember that they were lost in a mall in childhood? - experimenter interviewed relatives of participants - participants were given paragraphs describing four events from their childhood (three true events and one false--but plausible--event) e.g., false story about being lost at a mall at age 5 - participants wrote what they remembered about each event immediately after, and in two follow-up interviews - results: 68% of true events remembered 29% of participants “remembered” false events more words were used to describe the true events, and they were also rated as being somewhat more clear than the false events Other implanted false memories: accidentally spilling a bowl of punch on the parents of the bride at a wedding reception when they were 5 (integrating false event into actual events provided by parents of participants) (25% of people; Hyman & Billings, 1998) being a victim of a vicious animal attack in childhood, facilitated by an interviewer using “guided imagery” (26%; Porter et al., 1999)  believing they committed a crime (theft, assault, or assault with a weapon) that led to police contact (70%; Shaw & Porter, 2015) going on a hot-air balloon ride after seeing a Photoshopped photo (50%; Wade et al., 2002) meeting Bugs Bunny at Disneyland after seeing a fake print ad (16%; Braun et al., 2002) liking asparagus as a child (Laney et al., 2008) Eyewitness Identification Elizabeth Loftus (b.1944): won the Grawemeyer Award in Psychology in 2005 in 80,000 cases/year in USA, the only critical evidence is eyewitness identification 2,000-10,000 estimated wrongful convictions/year due to faulty eyewitness testimony The Innocence Project has helped exonerate over 200 wrongful convictions through DNA testing since 1989 (The Innocence Project, 2024) 63% involved eyewitness misidentification Innocence Canada has helped exonerate 29 people (Innocence Canada, 2024) Loftus & Palmer (1974): Expt. 1: participants saw film of a collision - were asked, “How fast were the cars going when they _______?” - word affected speed estimates: Descriptive word Speed estimates “Smashed” 40.8 mph  “collided” 39.3 mph “bumped” 38.1 mph “hit” 34.0 mph “contacted” 31.8 mph - could participants judge the speed properly? film of car crash at 20 mph estimated to be 37.7 mph 30 mph estimated to be 36.2 mph two films crash at 40 mph estimated to be 39.7 and 36.1 mph Expt. 2: participants saw film of a multiple-car crash - were asked about broken glass “smashed” = 32% yes “hit” = 14% yes Misinformation effect: exposure to misleading information after witnessing an event can lead people to believe that they have seen or experienced something they never did; possible causes: overwriting: misleading information replaces memory trace of the actual experience misinformation acceptance: people believe the post-event information is true because questioner is a person of authority source confusion: memory of the question is confused with memory of the experience Contrasting false memory with misinformation: an analogy: think of your memory like a document  false memories are like creating a new document from scratch, or editing an existing one with fictional details the misinformation effect is like someone taking your document, making changes to it, and then you mistakenly believing the altered version is the original Source Monitoring Framework reality monitoring: “processes by which people discriminate between memories derived from perception and those that were reflectively generated via thought, imagination, dreams, and fantasy” (Marcia Johnson, 1991) source monitoring: process of making attributions about the source of memories - memories are not “tagged” with source information - attribution occurs during retrieval; may be incorrect source monitoring errors: - source confusion (or source misattribution): believing the source of a memory is different than what it actually is - source failure: not remembering the source of a memory - cryptomnesia (“hidden memory”): remembering a previously forgotten memory, but believing it to be new and original e.g., George Harrison’s “My Sweet Lord” (1970) (inadvertently?) plagiarised the Chiffons’ “He’s So Fine” (1962) Lindsay & Johnson (1989): - showed picture of office, then read text describing the office - some text was misleading (described typical things in an office which were absent) e.g., “... a filing cabinet behind the woman...” - recognition test: was item in the picture? showed evidence of suggestibility - source-monitoring test: was item in picture, text, or both? much fewer errors; suggestibility effect gone - conclusion: source monitoring occurs at retrieval Affected by: contextual information e.g., where were you when you last had your wallet? sensory information e.g., do you remember the feel of your wallet when you put it into your pocket? fewer cognitive operations (i.e., memory retrieved with little effort) e.g., remembering where you put your wallet Although memory may be malleable, eyewitness testimony may be reliable under proper conditions (Wixted et al., 2018): an initial memory test is least likely to be contaminated (subsequent tests are more likely to be contaminated) police lineups should be fair (e.g., suspect does not stand out) eyewitness’ confidence should be recorded (it is correlated with accuracy) Misinformation and the Infodemic What is it? misinformation: misleading or incorrect information disinformation: intentionally misleading or incorrect information Why is it a problem? WHO declared a “massive infodemic” of misinformation about COVID-19 in March, 2020 this included false claims that: you could do a self-test for the SARS-CoV2 virus by holding your breath drinking large amounts of water protects against the virus gargling salt water prevents infection social media played a large role in this, with millions of posts of COVID-19 misinformation on online platforms including Twitter, Instagram, and YouTube (Cinelli et al., 2020) this has led to harm: misinformation has been blamed for vaccine hesitancy and a decrease in life expectancy in the USA of 3-5 years Why do people believe misinformation? when experiencing emotional motivators, including heightened fear or anger e.g., during a global pandemic if they are frequent users of social media, less educated, and on the political extremes motivated reasoning: tendency to accept information based on its desirability, rather than its accuracy, because of your emotions, values, and preferences e.g., students from two colleges were shown film of a football game between the colleges’ teams; they only agreed with the referees’ calls when the decision favoured their team confirmation bias: tendency to seek evidence that confirms one’s existing beliefs (and disregard contrary information) e.g., if you believe that Macs are better than Windows, you focus only on cases in which they are better illusory truth effect: repeated exposure to misinformation may make people more likely to believe it (partly due to processing fluency: when information is repeated, it becomes more familiar and easier to process) e.g., seeing false postings about a miracle cure on social media can increase belief in it How can misinformation be combated? debunking: correcting the effects of misinformation post-exposure; like fact-checking (Lewandowsky et al., 2022) - truth sandwich technique: 1. Fact: give the truth first 2. Warn about the myth: describe the misinformation 3. Explain fallacy: explain why the misinformation is wrong 4. Fact: repeat the truth again, with credible alternative explanation prebunking: developing immunity to misinformation in the future by using psychological inoculation techniques - based on inoculation theory (McGuire, 1961), which uses the metaphor of protection against disease by immunization - pre-exposure to weakened versions of a stronger, future threat can safeguard an attitude or belief from influence or persuasion - includes a forewarning about harmful misinformation, and pre-emptive disproof of the falsehood - prebunking found to be more effective than debunking (Jolley & Douglas, 2017) evidence (van der Linden et al., 2017): - asked people’s beliefs about topics including climate change - dependent variable: 0-100 agreement whether humans are causing global warming - over 2,000 participants then randomized into independent variable conditions: ▸ facts-only group: told there was 97% consensus among climate scientists about human- caused climate change ▸ misinformation only: shown “Global Warming Petition Project” ▸ false-balance: got facts and misinformation ▸ inoculation: given forewarning about misinformation, and facts ▸ full inoculation: received forewarning, facts, and prebunk about the bogus petition - latter two groups were then exposed to the misinformation - results: change in dependent variable pre- vs. post - conclusion: psychological inoculation (prebunking) is effective against misinformation How can this research be applied? inoculate other people (Garcia & Shane, 2021): - fact-based: correcting a specific false claim by communicating accurate information e.g., explaining that it is not possible to get the flu from the flu vaccines because most of them contain inactivated strains of the virus - logic-based: explaining misleading tactics used to manipulate people, and refuting logical fallacies e.g., pointing out inherently contradictory claims like “global temperatures cannot be accurately measured” and “temperature records show the climate has been cooling” - source-based: undermining the credibility of bad sources of information e.g., ridiculing people who believe that reptilian “lizard people” are controlling the world reduced others’ beliefs in it YouTube video campaign by Truth Labs for Education online games: Bad News (about “fake news”), Harmony Square (about political disinformation), and Go Viral! (about COVID-19 misinformation) Optional!^^ The “infodemic” of misinformation thrives on our emotions and biases, but we can fight back by learning to prebunk false claims and inoculate ourselves using an understanding of cognitive processes. Conceptual Knowledge Learning Outcomes 1. How are natural categories different from the classical approach? 2. Describe the components of the following approaches to knowledge representation: - prototypes - exemplars - feature comparison model - semantic network models - connectionist approach 3. Explain the pros and cons of each model. 4. What are some kinds of schemas? How do scripts and schemas affect memory? Categorization Category: class of objects, associated on the basis of some relationship Concept: mental representations of a category - depends on generalization: identifying features common to all members of a conceptual class - discrimination: notice differences between conceptual categories Bruner, Goodnow, & Austin (1956): benefits of categorization: reduces complexity of the environment means by which objects of the world are identified reduces need for constant learning allows us to decide what constitutes an appropriate action enables us to order and relate classes of objects and events Classical Approach Assumptions: categorization based on lists of defining features: those that are necessary to the meaning of the item e.g., a triangle is a closed, three-sided figure features are individually necessary and collectively sufficient Pros & cons: distinction between different categories is clear and logical works for logical categories: e.g., geometric shapes, prime numbers implies that all members are created equal Natural Categories Assumptions: groupings or clustering of objects or concepts that occur naturally in the real world fuzzy borders; membership may overlap e.g., what are the defining features for the concept vehicle? The Prototype Approach Eleanor Rosch [née Heider] (1973) Assumptions: items are comprised of a list of features or attributes concept organization based on prototype: abstract, idealized item that is most typical category member characteristic features: describe the prototype, but are not necessary categorization of items is based on similarity to the prototype concepts are hierarchically organized: ▸ Superordinate level (e.g., fruit, animals) - largest categories - members of a category have few attributes in common e.g., musical instruments - tends to be abstract ▸ Basic level (e.g., grapes, apples) - prototypes formed to represent the category - members of a category share many attributes, and are highly differentiated from other basic categories e.g., guitar, piano - shows priming effect: response is faster if item is preceded by a similar item e.g., judgments about specific apples are faster if preceded by apple, than by fruit - in general, people prefer to use basic-level names for things - children learn basic level objects first (dogs), before they categorize in superordinate (animal) or subordinate (collie, hound) ▸ Subordinate level (e.g., Thompson seedless, Concord) - narrowest categories - less distinct than basic categories e.g., classical guitar, folk guitar - experts prefer to use subordinate-level terms; greater expertise greater use of sub- subordinate terms (Johnson & Mervis, 1997) Evidence: Typicality effect: items differ in how well they represent a category e.g., rank order these items in their category: vehicles: car, elevator, sled, tractor, train clothes: jacket, mittens, necklace, pajamas, pants Family Resemblance: each item has at least one shared attribute with another item in the category; is a kind of typicality - good members of a category share many attributes with members of the same category - but share few attributes with members of other categories Rosch & Mervis (1975): - generated list of category members e.g., fruit: apple, banana, coconut, olive, orange, tomato - asked participants to rate typicality of each member within category e.g., apple as a member of fruit - others listed attributes of category members e.g., apple: red, sweet, crunchy, round - strong correlations found between typicality rating and number of attributes shared by other category members e.g., r =.85 for fruit (also, r =.88 for furniture, r =.91 for clothing) - Family resemblance is important to typicality brain activity Kosslyn, Alpert, & Thompson (1995): - participants placed in PET scanner - saw picture of an item, and heard a word (e.g., “toy,” “doll,” or “rag doll” - superordinate terms more likely to activate part of the visual cortex (language, associative memory) - subordinate terms more likely to activate visual attention areas Pros & cons: accounts for concepts representing loose groups e.g., games--merely share a family resemblance allows for variability explains how information is reduced to a single, idealized abstraction but we also store specific information about individual examples number of potential features is very large (infinite?) what determines feature weights? how does expertise change categories? counterintuitively implies categories have fuzzy boundaries The Exemplar Approach Assumptions: exemplars: members of a category that you have previously encountered (vs. prototypes, which are idealized) e.g., dog concept is based on actual dogs you’ve seen first you learn specific exemplars of a category, then you classify new items based on their similarity to the exemplars Heit & Barsalou (1996): - asked participants for exemplars of various categories of animals - measured exemplars’ typicality, and how typical the categories were of animals - strongly correlated: r =.92 (mammals are most typical; microorganisms the least) - implication: our concepts are based on the most typical (i.e., exemplary) items Pros & cons: accounts for typicality effect no lists of features needed no abstraction process needed allows for “exceptions” to categories e.g., penguins as birds good for categories with few members requires vast storage for individual members of large categories Feature Comparison Model (Smith, Shoben, & Rips, 1978) Assumptions: concepts represented as a set of features: Bird: Robin: wings wings feathers feathers … Red breast  … … defining features: essential, required features of a concept; are at the top of the feature list characteristic features: descriptive, but not essential (“loosely speaking”); are at the bottom of the list e.g., birds: - defining features = wings, feathers,... - characteristic features = fly, sing,... Sentence verification task: measure RT to correctly respond, “A robin is a bird.” relations between concepts are computed based on shared features; more features slower RTs two-stage model: Pros & cons: typicality effects: faster RT when item is a typical member of a category e.g., “A robin is a bird.” (faster than) “A penguin is a bird.” true/false effect: quick rejection of false sentences e.g., “A pencil is a bird.” (faster than) “A bat is a bird.” doesn’t explain category size effect: faster RT when item is member of a small category e.g., “A robin is a bird.” (faster than) “A robin is an animal.” - small categories have more defining features more stage 2 processing slower can account for violation of category size effect: e.g., “Scotch is a liquor.” (slower than) “Scotch is a drink.” not all defining features of a category are necessary e.g., “Is a robin with no wings still a bird?” features poorly defined; characteristic features have circular definition Semantic Network Models Common assumptions: concepts represented in network of interconnected nodes concept defined in terms of connection to other concepts Hierarchical-Network Model (Collins & Quillian, 1969) Assumptions: nodes represents a single concept concepts organized hierarchically pathways represent associations between concepts - “isa” pathways: express category membership - “hasa” pathways: express properties properties stored at the most general (“highest”) level possible, with no redundancy (cognitive economy) sentence verification via intersection search : - “A canary is a bird.” - “canary” and “bird” activated; activity spreads to neighbours - both spreads eventually intersect, allowing answer to be made Pros & cons: cognitive economy: e.g., “A bird has feathers.” (faster than) “A bird has skin.”  corresponds well with category size effect: e.g. “A canaryr...” Category Property “...is a canary.” 1,000 ms “...can sing.” 1,350 ms “...is a bird.” 1,200 ms “...can fly.” 1,400 ms “...is an animal.” 1,300 ms “...has skin.” 1,500 ms doesn’t explain violations of category size effect: e.g., “A dog is a mammal.” (slower than) “A dog is an animal.” typicality effects (model predicts they should not be obtained): e.g., “A robin is a bird.” (faster than) “An ostrich is a bird.” Spreading Activation Model (Collins & Loftus, 1975) Assumptions: not hierarchical link length represents degree of relatedness; search time depends on link length passive concepts not in working memory; active ones are spreading activation: degree of activation decreases over distance and time Pros & cons: accounts for: category size effects (& violations), typicality effects Meyer & Schvaneveldt (1976): - Lexical decision task (word/nonword):  Type of trial Prime Target RT Related prime “bread” “butter” 600 ms Unrelated prime “nurse” “butter” ____ ms - closely related concepts have shorter RTs - semantic priming effect: activation of a conceptual node facilitates retrieval of associated concepts or words difficult to falsify: what determines link length? Connectionist Approach - a.k.a. Parallel Distributed Processing (PDP) or artificial neural network models - classical approach: based on information & rules (serial approach) e.g., mind is like a computer: data & programs parallel: - many (simple) processors working simultaneously - models are artificial neural networks, analogous to human brain distributed: - memory and information processing occur in the connections, not in storage locations (“connectionism”) - information is distributed across the entire network, not localized processing: - processing units = neurons; are interconnected - active unit may pass along its activity via connections, excitatory vs. inhibitory - strength of connections may be modified--learning! - damage resistant: partial network may still solve problem Jets & Sharks Example (McClelland, 1981): - information can be stored in a table: name gang age education marital status occupation Art Jets 40s Jr. high single pusher Lance Jet 20s Jr. high married burglar Ralph Jets 30s Jr. high single pusher Rick Sharks 30s High school divorced burglar Sam Jets 20s college single bookie - or distributed in a network: Components: neuronally inspired nodes: processing units like neurons excitatory/inhibitory connections among nodes activation rules: specify conditions for activating a node learning rule: describes how connections can be changed, to improve performance Pros & cons: biologically plausible so far, fairly simple (but increasing in complexity) as a model of a complex system becomes more complete, it becomes less understandable (Bonini’s Paradox, 1963) Schemas & Scripts - contain generalized knowledge about things or events - allow organization of objects & experiences - kinds: script: sequences of actions for complex situations and events frame: for physical objects (room, desk, house) story schemas: setting, conflict, resolution, closing problem schemas: for typical structure of problems Scripts Bower & colleagues (1979): - participants were given texts over a dozen routine activities, including going to the dentist, a restaurant, or the doctor - were later given a recognition test that included: sentences included in the text (e.g., “The doctor was very nice to him.”) unstated sentences that fit the script (e.g., “John took off his clothes.”) sentences of false, but plausible actions (e.g., “The doctor was not very rude to him.”) - rated confidence that they had read the sentence on a 7-point scale - results: - scripts help guide actions; may fill in details Characteristics of Schemas Bransford & Johnson (1972): encoding - participants read (vaguely written) paragraphs e.g., about doing laundry, flying a kite - some were given topic sentence beforehand; had much better recall - schemas add meaning, which aids encoding & remembering Bartlett (1932): inferences - “The war of the ghosts” Native American story read by students in England - reproductive memory was poor, because story/structure were unfamiliar - recall included many reconstructive errors: story was combined with (or interpreted by) students’ existing schemas - schemas show inferences: logical interpretations/conclusions made that are not part of original stimulus material Brewer & Treyens (1981): memory selection - participants visited office for 35 seconds - free recall of “everything you can remember about the room you were just in” - good recall of objects consistent with “office” schema (e.g., desk, chairs) - poor recall of inconsistent objects (e.g., picnic basket, wine bottle) - some false recall of consistent objects that were not present (e.g., books, filing cabinet) - schemas provide enhanced memory--for schema-consistent items Schemas Summary based on our general world knowledge and experiences active, constructive process for comprehension (top-down) used at encoding, they increase details remembered may bias interpretation of ambiguous information may cause us to ignore inconsistent information Visual Imagery & Propositions Learning Outcomes 1. What is the basis of the analog coding approach to imagery? Describe the evidence to support this view. 2. What is the basis of the propositional coding approach? Describe the evidence to support this view. 3. What is the basis of Paivio’s Dual Coding Theory? Describe the evidence to support this view. 4. What are metrical and structural properties of cognitive maps? Describe the heuristics applied to cognitive maps. 5. What differences are their in male and female wayfinding? 6. What is eidetic memory? 7. Describe six mnemonic devices that can improve memory. Imagery: mental representation of stimuli (objects, events) that are not physically present What is the nature of the representation? analogical (“depictive”) propositional (“descriptive”) multiple coding Analog Coding (Roger Shepard, Stephen Kosslyn) Metaphor: - imagery = perception - images retain some sensory qualities; are pictorial Processes: - generation: images reconstructed from LTS stores - retention: images are maintained in STS - inspection: generated images can be scanned by “mind’s eye” - transformation: mental images can be manipulated like real objects Evidence: Shepard & Metzler (1971): rotation - same/different task using pairs of “3-D” objects: same (rotation in picture-plane) same (rotation in depth) different (no rotation possible) - RT to decide “same” is a linear function of angle of rotation: - RT for rotation in depth comparable to RT for rotation in picture-plane Kosslyn (1975): size - imagine a rabbit next to (a) a fly or (b) an elephant - participants asked “Does a rabbit have ears?” - RT 211 ms longer when target was beside larger animal; participants reported that the size of “imagined” rabbit was smaller - possible confounding produced by interest level - imagine target beside (a) giant fly or (b) tiny elephant - longer RT when target beside larger animal Kosslyn, Ball, & Reiser (1978): distance - participants memorized map: - task: scan from one location to another, and press a button when destination reached - results: - time to scan between imagined objects correlates with distance between those objects on a map Kosslyn (1973): distance - participants shown drawing: - visualized it from memory, and focused on one end - was a named component in the drawing? e.g., “motor” or “porthole” or “anchor” - named object farther from place of focus longer RTs - lag time implied scan of mental image Roland & Friberg (1985): physiological evidence - measured regional cerebral blood flow while participants: performed mental arithmetic  thought of a musical jingle visually imaged a walk through their neighbourhood - results: increased cerebral blood flow in visual cortex--only with imagery task Propositional Coding (Zenon Pylyshyn, 1973, 2006; Peter Slezak, 1995) - uses metaphor of language, not perception - images are represented by abstract, symbolic, “descriptions” of visual scenes - proposition: smallest piece of information that can be judged true or false. Described as a unit of meaning that is indivisible. e.g., “The dog is brown.” has 1 “The small dog is brown.” has 2 - propositions formed by connections between nodes of a network: - mind represents concepts; may form images as an epiphenomenal by-product e.g., book represents information in words, which generate images when read Evidence: - certain operations possible for real images are not possible for mental images e.g., mirror or negative images of faces (Phillips, 1972) e.g., rotation of images (Slezak, 1991) - images are limited Nickerson & Adams (1979): - showed 15 “pennies”, including an unmodified one - most people (84%) could not identify the real penny - images must be lacking in details - experimenter expectancy, demand characteristics, and participants’ expectations may be influencing results predictions of scanning time across island matched “mental scanning” (Mitchell & Richman, 1980) but this breaks down when predictions more difficult (scanning along a spiral vs. straight line) - part-whole relationships: finding particular shapes in a whole figure chance level (Reed, 1974) - semantic (verbal) information may distort recall of visual images Carmichael, Hogan, & Walter (1932): - name given to ambiguous figure affected remembering Multiple Codes Dual Coding Theory (Allan Paivio, 1986; 2006) - developed as a general theory of cognition Storage: two independent but interacting systems: imagery system: stores images (“imagines”); right hemisphere processing verbal system: stores linguistic information or verbal descriptions (“logogens”); left hemisphere processing Processing: in one or both systems: pictures: image processing; may also be labelled (verbal processing) concrete words: verbal processing; image may also be formed abstract words: verbal processing only Connections: a concept is connected to other related concepts in the same system (associative connections), and the other system (referential connections) activating any one concept also leads to activation of closely related concepts Pros & cons: Paivio & Csapo (1969): experimental support - stimuli: pictures, concrete words, and abstract words - varied rate of presentation:  fast = 5.3 items/s (limits participants to one code) slow = 2 items/s (opportunity for dual codes) - results on free recall task: - two codes increase likelihood of later retrieval - imagen system superior to logogen system Kounios & Holcomb (1994): EEG evidence - concrete words elicited greater activity than abstract words (dual coding) - concrete words activated both hemispheres equally (dual coding) - abstract words activated the left hemisphere more than the right (single code) criticism: there is no need for two representational systems; all memories are stored one way (common coding theory) Can using multiple coding interfere with memory? Schooler & Engstler-Schooler (1990): - procedure: participants shown videotaped robbery (creates visual memory) - task: imagine the robber describe the robber do a series of math problems - then given recognition memory test (pick robber out of line-up of 8 people) - results: Task: Recognition: Imagine 55% Describe 15% Math 42% - verbal overshadowing effect: verbalizing previously seen visual stimuli interferes with the original visual memory Cognitive Maps Have both metrical and structural components metrical: includes information about distances and directions; three types of knowledge: - landmark: information about particular features at a location - route-road: specific pathways for moving from one location to another - survey: estimated distances between landmarks  structural: includes information about regions and clusters Hirtle & Jonides (1985): - distance between landmarks judged - members of the same cluster judged to be closer than any member of another cluster e.g., Is it farther to Chapters in St. Albert or 99 Ave./170 St.? Milgram & Jodelet (1976): - asked 215 Parisians to draw a map of the Seine river - over 90% drew it straighter than it actually is - cognitive maps do not always correspond to physical locations Barbara Tversky (1981, 1993): - when navigating space, people use heuristic: methods that sometimes lead to a correct solution, but often oversimplify reality e.g., Which is farther north, Rome or Philadelphia? density heuristic: greater number of landmarks on a route makes it seem longer symmetry heuristic: representing shapes as more symmetrical than they actually are right-angle bias: representing intersections as forming 90° angles more than the angles actually do rotation heuristic: representing slanted figures or boundaries as more vertical or horizontal then they actually are  alignment heuristic: representing geographical features as more lined up with each other than they really are relative-position heuristic: representing relative positions of landmarks and boundaries by distorting cognitive maps to reflect conceptual knowledge, rather than actual spatial configurations Why does our navigational fallibility persist? Likely because we don’t often encounter feedback about our geographical or travel-time inaccuracy. Wayfinding Cornell, Sorenson, & Mio (2003): - measured self-reports of “sense of direction” (SOD) - also assessed wayfinding: pointing to nonvisible landmarks reversing a route with a detour devising a shortcut locate site within a building - found small-moderate correlations between these and SOD memory for past performance SOD - compared genders: males predominantly used survey knowledge: based on comparisons among landmarks; like bird’s eye view females relied on route knowledge: based on a series of directions; like “vectors” - females rated their SOD as worse than males, but no gender differences obtained in actual wayfinding - problems may arise if typical male is given route knowledge, or if typical female is survey knowledge Eidetic Memory - definition: detailed, vivid recollections of a complex visual scene; a.k.a. “photographic memory” Haber & Haber (1964, 1988): - tested elementary school age children, aged 7-12 - showed pictures on an easel for 30 s, then removed them and asked questions about them - 8% of children performed significantly differently from the rest - criteria for eidetic memory: images... must be reported must be positively coloured (not negative image) are projected onto space (not just “in the head”) described in the present tense associated with eye movements appropriate to the location of objects in the scene - images lasted at least 40 s - this ability is equally likely in males and females, but is less common in adults Stromeyer & Psotka (1970): - case study of eideteker named Elizabeth - could reproduce a poem written in a foreign language that she did not understand, from bottom to top, as fast as she could write--even years later - could fuse two-part random dot patterns, each containing 1,000,000 dots, presented 4 hours apart - could also fuse two patterns presented a day apart into stereoscopic images - however, no one else has passed such a test, Elizabeth refused to repeat any tests, and ended up marrying Stromeyer Miller & Peacock (1982): - compared eidetic 12-14 year old boys to control group eidetic control recall of picture details: 6.3 4.9 dot fusion (number correct out of 18) 13 3 - eidetic more susceptible to interference from a second stimulus than controls - eidetic memories lasted longer (25-180 s) than images of control group (0-13 s) Conclusions: - relatively rare ability to maintain vivid, detailed images; etiology unknown - quantitatively and qualitatively different from non-eidetic memory - not really “photographic memory”: recall far from perfect memories fade in a few minutes  images easily disrupted by new visual stimulation Mnemonic Devices - definition: techniques to aid memory, often by visually associating to-be-remembered items with a known series of images - in an anecdotal story, the poet Simonides of Ceos (c. 500 BC) was said to have escaped a collapsing feast hall; he later identified the victims by imagining where they were sitting Kinds of mnemonics: method of loci: associate to-be-recalled items with familiar locations (a “memory palace” or “Roman room”) peg word system: associate items with “pegs” e.g., 1 is a bun, 2 is a shoe, 3 is a tree, etc.5 key word: connect sound of word with familiar word e.g., pato (Spanish for “duck”) pronounced “pot-oh,” so picture a duck wearing a pot organization: hierarchically, or story, or rhyme e.g., chunking acronym: use first letter of each word to create pronounceable word e.g. POLKA (=Pegword, etc...)  acrostic: use phrase made with first letters e.g., Pa Observed Lice Kissing Ants Why do mnemonics work? provide structure for encoding and cues for retrieval apply multiple codes (verbal and imagery) form vivid, durable trace less subject to interference