Memory

Memory Copyright © Allyn & Bacon 2005 What is Memory? Memory = the ability of the brain to create, maintain, and retrieve information about the past   Issues of psychological study:  Encoding: processing and combining of received information (involves attention)  Storage: creation of a record of the encoded information  Retrieval/Recall: calling back the stored information in response to some cue for use in some process or activity Issues of neuroscientific study:  Connecting function to local brain structures (localization)  Contemporary= exposing patterns of activation (holistic processing) Copyright © Allyn & Bacon 2005 Early Psychological Studies of Memory  Hermann Ebbinghaus (1885)  1st systematic study of memory  Rehearsed lists of nonsense syllables  ZAT, BOK, QUJ 20 min, 1 hr, 8 hrs, 1 day, 2 day, 6 days, 31 days  Tried to recall after varied intervals   Studied the effect of list length, amount of practice, learning of serially ordered items  Ebbinghaus’s forgetting curve Copyright © Allyn & Bacon 2005 Copyright © Allyn & Bacon 2005 Copyright © Allyn & Bacon 2005 Early Psychological Studies of Memory  William James (1890)   Considered conscious experience too transitory to be considered ―Memory‖ Theorized dual memory stores based on introspection method but with little scientific support   very basic precursor to modern memory models Primary Memory:     Immediate memory Never leaves consciousness Gives a faithful rendition of events just perceived Now called Short-Term Memory Permanent memory-- experienced information not readily accessible ―Paths‖ etched into brain tissue—incorrect Now Called Long-Term Memory Copyright © Allyn & Bacon 2005  Secondary Memory:    Early Neurological Studies of Memory  Karl Lashley (1920)    Theorized that physical memory traces called engrams must be made in the brain when learning occurs (no evidence to support at his time but it is coming now!) Lashley removed portions of cortex, varying from 10-50% to study the role cortex played in learning of rats Led to 2 important learning/memory theories  Principle of Mass Action: proved that the amount of cortex removed was critical to the learning abilities of the rats= knowledge distributed throughout cortex  Equipotentiality: all areas of the cortex are equally important to learning/ no area was proven to be more important than any other area Copyright © Allyn & Bacon 2005 Early Neurological Studies of Memory  Donald Hebb (1949)– student of Lashley  Theory that changes that occur during learning develop among interconnections of neurons throughout wide areas of the brain  Cell Assemblies: loops of neurons become interconnected when learn something - memory is re-activation  Some neurological support for this theory Copyright © Allyn & Bacon 2005 Neurology of Memory  Memory functions are BOTH located in specialized areas AND distributed throughout the brain   PET scan of person deciding whether a word describes living or non-living object = high activation of frontal cortex (localization) but also activation to a lesser degree in other areas Research shows combination of specialization and generalization applies to many memory operations Copyright © Allyn & Bacon 2005 Neurology of Memory  Three main memory sites in the brain-  Cortex: thinking, problemsolving, remembering   Cerebellum: motor memory   Hippocampus: processes new information and routes it to parts of the cortex for permanent storage Copyright © Allyn & Bacon 2005 Early Psychological Studies of Memory  Evidence for James’s two memory stores (1890-1960)   Introspection suggests that some things are remembered for a short time, and others for a long time Physiological studies indicate that short term functions can be interrupted, whereas long term functions seem to remain intact  Electroconvulsive shock, head traumas  Psychological experiments suggest two memory stores  Large body of behavioral evidence Copyright © Allyn & Bacon 2005 1st Model of Memory  Waugh & Norman (1965)- developed 1st modern behavioral model of memory  Extended James’ early theory  Different than James because they quantified properties of primary memory   Idea of memory as stored in ―boxes-in-the-head‖ Basis of most contemporary models  Primary Memory:     Short-Term storage system Independent of secondary memory Has very limited capacity Forgetting due to  Decay -- memory fades over time  Interference – new information displaces older information Need rehearsal to send information to secondary memory Long-Term storage system Copyright © Allyn & Bacon 2005  Secondary Memory:  1st Model of Memory Model of Primary & Secondary Memory System-- Waugh & Norman (1960s) Rehearsal Stimulus Primary Memory Secondary Memory Forgotten Copyright © Allyn & Bacon 2005 Expanded Model of Memory  Atkinson & Shiffron Model of Memory (1968)   Also based on early dual store theory But presents a three stage informational processing model  Memory structures fixed– Sensory, Short, Long  Expanded earlier models to include processes of transfer  control processes variable– encoding, storage, retrieval Memory: data being retained Memory Store: structural component that contains the information Copyright © Allyn & Bacon 2005  Make distinction between   Components of Atkinson & Shiffron Model    Sensory Memory  Records information from the senses for up to three seconds  If information is attended to, transferred to STM otherwise it is lost Short-Term Memory  Holds information for up to thirty seconds  If the information is rehearsed/attended to, transferred to LTM otherwise it is lost Long-Term Memory  Relatively permanent  can hold vast amounts of information  Interference, decay, brain damage can cause information to be lost Copyright © Allyn & Bacon 2005 Classic Information-Processing Model of Memory Based on Atkinson & Shiffron Model of Memory (1968) Copyright © Allyn & Bacon 2005 Sensory Memory  Sensory information held briefly in cortical sensory areas so we can initially process it  Also called sensory register or sensory store  Function:  provides continuous flow of updated information available for processing  Types of codes:  Iconic Memory: A fleeting sensory memory for visual images that lasts only for a fraction of a second  Visual Cortex Auditory Cortex   Echoic Memory: A 2 to 3 second sensory memory for Auditory information  Other sensory memories for touch , smell, taste Copyright © Allyn & Bacon 2005 Classic Information-Processing Model of Memory Based on Atkinson & Shiffron Model of Memory (1968) Copyright © Allyn & Bacon 2005 Short-Term Memory (STM)  Transitory store of information  Held for 20-30 seconds then either transferred to Long-Term Memory or lost Intermediate storage area for attended-to environmental stimuli where rehearsal takes place Predominantly auditory  Even for non-auditory information So much of STM is reliant on your inner voice — you see the word ―MONEY‖ or ―$‖ and you say ―MONEY‖ in your head. Visual Semantic (meaning based) Copyright © Allyn & Bacon 2005  Function:   Types of codes:    Short-Term Memory (STM)  Limited processing capacity  held at most 20 seconds Based on early work of Peterson & Peterson (1959): •Subjects memorized nonsense syllables, (e.g., MJK, ZRW). •To prevent rehearsal, they were given a distracter task during the waiting period (count backwards by 3 from 506). •When a cue was given, subjects tried to recall the letters. •Short-term memories vanish within twenty seconds. Copyright © Allyn & Bacon 2005 Short-Term Memory (STM)  Limited processing capacity  Maintenance Rehearsal: sheer repetition allows information to remain in STM for an indefinite period of time and facilitates transferred to LTM Neurology = STM is a reverberating circuit of neural activity with a self-exciting loop of neurons   Note: Merely holding information in STM does not assure it will not be lost (forgotten)! Copyright © Allyn & Bacon 2005 Short-Term Memory (STM)  Limited storage capacity     STM can only hold a limited amount of ―items‖ Miller’s Magic number (1956) = 7 plus or minus 2 Once STM filled– information moved to LTM or lost Benefit of limit = helps us forget what is no longer useful (not cluttered with STM distracting trivia)  Q= WHAT MAKES AN ―ITEM‖? Copyright © Allyn & Bacon 2005 Short-Term Memory (STM)  STM storage capacity:  Chunking: Process of grouping distinct bits of information into larger wholes to increase short-term memory capacity    What is a chuck varies from person to person depending on their prior experiences # of chunks limited but people can learn to increase the size of chunks– ultimately increasing STM Chunking is one of the executive functions of the prefrontal association cortex ** Chunking makes STM capacity NOT an absolute measure!*** A & B Word List Copyright © Allyn & Bacon 2005 Short-Term Memory  Serial Position Curve  Indicates the tendency to recall more items from the beginning and end of a list than from the middle. Recency Effect:  The Serial-Position Effect  good recall of words at end of list   Considered to still be in STM ―pushed‖ earlier information out  Primacy Effect:  good recall words at beginning of list   more attention/rehearsal Gets to LTM  Middle words lost! Copyright © Allyn & Bacon 2005 From the work of Mary Whiton Calkins, early 1900s Working Memory  Most widely accepted and used model today! Baddeley (1986)  Re-modeling of STM   Short term memory is woven together with higher cognitive processes-- such as learning, reasoning, and comprehension Focuses on the maintenance of information through the speed of rehearsal  Challenges limited capacity viewed as not a matter of holding 7 +/- 2 items but can keep however much information we can rehearse in a fixed amount of time, so speed of processing is the limiting issue   Transitory area that is more than storage– conceptualized as an active, conscious work space where information is accessible for current use and processing occurs Differs from previous STM models in 2 major ways:   Bring information from Long-Term storage back into intermediate WM for active processing along with incoming information from senses Has 3 (plus 1) components (rather than just single storage bin)  Central Executive (master controller)  Articulatory Loop/Phonological Store (for verbal)  Visiospatial Sketchpad (for visual)  New: Episodic Buffer (integrates info so it makes sense) Copyright © Allyn & Bacon 2005 Working Memory •Central Executive: temporary memory about the goals active at the moment, expected inputs during sequences of action, and intermediate products of cognitive activities. Many of the inputs to this system are from cognitive processes rather than perceptual mechanisms. •Phonological Store and Articulatory Loop: a store that is dedicated to spoken information and a sub-vocal rehearsal mechanism. •Visual-Spatial Sketchpad: a separate working memory component used for imagery tasks •Episodic Buffer: an area that binds information from subsidiary systems and Long-term memory into a single representation (so the different parts make sense) Copyright © Allyn & Bacon 2005 Psychological Support for Working Memory  Articulatory Loop and word-length effect (Baddeley et al., 1975) Calculate in your head 37 x 28__ may use visual image may rehearse verbally to retain it use central executive to remember task is multiplication and keep track of where you are wit, sum, harm, bay, top university, opportunity, aluminum, constitutional, auditorium Crucial issue is how long it takes to say the word (speed of processing, rather than # of chunks) Copyright © Allyn & Bacon 2005 Neuroscientific Support for Working Memory  Many investigators have seen evidence supporting the idea of a central executive    have observed higher cognitive activity in an area in the Dorsolateral Prefrontal Cortex Object WM  ―where‖ system  spatial location  dorsolateral prefrontal cortex  ―what‖ system  object recognition  orbitofrontal prefrontal cortex Phonological Loop  Broca’s Area (frontal cortex left hemisphere) responsible for production of speech Activation of Right Prefrontal Cortex associated with Visuospatial Sketchpad Activation of Left Prefrontal Cortex associated with Phonological Loop Copyright © Allyn & Bacon 2005 Support for Working Memory   The Delayed Response Task  food placed in well  barrier appears  animal must remember which well after delay “A not B” Task  Jean Piaget  modified delayed response task for human children Monkeys with lesions in the frontal cortex and infants under 1 yrs old (immature frontal cortex) cannot perform the task  neurons in the Dorsolateral Prefrontal Cortex fire only during the delay portion of the task– as if they are keeping information active Copyright © Allyn & Bacon 2005 Classic Information-Processing Model of Memory Based on Atkinson & Shiffron Model of Memory (1968) Copyright © Allyn & Bacon 2005 Long Term Memory  Relatively permanent memory storage  Capacity is essentially limitless  Permanence is essentially endless   Permastore: very long term storage of information In terms of retrieval, research shows recognition better than recall must be rehearsed/attended to!  Holds information transferred form STM/WM   Function:  hold all information that is not currently being used but that is potentially retrievable Copyright © Allyn & Bacon 2005 Long Term Memory  Classes of information stored in LTM (Bower, 1975)  Spatial information  Symbolic structure corresponding to images of our world and where significant objects are located in that cognitive map Our knowledge of the properties of objects and things Of ourselves, others and how to behave in various social situations  Physical laws   Beliefs    Values and social goals Motor skills  For driving, riding a bike, playing pool, etc. Also problem-solving skills for various aspects of life and our plans for how to achieve various things For understanding language, interpreting art and music, etc. Copyright © Allyn & Bacon 2005  Perceptual skills  Long Term Memory  Codes:  Auditory  We remember meaning over exact wording  Wanner (1968):  LTM is better for changes in wording that results in a change of meaning rather than just a change in word style  Memory for meaning is equally good whether people are warned or not Droodles –Bower et al. (1975)  Visual  Shepard (1967):   LTM capacity is greater for visual information in a picture/scene than for verbal/read information  People have better memory for meaning of a picture than meaning of sentences We remember meaning over details (interpretation of the picture rather than the exact picture itself)  LTM is better for changes in images that effect the meaning of the picture  Semantic (meaning-based)  Predominant coding used!  Memory for detail is available initially but forgotten rapidly, whereas memory for meaning is retained. Copyright © Allyn & Bacon 2005 Neuroscience of LTM Storage Memory functions are distributed throughout the whole brain-Long-Term memory is stored throughout the Cerebral Cortex! Possible explanation for infantile amnesia (lack of memories before around age 3-4 yrs) the regions of cortex for long-term storage may not develop fully until later in childhood Motor memory is stored in the cerebellum Copyright © Allyn & Bacon 2005 Transferring information from STM/WM to Long-Term Memory Memory Consolidation: Process by which chunks of memory re-structured to go into LTM for permanent storage Crucial Brain areas for Memory Consolidation= •Medial Temporal Lobe •Hippocampus •Entorhinal Cortex •Thalamus Copyright © Allyn & Bacon 2005 Transferring information from STM/WM to Long-Term Memory     Case of H.M. Damage to temporal lobe and hippocampus inability to transfer explicit memories from short- to longterm memory  Hippocampus crucial to consolidation of explicit memories However, he could learn implicit perceptual and motor skills and retain that information over time.  Many brain areas crucial to consolidation of implicit memories– but all stored in cerebellum or cortex Copyright © Allyn & Bacon 2005 Classification of Long-Term Memories  Neurological and psychological research suggest different types of memory are held in Long-Term Memory Non Declarative Memory (Implicit) Procedural Memory Motor Perceptual Cognitive Declarative Memory (Explicit) Episodic Memory Autobiographical Flashbulb Semantic Memory Schemas Scripts Priming Condition Non-ing Associative Semantic Repetition Classical Operant Habituation Copyright © Allyn & Bacon 2005 Classification of Long-Term Memories  Declarative Memories: ―Knowing That‖ – stored long-term facts about ourselves and the world    Are Explicit - consciously retrievable memories Crucial for formation = medial temporal lobe structures & thalamus (major sensory relay) 2 types  episodic   memory of instances or episodes from events in our lives  What did you wear to prom? Memory consolidation = hippocampus memory of facts and general knowledge  What is prom? Memory consolidation = entorhinal cortex Copyright © Allyn & Bacon 2005  semantic   Classification of Long-Term Memories  Declarative Memories:   Episodic Sub-types- Autobiographical: recollections people have of their own personal experiences and observations (People’s memories are most vivid for times of transition)  Flashbulb: Particularly good (vivid) memories for events that are very important or traumatic Semantic Sub-types- Schemas: A mental framework of interrelated concepts in a meaningful organization based of the type of objects that they are, the parts that they tend to have and their typical properties  Scripts: A type of schema for a particular event (stereotypic sequences of actions) Copyright © Allyn & Bacon 2005 Neuroscience of Long-Term Memories  Declarative Memories and consolidation:  Neuroscience research suggests that memories are stored in the activity and change in neurons (remember Hebb’s Cell Assemblies & Lashley’s Engrams)  Long-Term Potentiation (LTP) Tendency of nerve cells that have been exposed to a rapidly repeating stimulus to enhance their response tendencies for an extended period of time  Belief is that some chemical and/or structural change occurs and the memory is permanently stored Long-Term Declarative memories are believed to begin as the cortex sends information to the hippocampus, a process that strengthens the memory by rapidly and repeatedly exciting the neural circuit in the cortex (New evidence suggests happens during REM sleep!) Copyright © Allyn & Bacon 2005   Classification of Long-Term Memories  Procedural Memories: ―Knowing How‖ – stored longterm knowledge of learned habits and skills     Are implicit - out of awareness Can be motor, perceptual, cognitive or combination examples:  riding a bike, driving, tie shoes, reading, mental math) uses many cortical areas depending on type of memory  visual priming - visual cortex  motor programs - basal ganglia (mid-brain structure important for motor control), cerebellum (motor memory)  does not involve temporal lobe and no hippocampus association Copyright © Allyn & Bacon 2005 Modern Information Processing Model of Memory STM/WM Dorsolateral Prefrontal Cortex Orbitofrontal Prefrontal cortex LTM Temporal lobe & Thalamus Visual/Auditory/ Somatosensory Cortex Prefrontal cortex Hippocampus Declarative Entorhinal Cortex Episodic Buffer Dorsolateral Prefrontal Cortex Non Declarative Many-depends on type Broca’s Area Temporal Cortex Dorsolateral Prefrontal Cortex Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Practice (Rehearsal)  Each time we use a memory, it increases in strength– can reach a higher level of activation so can be retrieved more rapidly  Retention is increased through ―overlearning‖  Power Law of Learning= linear relationship repeated research shows: the more practice, the faster the recall but with diminishing returns  Research shows LTM is better when practice is spread over a long period of time – distributed vs. mass (called the Spacing Effect)  new rat evidence relates hippocampal re-activation during REM and learning (suggests more REM more activation of hippocampal more consolidation into LTM better learning) Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Depth of Processing (Craik & Lockhart, 1972)    Theory that states that rehearsal improves memory only if the material is rehearsed (practiced) in a deep and meaningful way– attended to, fully analyzed, enriched by associations or images  PROCESSING IS THE KEY TO STORAGE! Deeper processing in terms of meaning (semantics) leads to better memory than shallow processing looking only at surface form. Emphasizes semantics as a primary code/organization of information in LTM Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Depth of Processing is supported by research!  Subjects were shown lists of words and asked to use one of three strategies:    Visual: Is the word printed in capital letters? Acoustic: Does the word rhyme with _____? Semantic: Does the word fit the sentence _________?  The more thought involved (deeper processing through semantic rehearsal), the better was their memory. Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Elaborative Processing    Embellishing an item to be remembered with additional information improves recall Elaborative processing results in better memory even if the processing is not focused on the meaning of the material The Generation Effect-- Having the person elaborate on the information themselves yields better memory than just presenting the information to them (taps into self-schema/ narcissistic trait rich and extensive network available for storing self-information)  The Self-Reference Effect– People tend to recall more information that refers to themselves than about other information Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Context    Encoding Specificity Principle: any stimulus encoded along with an experience can trigger one’s memory of the experience Context-Dependent Learning: easier to retrieve information from memory when in the same situation as when the information was first encoded (environment) State-Dependent Learning: easier to recall information if state of mind is the same as when the information was first encoded (mood) Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Incidental vs. Intentional Learning  Repeated research shows:    Whether a person intends to learn/later recall does not matter! What matters is how the person processes the information at its presentation People do tend to remember more when they intend to learn because they tend to engage in other activities   Rehearsal Elaborative processing Copyright © Allyn & Bacon 2005 Factors that effect Retrieval from Long-Term Memory  Memory is Constructive! Not simply a matter of consuming facts about the world and regurgitating them:  Prior experience, post-event information, perceptual factors and desire to recall certain events over others effect LTM  Research shows people have tendencies to distort memory:  Eye Witness testimony is susceptible to suggestion  Misattribution: recall of what we think we should have seen, rather than what we actually saw   Eye Witness Transference (mis-perception of an event where witness replaces one person with another in memory) Bias in recall  Tend to distort episodic memories more than semantic memories  we use more of our semantic memories in real-life situations--refresh your semantic memory (strengthen memory trace through activation) plus most semantic memories fit into pattern & corrected by others Copyright © Allyn & Bacon 2005 Long-Term Memory Forgetting  WHY???     Lack of Encoding Decay Interference Repression Information not stored in LTM Memory may exist but difficult (impossible) to retrieve Copyright © Allyn & Bacon 2005 Long-Term Memory Forgetting Can You Recognize a Penny?  One reason people forget is due to lack of encoding. Copyright © Allyn & Bacon 2005 Long-Term Memory Forgetting  Decay Theory:   Memory traces simply weaken in strength over time Some neurological evidence to explain decay– Long-Term Depression (LTD) = decrease in neural responsiveness Copyright © Allyn & Bacon 2005 Long-Term Memory Forgetting  Interference: Learning additional associations to a stimulus can cause old ones to be forgotten (lost)  Proactive Interference  The tendency for previously learned material to disrupt the recall of new information  Swimmer has problem learning new stroke because close to well-known stroke The tendency for new information to disrupt the memory of previously learned material  Difficulty skiing because recently learned to snowboard  Retroactive Interference  Copyright © Allyn & Bacon 2005 Long-Term Memory Forgetting  Repression:   Freud says that when we have memories, impulses, desires, and thoughts that are too difficult or unacceptable to deal with, we unconsciously exclude them from our consciousness (some people like to say we "push" them down from our consciousness to our unconsciousness). Not universally accepted (because of issues of suggestibility) Copyright © Allyn & Bacon 2005 Amnesic Effects of Brain Damage  Amnesia: Temporary or permanent loss of memory  depending on location of brain damage, could be an issue of  Anterograde Amnesia– can’t form new LTM (ex. Hippocampus damage) or  Retrograde Amnesia– can’t remember past events (damage to cortex where stored or area needed for retrieval (such as thalamus, cerebellum, etc.) Copyright © Allyn & Bacon 2005 Amnesic Effects of Brain Damage  Amnesia due to temporal lobe damage   case of H. M. - bilateral removal of medial temporal lobes (including hippocampus) to control for epilepsy inability to transfer declarative memories from short- to long-term memory  Amnesia due to thalamic damage   case of N. A. - fencing accident damaged left part of thalamus inability to learn new verbal material - can learn faces and locations Copyright © Allyn & Bacon 2005 Amnesic Effects of Brain Damage  Amnesia due to Korsakoff’s Syndrome      disease of chronic alcoholics related to loss of vitamin B1 suffer both anterograde amnesia (can’t form new memories) and retrograde amnesia (amnesia for past events) persevere in useless strategy that worked before show no improved learning when released from proactive inhibition (old memories interfere with recall of new) damage to thalamus, cerebellum, cerebral cortex (frontal lobes) Copyright © Allyn & Bacon 2005 Amnesic Effects of Brain Damage  Amnesia due to Electroconvulsive Treatment (ECT)    shock used to treat severe depression causes loss of memory, particularly for recent events (LTM intact) ECT induces seizures in temporal lobe - probably disrupts hippocampus thalamic damage (N.A.; Korsakoff’s) leads to normal rates of forgetting hippocampal and amygdala damage (H.M.; ECT) leads to rapid rates of forgetting  Differences in rates of forgetting   Copyright © Allyn & Bacon 2005