Episodic LTM -- Ch 61

Document Sample
Episodic LTM -- Ch 61 Powered By Docstoc
Long-Term Memory

    Chapter 6
 So what are we doing today?
 Types   of LTM

 Ebbinghaus

 Mnemonic   Devices

Sensory   Attention   Short-Term   Long-Term
Memory                 Memory       Memory
             Short-Term      Long-Term
              Memory          Memory

   Encoding: Getting info into LTM
   Retrieval: Getting info out of LTM
   Which is important for good memory?
Taxonomy of LTM (Squire, 1993)

     Explicit vs. Implicit Memory
   Explicit (Declarative): Knowledge that
    can be retrieved and then reflected on
    with conscious awareness

   Implicit (Nondeclarative): Knowledge
    that can influence thought and behavior
    without conscious awareness
    Episodic vs. Semantic Memory
   FACTS
       General world knowledge
       Stores concepts and categories (encyclopedia)
       e.g., what is a maiden name?

   EVENTS 
       Autobiographical memory
       Stores personally experienced events
       e.g., what is your mother’s maiden name?
                Ebbinghaus (1885)
   First scientific research on memory
       He was his only subject
       Learned over 1,200 lists!

   Studied nonsense syllables
       Excluded meaning to study the fundamentals
       CVC (Constant, Vowel, Constant)
            VAR, BEW, NIN, LOD, FAP, CAS
   Knowledge about one’s own memory, how
    it works and how it fails to work
       Ebbinghaus: Mnemonic and rehearsal
        processes take time so he presented the lists
        at rapid rate (2.5 items/second) to prevent
        rehearsal or mnemonic strategies

   Metacognition: Knowledge about one’s
    own cognitive system and its functioning
               Ebbinghaus (1885)
   Relearning Task
       A list is learned, set aside for some time, and then
        relearned to the same level of accuracy
       20 min, 1 hr, 9 hr, 1 day, 2 days…
   Savings Score
       The reduction in the number of trials necessary for
        relearning (compared to the number of trials for
        original learning)
       (t1 - t2) / t1        (10 - 6) / 10 = .40
The Forgetting Curve
            Ebbinghaus (1885)
   Forgetting is rapid at first
   In other studies…Repetition effects
     32 reps vs. 64 reps
     More frequently repeated lists show more
      savings upon relearning
     Over-learning prevents forgetting
          Evaluating Ebbinghaus
   Human memory relies heavily on meaning
   Participants are active learners  apply
    resources and strategies
   Ecological validity
       Results with meaningless stimuli are not
        representative of real-world situations
           Mnemonic Devices
   A strategic learning device or method  a
    rehearsal strategy

   Which ones do you use?
              Mnemonic Devices
   The Method of Loci:
     Memorize physical locations that can be
      recalled in order
     Create mental images of the to-be-
      remembered items in each location
     The more distinctive the image the more
         E.g.,   flag pole in the shape of a guitar
 The Peg Word Technique
 a pre-memorized set of words serve as mental “pegs”
on which the to-be-remembered material can be “hung”
     Three Mnemonic Principles
1.   Provides structure for learning the
     material by linking it with info already in
2.   By means of visual images and other
     associations they help form durable and
3.   Gives effective cues for recalling the
          Mnemonic Devices
   Time vs. Accuracy
     Trade  off
     Effective but time consuming
    Storage in Episodic Memory
   How do we store info in episodic memory?
       Get information into the system

   Factors that improve encoding:
     Rehearsal
     Depth of Processing
     Organization
     Imagery
   A deliberate recycling or practicing of
    STM’s contents
   Two effects of rehearsal:
     Maintains info in STM
     The longer info is held in STM the greater
      probability that it will be encoded into LTM
         Evidence   of this transfer function?
   Hellyer (1962)
       CVC task with a distractor task between study
        and recall
       CVCs spoken out loud 1, 2, 4, or 8 times
          Rehearsal
                                       Hellyer (1962)
                        The more rehearsed the item, the better it was retained

Probability of Recall




                                  0         3           9            18   27
                                                Retention Interval
Serial Position Curve
             Serial Position Curve
   Primacy Effect:
       Better memory for the first items in a list
       A LTM effect
       First items in a list get the most rehearsal

   Recency Effect:
       Better memory for the last items in a list
       A STM effect
       Last items still in STM when recall
            Your Data from Coglab
                         Serial Position

% Correct

                 1   2    3    4   5   6   7     8   9 10
                              Position in List
   Rundus (1971)
       20 item list (5s per word)
       Rehearse out loud (could see which words
        were rehearsed the most)
       Primacy effect: The items that got the most
        rehearsal (early items) were recalled better
     Levels of Processing (LOP)
   Memory is determined by how the person
    processes it
       NOT by how long it stays in the system

   Memory is a by-product of thinking  more
    thinking, better LTM
       Shallow processing leads to weak LTM
       Deep processing leads to strong LTM
          Two Kinds of Rehearsal
   Maintenance (Type I)
       Low-level, repetitive information recycling
       Once you stop rehearsal, it leaves no
        permanent record
       E.g., holding a phone # in memory until you
        dial it
         Two Kinds of Rehearsal
   Elaborative (Type II)
       More complex rehearsal using the meaning of
        the information to store it

       Stored more deeply and is permanently
        available for retrieval from memory
       E.g., imagery or mnemonic devices
         Which would be better for Midterm #2?
                Depth of Processing
Does maintenance rehearsal lead to LTM storage?

   Craik & Watkins (1973)
       Long list of words – remember the last word
        that began with a “g”
       Varied the number of items between the “g”
          Ifmaintenance rehearsal contributes to LTM then
           subjects should have better memory for words that
           were rehearsed more
             Depth of Processing
   Craik & Watkins (1973)
       Surprise recall of ALL the words at the end
       The amount of time an item had been in STM
        did NOT effect the subject’s recall
          Time   in STM only predicts later performance when
           the items have been elaborately encoded
          Supports LOP theory
Challenges to Depth of Processing
   Glenberg et al. (1977)
       Brown-Peterson task: Encode a number then
        recite words as a distractor task
       Recall tasks do not show any benefits of
        maintenance rehearsal
       Recognition tasks do show benefits of
        maintenance rehearsal
Challenges to Depth of Processing
   Maintenance rehearsal still leads to memory
    (just not as good)
   Doesn’t explain amnesia
       Good STM but no LTM

   A circular explanation for performance
       Higher recall then elaborate rehearsal was used
       What is deep processing?...anything that leads to
        better memory
             Why Does it Work?
   Deep processing definitely leads to better
   Why?
       Establishing stronger connections to existing
        memory structures
       Provides more and better retrieval
   The structuring or restructuring of
    information as it is being stored in memory
   Bousfield (1953)
       Randomized 60-word list for free recall
         4   categories (15 words each)
       Subjects recalled the words by category
          Cow,   dog, cat, bean, carrot, peas…
       Implies the operation of an organizing
        tendency during rehearsal
   Organization’s relationship to chunking?
   Bower et al. (1969)
     Presented words in random order or in
      hierarchical order
     112 words total
     4 trials to learn all the words
                                        Bower et al. (1969)
Average % of Words Recalled


                              40                                 Random


                                    1       2            3   4
   What about the organization of unrelated
   Tulving (1962)
     Same list of words is presented across
      several trials in different orders each time
     Looked for regularities in recall orders
         E.g.,   dog, lawyer, apple, brush…
   Even unrelated words become organized
    because subjects mental activity imposes
   Subjective Organization: Organization
    developed by the subject for remembering a list
    of items without experimenter-supplied
                 Visual Imagery
   The mental picturing of a stimulus that affects
    later recall or recognition
   Method of Loci
   Schnorr & Atkinson (1969):
       Subjects studied paired associates (dog-book) by
        forming visual images OR by repetition
       Imagery group was better remembering the second
        word (book) when cued with the first word (dog)
    Paivio’s Dual Coding Hypothesis
                            (Paivio, 1971)

   Concrete words (cigar, truck) can be stored
    twice in LTM
       as a word
       as a visual image

   Concrete words are remembered better than
    abstract words (justice, idea) because they can
    be retrieved in two ways
       abstract words can be retrieved only one way
Retrieving Episodic Information
   Decay or Interference?
       Decay – the older the memory, the more likely
        that it is forgotten
          Is   this true of LTM ?

       Interference – intervening activities disrupt
          Is   this true of LTM ?
              Retrieval Failure
   Cognitive psychology’s new theory of
    forgetting (loss from memory)
   There is no genuine forgetting from LTM
   Instead forgetting is a failure of retrieval
   Information is not lost from memory, it is
    lost in memory
     Tulving & Pearlstone (1966)
   Same task you just did
   Groups were treated identically until recall
       Assume both had acquired the same of info and
        retained the same amount in memory
   Free recall: 40%      Cued Recall: 62%
   Free recall group only recalled a portion of what
    they learned – lots of info available just not
        Availability vs. Accessibility
   Availability: The memory trace exists in LTM
       Like a book on the shelf

   Accessibility: Degree to which the memory
    trace can be retrieved from memory
   Retrieval failure occurs when the information is
    available, but not accessible
       A memory is lost in the system
              Retrieval Failure
   Tip of the Tongue (TOT): Temporarily
    unable to remember some information
    (e.g., a person’s name) that they know is
    stored in LTM
     Usually still have access to partial information
     E.g., Starts with, approximate length…
              Encoding Specificity
                   Tulving & Thompson (1973)

   Each item is encoded into a richer memory
    representation, one that includes any extra
    information about the item that was present
    during encoding
       E.g., “cat”  also encode “animal”

   Anything present during learning can serve as
    an effective retrieval cue for remembering
       E.g., “animal” is a cue for “cat”
Retrieval Cues & Encoding Specificity

   The way to increase access to information
    in memory is to reinstate the original
    learning context
       Congruence between current test conditions
        and the conditions during acquisition
       Access is increased by effective retrieval cues
Retrieval Cues & Encoding Specificity

   Thomson & Tulving (1970)
       Learn a list of words (uppercase)
       Cue words in lowercase
          Highassociates (e.g., hot – COLD)
          Low associates (e.g., wind – COLD)

       Tested with high or low associates
                  Encoding specificity predictions?
Retrieval Cues & Encoding Specificity

   Thomson & Tulving (1970)
       High associate cues benefited recall when
        high associates were given during learning
          STUDY:   hot – COLD  CUE: hot

       High associate cues did not benefit recall
        when low associates were given during
          STUDY:   hot – COLD  CUE: wind
Retrieval Cues & Encoding Specificity
   Thomson & Tulving (1970)
       Low associate cues benefited recall when low
        associates were given during learning
          STUDY:   wind – COLD  CUE: wind

       Low associate cues did not benefit recall
        when high associates were given during
          STUDY:   wind – COLD  CUE: hot

               Encoding specificity can even
                override existing associations
    Model of Memory

   The more thoroughly you encode something
    the more strands it will be connected to in the
   The more strands the easier the retrieval  if
    one strand is weak you can still reach the
    memory from another route
   Permanent loss of memory or memory abilities
    due to brain damage or disease
   Affects only conscious (explicit) episodic
    memory (recall and recognition)
       Working memory intact
       Semantic memory intact
       Implicit memory intact: Skill learning, Word-stem
        completion, Repetition priming
               Psychogenic Amnesia
                   (soap opera amnesia)
   Uncommon
   Sudden onset following physical or psychological stress
   Most commonly, personal information is forgotten
       Sometimes the patient is unable to recall anything about his or
        her past

   During the period of memory loss cognitive skills are
    entirely intact
   Recovery is usually complete with no residual memory
    impairment (within 48 hours)
              Amnesia (organic)
   Retrograde: Loss of memory for events before
    the injury
       Most common with neural degenerative disorders
        (e.g., Alzheimer's)

   Anterograde: Loss of memory for events after
    the injury
       Caused by damage to the hippocampus and temporal
        Explicit vs. Implicit Memory
   Controls
     Bonnebakker (1996)
     Jacoby & Dallas (1981)

   Patients with Amnesia
     Clive Wearing
     Scoville (1953)
   Hippocampus removed on both sides
   Attention, STM and Retrieval of LTM for events
    before the surgery were intact
   IQ 118
   Anterograde amnesia
   Implicit memory intact
       Mirror-drawing

Sensory   Attention    Short-Term   Encoding
Memory                  Memory                      Memory
      Taxonomy of LTM (Squire, 1993)


Semantic   Episodic
   Hippocampus is critical for transfer of
    explicit memories from STM to LTM
     Gateway to LTM
     Not a site of storage

   Amnesia is not due to injury of brain
    regions which process or store info
                    Implicit Memory
   Repetition Priming
       A previous encounter with info facilitates later
        performance, even unconsciously

   Jacoby & Dallas (1981)
       Subjects saw a list of words and answered questions
        about each word (manipulate LOP)
            Physical: “Does it contain a L”
            Sound: “Does it rhyme with purse”
            Semantic: “Is it an animal”
          Jacoby & Dallas (1981)
   Some subjects took an explicit memory test:
       “Yes or No, was this word in the list you saw?”

   Some subjects took an implicit memory test:
       The words were flashed briefly on a computer screen
        (e.g., 35 ms) and subjects reported the word they saw
Jacoby & Dallas (1981)
        Jacoby & Dallas (1981)
   Measures of explicit memory are
    sensitive to how the information is
   Measures of implicit memory show
    priming, regardless of how the information
    was processed
Semantic vs. Episodic Memory
   Patient K.C.
     Frontal lobe damage
     Episodic memory disrupted
         Does  not remember any autobiographical info from
          before or after his accident
         Retrograde and anterograde amnesia
       Semantic memory intact

        Dissociation between semantic and
            episodic memory systems
      Semantic & Episodic Memory

                       Summer vacation
                       50-years ago

History of astronomy

Shared By: