Metamemory: Knowledge and awareness of one's own memory processes and abilities
Egypt 4000 B.C. Thoth: god of learning, memory and wisdom
Greece 1000 B.C. Mnemosyn: goddess of memory
First text on improving memory: Dialexis 400 B.C.
The three most important aspects of memory are: remembering to pay attention, to rehearse and to use a formal mnemonic device (the method of loci)
Plato (427 - 347 B. C.) All essential truths are stored in memory and learning is the process of recollecting these truths.
* Wax tablet model: Memory is like impressions formed on a wax tablet
* Aviary model: Memories are like different species of birds.
* Scribe model: A personal scribe who records the events of ourlives
Aristotle ( 384-322 B. C.)
Recollection is based on associations
Associations are connections of mental events
Three laws of association: events tend to be associated when
* they occur together in time or space (law of contiguity)
* they are similar (law of similarity)
* they contrast with each other (law of contrast)
Cicero (106-43 B.C.) and Quitiliian (40-96 A.D.)
Emphasis on the role of memory in oratory:
The order of remembering is important
Physical and mental exercise helps
Augustine (354-430 A. D.)
Remembering is like exploring caverns.
Illusions of memory are possible when "we fancy we remember as though we had done or seen it, when we never did or saw at all".
Scientific approach to memory
studies his own learning of nonsense syllables
buv, weq, rol
Controlled: time of study, interval between study and recall
Measured amount of retraining needed based on:
interval, size of list, amount of studying
General theoretical apporaches to memory:
Biological: factors involved in the physiological structure and formation, storage and retrieval of memories
Associationism/connectionism:
Information-processing/cognitive: Memory viewed in terms of how information is processed.
Multimodal: Many different influences on memory (strategies, type of material, motivation, emotions, physiological factors, age)
Naturalistic Observation
Observing information in a non-experimental setting
When do people forget things? What factors are involved?
* Data from naturalistic observation.
* Rich, but hard to analyze.
Hypothesis: Students will remember more if I say everything twice Students will remember more if I say everything twice
Independent Variable: Variable that is manipulated to test the hypothesis.
Dependent Variable: Variable representing the behavior we want to measure
Control Variables: Other variables we need to control in order to see the effect of the independent variable
Subjects: Who is going to participate in the experiment?
Analysis: How do we know if there are differences between the manipulation and the control group?
How do we go about studying memory?
* Registration: transforming information into a form that can be
retained
* Retention: storage of information
* Remembering: Retrieval of the information
Implicit vs. explicit remembering
* Explicit: information explicitly remembered
* Implicit remembering: No recall, but still influences performance
Independent variables in Memory research
Organismic variables
The relativiely permanent characteristics of a person that affect memory performance
(ability to pay attention, motivation, intelligence, etc.)
Antecedent variables
Variables that alter a person's typical organismic level
(drugs, external motivators, sleep)
Task variables
1) Instructional variables
Implicit or explicit instructions given to the subject
(study hard, use a mnemonic)
2) Presentational variables
How the stimuli are presented
(Time of presentation, presented verbally or written)
3) Stimulus variables
Differences in stimuli used
Familiarity (green vs. chartreuse)
Frequency of occurence (home vs. abode)
Concreteness (dog vs. faith)
Other features: shock, meaningfullness
4) Context
place, mood can play a role in encoding and recall
Primary measures
Amount of information recalled
How likely is it that a subject guessed?
Secondary measures
Quality of information recalled
Speed of recall: reaction time (RT)
Serial learning: Learning to recall information in a particular order
Free recall learning: learning to recall information in any order
Cued recall: Recall given a cue
Paired-Associate Learning: Learn stimulus + response
dog-cat dog-plant dog-zik duck-pato duck-canard
Tests
Recognition: Yes/no Did you hear bed?
Which did hear?
street
pencil
toe
book
Study half the list:
Then see the whole list and decide if you way it or not
On the list
|
Not
on the list
| |
| Subject
claimed seeing it
|
Hit
|
False
alarm
|
| Subject
claimed not seeing it
|
Miss
|
Correct
rejection
|
Testing implicit memory
Amnesic subjects don't recall words from the list, but if you do a word completion
Fill in the rest of the word Ch____
Processes incoming information from the environment
* Individual sensory stores for each sense
* Information retained for a short duration
The visual sensory store (Iconic memory)
Experiments by Sperling (1960)
X M R K
C N J P
V F L B
The partial report technique
Auditory sensory store (echoic memory)
Experiment by Darwin, Turvey & Crowder (1972)
3 digits or letters auditorally presented to each ear and center at the same time
1) It maintains information long enough so that we can do additional processing to it.
2) Gives us time to choose what sensory information to give more attention to
3) Keeps information active long enough to get us to notice it
4) Provides perception of continuity in our environment
STM used to describe the fact that it holds information for a short time, while working memory refers to the processing capacity.
STM works as a temporary holding place for intermediate decisions.
Limited in size.
Chunking
Limited Time
Code: Acoustic
Working memory: there is a limited amount of processing capacity that you can use as you perform a problem
Brown-Peterson Distractor task
Task:
1) give 3 random letters and a 3 digit number.
2) Have subjects count backwards from the number by threes for X seconds
3) Recall
L W C 455
1
|
|||
| 2
|
|||
| 3
|
|||
| 4
|
|||
| 5
|
|||
| 6
|
|||
| 7
|
|||
| 8
|
|||
| 9
|
|||
| 10
|
|||
| 11
|
|||
| 12
|
|||
| 13
|
|||
| 14
|
|||
| 15
|
|||
| 16
|
Delay 6 Total # correct: ______
Delay 9 Total # correct: ______
Delay 12 Total # correct: ______
Delay 3 Total % correct(number correct/12): ______
Delay 6 Total % correct (number correct/12): ______
Delay 9 Total % correct (number correct/12): ______
Delay 12 Total %correct(number correct/12): ______
1
|
T
C X 584
|
3
| |
| 2
|
V
J B 832
|
6
| |
| 3
|
L
Q Z 387
|
9
| |
| 4
|
P
R H 256
|
12
| |
| 5
|
M
F S 492
|
3
| |
| 6
|
Y
Z P 217
|
6
| |
| 7
|
W
B T 456
|
9
| |
| 8
|
G
D V 782
|
12
| |
| 9
|
K
H W 353
|
3
| |
| 10
|
Z
Y W 587
|
6
| |
| 11
|
T
P G 548
|
9
| |
| 12
|
C
F S 772
|
12
| |
| 13
|
R
X W 232
|
3
| |
| 14
|
N
T B 663
|
6
| |
| 15
|
H
G D 924
|
9
| |
| 16
|
F
C L 345
|
12
|
Delay 6 Total # correct: ______
Delay 9 Total # correct: ______
Delay 12 Total # correct: ______
Delay 3 Total % correct(number correct/12): ______
Delay 6 Total % correct (number correct/12): ______
Delay 9 Total % correct (number correct/12): ______
Delay 12 Total %correct(number correct/12): ______
Sternberg Serial Scanning task
Example:
1) Memory set presented: 4, 8, 2, 5, 9
2) Subject encodes set into STM
3) Probe item presented: 5
4) Subject presses yes or no key
Implications:
1) We sequentially compare probe items
Evidence from the time increasing as a function of the number of items
2) We use exhaustive serial search.
Evidence in that there is no difference between the Yes and No responses, although, subjects could terminate the search once they see a Yes response.
Knowledge of: how to do things, things we have learned, grammar rules, personal memories.
All knowledge that is not active.
Information that becomes active is retrieved from LTM and put in STM.
Almost everything we learn is first processed in STM and some of it is put into LTM
Demonstration: The Serial Position Curve
Recency effect, Primacy Effect
Recency effect: Last items heard/seen are recalled well
Primacy effect: First items heard/senn are recalled well
Why?
* Recency effect: Due to STM
* Primacy effect: Due to LTM
There may be a single mechanism that could explain both STM and LTM
Some evidence
Evidence from Amnesia:
Evidence in Anterograde amnesia for patients with LTM intact, but inability to consolidate new information from STM into LTM.
Some patients have defective STM, but fine LTM.
Semantic Networks
Words can be represented as an interconnected network of sense relations
* Each word is a particular node
* Connections among nodes represent semantic relationships
How do these pieces of semantic information relate to each other?
Semantic verification task
An A is a B
An apple is a fruit
A robin is a bird
A robin is an animal
A dog has teeth
A fish has gills
A fish has feathers
An apple has teeth
NMSU is in New Mexico
Harvard is in California
Use time on verification tasks to map out the structure of the lexicon.
Feature comparison model (Smith, Shoben and Rips (1974)
Information (concepts) are stored as a set of features:
* Defining features (those features that it must have)
* Characteristic feature (those feature that it often has)
What are the features of a bird?
Comparing concepts through comparing features
Is a robin more like a swallow or or a penguin?
Lexical entries stored in a hierarchy, with features attached to the lexical entries
Representation permits cognitive economy
Testing the model
Sentence Verification time
Robins eat worms 1310 msecs
Robins have feathers 1380 msecs
Robins have skin 1470 msecs
A category size effect: Subjects do an intersection search
Problems with Collins and Quillian model
1) Effect may be due to frequency of association
2) Assumption that all lexical entries at the same level are equal
The Typicality Effect
Which is a more typical bird? Ostrich or Robin.
A whale is a fish vs. A horse is a fish
Major conclusions of the model:
1) If a fact about a concept is frequently encountered, it will be stored with that concept even if it could be inferred from a more distant concept.
2) The more frequently encountered a fact about a concept is, the more strongly that fact will be associated with the concept. And the more strongly associated with a concept facts are, the more rapidly they are verified.
3) Verifying facts that are not directly stored with a concept but that must be inferred takes a relatively long time.
* Words represented in lexicon as a network of relationships
* Organization is a web of interconnected nodes in which connections can
represent:
categorical relations
degree of association
typicality
Retrieval of information
* Spreading activation
* Limited amount of activation to spread
* Verification times depend on closeness of two concepts in a network
Context effect in spreading activation models
Present either: Murder is a crime or Libel is a crime
Then get verification time for Robbery is a crime
Subjects faster when they see Murder than Libel. Why?
Advantages of Collins and Loftus model
* Recognizes diversity of information in a semantic network
* Captures complexity of our semantic representation
* Consistent with results from priming studies
Semantic Priming
Meyer & Schvaneveldt (1971) Lexical Decision Task
Prime Target Time
Nurse Butter 940 msecs
Bread Butter 855 msecs
Evidence for associative spreading activation
Ratcliff and McKoon (1981)
Subjects study and memorize The doctor hated the book
Task: "Was this word from the sentence you memorized?"
Prime Target Time
None Book 667 msecs
Doctor Book 624 msecs
Sensory store
STM
LTM
Duration
500-2000ms
<30
Sec
minutes
to years
Format
(code)
Literal
copy
Acoustic
Semantic
Capacity
medium
(9 visual items)
small
(~5-7 chunks)
unlimited
Rehearsal
process
not
possible
maintenance
elaborative
Retrieval
process
Serial
Parallel
Explicit vs. incidental learning
Selective processing of information
We do not encode all information from stimuli
Effects of expectations
We encode information based on how we believe we will need to use it?
Effects of expections example: Frost (1972)
Subject's task: Encode 16 pictures
Instructions:
Group 1) You will need to recognize the actual object
Group 2) You will have to recall the names of all objects seen
Results:
Subjects in group 1 faster at responding to exact same drawings
Subjects in group 2 did not encode exact pictures
Conclusion: Evidence for explicit encoding strategies
Encoding of information in STM and LTM
Encoding of verbatim material vs. gist
Verbatim representation is lost very quickly (within a few seconds)
Johnson, Bransford & Solomon (1973)
John was trying to fixthe bird house. He was looking for the nail when his father came out to watch him and to help him do the work.
Subjects have false recognition on:
John was using a hammer to fix the bird house when his father came out to watch him and to help him do the work
We make automatic inferences
Does that mean we can't encode verbatim information into LTM?
Attention: we encode only what we pay attention to:
Incomplete sensory processes: Not all information gets in through the sensory processes
Motivation; We encode what we want/need to encode
Biases in encoding based on interests/background
Improving encoding
Organization: Organize the information in a meaningful way
Ellis & Hunt (1993): Items organized (blocked) in categories produced better recall than those not in categories
Bousfield (1953), Tulving (1962) subjects will impose structure or categories on information to help with encoding.
Chunking: Chunk information into meaningful units
Give examples of memory experts. S. F., Rajan
Warmup: Practice with similar material.
Distributed practice
The Spacing Effect and Spaced vs. Massed practice
Overlearning increases speed of access, probability of correct retrieval
Large knowledge base: Encoding depends on having the appropriate prior knowledge
The llama's hemoglobin has a left-shifted oxygen-binding curve, which means that it can become 100 percent saturated with O2 at the low P02 values at high altitudes.
What "monetarists" do believe is that controling the behavior of M will help much to control GNP - for the reason that the changes in V will be so predictabe as to make one confident that dollar GNP will still move in the same direction as M.
Ebbinghaus found that the spacing of practice is a potent task variable, with a brief exposure providing for minimal encoding operations, but continued exposure allowing for more and deeper encoding.
Meaning: Encode information so that it ties into other parts of your knowledge.
Levels of processing emphasizes one memory store
Conscious portion of memory called primary memory
Assumption: All incoming stimuli can be processed at different levels of analysis
The strength of a memory association is based on how "deep" it is encoded.
Maintenance vs. Elaborative rehearsal
* Maintenance rehearsal: Maintain information in primary
memory
* Elaborative rehearsal: Deeper analysis of the information.
Develop meaningful associations
Evidence for levels of processing
Incidental learning task
Rate the words on this feature:
Structure: Is it in capital letters?
Phomemic: Does it rhyme with bag?
Pleasantness: Is this a pleasant word?
Semantic: Does the word fit into this sentence:
John went to the store and bought a ___.
Self Reference: Does this word describe you?
Then surprise recall test.
Generating information results in better retention than just studying it.
Subject generates rhymes to words:
TAME -----> NAME
BANK ----> RANK
FABLE -----> TABLE
....
other subject studies same list of pairs
Recall of second list of words is better for subjects who generate list than those who just studied list
Effect for math problems as well (Crutcher & Healy, 1993)
Why is there a generation effect?
Generating items results in deeper processing
Implications for studying: Better retained if you have to generate material (e.g., paraphrasing, putting in your own words)
Hyopthesis 1: Elaboration: We generate a rich elaborate encoding with many associations
Hypothesis 2: Distinctiveness: Deep processing makes that memory trace more distinctive from others.
The von Restorff effectl Items that stick out are easier to remember.
Criticisms of levels of processing
1) What do we mean by depth? Hard to measure depth
2) Some evidence that maintenance rehearsal leads to long-term retention of information (Nairne 1983)
3) Transfer-appropriate processing
Words encoded for rhymes are better recalled than words encoded semantically if a subject is asked to use them for a rhyming-related recall task.
Moral: The best type of encoding for information is to encode information in the way that you will need to retrieve it.
Task: Encode the following 32 digits
23157548590183725993762497088695
Why do we forget?
* Failure to store/encode
* Decay of information in memory
* Failure to retrieve information from memory
Do we have permanent memory?
Ebbinghaus' Forgetting Curve
Ebbinghause studied 1200 nonsense syllables (13/list)
* Learned first to two correct recitations
* Measured how much he retained (savings)
* Examined the number of trials/time to relearn list
* Looked at different retention intervals
Decay of recall follows a power function
Does forgetting curve hold for other types of information?
* Meaningful vs. meaningless information
Permastore: a relatively permanent storage of knowledge
Bahrick (1984) Retention of unused information over long periods of time
* Memory for vocabulary from high school/college spanish courses
* 773 students over 50 years
* Measured performance as percent of original score
Linton (1975) recall of autobiographical memories.
Why? are autobiographical memories recalled better?
Retention of non-declarative memories
Declarative vs. non-declarative memory
Non-declarative memory: not accessible to consious awareness
Declarative memory: facts that can be learned, often in one trial
Not as much loss in non-declarative memories.
It can depend on number of trial/ptractice, though
Organic disorders
Disease, damage, drugs
Disuse (Decay Theory)
Thorndike (1914) law of disuse: Unless memory is attended to, it decays
New theory of Disuse(Bjork & Bjork, 1992)
Assumptions:
* No limit on how much information can be stored but... limits on ability
of retrieving information
* Information becomes increasingly inaccessible unless periodically
retreived (even if well learned)
Forgetting is due to new information competeing to come to mind over the old information.
Approach is biologically adaptive.
We tend to need new information, not old information.
Interference Theory
Intervening information/events interferes with previously learned information
Recall of information before/after sleep (Jenkins & Dallenbach, 1924)
* Studied nonsense syllables then recalled after full 8 hour day, or 8 hour
sleep
Two interpretations:
* 1) Sleep helps with consolidation of information
* 2) Events during the day interfere with recall
Other evidence for consolidation:
1) Effects of REM sleep
2) Experiment on rats and amnesia Duncan (1949)
Trained rates on using avoidance procedure.
Varied interval between training and delivery of electroconvulsive shock (ECS)
The longer time between learning and ECS, the better learned.
Types of interference: Retroactive vs. proactive interference
Retroactive: new info interferes with old info
Proactive, old info intereferes with new info
Effects of stimulus similarity on interference
There is a distinction between whether something is accessible or available
Memories in LTM are often available, but not always accessible
Evidence for accessibility/availability distinction
1) Motivated forgetting
When we want to make a memory less available
Types of motivated forgetting:
Supression: Consciously tries to forget information
Laboratory experiments: of directed forgetting (Bjork, LaBerg & Legrand)
Represession: Unconsciously lessens accessibility of information
Freud's view
Dissociative disorders
2) Cue-dependent forgetting
Memory failures can be due to not having the right cue
Example: Tulving & Pearlstone (1966)
Two groups of subjects learn list of words from categories
Furniture: chair, bed, table, sofa, ...
Fruits: banana, apple, orange, kiwi, ...
One group given category name before free recall, the other not.
Result: Improved recall if given category name
Implications: Items were probably available to both groups, but without category name, they were not accessible.
Suggests that proper cues help recall.
A person's failure to recall information that the person knows he/she knows.
What is the word designating the small boat used in the river and harbor traffic in China and Japan?
Evidence for availability but not accessibility
Encoding specificity principle: Information will most likely be retreived if it is paired with the same contextual cues as when it was learned. (Tulving, 1974)
Similar to transfer- appropriate processing: The most appropriate encoding for information is to encode it in the way that you will need to retrieve it.
Evidence: Light & Carter-Sobell (1970)
"The STRAWBERRY JAM tasted great"
S's told they would have to recognize underlined phrases
Recognition test, did you see this noun before?
STRAWBERRY JAM recognized 65% of the time
TRAFFIC JAM recognized 27% of the time
Other context effects
* Abernath (1940) Subjects who study in the same classroom, do
better on tests
* Schab (1900) subjects study in room with or without chocolate odor
24 hours later, subjects perform better on surprise test with matching odor
* Godden and Baddeley (1975) Underwater vs. above water context effects
State-dependent memory
Mood-dependent memoryRetrospective and Prospective Memory
Prospective memory: ability to plan/remember future actions
* Take a pill at 5:00
* Take your umbrella today
* Call you mom and say that you got home safely
There tends to be a zero or negative correlation between prospective and retrospective memory performance
Difference of remembering what vs. when.
Types of Prospective memory
Habitual vs. Episodic
Much of memory requires use of both prospective and retrospective memory
What factors affect prospective memory?
* Having appropriate cues and reminders
* Compliance, motivation and commitment