Transcript 9 - smw15.org

PSY 445: LEARNING & MEMORY
Chapter 9:
Encoding
SEPARATING ENCODING FROM RETRIEVAL
Can be tough to separate the two; arbitrary
• Brain imaging devices have identified different
brain areas
• Typical procedure is to inject a tracer element into
an individual’s blood and use positron emission
topography (PET) scan maps to see where the
element travels in the brain when someone is
performing different cognitive tasks related to
either encoding and retrieval
• Brain activity is measured during study and then
later during recall; correlations are made to
retrieval (remembering and forgetting)
SEPARATING ENCODING FROM RETRIEVAL
Wagner et al. (1998)
• Certain areas of the left hemisphere are more
active during the study of verbal items that were
recalled than when studying items that were later
not recognized
Brewer et al. (1998)
• Certain areas of the right hemisphere are more
active during encoding of photographs of scenes
that were later recognized than to photographs
that were not recognized
SEPARATING ENCODING FROM RETRIEVAL
Encoding
The acquisition of information
The initial formation of a memory trace
Encoding/Failure
Retrieval? (LTM)
Retention
time
BASIC VARIABLES IN ENCODING
• Elaborate Rehearsal
• Imagery & Memory for Pictures
• Meaningfulness
ELABORATE REHEARSAL
Involves thinking about how new material relates
to information already stored in memory
• More than just a recycling of information as is done
in maintenance rehearsal
What exactly is Elaborate Processing?
• Three hypotheses:
•
Elaboration
•
Distinctiveness
•
Effort
WHAT EXACTLY IS ELABORATE
PROCESSING
Elaboration
• Expanding a newly formed memory trace
• Something is remembered better if it is related to
other known facts; making associations
• Deep processing
WHAT EXACTLY IS ELABORATE PROCESSING
Distinctiveness
• If it stands out, then its better remembered; less
interference during retrieval
• Appears to tap into both shallow and deep processing
Winograd (1981)
• Pictures of faces were recalled better if asked about
their features; shallow processing
Hunt & Elliott (1980)
• Orthographically unusual words (afghan) are recalled
better than the more common ones (leaky); shallow
processing
ELABORATE PROCESSING:
DISTINCTIVENESS
Referencing yourself
Rogers et al. (1977)
Procedure 
Results 
• Self-referencing is an example of distinctiveness and deep
processing
9
Interpretation
WHAT EXACTLY IS ELABORATE PROCESSING
Effort
•
The amount of effort expended during encoding is
correlated with retention
•
Deep processing
Eysenck & Eyseneck (1979)
Procedure
• List of words to be remembered are presented;
secondary task is to push a button every time a tone
sounds
• Different types of processing tested: shallow vs. deep
Elaborate Processing: Effort
Results
•
More words recalled when participant needs to make a
deep judgment; but it takes longer
Interpretation
•
We have better memory when we take our time to
think things through
IMAGERY & MEMORY FOR PICTURES
Concrete words
• Words with tangible characteristics (boy, girl)
Abstract words
• These words refer to ideas (peace, love)
IMAGERY & MEMORY FOR PICTURES
Concrete or abstract adj-noun pairs
•Square-door, rusty-engine
•Subtle-fault, absolute-truth
Free-recall (recall as many as possible)
•Concrete > Abstract
Paired-Associate Learning (Square ____?)
•Concrete performance improved over free-recall
•Abstract performance did not
Begg (1972)
IMAGERY & MEMORY FOR PICTURES
Standing, Conezio, & Haber (1970)
Procedure
Acquisition
• Participants were shown 2200 photos over four days
• Two-hour sessions each day
Testing
• 280 pairs of slides are presented and participants must
decide which picture in each pair had been seen in the
acquisition phase
Results
• Correct choices were made over 90% of the time
Interpretation
• Distinctiveness of photos makes for efficient recall
IMAGERY & MEMORY FOR PICTURES
Much evidence: Memory for pictures is strong;
better than for words. Why?
• Possibly dual-coding is taking place
Paivio (1969)
• Demonstrated imagery had functional role – as
seen in memory tasks like paired associate
learning
• Found concrete words easier to recall than
abstract words due to imagery
• Key point of Paivio’s work – imagery impacts
memory
PAIVIO’S DUAL-CODING THEORY
Memory is served
by two systems:
• Verbal
• Nonverbal (visual)
Concrete nouns create mental images
other words can “hang onto” 
MEANINGFULNESS
• Factors involved in meaningful stimuli:
•
•
•
•
Frequency
Pronunceability
Imaginable
Associations
• Suggests a domain-specific memory
PRESENTATION VARIABLES
Roediger & Karpicke (2006): Experiment 1
Procedure
• Participants read prose passages for 7 minutes
followed by a 2-minute break during which they
solved math problems
Group 1: Rereading group
Group 2: Testing group
• After various delays, all participants were given
recall test
RETRIEVAL PRACTICE
TESTING EFFECT
Results 
Roediger & Karpicke (2006)
PRESENTATION VARIABLES
Roediger & Karpicke (2006): Experiment 2
Procedure
Students were asked to memorize a prose passage
through one of three conditions
Repeated Study (SSSS)
• Passage was read four times without a test
Single Test (SSST)
• Passage was read three times, followed be a recall test
Repeated Test (STTT)
20
• Passage was read once, followed by three recall tests
TESTING EFFECTS
Roediger & Karpicke (2006): Experiment 2
Results
• Repeated study was most effective at the short retention
interval; they also expected the best long-term memory
• However, repeated testing was most effective at the long
retention interval
• Practice effect; more effortful, mimics test
21
Interpretation
TESTING EFFECTS
Pashler et al. (2005)
Procedure
• Learn Luganda–English translations
• Some participants got corrective feedback on incorrect test trials
during study
• Others did not receive feedback
• Tested 1 week later on the vocabulary
Results
• Recall for the words they had gotten wrong a week before was
about five times better if they had received immediate feedback
•
Testing yourself with feedback is best for long-term retention
22
Interpretation
TESTING EFFECTS
Three Explanations
• Practice test is another version of learning
• Practice test provides a more challenging restudying of the material; more active manner of
studying; more effort
• Practice test mimics what students will be doing
when they actually take the real test (transferappropriate processing)
23
Roediger & Karpicke (2006)
ISOLATION EFFECTS (VON RESTORFF EFFECT)
Our tendency to remember unusual items better than
more common ones
Hastie & Kumar (1979)
Procedure
Acquisition
• Hypothetical individuals were described by trait labels to induce a personality
impression in the participants
• List of words to be remembered were presented; some words were congruent with
the initial description; others were incongruent
Testing
• Recall after short delay
Results
• Incongruent words (77%) were remembered better than the congruent
words (50%)
Interpretation
• Another case of distinctiveness helping with recall
ENCODING: ISOLATION EFFECTS
Both retrograde amnesic and anterograde
amnesic effects have been found
ENCODING:
SEDUCTIVE DETAILS
Remembering the distinctive tangents (for
example, seductive details) rather than the main
facts
• Recall of important information seems to be
impaired rather than strengthened
ENCODING: SPACING EFFECT
Spacing leads to better retention than does massed
practice
Three Hypotheses:
1. Retrograde Amnesia Hypothesis: A second massed
presentation of an item interferes with ongoing processing
of the first presentation
2. Attention-Deficit Hypothesis: We seem to attend less to
a second presentation when it is repeated too soon
• Shaughnessy, Zimmerman, & Underwood (1972):
students used remote control to decide which
presentation to attend to; most preferred to go back to
old material and did better than those who did not
3. Encoding Variability Hypothesis: Spacing repetitions of
a target item leads to slightly different encodings; thus
better retention
ENCODING: SPACING EFFECT
Limitations:
• If retention is tested after a short delay interval after
the second presentation, massed practice is
sometimes better
Balota, Duchek, & Paullin (1989)
•
College-aged and elderly participants recalled more
massed practiced items on an immediate test and
more spaced practiced items on a delayed test
ENCODING: SPACING EFFECT
Is there an optimal spacing interval?
• In general, the longer interval before the test
occurs, the wider the spacing should be
ENCODING: GENERATION EFFECT
The increased memory performance on study
material that is generated by the participant rather
than being externally provided
• Effortful; deep processing
ELABORATE PROCESSING
Limitations:
Implicit Memory does not seem rely on elaborate
processing
• For example, deep processing does not seem to
be a relevant factor in implicit memory testing
Verbal Overshadowing interferences
• Verbal description of nonverbal information such
as pictures and faces can lead to less accurate
retention
•
Occurs in experiments on faces, wine tasting, etc.
LEARNING VARIABLES
Incidental vs. Intentional Learning
•
Spontaneous memory can occur; at times we do not need
to deliberately study material to be effective when recalling
Incentives
•
Can rewards increase performance on memory tasks?
•
Once a participant has committed to learn stimuli, reward
has little additional effect
•
Nelson (1976); Nilsson, 1987): Both studies found no
significant differences in reward vs. non-reward groups
•
One limitation is that if certain words are given a reward to
remember vs. those not given a reward to remember (also a
finding of Nelson, 1976)
LEARNING VARIABLES
Interest
• A personal interest in the material to be studied
will usually increase retention
• Attention and contact with material will usually
increase as a result; this is likely to lead to deeper
cognitive processing of the material
Morris, Tweedy, & Gruneberg (1985):
•
Study of British football (soccer) fans who attempted
recall of real vs. simulated games
•
Emotional involvement with real teams lead to better
recall
LEARNING VARIABLES
Arousal
• Performance is usually better at some
intermediate level of arousal and is less efficient at
lower and higher levels
Yerkes-Dodson Law (1908)
• Their hypothesis applied to sport, which states
that performance improves as arousal levels
increase up to an optimum point, beyond which it
deteriorates
See next slide 
INVERTED-U HYPOTHESIS
PERFORMANCE
LEVEL
OPTIMUM
LEVEL
LEVEL OF AROUSAL
AROUSAL
Bahrick, Parker, Fivush, & Levitt (1998)
Procedure
•
Preschoolers who were caught in a hurricane were
asked to recall events
•
Three groups:
•
•
Low Arousal (minimal damage)
Moderate Arousal (extensive damage done to outside
of house)
• High Arousal (extensive damage – house penetrated)
Results
•
Moderately stressed children recalled the most
See next slide 
AROUSAL
Bahrick, Parker, Fivush, & Levitt (1998)
Results 
Interpretation
•
There is an optimal amount of arousal required to
achieve optimal performance
LEARNING VARIABLES
Arousal: Stimulant drugs
• These drugs increase CNS arousal
• Mixed results depending on the study
Arousal: Circadian Rhythms
•
Arousal varies during the day
Blake (1967): STM digit span
task - performance peaked at
10:30 AM; declined steadily into
early evening
Folkard & Monk (1980):
immediate recall of 1500 word
science article – optimal recall at
8:00 AM with a steady decline
until after lunch (slight increase)
LEARNING VARIABLES
Emotions and Encoding
Three factors:
1. Emotional arousal will focus attention on certain
aspects of a situation
• Peripheral details may go unnoticed
• Can be a detriment to recall
2. Bodily arousal may produce LTM formation
• Epinephrine
• Can enhance recall
3. Emotional events are distinctive
•
Can enhance recall
FLASHBULB MEMORIES
Brown & Kulick (1977)
• Proposed idea that we tend to remember
important, shocking, and stunning events more
vividly like a mental photography
Are flashbulb memories really accurate?
McClosky, Wible, & Cohen (1988)
Procedure
• Repeated recall experiments
• Let participants recall the special event repeatedly at
different times after the event
• Tested the consistency of their recall
Results
• Very consistent (94% accuracy in specific questions
asked); however, memories became more general and
contained fewer details; 9% of facts changed; 6% of facts
were forgotten
Interpretation
• Fairly robust memories but not immune to forgetting
Are flashbulb memories really accurate?
Talarico & Rubin (2003)
Procedure
• College students asked a number of questions on September
12, 2001
• Some of these questions were about the terrorist attacks
• Others were similar questions about an everyday event in the
person’s life that occurred in the days just preceding the
attacks
• After picking the everyday event, the participant created a
two- or three-word description that could serve as a cue for
that event in the future
• Some participants were retested 1 week later, some 6 weeks
later, and some 32 weeks later
Are flashbulb memories really accurate?
Talarico & Rubin (2003)
Results 
Interpretation
o In reality, no differences between Flashbulb Memories
and regular memories appear to exist; however, we have
a perception that they are very different
LEARNING VARIABLES: EMOTIONS &
ENCODING
Eyewitness Memory and Emotions
• Emotional arousal during these instances can lead
to focusing and retention of certain details but
poorer retention for peripheral details and
unattended details
• Weapon’s focus effect: series of studies by
Elizabeth Loftus found that participants were not
as successful in a facial recognition test if a
weapon was present
EYEWITNESS MEMORY AND EMOTIONS
Loftus (1979)
• Participants in a waiting room hear a heated argument in the
next room
• In one condition a person with a pen in hand emerges; in the
other condition the person appears to have a bloody letter
opener
• Results: More were able to identify the person when he had a
pen
Loftus & Burns (1982)
• Participants shown two-minute video of bank robbery
• One version bank teller is shot; control condition this does not
happen
• Results: 28% recalled number of football jersey on robber if no
gun (control condition); however, only 4% knew the number
when teller was shot
METAMEMORY
Refers to our knowledge about memory
•
Includes estimates of the difficulty of learning certain
materials, the strategies we use, monitoring of
progress, awareness of what is known and not known,
and beliefs about how our memories differ from others
•
Judgments of Learning (JOL) are ratings taken after
the participant has studied but before they have been
tested
•
JOL’s are estimates or predictions of how well
participants think they will do on the upcoming
memory test
CREDITS
Some of the slides in this presentation prepared with the
assistance of the following web sites:
• psychology.illinoisstate.edu/jccutti/.../24.memory12.ppt