Theories of Forgetting
I. Introduction
II. Decay Theory
III. Consolidation Theory
IV. Interference Theory
V. Retrieval Failure
VI. Repression
VII. Conclusions
I. Introduction
A. Focusing on Forgetting from LTS
Forgetting from SIS: decay, masking
Forgetting from STS: displacement
Forgetting from LTS?
B. Multiple Factors
II. Decay Theory
http://www.youtube.com/watch?v=A2Pm3vJycWY
A. Definition: Spontaneous loss of information over time.
Classic Shape of the forgetting curve (Woodworth, 1938)
100
90
80
70
60
50
40
30
20
10
0
0
20
40
Retention
60
80
100
Interval
Page 1
Decay Theory
B. Supporting Evidence: Woodworth & Schlosberg (1961)
II. Decay Theory
C. Challenges: Jenkins & Dallenbach (1924)
Two subjects studied non-sense syllables
They were tested either immediately, 1, 2, 4 or 8 hours
later
During the retention interval they either were awake
doing daily activities, or they slept.
If decay is the primacy source of forgetting, then the
rates of forgetting should be similar in the awake and
the sleep conditions.
Jenkins & Dallenback (1924)
10
9
8
7
Number
Recalled 6
5
4
3
2
1
0
Sleeping
Awake
These result suggest that
factors other then time
significantly contribute to
forgetting.
0
1
2
3
4
5
6
7
8
Retention Interval (hours)
Page 2
II. Decay Theory (cont)
D. Further Challenges: Studies of Very-Long-Term
Memory
Is there a permanent memory that doesn’t decay?
1) Bahrick et al. (1975)
Studied retention of names and faces of high school
classmates.
Percent Correct
Bahrick et al. (1975) Results
100
90
80
70
60
50
40
Free Recall
Name Recognition
30
Picture Recognition
20
10
0
0.25 0.75 2 3.8 7.5 14.5 25.8 34.1 47.6
Years Since Graduation
Studies of Very-Long Term
Memory (cont)
2. Bahrick (1984)
Studied retention of Spanish Vocabulary
Page 3
II. Decay Theory (cont)
E. Jost’s Law:
Jost's Law
(from Wixted, 2004)
100
Percent Recall
If two memories are of
equal strength but different
ages, the older memory will
be forgotten less quickly.
90
Old
80
New
70
60
50
40
30
20
10
0
0
20
40
60
80
100
Time since learning Old information
F. Conclusions on Decay Theory
1) Shape of the forgetting curve is greatly influenced by
the activities during the retention interval (sleep
study).
2) A simple, constant decay model, cannot fit the data
(studies of very-long-term memory).
3)  Memories appear to be increasingly resistant to loss
with increasing passage of time (Jost’s Law).
Page 4
III. Consolidation Theory
A.  Consolidation: As a result of experience, certain neural
activities responsible for permanent memories are set into
motion. Disruption of these activities leads to poorly formed
memories, and thus, forgetting.
B.  Consolidation processes may continue for long periods of
time, providing an explanation for Jost’s law.
C. Evidence: temporally graded retrograde amnesia (older
memories are less likely to be disrupted)
Brain Trauma
Retrograde Amnesia
Anterograde Amnesia (common)
time
III. Consolidation Theory
C. Retrograde Amnesia: Anecdotal Evidence
Reports of amnesia following head trauma
By NICOLAS MARMIE .c The Associated Press
PARIS (Sept. 19, 1997) - Bodyguard Trevor Rees-Jones, the sole survivor of
the accident that killed Princess Diana, does not remember the
circumstances of the crash, a judicial source said today.
Doctors had said all along that the large doses of anesthesia given to ReesJones during surgery, along with the trauma of the accident, might affect
his memory.
D. Retrograde Amnesia: Laboratory Evidence
Passive avoidance training - training animals not to
make a response
Passive avoidance
apparatus
Page 5
Electroconvulsive Shock
http://www.youtube.com/watch_popup?v=NAzc95e70ZA
Passive avoidance
Results: Chorover & Schiller (1965)
100
90
80
Percent of 70
Subjects
Showing 60
Avoidance 50
Learning 40
30
20
10
0
Control (no ECS)
0
10
20
30
Time between learning and ECS (sec)
Electroconvulsive Therapy and
Reconsolidation
Kroes et al. (2014)
Participants: Depressed individuals undergoing ECT.
Materials: Two emotional stories
1)  Mother & son (see example)
2)  Sisters, one taken hostage and
physically molested
Page 6
Electroconvulsive Therapy and
Reconsolidation
Kroes et al. (2014) Design:
.25-5 msec,
.9 A, 450v
The memory test was series of multiple choice items on content of each of the slides.
ECT
http://www.youtube.com/watch_popup?v=9L2-B-aluCE&feature=youtu.be
Kroes et al. (2014) Results
ECT disrupts
reconsolidation.
Memory scores were
not different from
chance (group A)
after reactivation
and ECT.
Reactivation without
ECT led to improved
memory (group C)
Story and Group
Page 7
III. Consolidation Theory (cont)
E.  Consolidation and Sleep
Kelly, Kelly, & Clanton (2001)
College GPA associated with self reported sleep
Short sleepers (6 or fewer hours) gpa M = 2.74
Long sleepers (9 or more hours) gpa M = 3.24
Stickgold et al (1999)
Optimal learning of laboratory tasks required 6 hours of sleep
Plihal & Born (1997) Sleeping during a retention interval led to
better memory than wakefulness:
early sleep (non-REM slow wave sleep)
aids declarative memory
late sleep (REM)
aids procedural memory
Consolidation and Sleep
Rasch, Buchel, Gais, & Born (2007, see Rasch & Born, 2008)
Participants learned location of card pairs in “concentration” while smelling
the scent of roses. During the retention interval they sleep with either the
odor or a control odorless oil during slow wave sleep (S3 & S4). They were
then tested (w/out the odor) for the location memories.
Consolidation and Sleep
The presence
of odor during
slow wave
sleep
improved
memory for
card location.
Page 8
Consolidation and Sleep
Later fMRI study
demonstrated
that the odor
during slow
wave sleep
activated the
left
hippocampus.
F. Conclusions on Consolidation
theory
1.  Memories become increasingly resistant to loss with
the passage of time
2.  Brain trauma - disrupts consolidation
3.  Sleep - aids consolidation
4.  Problem: consolidation does not address the role of
the “content” of the experiences during a retention
interval.
IV. Interference Theory
A. Basic Assumptions
1. Forgetting is caused by interference between
information being tested and other information that
has been learned
2. Analysis of learning in terms of associations
between stimulus and response terms.
S
R
Page 9
IV. Interference Theory (cont)
A. Assumptions (cont)
3. a) Interference occurs when the same stimulus is
associated with more than one response.
S
R1
R2
b) or, when similar stimuli (S and S’) are associated
with different responses.
S
R1
S’
R2
IV. Interference Theory (cont)
A. Assumptions (cont)
4. Two mechanisms are responsible for interference:
response competition: multiple responses are
elicited by a stimulus
unlearning: new associations cause extinction of
the previous responses
S
R1
R2
IV. Interference Theory (cont)
B. Examples of interference in action:
phone numbers
tennis
Page 10
IV. Interference Theory (cont)
C. Experimental Designs for Studying Interference
Retroactive Interference:
Interference Group
Control Group
Study A Study B
Study A
-
Test A
Test A
Study A Study B
Study B
Test B
Test B
Proactive Interference:
Interference Group
Control Group
IV. Interference Theory (cont)
D. Illustrative Experiment: Barnes & Underwood (1959)
1) Participants learned a list of eight paired-associates until
perfect recall: (nonsense sylabols paired with adjectives)
A-B (e.g., dax-fruitless)
2) Ss then studied a second list with the first terms paired with
new words:
A-C (e.g., dax-double)
They studied this second list 1, 5, 10, or 20 times
3) Ss then tested on memory for both response terms:
A- B , C (e.g., dax- fruitless, double )
IV.
Interference
Theory (cont)
D. Barnes &
Underwood (1959)
Results:
Page 11
IV. Interference Theory (cont)
E. Problems with Interference Theory
1. According to Interference Theory, Proactive
Interference (PI) is the result of response competition.
However, PI continues even when response
competition is removed.
2. Interference theory did not satisfactorily explain
retrieval failures.
V. Retrieval Failure Theory of
Forgetting
A. Basic Assumption:
Forgetting is caused by the inability to access
information that is represented in memory.
availability: the information is represented in memory
accessibility: the information that is available can be
retrieved at a specific time/place.
V. Retrieval Failure Theory of
Forgetting
B. Demonstration:
1. occupation
2. dairy product
3. circus performer
4. type of seafood
5. part of the body
6. a kind of bedroom furniture
7. a writing instrument
8. a kind of animal
9. a part of the house
10. a kind of flower
11. an art form
12. a kind of vehicle
Page 12
V. Retrieval Failure Theory of
Forgetting
B. Demonstration:
1. lawyer
2. butter
3. acrobat
4. shrimp
5. heart
6. mirror
7. pencil
8. insect
9. window
10. daisy
11. music
12. train
V. Retrieval Failure Theory of
Forgetting
C. Encoding Specificity Principle (Tulving)
Specific encoding operations determine the type of
memory trace stored in memory. The type of memory
trace determines what retrieval cues will be
successful at gaining access to the memory trace.
Overlap determines
success at retrieval
Operations
at
Encoding
Operations
at
Test
V. Retrieval Failure Theory of
Forgetting
D. Generate Edit Theory: An alternative explanation of retrieval
processes:
Generate: retrieval cues used to generate associates
Edit: recognize items generated based on familiarity
E. Encoding Specificity vs. Generate Edit theories
of retrieval.
encoding specificity: stresses study-test cue overlap
generate edit: stresses strength of cue-item association
Page 13
Encoding Specificity
vs. Generate Edit
Thomson & Tulving (1970) experiment:
a) Subjects studied either:
strong associates: e.g., white-black
or weak associates: e.g., train-black
b) memory test contained either:
strong cues: e.g. white
weak cues: e.g. train
Example Materials
TargetWords
loft
secret
plane
clock
ship
money
knife
beer
Strong Associates
attic
agent
air
alarm
anchor
bill
blade
bottle
Weak Associates
pigeon
societies
elevation
antique
nelson
dealer
arrow
tanks
Encoding Specificity
vs. Generate Edit
Thomson & Tulving (1970) experiment (cont)
Results
Test Cues
Strong (white ?) Weak (train ?)
Study
Cues
Strong
(white-black)
Weak
(train-black)
20.2
9.2
13.9
15.7
Page 14
V. Retrieval Failure Theory of
Forgetting
E. Applications of Encoding Specificity
1) Context Effects on Memory
water
music
odor
2) State Dependent Effects on Memory
drugs
Mood
Look for congruence effects: when a match between
study and test leads to best memory performance.
Context Effects
Godden & Baddeley (1975) wet/dry study
Contexts: Wet: under 20 ft water
Dry: sitting on the dock
Results
Test Environment
Dry
Wet
Study
Dry
13.4
8.5
Environment
Wet
8.4
11.5
Context Effects (cont)
Smith (1985) music and memory
Results
Music
at
Study
Mozart
Jazz
Quiet
Mozart
18.2
11.2
16.3
Music at Test
Jazz
Quiet
12.7
13.3
20.8
8.5
15.3
11.7
Page 15
Context Effects (cont)
Schab (1990) Chocolate Study
Odors (chocolate, apple-cinnamon, moth ball)
Results
Study
Odor
Cues
No Odor
Odor
.21
Test Cues
No Odor
.17
.13
.14
Herz (1997) Odor & Memory
Osmanthus
Study
Odor
Cues
No Odor
Pine Cleaner
Study
Odor
Cues
No Odor
Test Cues
Odor
No Odor
.62
.39
.43
.42
Test Cues
Odor
.56
.46
No Odor
.47
.45
2) State Dependent Memory
Eich et al (1975) marijuana study
Results
Study
No Drug
Drug
No Drug
11.5
6.7
Test
Drug
9.9
10.5
Page 16
V. Retrieval Failure Theory of
Forgetting
F. Conclusions:
Very rich description of memory failures
May be circular:
How could this theory be proved wrong?
VII. Conclusions on Forgetting
Four different explanations:
decay: minimal role in LTS forgetting
consolidation: disruption occurs under special
circumstances
interference theory: explains some forgetting, but the
mechanisms need work
retrieval failure: richest, most complete explanation,
but may be circular.
Page 17