Carlyle Smith

History and Background
Sleep States and Memory
• Two experimental designs have been used
extensively to study the relationship of
sleep to memory processes.
• 1) Learn Task - Observe Sleep - Retest
• 2) Learn Task - Deprive Sleep - Retest
Dr. Carlyle Smith
Dept. of Psychology
Trent University
Peterborough, Ontario
50 TRIALS PER DAY FOR 2 DAYS
Animal Studies
Learning Animals
Non-learning Controls
Baseline
30
20
10
0
REM in Minutes from Baseline
• Training was done with rats in a 2-way
shuttle shock avoidance task.
• Animals were given 50 trials/day for two
consecutive days.
• Sleep recording took place both before
and for many days after the end of
training
MINUTES OF REM SLEEP RELATIVE TO BASELINE
FOR LEARNING AND NON-LEARNING ANIMALS
-10
TD1
TD2
PTD1
PTD2
PTD3
PTD4
PTD5
PTD6
PTD7
Days
Early Human Work
REM Sleep Changes After Exams
• In a study done in a practical setting, we
examined the changes in sleep
architecture that might occur after
substantial learning in a real life
situation.
• We examined the sleep of students before
and after exams without typifying
memory systems
• We sleep recorded 4th year Honours
Psychology students in the summer, when
they were not taking classes of any kind.
• We recorded these same subjects 3-5
days after their last Christmas exam in
Dec.
• We then recorded them again the next
summer, when they again were not
taking courses of any kind.
1
MEAN TOTAL NUMBER OF REMS FOR LEARNING
STUDENTS AND FOR NON-LEARNING CONTROLS
Exam Study Con’t ….
**
DISTRIBUTION OF REMS IN EACH REM PERIOD
FOR LEARNING AND CONTROL SUBJECTS AFTER
CHRISTMAS EXAMS
1000
Learning
Controls
800
600
400
200
Mean Total Number of REMs
• Control subjects were same age
individuals that had been in Honours,
and were now in tech or teaching
assistant positions.
• Their intent was eventually to go on to
grad school, but they were consolidating
debts, waiting for a partner to finish, etc.
• They were sleep recorded in Summer 1,
at Christmas and during Summer 2 as
well.
Summer 1
Xmas
Summer 2
REMS DENSITIES FOR LEARNING AND NON-LEARNING CONTROL GROUPS
FOLLOWING CHRISTMAS EXAMS FOR THE LEARNING GROUP
Learning
Controls
300
*
8
Learning
Controls
*
250
6
200
150
4
100
2
50
0
0
-2
-50
1
2
3
4
5
REM Period
Mean REM Densities (from baseline)
Number of REMs (from Baseline)
-100
1
2
3
4
5
REM Period
Appetitive Bar Press Task
Con’t...
• We examined (Smith&Wong, 1991)
changes in sleep after extended training in
Skinner box with increasing levels of
difficulty
• 1 Magazine (M), 3 Simple Bar press (A),
FR -10 sched (2 days) and then 3 days with
a difficult sequential operant (SO).
• There were two bars to press.
• During the Sequential Operant, reward was
only given when rats could “move” the light
on a 3 x 3 matrix from upper left to lower
right - after which they got a pellet
• This required alternate bar presses or 2 left
bar, 2 right bar.
• Deviations reset the program.
2
Minutes of REM Sleep For Each of the Groups
from Pre - Training Mean Baseline REM
Con’t...
FL - fast learning
FRC - FR Control
SL - Slow learning
NL - Did not learn
Mean Baseline REM
30
Human Studies
– 1) declarative - often explicit
– 2) procedural - often implicit
Effect of Sleep Deprivation on Memory
for Neuroanatomical Structures
No differences
Controls
(H)
Controls
(L)
Stage 2
Aw
Training
Re-Test
REMD
% Correct
10
0
-10
M
A1
A2
A3
FR1
FR2
SO1
SO2
SO3
BAR PRESS CONDITION
Declarative Memory and Sleep
• The progress in using human subjects
has paralleled the progress in the
understanding of memory itself.
• Different kinds of memory are now
believed to exist.
• The two main types of memory are -
90
80
70
60
50
40
30
20
10
0
20
REM Sleep in minutes
• Only some rats learned the task. We called
them Fast Learning
• Some rats could not learn and extinguished.
Called Slow Learners
• Non-learning rats were not trained.
• FRC controls pressed at the FR-10 level
instead of changing to SO task
• Declarative learning does not seem to be
related to any stage of sleep.
• Sleep deprivation following acquisition of
a paired associates task did not impair
memory for this task 1 week later
• Sleep deprivation did not impair recall
for of a set of memorized
neuroanatomical structures.
Cognitive Procedural Learning
• Cognitive procedural learning is one type
of procedural learning.
• This type of learning requires the learner
to develop a new cognitive strategy in
order to solve the problem (e.g. Tower of
Hanoi).
3
Cognitive Procedural Memory
and Sleep
• Cognitive procedural tasks appear to
require REM sleep for most efficient
learning
• Sleep following acquisition of the Tower
of Hanoi and Mirror Trace tasks shows a
marked increase in number of REMs and
REM density.
• There is no change in the amount of
REM sleep (minutes) or any other stage.
REM DENSITIES
Total Number of REMs
REMs
1800
1600
1400
1200
1000
800
600
400
200
0
*
p < .0001
*
18
16
p < .0002
14
TEST
CONTROL
PRE
Night 2
POST
Night 3
12
10
Test
Control
8
6
4
2
0
PRE
Night 2
POST
Night 3
4
% REM Sleep
Total REM Sleep
120
110
100
90
80
70
60
50
40
30
20
10
0
Minutes
No Differences
No Differences
25
20
%
TEST
CONTROL
15
TEST
CONTROL
10
5
PRE
Night 2
0
POST
Night 3
PRE
POST
Night 2
Night 3
IQ Groups
IQ Con’t...
• The test groups were separated in terms
of High (HiQ; X=125.8), Medium (MedQ;
X=114.7) and Low IQ (LoQ = 102.5).
• Controls were a mix of all levels @ X=
108.3 (n=6/group, N=24)
• The higher the IQ score, the better were
the learning scores for the tasks.
• The higher the IQ score, the greater was
the number of post training REMs and
the larger was the REM Density measure.
(HiQ>MedQ>LoQ).
• The intensity of REM sleep was also
correlated with success in task
acquisition and re-test.
TOTAL NUMBER OF REMS BEFORE AND AFTER TRAINING
ON THE TOWER OF HANOI AND MIRROR TRACE TASKS
**
2500
**
2000
*
1500
1000
Mean Number of REMs
500
0
Baseline
Post-training
LoQ
MedQ
HiQ
Controls
IQ Con’t...
• There was no correlation between
“native” REM intensity and IQ or task
scores.
• The largest REM intensity responses
were exhibited by Ss with the highest IQ
and the highest acquisition and re-test
scores.
• Post-training REM intensity may be a
biological marker for learning potential
Nights
5
NUMBER OF RAPID EYE MOVEMENTS FOR COMBINED TRAINED
AND NON-TRAINED CONTROLS AT EACH REM PERIOD
PERCENTAGE REM FOR TEST AND CONTROLS
AS A FUNCTION OF NIGHT
Combined Trained
Non -Trained Controls
30
TEST
CONTROLS
28
800
26
24
700
22
20
600
18
16
500
14
400
12
10
300
8
6
Number of Rapid Eye Movements
100
1
2
3
4
REM PERIOD
PERCENTAGE REM
200
4
2
NIGHT 1 AND NIGHT 2
Tower of Hanoi Task
Improvement after sleep deprivation
Selective REM Deprivaiton
25
20
15
*
*
10
5
0
-5
Pre - Post (# of Moves to Solution)
Savings Scores
• Selective REM sleep deprivation
following acquisition of the Tower of
Hanoi or a complex Logic Task results in
marked memory loss on re-test one week
later.
• NREM awakened controls were not
impaired on these tasks.
*
-10
TS
D
Co
nt
ro
ls
(H
C)
RE
M
D
Groups
Co
nt
ro
ls
N
(R REM
)
Aw
ak
en
ing
Alcohol Ingestion
Alcohol Con’t...
• One common type of behaviour on
campus is to work hard during the day,
drink hard in the evening and then fall
into bed.
• Alcohol is know to induce a reduction in
number and density of REMs when
ingested in moderate amounts (as few as
3 drinks) just prior to bedtime.
• After an acclimatization night in the sleep
lab., we trained students in both a
declarative (Paired Associates) and a
cognitive procedural (Logic) task
(Alcohol Night 1) .
• They then drank 3-5 drinks of vodka
(1 1/4 oz/drink) and orange juice and
then went to bed (0.7g/Kg.).
6
% CORRECT PAIRED ASSOCIATE
RESPONSES BEFORE AND AFTER
ALCOHOL CONSUMPTION
Alcohol con’t...
100
90
70
60
50
40
30
20
10
PRE
Experiment 2
• Was this an alcohol or a sleep effect?
• Trained subjects on Tower of Hanoi and
Pursuit Rotor at 4 PM in the afternoon.
• One group (Aft - Alcohol) then drank
immediately (0.9g/Kg) after learning and
went to bed 5 hours later.
• One group drank just before bed (Eve Alcohol).
100
80
ALN1
ALN3
Controls
*
60
40
% Correct Wffs Constructed
20
PRE
POST
Before and After Alcohol Consumption
% CORRECT WFFs BEFORE AND AFTER
ALCOHOL CONSUMPTION
*
ALN1
ALN3
Controls
80
% Correct Paired Associates
• Another group drank orange juice alone
after learning (OJ Night 1)
• A third Control group drank nothing (C
Night 1).
• All groups slept a second night with no
drinking.
• On Night 3, the OJ group of Night 1
drank alcohol, while the alcohol group of
Night 1 drank OJ.
POST
Pursuit Rotor Target Times for AFT- EtOH and EVE- EtOH
Groups Before and After Alcohol Ingestion
NUMBER OF MOVES TO SOLUTION
ON THE TOWER OF HANOI TASK
*
160
*
140
100
EVE-EtOH
AFT- EtOH
120
Evening - Alcohol
Afternoon - Alcohol
80
100
60
80
40
60
0
PRE
POST
Mean Number of Seconds on Target
Number of Moves to Solution
20
PRE
POST
Alcohol Condition
7
Sleep Effects
Conclusions
• REM sleep was similarly modified in
both experiments.
• Both number of minutes of REM sleep
and number of REMs were reduced in
the first half of the night in study 1.
• The density of REMs was reduced in the
first half of the night in study 2.
• Even moderate alcohol ingestion prior to
bedtime can modify sleep architecture,
particularly REM sleep in the first half of
the night.
• Moderate alcohol ingestion at bedtime
can induce memory loss the night of
learning AND/OR 2 nights after the
original learning has taken place.
Conclusions con’t...
Conclusions Con’t ...
• Interference with normal REM sleep
either the same night, or two nights after
acquisition of procedural material can
result in memory deficits of 20-40%.
• Alcohol at bedtime can impair memory
for both cognitive procedural and motor
skills tasks.
• The REMs results implicate the PGO
system of the pons in efficient memory
consolidation
• Animal studies (Datta et. al., 2000)
support this theoretical position, since
they have observed increases in P wave
(PGO) activity in the rat following
successful acquisition of the 2-way shuttle
avoidance task
8

Download Report