Transcript Slide 1

The Effects of Stimulant Countermeasures
on Executive Functioning and DecisionMaking During Sleep Deprivation
William D. S. Killgore, Ph.D.
Walter Reed Army Institute of Research (WRAIR)
McLean Hospital, Harvard Medical School
1
Hypotheses
Reduced metabolic activity will lead
to deficits in “executive functioning”
including planning, judgment, and
decision-making
Modified from Thomas et al. (2000)
2
Hypotheses
Reduced metabolic activity will lead
to deficits in “executive functioning”
including planning, judgment, and
decision-making
Stimulants may “activate” these
areas and “reverse” these deficits
Modified from Thomas et al. (2000)
3
Hypotheses
Reduced metabolic activity will lead
to deficits in “executive functioning”
including planning, judgment, and
decision-making
Stimulants may “activate” these
areas and “reverse” these deficits
or
Modified from Thomas et al. (2000)
Stimulants may activate other areas
involved in alertness, while leaving
prefrontal areas and associated
executive functions degraded
4
General Study Design
In-Residence Sleep Deprivation Studies
Training
Friday
Thursday
Wed
Sleep
2400
Sunday
Saturday
Sleep
1200
2400
1200
2400
1200
2400
1200
5
Executive Function Test Battery
Dorsolateral Prefrontal Cortex
44 HOURS
3:00 AM
Caffeine 600 mg
D-AMPH 20 mg
Modafinil 400 mg
Placebo
Training
Friday
Thursday
Wed
Sleep
2400
Sunday
Saturday
Sleep
1200
2400
Killgore et al., Journal of Sleep Research (2008)
1200
2400
1200
2400
1200
6
Wisconsin Card Sorting Test
Match each card to the appropriate “key” card:
WRONG!
RIGHT!
Concept Formation, Set Shifting,
& Mental Flexibility
7
Wisconsin Card Sorting Test
60
Perseverative Errors (T-Score)
Worse
Better
Perseverative Errors (T-Score)
58
56
54
52
Age Norm
50
48
46
44
42
40
Placebo
Caffeine 600
D-AMPH 20
Modafinil 400
Modafinil significantly improved WCST performance (i.e., reduced
perseveration during sleep deprivation) relative to other stimulants
or placebo.
Killgore et al., Sleep (2009)
8
Tower of Hanoi
Number of Moves
Time to Solve
Planning & Sequencing
9
Tower of London
Your Tower
Goal
Number of Moves beyond Optimal
Time Before Each Move
Total Time
Planning & Sequencing
10
Stimulant Effects: TOH & TOL
Tower of London
Tower of Hanoi
90
120
p < .05
75
Below
Average
70
65
60
Total Number of Moves
Impaired
80
p = .02
110
p = .05
105
100
95
90
85
80
Average
55
75
50
70
Placebo
Caffeine 600
D-AMPH 20
Modafinil 400
Placebo
Total Moves
Caffeine 600
D-AMPH 20
Modafinil 400
Total Moves
4
5
p = .009
p = .006
4.5
3.5
p = .017
4
Mean Pickup Time
3
Mean Pickup Time
Average
Total Number of Moves
Below
Average
p = .004
115
85
2.5
2
1.5
3.5
3
2.5
2
1.5
1
1
0.5
0.5
0
0
Placebo
Caffeine 600
D-AMPH 20
Modafinil 400
Mean Pickup Time
Killgore et al., Sleep (2009)
Placebo
Caffeine 600
D-AMPH 20
Modafinil 400
Mean Pickup Time
11
Two Different Executive “Brain Systems”
Dorsolateral vs. Ventromedial
Dorsolateral
Behavioral Inhibition
Planning
Working Memory
Mental Control
Attention
Set-Shifting
Ventromedial
Affective Processing
Reward/Punishment Learning
12
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
Choose cards from any deck and switch as often as you like.
Try to win as much money as possible.
Some decks are “worse” than others.
You will win the game if you avoid the worst decks.
13
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
WIN
$100
Net: +$100
Risk Taking and Judgment
14
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
WIN
$75
Risk Taking and Judgment
15
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
WIN
$75
LOSE
$50
Net: +$25
Risk Taking and Judgment
16
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
WIN
$300
Net: +$300
Risk Taking and Judgment
17
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
WIN
$500
Risk Taking and Judgment
18
Iowa Gambling Task
Total $$$:
-300
-200
-100
0
100
200
300
400
500
WIN
$500
LOSE
$1200
Net: -$700
Risk Taking and Judgment
19
Iowa Gambling Task
Goal: Learn to choose good decks & avoid bad ones
Net
Win
Good Decks
Net
Win
Net
Loss
Net
Loss
Bad Decks
Risk Taking and Judgment
20
Iowa Gambling Task
Damage to Ventromedial Prefrontal Cortex Impairs Emotional Decision Making
Ventromedial Prefrontal
Lesion (n = 5)
Normal Controls (n = 13)
Good Decks
Bad Decks
Bad Decks
Good Decks
Bechara et al., 1999
Bechara, Damasio, & Damasio, 2000
21
Iowa Gambling Task
Similar Impairment?
Thomas et al., 2000
Sleep Deprivation
Bechara, Damasio, & Damasio, 2000
VMPFC Lesions
We predict a similar pattern of impaired
decision-making performance…
22
49 Hours of Sleep Deprivation Mimics VMPFC Damage
At rested baseline,
participants learn to
avoid risky choices
10
Baseline
Sleep Deprived
Composite of VM Lesions
Net Score (C'+D')-(A'+B')
8
Good Choices
Iowa Gambling Task
6
4
2
0
-2
-4
-8
1-20
N = 34
Killgore et al., Journal of Sleep Research (2006)
21-40
41-60
Blocks of Cards
61-80
81-100
Bad Choices
-6
23
49 Hours of Sleep Deprivation Mimics VMPFC Damage
At rested baseline,
participants learn to
avoid risky choices
Baseline
Sleep Deprived
Composite of VM Lesions
8
Net Score (C'+D')-(A'+B')
When sleep deprived,
these same participants
failed to learn to avoid
risky choices—actually
preferring to choose
from the worst decks
10
Good Choices
Iowa Gambling Task
6
4
2
0
-2
-4
-8
1-20
N = 34
Killgore et al., Journal of Sleep Research (2006)
21-40
41-60
Blocks of Cards
61-80
81-100
Bad Choices
-6
24
49 Hours of Sleep Deprivation Mimics VMPFC Damage
At rested baseline,
participants learn to
avoid risky choices
Baseline
Sleep Deprived
Composite of VM Lesions
8
Net Score (C'+D')-(A'+B')
When sleep deprived,
these same participants
failed to learn to avoid
risky choices—actually
preferring to choose
from the worst decks
10
Good Choices
Iowa Gambling Task
6
4
2
0
-2
-4
-6
-8
1-20
N = 34
Killgore et al., Journal of Sleep Research (2006)
21-40
41-60
Blocks of Cards
61-80
81-100
Bad Choices
Sleep deprived
performance was very
similar to that seen in
patients with actual
brain lesions to VMPFC
25
Iowa Gambling Task
10
Good Choices
49 Hour Study (n = 34)
Baseline
Sleep Deprived
Composite of VM Lesions
8
75 Hour Study (n = 23)
14
12
Net Score (C'+D')-(A'+B')
Net Score (C'+D')-(A'+B')
10
6
4
2
0
-2
-4
Baseline
51 hours
75 hours
8
6
4
2
0
-8
1-20
21-40
41-60
61-80
Blocks of Cards
81-100
Bad Choices
-2
-6
-4
-6
1-20
21-40
41-60
61-80
81-100
Block of Cards
Sleep loss affects emotional decision making in a dose-dependent manner
Less sleep = poorer decisions
Killgore et al., Aviation Space and Environmental Medicine (2007)
26
Iowa Gambling Task
Caffeine (n = 12) vs. Placebo (n = 11)
75 Hours Awake
51 Hours Awake
14
14
Baseline
12
Baseline
12
Placebo
Caffeine
10
Placebo
10
8
8
6
6
4
4
2
2
0
0
-2
-2
-4
-4
-6
Caffeine
-6
1
2
3
4
5
1
2
3
4
5
Caffeine was not significantly better than placebo at enhancing decisionmaking, suggesting that it may have little effect on VMPFC regions
Killgore et al., Aviation Space and Environmental Medicine (2007)
27
Iowa Gambling Task
Caffeine (600 mg), D-AMPH (20 mg), Modafinil (400 mg), vs. Placebo
Across 4 Test Sessions
16
16
Net Score
Net
Score (Good
(Good -- Bad
BadDecks)
Decks)
14
14
12
12
10
10
Baseline: Normal Learning
88
66
44
22
00
Rested Baseline
23 Hours Sleep Deprived
46 Hours Sleep Deprived + Drug
Post-Recovery Sleep
-2
-2
-4
-4
-6
-6
11
22
33
44
5 5
Block
Block
Killgore et al., Sleep (2006)
28
Iowa Gambling Task
Caffeine (600 mg), D-AMPH (20 mg), Modafinil (400 mg), vs. Placebo
Across 4 Test Sessions
16
16
Net Score
Net
Score (Good
(Good -- Bad
BadDecks)
Decks)
14
14
12
12
10
10
Baseline: Normal Learning
88
66
23 Hrs Awake: Mild Impairment
44
22
00
Rested Baseline
23 Hours Sleep Deprived
46 Hours Sleep Deprived + Drug
Post-Recovery Sleep
-2
-2
-4
-4
-6
-6
11
22
33
44
55
Block
Block
Killgore et al., Sleep (2006)
29
Iowa Gambling Task
Caffeine (600 mg), D-AMPH (20 mg), Modafinil (400 mg), vs. Placebo
Across 4 Test Sessions
46 Hours Sleep Deprived + Drug
P = .002
16
16
16
P < .001
110
14
14
Placebo
P < .001
12
12
10010
Net Score (Good - Bad Decks)
PVT Speed (% Baseline)
Net
Net Score
Score (Good
(Good--Bad
BadDecks)
Decks)
105
14
Caffeine
12
D-Amphetamine
10
Modafinil
Baseline:
Normal Learning
8
10
95 8
8
90 6
6
8544
6
4
23 Hrs
Awake: Mild Impairment
8022
7500
Rested Baseline
23 Hours Sleep Deprived
46 Hours Sleep Deprived + Drug
Post-Recovery Sleep
70
-2
-2
65
-4
-4
-6
60
-6
11
Placebo
2
4
2
Caffine
600 3 3 D-AMPH 420
Block
Block
5 5 400
Modafinil
2
0
46 Hrs
Awake: Moderate Impairment
-2
-4
-6
1
2
3
4
5
Block
Sleep deprivation impaired decision making abilities. These deficits were not
reversed by dextroamphetamine, caffeine, or modafinil at doses sufficient to
restore alertness and vigilance.
Killgore et al., Sleep (2006)
30
Iowa Gambling Task
Caffeine (600 mg), D-AMPH (20 mg), Modafinil (400 mg), vs. Placebo
Across 4 Test Sessions
16
16
Net
Net Score
Score (Good
(Good--Bad
BadDecks)
Decks)
14
14
12
12
Post Recovery: Normal Learning
10
10
88
66
44
22
00
Rested Baseline
23 Hours Sleep Deprived
46 Hours Sleep Deprived + Drug
Post-Recovery Sleep
-2
-2
-4
-4
-6
-6
11
22
33
44
55
Block
Block
Killgore et al., Sleep (2006)
31
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
5
Risk Taking and Judgment
32
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
5
Risk Taking and Judgment
33
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
5
Risk Taking and Judgment
34
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
5
Risk Taking and Judgment
35
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
POW
5
Risk Taking and Judgment
36
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
5
Risk Taking and Judgment
37
Balloon Analog Risk Task—BART
Win money for each pump…but don’t pop the balloon:
Correlates with:
Sensation Seeking
Impulsivity
Disinhibition
Risk Taking
Addictive Behavior
$0.10
$0.05
$4.65
$0.30
$3.80
$2.25
$6.55
$5.20
$7.95
$1.50
$0.90
5
Risk Taking and Judgment
38
Balloon Analog Risk Task—BART
Placebo:
1.3
1.2
• Significant disinhibition at 75
hours
1.1
Caffeine:
• No significant difference from
placebo at 51 hours
• Performance sustained across 51
and 75 hours of wakefulness
BART Cost/Benefit Ratio
• No effect of sleep loss at 51 hours
Placebo
Release
Caffeine of
Inhibition?
1.0
0.9
P = .02
0.8
0.7
0.6
0.5
0.4
3 hrs
51 hrs
75 hrs
Hours of Wakefulness
Differs from IGT – Suggests an “unfolding” of behavioral processes:
Decision-Making Impaired Early – Inhibition Impaired Later
Inhibition Sustained by Caffeine
Killgore et al., Sleep (2009)
39
Conclusions
Does Sleep Deprivation
Do Stimulants Reverse the
Impair Executive
Deficits?
Functioning/Decision Making?
Yes—On Some Tasks
Particularly those involving
Ventromedial PFC
Systems (IGT; BART)
Minimal Effects on
Dorsolateral PFC Systems
(WCST; TOH; TOL)
Sometimes—
All Stimulants were
Effective for PVT
Not Effective for Emotion
Based Ventromedial Tasks
Differential Effects on
Dorsolateral Tasks
Conclusion
Sleep Loss May Affect
some Prefrontal Systems
More than Others
Stimulants may restore
some aspects of
functioning while leaving
others impaired
40
Walter Reed Army Institute of Research
McLean Hospital
Harvard Medical School
Collaborators:
Thomas J. Balkin
Nancy J. Wesensten
Gary H. Kamimori
Nancy L. Grugle
Email: [email protected]
The opinions expressed are those of the author and do not reflect the official position of
the U.S. Army, DoD, or any of the institutions with which the author may be affiliated.
41
42
Future Directions / Challenges
Sleep Deprivation and Judgment/Decision-Making:
• How does sleep loss affect the ability to control/modulate affect?
Stimulant Countermeasures:
• Are there novel stimulants or combinations of stimulants that might
prove more effective at reversing deficits in judgment and decisionmaking?
Individual Differences:
• Role of social stimulation and its interaction with individual traits (e.g.,
introversion/extraversion) on the ability to resist sleep loss
• Individual differences in baseline levels of brain chemistry that might be
predictive of the ability to resist sleep loss (e.g., GABA, glutamate).
43
Humor Appreciation
Joke is “Funny”
Goel & Dolan (2001)
Active
Sleep Deprived
Ventromedial
Prefrontal
Cortex
Inactive
Thomas et al., 2000
Humor appreciation is one of the most complex
cognitive capacities
44
Humor Appreciation
Cartoons
Newspaper Headlines
1
Veterinarian
Investigates
Failed Panda
Mating
2
Panda
Mating Fails;
Veterinarian
Takes Over
What’s so
Funny…
Killgore et al., Sleep (2006)
45
Humor Appreciation Score
100
***
Percent Correct
95
*
*
Verbal Norm
90
85
Visual Norm
80
75
70
Placebo
Caffeine 600
Visual
D-AMPH 20
Modafinil 400
Verbal
Caffeine (600 mg) was ineffective at restoring humor appreciation after 49.5
hours of sleep deprivation
Killgore et al., Sleep (2006)
46
Psychomotor Vigilance
Psychomotor Vigilance Speed (1/RT*1000)
Mean (+SE) % of Baseline
120
110
100
90
80
Sleep
Placebo
70
Caffeine
60
Modafinil
Dextroamphetamine
50
Training
Friday
Thursday
Wed
Sleep
2400
Sunday
Saturday
Sleep
1200
2400
Killgore et al., Journal of Sleep Research (2008)
1200
2400
1200
2400
1200
47
Mean (+SE) % of Baseline
Minor Lapses (0.5 to 3 seconds)
850
Placebo
750
Caffeine
650
Modafinil
Dextroamphetamine
550
Sleep
450
350
250
150
50
Training
Friday
Thursday
Wed
Sleep
2400
Sunday
Saturday
Sleep
1200
2400
Killgore et al., Journal of Sleep Research (2008)
1200
2400
1200
2400
1200
48
Major Lapses (≥ 3 seconds)
500
Mean (+SE) % of Baseline
450
Caffeine
400
Modafinil
350
Dextroamphetamine
300
Sleep
250
200
150
100
50
Friday
Thursday
Wed
Training
Placebo
Sleep
2400
Sunday
Saturday
Sleep
1200
2400
Killgore et al., Journal of Sleep Research (2008)
1200
2400
1200
2400
1200
49
Balloon Analog Risk Task—BART
Scoring Method: COST / BENEFIT RATIO
COST = % of Popped
Balloons (Lost Opportunity)
(10 Pops / 30 Balloons) = 33%
BENEFIT = % of Max
Possible Money Won
($45 Won / $94.50 Possible) = 48%
Cost / Benefit Ratio = 33% / 48% = .69
50
Evaluation of Risks Scale—EVAR
60
Interaction:
58
56
EVAR Total Risk
• F2,22 = 5.19, P = .01
Placebo:
• Main Effect, P = .006
Caffeine:
• Main Effect, P = .04
54
*
52
50
48
*
46
44
Placebo
42
Caffeine
40
3 hrs
65
60
55
Placebo
80
55
75
50
70
45
40
35
Caffeine
50
Caffeine
30
3 hrs
51 hrs
75 hrs
Hours of Wakefulness
65
55
Placebo
51 hrs
75 hrs
Hours of Wakefulness
65
35
30
25
Caffeine
50
3 hrs
70
40
60
Placebo
*
45
Invincibility
Danger Seeking
Self Control
70
60
Energy
75
Invincibility
Impulsivity
Energy
Impulsivity
Danger Seeking
Self Control
51 hrs
75 hrs
Hours of Wakefulness
Placebo
Caffeine
20
3 hrs
51 hrs
75 hrs
Hours of Wakefulness
*
*
60
55
*
50
45
Placebo
Caffeine
40
3 hrs
51 hrs
75 hrs
Hours of Wakefulness
3 hrs
51 hrs
75 hrs
Hours of Wakefulness