Measurement of impulsive choice in rats:
I. Preliminary assessment
1
Hill* ,
2
Peterson ,
2
Kirkpatrick
Catherine
Jennifer R.
Kimberly
1University of the Incarnate Word, 2Kansas State University
Introduction
•
•
•
•
Mazur (M)
•
•
•
•
12 male Sprague-Dawley rats
20 4-trial blocks each day
2 Forced Choice and 2 Free Choice per block
LL delay increased systematically across blocks within each session
80
60
40
As the LL delay increased, the
rats made fewer LL choices
20
0
0
20
40
12 male Sprague-Dawley rats
20 4-trial blocks each day
2 Forced Choice and 2 Free Choice per block
LL delay adjusted depending on the choices in each block
Mean Adjusting Delay (s)
100
% LL Choices
• Impulsive choice underlies maladaptive behaviors
such as gambling, substance abuse, and obesity
• Different procedures are used to assess choice
behavior between a smaller-sooner (SS) versus a
larger-later (LL) reward, with choices of the SS
indicating impulsivity
• The procedures are all assumed to measure the
same underlying construct, despite differences in
task demands
• The current experiment compared Green & Estle
(2003), Evenden & Ryan (1996) and Mazur (1987)
procedures
• Rats were initially trained on Green & Estle (G&E)
• Then, they were trained with either Evenden &
Ryan (E&R) or Mazur (M)
Evenden & Ryan (E&R)
60
30
20
10
The rat’s choices stabilized at an
indifference point of
approximately 24 s.
0
35
37
39
41
Session
LL Delay (s)
Green & Estle (G&E)
24 male Sprague-Dawley rats
20 4-trial blocks each day
2 Forced Choice and 2 Free Choice per block
SS delay increased systematically across phases
Comparison of Procedures
10
5
20
40
PSE (E&R)
60
As the SS delay
increased, the rats
made more LL choices
40
20
20
15
10
5
0
0
80
Mean Correlation
Coefficient
15
0
100
R² = 0.0937
R² = 0.3048
PSE (G&E)
PSE (G&E)
25
20
Median response
time computation
G&E vs. M
G&E vs. E&R
25
% LL Choices
•
•
•
•
60
0
20
40
60
PSE (M)
• A comparison of choice behavior across procedures used the point of
subjective equality (PSE) for each rat
• The PSEs were positively correlated, and the two systematic
procedures were significantly correlated
• M and E&R had higher mean PSEs than G&E
Discussion
SS
LL
• This may have been because the G&E procedure
manipulated SS delay whereas the E&R and M
procedures manipulated LL delay
1
0.8
0.6
0.4
0.2
0
G&E E&R
M
• A median response time
measure was correlated with the
SS or LL delay on each trial
• Both systematic procedures had
significantly higher correlations
than the adjusting procedure
0
0
5
10
15
SS Delay (s)
20
• The G&E procedure resulted in lower PSEs
compared to the other two procedures
25
References
Evenden, J. L., & Ryan, C. N. (1996). The pharmacology of impulsive behavior in rats: the effects of drugs on response choice with varying delays to reinforcement. Psychopharmacology (Berlin), 128, 161-170.
Green, L., & Estle, S. J. (2003). Preference reversals with food and water reinforcers in rats. Journal of the Experimental Analysis of Behavior, 79(2), 233-242.
Mazur, J. E. (1987). An adjusting procedure for studying delayed reinforcement. In M. L. Commons, J. E. Mazur, J. A. Nevin & H. Rachlin (Eds.), Quantitative analyses of behavior. Vol. 5. The effect of delay and of intervening events on reinforcer
value (pp. 55-73). Hillsdale, NJ: Erlbaum.
Acknowledgements
Thank you to the members of the Kirkpatrick RTD lab, especially Andrew Marshall and Aaron Smith for your help with this project. This research was supported by National Institutes of Mental Health grant R01-MHO85739.
• The two systematic procedures were more
highly correlated at the individual differences
level, suggesting greater shared task variance
• The difference in the correlation between LL or
SS delay and median response time suggests
that:
• Rats on both systematic procedures tracked the
delays to reward
• Rats on the adjusting procedure did not track the
LL delays as strongly, perhaps due to the frequent
changes in delay
• Therefore, different mechanisms could be
influencing choice and timing measurements
across the three procedures
• Further research should examine factors that
may lead to shared versus different processes
(See Poster 44)
*Email: cchill@student.uiwtx.edu