Effect of drugs on response-duration differentiation VII: response-force requirements.

Abstract

Rats were trained to press a lever for at least 1 s but for less than 1.3 s. The force required to press the lever was then increased or decreased by 10, 15, or 20 g. Increases in the force requirements for lever pressing decreased timing accuracy, but decreases in the force requirement had the opposite effect. Accuracy decreases at increasing force requirements were characterized by an increase in the relative frequency of responses that were too short to meet the reinforcement criterion. In contrast, increases in accuracy when the force requirements were decreased were characterized by increases in response durations that met the reinforcement criterion and decreases in the relative frequency of responses that were too short to produce the reinforcer. Phencyclidine (PCP) and methamphetamine produced dose-dependent decreases in accuracy that were associated primarily with increases in the relative frequency of short response durations, although methamphetamine also produced increases in long response durations at some doses. When the effects of PCP were determined with the force requirement increased by 10 g or decreased by 15 g, the cumulative response-duration distribution shifted toward even shorter response durations. When the effects of methamphetamine were determined with the force requirement on the lever increased by 10 g, the cumulative frequency distribution was shifted toward shorter response durations to about the same extent as it had been before force requirements increased; however, when the force required to press the lever was decreased by 15 g, these shifts toward shorter response durations almost completely disappeared. These results show that increases and decreases in the force requirements for lever pressing have different effects on the accuracy of temporal response differentiation.

Full text

Page 1

295
JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR
2000, 74, 295–309
NUMBER 3 (NOVEMBER)
EFFECT OF DRUGS ON
RESPONSE-DURATION DIFFERENTIATION
VII: RESPONSE-FORCE REQUIREMENTS
G. Y. H. MCCLURE, W. C. HARDWICK, AND D. E. MCMILLAN
UNIVERSITY OF ARKANSAS FOR MEDICAL SCIENCE
Rats were trained to press a lever for at least 1 s but for less than 1.3 s. The force required to
press the lever was then increased or decreased by 10, 15, or 20 g. Increases in the force require-
ments for lever pressing decreased timing accuracy, but decreases in the force requirement had
the opposite effect. Accuracy decreases at increasing force requirements were characterized by an
increase in the relative frequency of responses that were too short to meet the reinforcement
criterion. In contrast, increases in accuracy when the force requirements were decreased were
characterized by increases in response durations that met the reinforcement criterion and de-
creases in the relative frequency of responses that were too short to produce the reinforcer. Phen-
cyclidine (PCP) and methamphetamine produced dose-dependent decreases in accuracy that were
associated primarily with increases in the relative frequency of short response durations, although
methamphetamine also produced increases in long response durations at some doses. When the
effects of PCP were determined with the force requirement increased by 10 g or decreased by 15
g, the cumulative response-duration distribution shifted toward even shorter response durations.
When the effects of methamphetamine were determined with the force requirement on the lever
increased by 10 g, the cumulative frequency distribution was shifted toward shorter response du-
rations to about the same extent as it had been before force requirements increased; however,
when the force required to press the lever was decreased by 15 g, these shifts toward shorter
response durations almost completely disappeared. These results show that increases and decreases
in the force requirements for lever pressing have different effects on the accuracy of temporal
response differentiation.
Key words: temporal response differentiation, methamphetamine, phencyclidine, response force,
lever press, rat
Many drugs alter timing behavior and
thereby potentially affect behavior that re-
quires accurately timed responses. Some in-
vestigators have studied the effects of drugs
on timing behavior using schedules of rein-
forcement that reinforce temporally spaced
responding, but in general these investigators
have used schedules that require that re-
sponses be spaced more than 10 s apart (Li,
Marek, Hand, & Seiden, 1990; Marek, Li, &
Seiden, 1989; Marek & Seiden, 1988; Mc-
Guire & Seiden, 1980; O’Donnell & Seiden,
1983). Fewer investigators have studied the
effects of drugs on shorter, precisely timed
responses. Some years ago, a procedure for
studying temporal response differentiation
(TRD) was developed (McMillan & Patton,
This research was supported by NIDA Grant DA-02251.
McClure is now at the Department of Surgical Oncology,
University of Arkansas for Medical Sciences.
Address correspondence to D. E. McMillan, Depart-
ment of Pharmacology and Toxicology, University of Ar-
kansas for Medical Sciences, 4301 W. Markham St., Slot
638, Little Rock, Arkansas 72205 (E-mail: mcmillan@
biomed.uams.edu).
1965). These TRD schedules require not only
that the subject hold down a lever for a min-
imum time period but also that the lever be
released before a maximum time period has
elapsed. Temporal response differentiation
schedules that reinforce only these precisely
timed responses recently have been used to
study the effects of the drugs of abuse on tim-
ing behavior (Hudzik & McMillan, 1994a,
1994b; McClure, Wenger, & McMillan, 1997;
McMillan, Adams, Wenger, McClure, & Hard-
wick, 1994; Schulze et al., 1988; Schulze &
Paule, 1990, 1991; Schulze, Slikker, & Paule,
1989).
Under the TRD schedule used in this study,
the reinforcer was delivered following a con-
tinuous lever press that was at least 1.0 s in
duration but not longer than 1.3 s. In previ-
ous studies using a TRD 1–1.3-s schedule in
which the reinforcer was delivered following
a continuous lever press that was at least 1.0
s in duration but not longer than 1.3 s, both
methamphetamine and phencyclidine (PCP)
decreased the percentage of responses that

Page 2

296G. Y. H. MCCLURE et al.
were reinforced. However, methamphet-
amine flattened the relative response-dura-
tion distributions at doses of 3.0 mg/kg or
greater (Hudzik & McMillan, 1994a; McClure
et al., 1997; McMillan et al., 1994) by increas-
ing the proportions of response durations
that were either too short or too long to pro-
duce a reinforcer. In contrast to metham-
phetamine, PCP increased the relative fre-
quency only of response durations that were
too short to produce the reinforcer (Hudzik
& McMillan, 1994a; McClure et al., 1997; Mc-
Millan et al., 1994).
Methamphetamine and PCP also de-
creased accuracy under TRD schedules re-
quiring longer response durations for food
presentation (McClure et al., 1997). Under
these two separate TRD schedules, the rein-
forcer was delivered following a continuous
lever press that was at least 4.0 s in duration
but not longer than 5.2 s, or following a con-
tinuous lever press that was at least 10 s in
duration but not longer than 13 s. In contrast
to the effects of these drugs on responding
under the TRD 1–1.3-s schedule, the effects
of methamphetamine and PCP on the rela-
tive frequency distributions of response du-
rations under longer TRD schedules were
very similar. Because these drugs produce dif-
ferent effects under the TRD 1–1.3-s schedule
than under TRD schedules with longer tim-
ing requirements, performance under these
schedules may involve different behavioral
mechanisms. Under a TRD 1–1.3-s schedule,
the animal must release the lever within a
300-ms period after 1 s has elapsed, and these
precisely timed responses may require motor
skills not required under longer TRD sched-
ules.
As a continuation of studies of the effects
of methamphetamine and PCP on respond-
ing under TRD schedules, we determined
whether the differential effects of these
drugs seen under the TRD 1–1.3-s schedule
were altered by changes in lever-force re-
quirements. It was reasoned that under the
TRD schedule, the animal makes a precise
motor response to meet the timing require-
ments of the schedule. Changes in the force
requirements for lever presses would be like-
ly to disrupt performance under the TRD
schedule and perhaps differentially modify
the effects of drugs on responding under
this schedule.
METHOD
Subjects
Four male Sprague-Dawley rats that had
been used in a previous experiment (McMil-
lan et al., 1994) were used. The rats weighed
315 to 320 g at 80% of free-feeding weights.
They were maintained at these weights by
food presented during the session and by
supplemental feeding after the test sessions.
Rats were housed individually in suspended
stainless-steel cages in a colony room main-
tained at 70 to 74 ?F with a 12:12 hr light/
dark cycle (lights on 6:00 a.m. to 6:00 p.m.)
during the initial training period. Water was
available at all times except during the ex-
perimental sessions.
Apparatus
Each rat was trained and tested in a differ-
ent two-lever chamber (Gerbrands Model
G7410) encased in a different sound-attenu-
ating Gerbrands enclosure (Model G7210). A
Gerbrands feeder delivered 97-mg Noyes
food pellets into a food cup mounted be-
tween the levers when schedule contingen-
cies had been met. A houselight and a stim-
ulus light consisting of 28-V DC bulbs were
mounted in the ceiling of the experimental
chamber and above the right lever. A down-
ward force sufficient to close the microswitch
contact on the right lever (Gerbrands Model
G6312) activated the stimulus light and pro-
duced a continuous tone (Sonalert Model
SC628H in series with a 15-k? resistor) when
microswitch contacts were closed. The forces
in newtons (measured in grams on a dyna-
mometer) that were required to close the mi-
croswitch when the lever was pressed were
0.37 N (38 g) for Rat 402 in Chamber 1, 0.25
N (25 g) for Rat 405 in Chamber 2, 0.30 N
(31 g) for Rat 408 in Chamber 3, and 0.32 N
(33 g) for Rat 413 in Chamber 4. Because the
dynamometer readings were in grams, lever-
force requirements will be specified in grams
hereafter. The force required to close the mi-
croswitch was not equalized across chambers
because the experiments being conducted by
other investigators using the same chambers
might be disrupted. Events in the chambers
were controlled and data collected using a
Firestar 386 computer with a Med Associates
interface housed in a separate room.

Page 3

297DRUG EFFECTS AND FORCE REQUIREMENTS
Table 1
Sequence order of force changes for each rat.
Rat
Force changes (in grams)
123456
402
405
408
413
?10
?10
?10
?10
?10
?10
?10
?10
?20
?20
?20
?20
?20
?20
?20
?20
?15
?15
?15
?15
?15
?15
?15
?15
Table 2
Sequence of drug–force combinations for each rat.
Rat 402 Rat 405 Rat 408Rat 413
Series 1 (?15 g force)
1
2
3
4 0.3 MAP
Series 2 (?15 g force)
1
2
3
4 3.0 PCP
Series 3 (?10 g force)
1
2
3
4 0.3 PCP
Series 4 (?10 g force)
1
2
3
4 5.6 MAP
5.6 MAP
3.0 MAP
1.0 MAP
5.6 MAP
3.0 MAP
1.0 MAP
0.3 MAP
0.3 MAP
1.0 MAP
3.0 MAP
5.6 MAP
0.3 MAP
1.0 MAP
3.0 MAP
5.6 MAP
0.3 PCP
1.0 PCP
1.7 PCP
0.3 PCP
1.0 PCP
1.7 PCP
3.0 PCP
3.0 PCP
1.7 PCP
1.0 PCP
0.3 PCP
3.0 PCP
1.7 PCP
1.0 PCP
0.3 PCP
3.0 PCP
1.7 PCP
1.0 PCP
3.0 PCP
1.7 PCP
1.0 PCP
0.3 PCP
5.6 MAP
3.0 MAP
1.0 MAP
0.3 MAP
5.6 MAP
3.0 MAP
1.0 MAP
0.3 MAP
0.3 MAP
1.0 MAP
3.0 MAP
0.3 MAP
1.0 MAP
3.0 MAP
5.6 MAP
0.3 PCP
1.0 PCP
1.7 PCP
3.0 PCP
0.3 PCP
1.0 PCP
1.7 PCP
3.0 PCP
Note. MAP ? methamphetamine; PCP ? phencyclidine.
Training Procedure
The training of these rats has been de-
scribed in detail previously (McMillan et al.,
1994). Briefly, the rats were trained to press
the lever on the right side of the chamber
with sufficient force to close the microswitch
to deliver a food pellet (hereafter the term
lever press will imply a response with sufficient
force to close the microswitch). Once the rat
consistently pressed the lever, the response
duration required to produce a pellet was
gradually lengthened to 1.0 s in 0.3-s incre-
ments, with a final increment of 0.1 s. Under
this procedure, responses were reinforced
only if their duration exceeded the minimum
time requirement required by the schedule.
Once this 1.0-s minimum hold had been es-
tablished, the upper limit of the response du-
ration below which the reinforcer could be
produced was gradually reduced. Under the
final TRD schedule, only lever presses at least
1 s but less than 1.3 s in duration were rein-
forced. The session ended with the delivery
of 50 reinforcers or after 40 min, whichever
occurred first.
Force Variation Procedure
The baseline force on the lever that was
necessary to close the microswitch as mea-
sured by the dynamometer was 25 g to 38 g
depending on the chamber. In each chamber
during six different test sessions, washers
were added or removed on the side of the
fulcrum located behind the test chamber wall
to change the force required to activate the
microswitch by ?10, ?15, ?20, ?10, ?15, or
?20 g. All rats were tested once at each of six
different force requirements in the order
shown in Table 1. Sessions were conducted
Mondays through Fridays between 7:30 a.m.
and 9:30 a.m., with force requirements
changed only on Tuesdays and Fridays. The
force required during training sessions will be
referred to as the baseline force requirement
in these experiments.
Drugs and Testing
For determination of dose–effect curves
with baseline force requirements, doses of
methamphetamine sulfate (Sigma Chemical
Co.) and phencyclidine hydrochloride (Na-
tional Institute on Drug Abuse) were given in
ascending dose order, except that Rats 405
and 413 received the lowest dose last. Both
PCP (0.3, 1.0, 1.7, or 3.0 mg/kg) and meth-
amphetamine (0.3, 1.0, 3.0, or 5.6 mg/kg)
were dissolved in physiologic saline and ad-
ministered intraperitoneally in a volume of 1
ml/kg, 10 min prior to the beginning of ses-
sions. All dose levels are expressed as the
salts. Methamphetamine, PCP, or saline was
administered on Tuesdays and Fridays. Table
2 shows the order of exposure to the differ-
ent force requirements when the effects of
drugs were studied.
Data Analysis
The total number of responses emitted, re-
sponse rate (responses per second), accuracy
(reinforced responses divided by total re-
sponses), and mean response duration were
calculated for every session for each rat.

Page 4

298G. Y. H. MCCLURE et al.
Table 3
Performance indicators for responding under the TRD 1.0–1.3 s schedule at different force
requirements.
Treatment
Rat 402
ResponsesResponse rateAccuracy
Rat 405
ResponsesResponse rateAccuracy
?20 g
?15 g
?10 g
Baseline
?10 g
?15 g
?20 g
57
68
73
101.3 ? 7.9
116
167
207
0.06
0.06
0.07
0.11 ? 0.01
0.10
0.12
0.17
87.7
73.5
68.5
49.6 ? 4.0
43.1
29.6
24.2
64
67
75
75.7 ? 3.5
83
110
125
0.08
0.09
0.08
0.09 ? 0.01
0.09
0.12
0.14
78.1
74.6
66.7
66.2 ? 3.0
60.2
45.5
40.0
Note. The total number of responses emitted, accuracy (% reinforced responses), and response rate (responses per
second) are based on single observations in each of 4 rats. The baseline level is an average of 10 daily training sessions
during the experimental periods. Baseline data include standard deviations.
Table 4
Effect of phencyclidine and force changes on performance indicators in individual rats. The
number of responses emitted, response rate (responses per second), and accuracy (% rein-
forced responses) for PCP alone and PCP combined with force changes. Each value for the
saline days represents an average of three sessions with standard deviations. Drug data rep-
resent single observations in each rat.
Dose
Rat 402
ResponsesResponse rate Accuracy
Rat 405
ResponsesResponse rateAccuracy
Baseline
Saline
PCP 0.3
PCP 1.0
PCP 1.7
PCP 3.0
?15 g force
Saline
PCP 0.3
PCP 1.0
PCP 1.7
PCP 3.0
?10 g force
Saline
PCP 0.3
PCP 1.0
PCP 1.7
PCP 3.0
88.7 ? 20.5
110
91
131
334
0.09 ? 0.02
0.10
0.11
0.13
0.24
58.5 ? 14.0
45.5
55.0
38.2
15.0
88.0 ? 7.5
88
174
322
420
0.09 ? 0.01
0.09
0.17
0.22
0.23
57.1 ? 4.8
56.8
28.7
15.5
10.5
70.0 ? 11.3
106
127
124
290
0.08 ? 0.02
0.11
0.13
0.14
0.16
72.6 ? 10.7
47.2
39.4
40.3
4.5
60.7 ? 3.5
80
126
316
1
0.08 ? 0.002
0.11
0.13
0.18
0.001
82.6 ? 4.8
62.5
39.7
10.8
—a
127.0 ? 10.8
148
174
168
58
0.12 ? 0.02
0.15
0.19
0.18
0.03
39.0 ? 3.3
33.8
28.7
29.8
0.0
85.3 ? 9.1
86
159
129
43
0.10 ? 0.01
0.10
0.18
0.16
0.02
59.0 ? 6.2
58.1
31.5
38.8
7.0
aThese responses were not used for relative cumulative frequency distributions because fewer than 25 responses
were emitted.
Mean curves were plotted for accuracy as a
function of the force required to press the
lever, and data from individual animals were
presented in tables. Mean dose–response
curves for the effects of PCP and metham-
phetamine on accuracy were also plotted, and
the data from individual animals were pre-
sented in tables. Pearson product-moment
correlation coefficients (r) relating accuracy
to response rate were calculated by correlat-
ing accuracy with response rates in individual
animals from the data in Tables 3 through 5.
Each response, depending on its duration,
was collected into one of 24 consecutive time
bins. Each of the first 23 time bins was 0.1 s
wide. All response durations less than 0.1 s
were cumulated in Bin 1, all response dura-
tions of at least 0.1 s but less than 0.2 s were

Page 5

299DRUG EFFECTS AND FORCE REQUIREMENTS
Table 3
(Extended)
Rat 408
ResponsesResponse rateAccuracy
Rat 413
ResponsesResponse rateAccuracy
66
79
93
96.1 ? 6.5
120
364
121
0.07
0.10
0.10
0.10 ? 0.01
0.12
0.23
0.07
75.8
63.3
53.8
52.3 ? 3.7
41.7
13.7
2.5
76
82
104
100.2 ? 10.5
124
239
331
0.09
0.10
0.10
0.11 ? 0.02
0.14
0.13
0.18
65.8
61.0
48.1
50.4 ? 5.2
40.3
20.1
15.1
Table 4
(Extended)
Rat 408
ResponsesResponse rateAccuracy
Rat 413
Responses Response rateAccuracy
101.0 ? 5.3
124
132
188
241
0.12 ? 0.01
0.19
0.16
0.27
0.26
49.6 ? 2.7
40.3
37.9
26.6
20.8
91.0 ? 1.0
96
105
125
161
0.14 ? 0.01
0.14
0.15
0.17
0.18
55.0 ? 0.6
52.1
47.6
40.0
31.1
67.0 ? 1.7
62
97
202
396
0.08 ? 0.01
0.08
0.12
0.21
0.22
74.7 ? 1.9
80.7
51.6
24.8
10.6
78.0 ? 11.5
73
109
382
226
0.08 ? 0.04
0.10
0.15
0.24
0.13
61.6 ? 6.3
68.5
45.9
13.1
15.0
100.7 ? 19.9
159
440
560
34
0.12 ? 0.01
0.18
0.35
0.31
0.02
51.1 ? 10.9
31.5
11.4
1.4
2.9
114.7 ? 8.6
115
157
166
264
0.15 ? 0.02
0.18
0.22
0.23
0.15
43.8 ? 3.2
43.5
31.9
30.1
9.5
cumulated in Bin 2, and so on. Response du-
rations of 2.3 s and longer were collected in
the 24th bin. Bins 11, 12, and 13 collected
response durations that produced the rein-
forcer. The relative frequency of responses in
each time bin (the number of responses col-
lected in each bin divided by the total num-
ber of responses made during the session)
was calculated for each individual rat for each
session. These cumulative frequencies were
plotted as sigmoidal curves to analyze the re-
sponse-duration distributions. If animals
failed to respond at least 25 times within a
session, the data were used only to calculate
response rates and were not used to plot re-
sponse-duration distributions.
RESULTS
Baseline accuracy (percentage of total re-
sponses with durations of at least 1.0 s but less
than 1.3 s) was close to 50% for 3 of the rats

Page 6

300G. Y. H. MCCLURE et al.
Table 5
Effect of methamphetamine (MAP) and force changes on performance indicators in individ-
ual rats. The number of responses emitted, response rate, and accuracy for MAP doses alone
and MAP combined with force changes. Each value for the saline days represents an average
of three sessions with standard deviations. Drug data represent single observations in each rat.
Dose
Rat 402
ResponsesResponse rate Accuracy
Rat 405
ResponsesResponse rateAccuracy
Baseline
Saline
MAP 0.3
MAP 1.0
MAP 3.0
MAP 5.6
?15 g force
Saline
MAP 0.3
MAP 1.0
MAP 3.0
MAP 5.6
?10 g force
Saline
MAP 0.3
MAP 1.0
MAP 3.0
MAP 5.6
88.7 ? 20.5
112
122
150
0
0.09 ? 0.02
0.10
0.15
0.08
0.00
58.5 ? 14.0
44.6
41.0
11.3
—a
88.0 ? 7.5
126
116
127
0
0.09 ? 0.01
0.13
0.17
0.07
0.00
57.1 ? 4.8
39.7
43.1
34.7
—a
70.0 ? 11.3
87
74
72
0
0.08 ? 0.02
0.11
0.09
0.08
0.000
72.6 ? 10.7
57.5
67.6
69.4
—a
60.7 ? 3.5
76
61
70
0
0.08 ? 0.002
0.13
0.09
0.09
0.000
82.6 ? 4.8
65.8
82.0
71.4
—a
127.0 ? 10.8
145
166
144
0
0.12 ? 0.02
0.16
0.24
0.08
0.000
39.6 ? 3.3
34.5
30.1
22.9
—
85.3 ? 9.1
76
110
9
1
0.10 ? 0.01
0.09
0.14
0.01
0.001
59.0 ? 6.2
65.8
45.5
11.1a
—a
aThese responses were not used for relative cumulative frequency distributions because fewer than 25 responses
were emitted.
Fig. 1.
accuracy. Abscissa: increases or decreases in baseline
force requirements in grams. Ordinate: percentage of re-
sponses between 1.0 and 1.3 s in duration (percentage
of correct responses or accuracy). Each point is a single
observation in 1 rat as follows: circles ? Rat 402; squares
? Rat 405; triangles ? Rat 408; inverted triangles ? Rat
413. Baseline points at 0 show means from 10 sessions.
Brackets show ?1 SD.
Effects of changes in force requirements on
but was 66% for Rat 405. Rat 405 had the
lowest baseline force requirement of the 4
rats. Figure 1 illustrates changes in accuracy
of individual rats as the baseline force re-
quirements changed. Increases in force re-
quirements decreased the timing accuracy of
all 4 rats, with larger increases in force pro-
ducing larger decrements in accuracy. How-
ever, decreases in force requirements in-
creased accuracy in all rats. Three of the 4
rats showed only slight changes in accuracy
with both 10-g increases and 10-g decreases
in the force requirement. Rat 402 (circles in
Figure 1) was more sensitive to the 10-g de-
crease in force requirement than the other
rats. For each rat, increasing the force re-
quirements by more than 10 g decreased ac-
curacy and decreasing the force require-
ments by more than 10 g increased accuracy.
The correlation between force requirement
and percentage of correct reinforced re-
sponses calculated across all 4 rats was high
(r ? ?.94). This correlation occurred across
a range of force requirements and across a
range of percentage of reinforced responses.
Table 3 shows the effects of changing the

Page 7

301DRUG EFFECTS AND FORCE REQUIREMENTS
Table 5
(Extended)
Rat 408
ResponsesResponse rateAccuracy
Rat 413
ResponsesResponse rateAccuracy
101.0 ? 5.3
131
157
240
1
0.12 ? 0.01
0.19
0.24
0.22
0.001
49.6 ? 2.7
38.2
31.9
20.8
—a
91.0 ? 1.0
118
121
151
0
0.14 ? 0.01
0.20
0.22
0.13
0.000
55.0 ? 0.6
42.4
41.3
33.1
—a
67.0 ? 1.7
74
78
75
71
0.08 ? 0.01
0.10
0.010
0.10
0.04
74.7 ? 1.9
67.6
64.1
66.7
19.7
78.0 ? 11.5
99
110
84
0
0.08 ? 0.04
0.14
0.09
0.06
0.00
61.6 ? 6.3
50.5
45.5
59.5
—a
100.7 ? 19.9
207
258
219
0
0.12 ? 0.01
0.18
0.29
0.26
0.000
51.1 ? 10.9
24.2
19.4
22.8
—a
114.7 ? 8.6
254
176
2
0
0.15 ? 0.02
0.21
0.10
0.001
0.00
43.8 ? 3.2
19.7
21.0
—a
—a
force requirements on the total number of
responses made during the session and on
the response rate for each rat. Increasing the
force requirement decreased accuracy, which
increased the number of responses emitted,
because the session usually continued until
all 50 reinforcers had been delivered. Except
for Rat 408, the number of responses emitted
increased as the force requirement increased.
Increasing the force requirement generally
increased the response rate (except Rat 408
with a 20-g increase). Accuracy was negatively
correlated with the response rate (r ? ?.48)
when the force requirements were increased.
When the force requirements for lever
pressing were decreased, accuracy usually in-
creased, thereby decreasing the number of
responses made by all rats. Decreasing the
force requirements produced small decreases
in the response rate of most rats. Accuracy
was negatively correlated with response rate
(r ? ?.85) when the force requirements on
the lever were decreased.
Figure 2 shows the relative cumulative fre-
quency distributions of response durations
for each rat during saline control sessions. A
sigmoidal curve with many response dura-
tions accumulating in Bins 11 through 13 in-
dicates accurate responding, with many re-
sponses having durations between 1.0 and 1.3
s and producing the reinforcer. Leftward
shifts of the cumulative curves outside the
dashed lines that encompass Bins 11 through
13 indicate relative increases in response du-
rations too short to produce the reinforcer,
and rightward shifts indicate increases in the
relative frequency of response durations too
long to produce the reinforcer. A decrease in
slope of the sigmoidal curve (seen in later
figures) occurs when the relative frequency of
both longer and shorter response durations
increases.
As the force requirement on the response
lever increased (uppercase letters in Figure
2), the cumulative frequency distribution
curves shifted upward and leftward in all rats
due to the relative increase in short response
durations (see Curves B and C). In contrast,
decreases in the force requirements shifted
the cumulative frequency distributions right-
ward (lowercase letters, Curves a, b, c), with
more response durations falling within the
vertical dashed lines indicating a higher pro-
portion of reinforced responses. Rat 408 also

Page 8

302G. Y. H. MCCLURE et al.
Fig. 2.
1–1.3-s schedule. Abscissa: response durations in 0.1-s time bins. Ordinate: cumulative number of responses in each
bin as a percentage of the total number of responses. Dashed vertical lines encompass the reinforced response
durations. Each drug curve represents a single observation in each rat. Each point for the control curve represents
a mean based on 10 observations in each rat. Brackets show ?1 SD. The filled points show data from sessions at
baseline force and the letters show force changes as follows: A ? ?10 g, B ? ?15 g, C ? ?20 g, a ? ?10 g, b ?
?15 g, c ? ?20 g.
Cumulative frequency distributions of response durations at different force requirements under the TRD
showed a slight increase in the frequency of
some very long response durations when the
force requirements were decreased by 15 g
(see Curve b), which caused the cumulative
frequency curve to reach asymptote much
more slowly than for other rats.
Figure 3 shows the effect of PCP on accu-
racy with the force required for lever pressing
at the training level (baseline force), and with
the requirement of either an increase in 10
g or a decrease in 15 g of force (individual-
animal data are shown in Table 4). PCP pro-
duced a dose-dependent decrease in accuracy
when the force requirements were the same
as during training. When the force require-
ment was increased, accuracy after saline ad-
ministration showed little change, and PCP
reduced accuracy with a dose–response curve
that fell slightly below and was approximately
parallel to the original PCP dose–response
curve before the force requirements were
changed. When the force requirement for le-
ver pressing was decreased by 15 g, the base-
line accuracy after saline was higher, but the

Page 9

303DRUG EFFECTS AND FORCE REQUIREMENTS
Fig. 3.
on accuracy under different force requirements. Abscis-
sa: mg/kg dose of PCP, log scale. Ordinate: percentage
of responses between 1 and 1.3 s in duration. Each point
is an average single response in each of 4 animals. Tri-
angles indicate baseline force requirements, circles indi-
cate a 10-g increase in force requirements on the lever,
and squares indicate a 15-g decrease in force require-
ments on the lever. All saline points are mean of three
observations immediately preceding or following the
drug administrations. Brackets show ?1 SD.
Dose–response curves for the effects of PCP
PCP dose–response curve for accuracy was
very similar to the original dose–response
curve.
Table 4 shows the effects of PCP on accu-
racy, on the number of responses emitted
during the session, and on the response rate
for each rat. With the force requirements for
the lever press at the baseline level, PCP de-
creased accuracy, increased the total number
of responses emitted, and increased the re-
sponse rate. In general, these effects of PCP
increased with increasing doses. Accuracy was
negatively correlated with response rate (r ?
?.85) as the dose of PCP increased.
When the PCP dose–response curve was re-
determined with 15-g less force required to
press the lever (middle section of Table 4),
the effects of PCP were very similar to those
when the lever was at the baseline force, until
the highest dose was reached. That is, increas-
ing doses of PCP decreased accuracy, in-
creased the number of responses emitted,
and increased the response rate. After the 3.0
mg/kg dose of PCP, however, the number of
responses emitted and the response rate de-
creased for Rat 413, and especially for Rat
405. There continued to be a negative cor-
relation between accuracy and response rate
(r ? ?.80) as the dose increased, if the data
for Rat 405 after the highest dose of PCP
were not entered into the calculation of these
correlations because only a single response
was made.
When the PCP dose–response curve was re-
determined with an increase in 10 g of force
required to press the lever (bottom section of
Table 4), the effects of PCP were slightly dif-
ferent. Low doses of PCP (0.3 and 1.0 mg/
kg) again decreased accuracy, increased the
number of responses emitted, and increased
the response rate, but at higher doses the to-
tal number of responses and thus the re-
sponse rate decreased for 3 of the 4 rats (1.7
mg/kg for Rats 402 and 405; 3.0 mg/kg for
Rat 408). These decreases after higher doses
of PCP decreased the negative correlations
between accuracy and response rate as the
dose of PCP increased (r ? ?.03).
Figure 4 shows the individual cumulative
response distributions after PCP administra-
tion for each rat at each lever-press force re-
quirement. The top row shows the effects of
PCP on the cumulative relative frequency dis-
tributions of response durations in these rats
when the force requirements were the same
as during training. The middle row shows the
effects of decreasing the force requirements,
and the bottom row shows the effects on in-
creasing the force requirements on the rela-
tive frequency distributions of response du-
rationsafterPCP.
requirements after PCP administration, the
curves shift leftward showing an increase in
the relative frequency of shorter response du-
rations which become more pronounced with
increasing doses of PCP. One exception to
this finding is that at the lowest dose of PCP
(0.3 mg/kg), there was a decrease in fre-
quency of short response durations for Rat
405. In addition to an increase in the relative
frequency of short response durations, Rat
402 produced an increase in longer response
durations at the 1.7 mg/kg dose, causing the
flattening of the slope of the cumulative re-
sponse-duration distribution curve (Curve
C). Also in this rat, at the 3.0 mg/kg dose a
large increase in shorter responses shifted the
curve leftward to such an extent that almost
Atbaseline force

Page 10

304G. Y. H. MCCLURE et al.
Fig. 4.
different force requirements. Abscissa: response durations in 0.1-s time bins. Ordinate: cumulative number of re-
sponses in each bin as a percentage of the total number of responses. Each PCP curve represents single observations
in each rat. Each saline curve represents a mean of three observations. Brackets show ?1 SD. Doses are indicated in
the key. Dashed vertical lines encompass reinforced response durations. Squares indicate the control baseline, in-
verted triangles indicate the effects of saline with the 15-g decrease in the force requirement, and triangles indicate
the effects of saline with the 10-g increase in the force requirement. PCP doses are as follows: ● ? saline, A ? 0.3,
B ? 1.0, C ? 1.7, D ? 3.0 mg/kg.
Effects of phencyclidine on the cumulative relative frequency distribution of response durations with
no responses fell in the reinforced zone
(Curve D).
When the force requirements were de-
creased by 15 g (Figure 4), distributions also
shifted toward shorter response durations af-
ter PCP administration. The magnitude of
the increase in short response durations was
greater than under baseline force require-
ments. These effects became more pro-
nounced after higher doses of PCP, at which
short response durations became more fre-
quent. Thus PCP shifted the cumulative rel-
ative response-duration distribution to the
left.
When the force requirements were in-
creased by 10 g (Figure 4), the response-du-
ration distribution also shifted toward shorter
durations after PCP aministration. Again, the
magnitude of these shifts was greater than un-
der the baseline force requirments. The de-
gree to which the distribution shifted to the
left generally increased with the dose, and af-
ter the 3.0 mg/kg dose (D curves) the shifts
were particularly large.
Figure 5 shows the effect of methamphet-
amine on accuracy under the same three
force requirements for lever pressing (indi-
vidual data are shown in Table 5). There was
a dose-dependent decrease in accuracy when
the required force to press the lever was the

Page 11

305DRUG EFFECTS AND FORCE REQUIREMENTS
Fig. 5.
amphetamine on accuracy at different lever force re-
quirements. Details as in Figure 3.
Dose–response curves for the effects of meth-
same as during training, although the dose–
response curve was not very steep. The ad-
dition of the 10-g force requirement pro-
duced minimal change in the performance
after saline administration (Figure 5), and
the increased force requirement decreased
accuracy only slightly after doses of metham-
phetamine relative to the original metham-
phetamine dose–response curve. However,
doses of 0.3 to 3.0 mg/kg methamphetamine
had little effect on accuracy when force re-
quirements were decreased by 15 g. The ef-
fect of the highest dose of 5.6 mg/kg was
omitted from this figure because this dose
eliminated responding in all but 1 rat.
Table 5 shows the effects of methamphet-
amine on accuracy, number of responses, and
response rate. With the force requirements
for the lever press at the baseline level, meth-
amphetamine produced a dose-dependent
decrease in accuracy and an increase in the
number of responses emitted until the 5.6
mg/kg dose was given, which almost elimi-
nated responding in all rats. Doses of 0.3 and
1.0 mg/kg methamphetamine increased the
response rate, but at the 3.0 mg/kg dose, the
response rate began to decline. The correla-
tion between accuracy and response rate was
low (r ? ?.15).
When the methamphetamine dose–re-
sponse curve was redetermined with 15-g less
force required to press the lever (middle sec-
tion of Table 5), accuracy decreased only
slightly at doses lower than 5.6 mg/kg. The
number of responses emitted showed some
small increases after most doses, except that
the highest dose (5.6 mg/kg) decreased the
number of responses in all rats except Rat
408. Response rates increased slightly at the
0.3 mg/kg dose and then showed a gradual
decrease with increasing doses in all rats. Ac-
curacy was only moderately correlated with
the response rate (r ? ?.38).
When the methamphetamine dose–re-
sponse curve was redetermined with 10 g of
additional force required to press the lever
(bottom section of Table 5), accuracy de-
creased as the dose increased (Rats 402 and
405) or decreased at a low dose and re-
mained decreased until responding was elim-
inated (Rats 408 and 413). The number of
responses increased after the 0.3 and the 1.0
mg/kg doses and then decreased at higher
doses, with essentially no responding occur-
ring after the 3.0 mg/kg dose for Rat 413 and
after the 5.6 mg/kg dose for all rats. Rates of
responding showed a similar pattern. Accu-
racy was negatively correlated with the re-
sponse rate (r ? ?.57).
Figure 6 shows the individual cumulative
response distributions for all 4 rats after
methamphetamine administration. At the
baseline training force requirement, each rat
showed increases in the relative frequency of
short response durations that became more
pronounced as the dose increased, although
the effects were small for Rats 405 and 413.
The 3.0 mg/kg dose also increased the fre-
quency of response durations that were too
long to produce the reinforcer. These in-
creases in longer response durations were
particularly pronounced for Rats 402 and
413. Thus, the 3.0 mg/kg dose flattened the
distribution, increasing the relative frequency
of response durations too short and too long
to produce the reinforcer.
When the force requirements on the lever
were decreased by 15 g (Figure 6), the effects
of methamphetamine were attenuated in all
animals except Rat 413, for which there was
little effect of methamphetamine to be atten-
uated under baseline conditions. Metham-
phetamine had little effect on the cumulative
response-duration distributions of Rat 413
when the lever-press requirement was at the
baseline force or when an increase in force

Page 12

306G. Y. H. MCCLURE et al.
Fig. 6.
different force requirements. Details as in Figure 4, except that doses are as follows: ● ? saline, A ? 0.3, B ? 1.0, C
? 3.0, D? 5.6 mg/kg.
Effects of methamphetamine on the cumulative relative frequency distribution of response durations with
of 15 g was required. For Rats 402 and 405
the decreased force requirement resulted in
an increased accuracy after the 3.0 mg/kg
dose (C curves).
When the force requirements for respond-
ing were increased by 10 g, for all rats meth-
amphetamine increased the frequency of
short response durations. The increases in
the relative frequency of short response du-
rations were greater than when the baseline
force requirement was in effect for Rats 402,
408, and 413, and about the same for Rat 405.
Rat 408 also showed an increase in the fre-
quency of long response durations after the
3.0 mg/kg dose. Thus methamphetamine
produced increases in the relative frequency
of short responses when the lever-press force
requirements were the same as during train-
ing. When less force was required to press the
lever, these effects were attenuated. When
greater force was required to press the lever,
the direction of the effects depended on the
dose and the rat.
DISCUSSION
Fifty to 66% of the responses of rats trained
to hold a lever down for at least 1.0 s but less
than 1.3 s were reinforced. Most of the un-
reinforced responses were too short to pro-
duce the reinforcer. This finding is similar to
previous reports on the patterns of respond-
ing under this TRD schedule (Hudzik & Mc-
Millan, 1994a, 1994b; McMillan et al., 1994;
McMillan & Patton, 1965). It was suggested
that these precisely timed responses might re-
quire the acquisition of precise motor re-
sponses to produce the reinforcer. If this is

Page 13

307DRUG EFFECTS AND FORCE REQUIREMENTS
the case, changing the force requirements on
the lever might be expected to disrupt per-
formance, perhaps by changing the proprio-
ceptive feedback from lever presses. As ex-
pected, increases in the force requirement
for lever pressing did decrease accuracy. Fur-
thermore, this decrease in accuracy became
more pronounced as the force required to
press the lever increased. However, when the
force requirements for lever pressing were
decreased, a disruption of accuracy in re-
sponding was also expected for similar rea-
sons. This did not occur. Decreasing the force
requirements to operate the lever increased
accuracy in performance under the TRD 1–
1.3-s schedule. Therefore, accuracy on this
timing task cannot be explained entirely on
the basis of the conditioning of a precise mo-
tor response controlled by proprioceptive
feedback from lever pressing.
It is also difficult to explain these results in
terms of a central ‘‘clock’’ or timing mecha-
nism for the same reasons. If the animal is
producing a timed response, the duration of
which is controlled by a timing mechanism
presumably located in the central nervous sys-
tem, it is not clear why increasing the force
requirements should disrupt performance
but decreasing the force requirements should
improve performance.
In these experiments, accuracy appeared to
be inversely proportional to the force re-
quirements for lever pressing. Such data
might suggest a relation between response ef-
fort and performance on this timing task.
Perhaps a requirement for increasing effort
at higher force requirements caused the lever
to be released sooner, and a requirement for
decreasing effort with lower force require-
ments caused the lever to be held down lon-
ger. This supposition is consistent with the
data in Figure 2 that suggest an increase in
the relative frequency of short responses that
were not reinforced as the force requirement
increased and a decrease in the relative fre-
quency of these short responses as the force
requirement decreased.
Additional data also suggest a relation be-
tween lever pressing and effort in making
these precisely timed responses. Performance
at baseline appeared to be loosely related to
the force requirement in different rats. The
rat that had the lowest force requirement had
the highest percentage of correct responses.
The other 3 rats showed nearly identical base-
line accuracy, although the force required to
press the lever ranged from 25 to 32 g.
It is also possible that the changes in ac-
curacy as the force requirements changed
were influenced by changes in the physical
characteristics of the lever. For example, add-
ing weight to the lever on the side of the ful-
crum away from the animal might cause the
microswitch to open more rapidly when the
lever was released, but subtracting weight
might slow the return of the microswitch to
the normally open position. This might ex-
plain why the relative frequency of short re-
sponses increased with increasing force re-
quirements and decreased with decreasing
force requirements. However, it is difficult to
explain why a ‘‘sluggish’’ lever return would
lead to an increased reinforcement frequency
after extended training with lower force re-
quirements, and such considerations do not
explain the drug effects that were observed.
Furthermore, the Gerbrands lever is con-
structed such that the microswitch spring
pushes in the same direction as the weights
on the lever. Even without weights on the le-
ver, the microswitch spring is strong enough
to return to its normally open position.
Increases in force requirements did not
cause the rats to decrease the total number
of responses emitted or the response rate. In
fact, total responses and response rate both
increased at higher force requirements. The
negative correlations between accuracy and
both responses emitted and response rate,
along with the shifts toward shorter response
durations, suggest that as the force require-
ments increased, the proportion of responses
too short to produce the reinforcer in-
creased. This caused the total number of re-
sponses to increase, probably not so much be-
cause the response durations were shorter,
but rather because the decreased frequency
of reinforced responses resulted in rats hav-
ing to emit more responses to obtain the 50
reinforcers within the 40-min session. Re-
sponse rate might also increase due to the
increased frequency of shorter responses,
which results in a decreased rate of reinforcer
delivery and thus less pausing to eat food pel-
lets.
With baseline force requirements in effect,
PCP decreased accuracy as a function of in-
creasing dose. This effect occurred primarily

Page 14

308 G. Y. H. MCCLURE et al.
because of increases in the relative frequency
of response durations too short to produce
the reinforcer. Hudzik and McMillan (1994a)
and McClure et al. (1997) also found that
PCP decreased accuracy by increasing the rel-
ative percentage of short response durations
using this TRD schedule.
The tendency for PCP to increase the rel-
ative frequency of short responses was poten-
tiated both by increasing and decreasing the
force required to operate the lever, although
the magnitude of the effect was somewhat
larger when force increases were required.
Because both increasing and decreasing the
force required to operate the lever increased
the proportion of short response durations
after PCP administration, it seems unlikely
that PCP produced its effect by interacting
with effortfulness of the response. Instead,
PCP may have produced its effects by inter-
acting with central nervous system timing
mechanisms, such as the internal clock pro-
posed by Church and Meck (Church, 1984;
Meck, 1983; Meck & Church, 1983, 1987).
According to the internal clock concept,
the ‘‘speeding up’’ of the internal clock
would result in release of the lever before the
minimum duration required for reinforce-
ment had elapsed. This should occur under
all force requirements, which is what was ob-
served. Why there should be a potentiation
of the increased frequency of short response
durations produced by PCP with both in-
creases and decreases in force requirements,
however, is not readily explained by reference
to an internal clock.
Whatever mechanism produced the effects
of PCP, it was not that the total number of
responses was being suppressed by PCP or
that responding was extinguishing due to in-
frequent delivery of the reinforcer after drug.
The low doses of PCP that decreased rein-
forcement frequency increased the total
number of responses and the response rate.
Changes in the reinforcement rate after drug
do not appear to be a major mechanism in
disrupting timing behavior under TRD sched-
ules. For example, reinforcing only 50% of
the correct responses increased accuracy un-
der the TRD schedule, rather than decreas-
ing it (McMillan et al., 1994).
Another possible explanation for these
data is that the effects of PCP are determined
by the baseline accuracy. The dose–response
curves for PCP determined at different lever
weights had similar shapes and slopes. The
primary differences in the PCP curves are
that they begin their descents from different
levels of accuracy, which in turn are largely a
function of the force required to operate the
lever. According to this description, baseline
performance is modified by changes in force
requirements, but the effects of PCP are very
similar at each force requirement.
Under baseline force requirements, meth-
amphetamine also decreased the frequency
of reinforced responding as a function of
dose. This effect also was caused by an in-
creased frequency of short response dura-
tions, although at high doses there was also a
tendency for an increase in longer durations,
indicative of a flattening of the response-du-
ration distribution. These effects of meth-
amphetamine are also similar to those re-
ported previously (Hudzik & McMillan,
1994a; McClure et al., 1997). These effects of
methamphetamine were attenuated when the
force requirements for lever pressing were re-
duced. The shifts toward shorter response du-
rations that methamphetamine produced
were attenuated, and the curves relating the
percentage of correct responses to the dose
of methamphetamine remained relatively flat
in all rats. The decrease in force requirement
appeared to be protecting the rats against the
methamphetamine effect. In contrast, when
the force requirements for lever pressing
were increased, methamphetamine again re-
duced the frequency of reinforcer delivery.
Although the decrease was related to an in-
crease in short response durations, the in-
creases in long response durations were elim-
inated. The differences in the effects of
methamphetamine as a function of lever
force requirements and the individual differ-
ences among animals make a simple expla-
nation based on the speeding up of an inter-
nal timing mechanism difficult to reconcile.
Although we are not able to explain pre-
cisely the mechanism by which the timing of
these short responses occurs, nor are we able
to specify exactly how the drug effects are me-
diated, the data were consistent across ani-
mals. The experiments illustrate the complex-
ity of timed responding and the effects of
drugs on this responding. We have shown
previously that the effects of drugs on re-
sponding under TRD schedules depend on

Page 15

309DRUG EFFECTS AND FORCE REQUIREMENTS
the drug (Hudzik & McMillan, 1994a, 1994b;
McClure & McMillan, 1997; McClure et al.,
1997), the duration of the response being
timed (McClure & McMillan, 1997; McClure
et al., 1997), what the animal is required to
do during timing (McClure & McMillan,
1997; McClure et al., 1997), and the schedule
under which timed responses are reinforced
(McMillan et al., 1994). Mechanisms such as
changes in the speed of a hypothetical inter-
nal clock or changes in proprioceptive re-
sponse feedback are unlikely to provide a full
explanation for all of the complex interac-
tions between drugs and timing behavior.
REFERENCES
Church, R. M. (1984). Properties of the internal clock.
New York Academy of Sciences Conference on Tim-
ing and Time Perception. Annals of the New York Acad-
emy of Sciences, 423, 566–582.
Hudzik, T. J., & McMillan, D. E. (1994a). Drug effects
on response-duration differentiation: I. Differential
effects of drugs of abuse. Psychopharmacology, 114,
620–627.
Hudzik, T. J., & McMillan, D. E. (1994b). Drug effects
on response-duration differentiation. II: Selective ef-
fects of antidepressant drugs. Journal of Pharmacology
and Experimental Therapeutics, 268, 1335–1342.
Li, A. A., Marek, G. J., Hand, T. H., & Seiden, L. S.
(1990). Antidepressant-like effects of trazodone on a
behavioral screen are mediated by trazodone, not the
metabolite m-chlorophenylpiperazine. European Jour-
nal of Pharmacology, 177, 137–144.
Marek, G. J., Li, A. A., & Seiden, L. S. (1989). Selective
5-hydroxytryptamine2antagonists have antidepres-
sant-like effects on differential-reinforcement-of-low-
rate 72-second schedule. Journal of Pharmacology and
Experimental Therapeutics, 250, 52–59.
Marek, G. J., & Seiden, L. S. (1988). Selective inhibition
of MAO-A, not MAO-B, results in antidepressant-like
effects on DRL 72-s behavior. Psychopharmacology, 96,
153–160.
McClure, G. Y., & McMillan, D. E. (1997). Effects of
drugs on response duration differentiation. VI: Dif-
ferential effects under differential reinforcement of
low rates of responding schedules. Journal of Pharma-
cology and Experimental Therapeutics, 281, 1368–1380.
McClure, G. Y., Wenger, G. R., & McMillan, D. E. (1997).
Effects of drugs on response duration differentiation.
V: Differential effects under temporal response dif-
ferentiation schedules. Journal of Pharmacology and Ex-
perimental Therapeutics, 281, 1357–1367.
McGuire, P. S., & Seiden, L. S. (1980). Differential ef-
fects of imipramine in rats as a function of DRL
schedule value. Pharmacology Biochemistry and Behavior,
13, 691–694.
McMillan, D. E., Adams, S. L., Wenger, G. R., McClure,
G. Y., & Hardwick, W. C. (1994). Effects of drugs on
response duration differentiation. III. Acute variation
of reinforced duration. Pharmacology Biochemistry and
Behavior, 48, 941–957.
McMillan, D. E., & Patton, R. A. (1965). Differentiation
of a precise timing response. Journal of the Experimental
Analysis of Behavior, 8, 219–226.
Meck, W. H. (1983). Selective adjustment of the speed
of internal clock and memory processes. Journal of Ex-
perimental Psychology: Animal Behavior Processes, 9, 171–
201.
Meck, W. H., & Church, R. M. (1983). A mode control
model of counting and timing processes. Journal of
Experimental Psychology: Animal Behavior Processes, 9,
320–334.
Meck, W. H., & Church, R. M. (1987). Cholinergic mod-
ulation of the content of temporal memory. Behavioral
Neuroscience, 101, 457–464.
O’Donnell, J. M., & Seiden, L. S. (1983). Differential-
reinforcement-of-low-rate 72-second schedule: Selec-
tive effects of antidepressant drugs. Journal of Phar-
macology and Experimental Therapeutics, 224, 80–88.
Schulze, G. E., McMillan, D. E., Bailey, J. R., Scallet, A.,
Ali, S. F., Slikker, W., Jr., & Paule, M. G. (1988). Acute
effects of delta-9-tetrahydrocannabinol in rhesus mon-
keys as measured by performance in a battery of com-
plex operant tests. Journal of Pharmacology and Experi-
mental Therapeutics, 245, 178–186.
Schulze, G. E., & Paule, M. G. (1990). Acute effects of
d-amphetamine in a monkey operant behavioral test
battery. Pharmacology Biochemistry and Behavior, 35,
759–765.
Schulze, G. E., & Paule, M. G. (1991). Effects of mor-
phine sulfate on operant behavior in rhesus monkeys.
Pharmacology Biochemistry and Behavior, 38, 77–83.
Schulze, G. E., Slikker, W., Jr., & Paule, M. G. (1989).
Multiple behavioral effects of diazepam in rhesus
monkeys. Pharmacology Biochemistry and Behavior, 34,
29–35.
Received June 15, 1999
Final acceptance June 27, 2000