QUANTIFICATION OF ETHANOL'S ANTIPUNISHMENT EFFECT IN HUMANS USING THE GENERALIZED MATCHING EQUATION

JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR
2009, 92, 161–180
NUMBER 2 (SEPTEMBER)
QUANTIFICATION OF ETHANOL’S ANTIPUNISHMENT EFFECT IN HUMANS USING THE
GENERALIZED MATCHING EQUATION
ERIN B. RASMUSSEN
IDAHO STATE UNIVERSITY
AND
M. CHRISTOPHER NEWLAND
AUBURN UNIVERSITY
Increases in rates of punished behavior by the administration of anxiolytic drugs (called antipunish-
ment effects) are well established in animals but not humans. The present study examined
antipunishment effects of ethanol in humans using a choice procedure. The behavior of 5 participants
was placed under six concurrent variable-interval schedules of monetary reinforcement. In three of the
six concurrent schedules, punishment, in the form of monetary loss, was superimposed on one
alternative. Data were analyzed according to the generalized matching equation which distinguishes
between bias (allocation of behavior beyond what matching to relative reinforcer densities would
predict) and sensitivity to reinforcement (how well behavior tracks relative reinforcer densities). In
addition, participants completed a pencil-tapping test. Under placebo punishment conditions, all
participants demonstrated low response rates and a bias against the alternative associated with
punishment, despite a resultant loss of available reinforcers. Bias against the punished alternative was
dose-dependently reduced in participants shown to be most sensitive to ethanol (0.6, 1.2, and 1.8 g/kg)
in measures of overall responding and on the pencil-tapping test. No ethanol-induced change in bias
was noted when punishment was not imposed. Sensitivity to reinforcement also decreased for
participants shown to be sensitive to ethanol. In addition to extending antipunishment effects to
humans, these results also show that antipunishment effects can be quantified via the matching
equation.
Key words: antipunishment, choice, concurrent schedules, ethanol, generalized matching equation,
humans, matching, punishment
_______________________________________________________________________________
Anxiolytic drugs can be characterized by
effects has been studied. Results of several
their ability to increase rates of responding
studies have indicated that analgesia is unlikely
suppressed by punishment, and a vast litera-
to play a role since antipunishment effects
ture documents these effects with a variety of
have not been noted with potent analgesics
anxiolytic compounds in nonhuman species
like morphine (e.g., McCloskey, Paul, &
(see Commissaris, 1993, Pollard & Howard,
Commissaris, 1987; McMillan & Leander,
1990, and Rasmussen, 2006, for reviews.)
1975). Moreover, antipunishment effects to
These reports also demonstrate the specificity
anxiolytics are observed with several different
of antipunishment effects to mostly benzodi-
types of punishers, including electric shock,
azepines and barbiturates. Stimulants, opiates
pressurized air (Spealman, 1979), and timeout
and other compounds do not reliably increase
from reinforcement (van Haaren & Anderson,
behavior that has been suppressed by punish-
1997). The phenomenon is most consistently
ment. Because most punishing stimuli induce
observed with drugs that act on GABAergic
pain, the role of analgesia in antipunishment
systems, though it has also been noted with
serotonergic anxiolytics (McCloskey, Paul, &
Commissaris, 1987; van Haaren & Anderson,
Funding for these studies was provided by the Graduate
1997), but with less consistency (Sanger,
School and Department of Psychology at Auburn Univer-
sity and Sigma Xi. Portions of this manuscript were
1990).
presented at the Association for Behavior Analysis and
Several studies document antipunishment
the Southeastern Association for Behavior Analysis.
effects with ethanol, another GABAergic com-
Address correspondence to: Erin B. Rasmussen, Ph.D.,
pound, in nonhuman species (Barrett, Brady,
Dept. of Psychology, Campus Box 8112, Idaho State
University, ID 83206 (e-mail: rasmerin@isu.edu).
& Witkin, 1985; Glowa & Barrett, 1976; Koob,
doi: 10.1901/jeab.2009.92-161
Braestrup & Britton, 1986; Vogel, 1980). In a
161

---

162
ERIN B. RASMUSSEN and M. CHRISTOPHER NEWLAND
study by Vogel (1980), for example, licking of
the two alternatives approximately matches the
a water tube by rats was suppressed when each
relative rates of reinforcement that they
twentieth lick produced shock (FR 20). Etha-
produce
(Herrnstein,
1961;
Davison
&
nol (0.5–2 g/kg) increased punished licking
McCarthy, 1988). This relationship also has
in a manner that resembled the increases
been described quantitatively using the gener-
observed with chlordiazepoxide, a drug that
alized matching equation (Baum, 1974), a
has well established antipunishment effects.
power-law formulation that partitions two
Licking the tube when punishment was not
sources of deviation from matching: bias and
scheduled was not increased by these same
sensitivity.
doses, so rate increases were specific to the
The generalized matching equation is often
presence of a punishment contingency. Anti-
expressed as:
punishment effects with ethanol have also
been reported with rats (Barrett et al., 1985)
log B
ð a=BbÞ~logðkÞzc logðRa=RbÞ
ð1Þ
and squirrel monkeys (Glowa & Barrett, 1976)
in a variety of punishment procedures.
where the ratio of behavioral responses allo-
There is indirect evidence that antipunish-
cated to two reinforcer alternatives (Ba and Bb)
ment effects are relevant to the use of
is related to the ratio of reinforcers earned by
anxiolytics in human clinical psychopharma-
the two alternative responses (Ra and Rb). The
cology. As noted earlier, the effect is obtained
two free parameters, log k and c, describe bias
with drugs that are used clinically for manag-
and sensitivity to reinforcement, respectively.
ing anxiety. Doses of these drugs that produce
Bias (log k) appears as a preference for one
anxiolytic effects in animals are closely corre-
alternative and may come from characteristics
lated with doses that are clinically efficacious
of the experimental setting, such as difficulty
with humans (Cook & Davidson, 1973; Kleven
& Koek, 1999). The evidence for generaliza-
operating one response device (Davison &
tion of antipunishment effects from animals to
McCarthy, 1988) or, here, punishment of
humans is indirect, but intriguing. Socially
responding on one alternative. A log k . 0
inappropriate behavior, such as aggression
means behavior is biased toward the numera-
(e.g., Cherek, Steinberg, & Manno, 1985;
tor (alternative a in equation 1); log k , 0
Cherek, Steinberg, & Vines, 1984; Dougherty,
means that behavior is biased toward the
Cherek, & Bennett, 1996) or observing sexu-
denominator (alternative b). Sensitivity to
ally explicit pictures (Kallmen & Gustafson,
reinforcement refers to the manner in which
1998) is increased by alcohol in humans.
the ratio of responding on the two alternatives
Under nondrug conditions, aggression or
tracks the reinforcer ratio delivered by them. A
viewing sexually explicit pictures occurred at
a low rate, and it may be the case that these
value of c 5 1, which is not commonly seen, is
behaviors were suppressed by punishers from,
matching, and the response ratios equal rein-
for example, social or cultural sources prior to
forcer ratios. If c , 1, then response ratios are
the experiment. The direct demonstration of
less sensitive to reinforcer ratios, which is
antipunishment effects, however, requires the
called undermatching and is a common finding
suppression of baseline behavior by a re-
in animals (Baum, 1974) and humans (Kollins,
sponse-contingent punisher, and the subse-
Newland, & Critchfield, 1997; Pierce & Epling,
quent increase in punished behavior pro-
1983). A value of c . 1, called overmatching,
duced by the drug. The present study was
indicates that changes in response ratios are
designed to study ethanol using this approach.
highly sensitive to changes in reinforcer ratios.
Punishment has been studied and quanti-
fied in humans by applying the generalized
Overmatching occurs, for example, when
matching equation, a model of choice, to
there is a high cost for switching alternatives
behavior under concurrent schedules of rein-
or if changing from one alternative to another
forcement. Under concurrent schedules, two
is punished (Todorov, 1971).
response alternatives are available simulta-
A schedule of punishment superimposed on
neously and the two responses are often
one response alternative of a concurrent
maintained under separate variable-interval
schedule shifts the distribution of behavior
schedules. Allocation of responses or time to
toward the alternative associated with no

---

ETHANOL’S ANTIPUNISHMENT EFFECTS
163
punishment even when there is a net loss of
the earlier findings by examining the degree
reinforcement. This shift has been reported
to which ethanol increases rates of punished
with animals (Deluty & Church, 1978; de
responding and diminishes bias toward the
Villiers, 1980; Farley & Fantino, 1978; Wojnicki
alternative associated with no punishment.
& Barrett, 1993) and humans (e.g., Bradshaw,
Based on previous studies (e.g., Carlton et
Szabadi, & Bevan, 1979; Carlton, Siegel,
al., 1981; Rasmussen & Newland, 2008), it was
Murphee, & Cook, 1981; Critchfield, Paletz,
hypothesized that bias of 0 would appear
& MacAleese, 2003; Katz, 1973; Rasmussen &
under the no-punishment condition, and bias
Newland, 2008). A global shift in preference is
would be less than 0 (toward the unpunished
captured by the bias parameter (log k) in the
component) in the punishment condition
generalized matching equation (Carlton et al.,
during placebo conditions. It was further
1981; Rasmussen & Newland, 2008). In some
hypothesized that ethanol would, in a dose-
cases, punishment also lowers sensitivity to
related fashion, shift bias toward 0 and
reinforcement. For example, a study by Ras-
increase response rate under punishment.
mussen and Newland (2008) reported on
punishment effects with humans in a concur-
rent schedule arrangement with monetary gain
METHOD
as a reinforcer and monetary loss as a punisher.
Participants
When no punisher was present, both matching
Five male college students, at least 21 years
and undermatching occurred and no bias was
of age and weighing between 74–83 kg (165–
demonstrated toward either alternative. Under
185 pounds) participated. Participants were
the punishment condition, all participants
screened to ensure that there was no history of
exhibited a bias (log k range 20.20 to 20.76)
problem drinking patterns or other medical
toward the alternative associated with no
problems by completing the Rutgers Ethanol
punishment, and this was true whether ob-
Problem Inventory and a modified version of
tained reinforcers (total reinforcers earned on
the Daily Drinking Questionnaire (Collins,
each alternative), or net reinforcers (reinforc-
Parks, & Marlatt, 1985), which includes ques-
ers on each alternative minus those lost due to
tions about medical history. Male participants
punishment on an alternative) were used in the
were used so gender differences in alcohol
analysis. Moreover, all participants showed
metabolism did not introduce variability. Over
greater undermatching, as indicated by shal-
the course of the subjects’ participation in the
lower slopes, under punishment conditions
study, they were asked to refrain from ethanol
than under no-punishment conditions, suggest-
consumption outside of the experiment, such
ing punishment also reduced sensitivity to
that tolerance to ethanol would be minimized.
reinforcement densities of the two alternatives.
All experiments were approved by the Auburn
The present study examined the effects of
University Institutional Review Board.
ethanol on punished responding in human
participants in order to assess further the
Materials
extension of antipunishment effects to the
behavior of human participants. Behavior was
A DOS-based computer was used to present
maintained under concurrent schedules of
visual images, transduce responses from the
reinforcement with a conjoint schedule of
participants, present stimuli, and record data.
punishment superimposed on one response
All code was written in VisualBasicH. Participants
alternative. Ethanol’s antipunishment effects
were placed individually in small, separate,
were examined using the generalized match-
office-sized rooms containing a desk, chair,
ing equation. A detailed description of behav-
computer monitor, mouse, and mouse pad. A
ior under nondrug conditions with the same
Breathalyzer was used to monitor breath-alcohol
participants as the present study and the ability
concentration (BAC). BAC is a reliable, valid,
of the matching equation to characterize
and noninvasive predictor of blood-ethanol
punished responding has been described
concentration (Jones & Andersson, 1996).
previously (Rasmussen & Newland, 2008). In
that study, punishment created a bias toward
Procedure
the alternative associated with no punishment.
Procedures were similar to those reported in
In the present study, however, we extended
Rasmussen and Newland (2008), and the same

---

164
ERIN B. RASMUSSEN and M. CHRISTOPHER NEWLAND
5 participants from that report were used in
Programmed reinforcement ratios for these
the present study. A participant was escorted
schedules were 5:1, 1:1, and 1:5, respectively
into a 400-square-foot room and seated at a
and the specific values used resulted in the
desk with a computer monitor. On the first day
same overall programmed reinforcement rate
of the study, before participation began, each
for all conditions. Each concurrent schedule
participant completed a consent form and was
was in effect until stability occurred. Stability
given a set of instructions to read before
was defined as three consecutive sessions in
beginning the computer task. After the partic-
which response allocation (percent of respons-
ipant mouse-clicked a START icon on the
es on the left alternative) differed by no more
monitor, an 8-min concurrent schedule ses-
than 5% of the mean of the last three sessions,
sion began. In the 8-min session, the monitor
with no trends apparent. Table 1 summarizes
screen was split vertically into halves and a
the order of conditions that each participant
small colored box was located in the middle of
experienced. The order of schedules was
each half of the screen. For each participant,
counterbalanced across participants.
the left box (alternative A) and right box
In the current study, participants performed
(alternative B) were different colors. A mouse-
five to seven 8-min sessions within an approx-
click on either box started both boxes moving
imate 1-hr block of sessions. (Session blocks
in a random pattern at a constant rate
sometimes ran up to 65 min, depending on
throughout each half of the screen. Partici-
whether a participant needed a short break
pants could mouse-click (the response of
between sessions at some point during the
interest) on one of the two moving colored
session block to, for example, use the re-
box icons, but could not click on both boxes
stroom.) One session block (or visit to the
simultaneously. Each box was associated with
laboratory) was conducted per day, and at least
one component of a VI schedule of reinforce-
2 days separated each session block. Therefore
ment in which a reinforcer (a flashing ‘‘+4¢’’
a participant could make up to four visits a
icon that appeared on the screen for 2 s,
week for a total of four session blocks per
representing that 4¢ had been earned) was
week. If stability under one concurrent sched-
delivered on each alternative after the first
ule was reached before the end of a session
response after a varying amount of time
block (e.g., stability under conc VI 12-s VI 60-s
elapsed (described below). Participants could
was reached within three sessions), the next
switch alternatives throughout the session; a
concurrent schedule was placed in effect in
2-s changeover delay during which points
the next session and remained in effect until
could not be delivered or subtracted was
stability occurred (e.g., conc VI 20-s VI 20-s
imposed after each switch.
would then be implemented). If stability
No-punishment condition. Each participant’s
within a concurrent schedule did not occur
mouse-clicking was reinforced under a con-
by the end of a session block, the same
current VIA VIB (conc VIA VIB) schedule (A
schedule was implemented again at the begin-
represented the value on the left alternative,
ning of the next session block.
or alternative A; B represented the value on
Punishment condition. After behavior stabi-
the right alternative, or alternative B). The
lized under the three concurrent schedules of
following schedules were used: conc VI 12-s VI
the no-punishment condition, a VI schedule of
60-s, conc VI 20-s VI 20-s, and conc VI 60-s VI
punishment (a flashing ‘‘24¢’’ that appeared
12-s. For example, under conc VI 12-s VI 60-s,
on the screen for 2 s and represented the loss
the first response on the alternative A was
of 4 cents) was superimposed on one alterna-
reinforced after the lapse of an average of 12 s;
tive of the concurrent schedule, forming a
for alternative B, the first response after the
conjoint schedule for that response alterna-
lapse of an average of 60 s on an independent,
tive. The average interval for the punishment
but continuously running, clock produced a
VI was programmed to be 25% greater than
reinforcer. The VI clock reset at the onset of
the average interval in the schedule of
the 2-s reinforcer interval. The VI schedules
reinforcement on which it was superimposed.
were arranged using Fleshler and Hoffman’s
For example, if a VI 20-s schedule of rein-
(1962)
constant
probability
distributions.
forcement was programmed for alternative A
There were no counters available to the
responses, then the punishment schedule, also
participant that tallied earnings or losses.
on the alternative A responses, would be VI 25-

---

ETHANOL’S ANTIPUNISHMENT EFFECTS
165
Table 1
Order of VIA VIB schedules for each participant. Reinstatement of no-punishment conditions is
omitted for brevity.
Participant 02
Participant 04
Participant 05
Participant 06
Participant 08
Placebo
0.6 g/kg
0.6 g/kg
Placebo
0.6 g/kg
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 20-s VI 20-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 60-s VI 12-s
VI 60-s (PUN) VI 12-s VI 12-s VI 60-s (PUN) VI 20-s VI 20-s (PUN) VI 20-s (PUN) VI 20-s VI 12-s (PUN) VI 60-s
VI 20-s (PUN) VI 20-s VI 20-s VI 20-s (PUN) VI 60-s VI 12-s (PUN) VI 12-s (PUN) VI 60-s VI 60-s (PUN)VI 12-s
VI 12-s (PUN) VI 60-s VI 60-s VI 12-s (PUN) VI 12-s VI 60-s (PUN) VI 60-s (PUN)VI 12-s
VI 20-s (PUN) VI 20-s
0.6 g/kg
1.2 g/kg
1.2 g/kg
0.6 g/kg
Placebo
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 20-s VI 20-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 60-s VI 12-s
VI 60-s (PUN) VI 12-s VI 12-s VI 60-s (PUN) VI 20-s VI 20-s (PUN) VI 20-s (PUN) VI 20-s VI 12-s (PUN) VI 60-s
VI 20-s (PUN) VI 20-s VI 20-s VI 20-s (PUN) VI 60-s VI 12-s (PUN) VI 12-s (PUN) VI 60-s VI 60-s (PUN)VI 12-s
VI 12-s (PUN) VI 60-s VI 60-s VI 12-s (PUN) VI 12-s VI 60-s (PUN) VI 60-s (PUN)VI 12-s
VI 20-s (PUN) VI 20-s
1.2 g/kg
1.8 g/kg
Placebo
1.2 g/kg
1.2 g/kg
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 20-s VI 20-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 60-s VI 12-s
VI 60-s (PUN) VI 12-s VI 12-s VI 60-s (PUN) VI 20-s VI 20-s (PUN) VI 20-s (PUN) VI 20-s VI 12-s (PUN) VI 60-s
VI 20-s (PUN) VI 20-s VI 20-s VI 20-s (PUN) VI 60-s VI 12-s (PUN) VI 12-s (PUN) VI 60-s VI 60-s (PUN)VI 12-s
VI 12-s (PUN) VI 60-s VI 60-s VI 12-s (PUN) VI 12-s VI 60-s (PUN) VI 60-s (PUN)VI 12-s
VI 20-s (PUN) VI 20-s
1.8 g/kg
Placebo
1.8 g/kg
1.8 g/kg
1.8 g/kg
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 12-s VI 60-s
VI 60-s VI 12-s
VI 12-s VI 60-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 20-s VI 20-s
VI 20-s VI 20-s
VI 60-s VI 12-s
VI 60-s VI 12-s
VI 60-s (PUN) VI 12-s VI 12-s VI 60-s (PUN) VI 20-s VI 20-s (PUN) VI 20-s (PUN) VI 20-s VI 12-s (PUN) VI 60-s
VI 20-s (PUN) VI 20-s VI 20-s VI 20-s (PUN) VI 60-s VI 12-s (PUN) VI 12-s (PUN) VI 60-s VI 60-s (PUN)VI 12-s
VI 12-s (PUN) VI 60-s VI 60-s VI 12-s (PUN) VI 12-s VI 60-s (PUN) VI 60-s (PUN)VI 12-s
VI 20-s (PUN) VI 20-s
s schedule. For each participant, the punish-
The results of this baseline phase are discussed
ment schedule was always on the same
in detail in Rasmussen and Newland (2008)
response alternative (A or B). As in the earlier
and will not be repeated here. The placebo
phase, each of the three concurrent schedules
and drug conditions for the present study
with punishment was in effect until stable
began after the baseline phase was completed.
responding was obtained. After stability was
Three volumes of vodka (40% ethyl alcohol)
obtained with punishment, one of the three
were mixed in orange juice so that the total
no-punishment
concurrent
schedules
was
solution administered was 16 oz, yielding
placed in effect again until stability was
ethanol doses of 0.6, 1.2 and 1.8 g/kg body
obtained, such that each concurrent schedule
weight. Doses were administered to all partic-
with punishment alternated with a no-punish-
ipants in increasing order. These doses yield
ment concurrent schedule. This was done to
approximately 0.64, 1.28, and 1.95 oz of
reestablish matching and to ensure that
absolute ethanol, respectively, for a 170 pound
punishment did not affect subsequent match-
male, and produced respective mean BACs of
ing. The reestablished conditions yielded data
approximately 0.02 (SD 5 0.003), 0.04 (SD 5
that were nearly identical to the initial
0.01), and 0.07 (SD 5 0.01) for our partici-
matching sessions, suggesting punishment
pants. Peak BACs did not exceed 0.09 under
did not affect subsequent matching (see data
any condition. A placebo condition was imple-
presented in Rasmussen & Newland, 2008, for
mented in which 1 mL of ethanol was floated
details).
on top of the mixer and rubbed on the sides of
Drug conditions. A complete series of condi-
the drinking glass. The placement of the
tions, including concurrent schedules with
placebo condition within the dose series was
punishment and no-punishment, were con-
counterbalanced across participants (see Ta-
ducted before the present study commenced.
ble 1). Doses (including placebo), and there-

---

166
ERIN B. RASMUSSEN and M. CHRISTOPHER NEWLAND
fore session blocks, were separated by at least 2
06, who completed 40 session blocks. Under a
days.
particular dose of ethanol, if a participant
Participants were asked to refrain from
completed six of the concurrent schedules
eating or drinking for 2 hr prior to the
across several session blocks, but reached
experiment. When a participant arrived, he
stability before the hour was up, the session
was given the drink with ethanol as described
block ended early, i.e., he was not given the
above, 1 hr before the session commenced.
next dose of ethanol. On the next visit, he was
The drink was given in thirds that were spaced
given the next dose and the concurrent
apart by 5 min to control for rate of consump-
schedule sequence continued. Table 2 shows
tion. BAC data were collected 1 hr after the
the number of sessions and session blocks
first third was administered. During this 1-hr
completed for each participant under each
waiting period, the participant was allowed to
schedule and dose of ethanol (Note: Reestab-
engage in quiet activities in the waiting area,
lishment data are not included in this table, but
such as reading a magazine or studying.
typically increased the total session blocks by 2–
To measure other effects of ethanol (e.g.,
4 per individual per drug dose).
motor effects), a pencil-tapping task was used.
After a session block was completed, partic-
Tapping has been used as a measure of motor
ipants were required to stay in the laboratory
function, and has assessed motor dysfunction
until BACs were 0.02 or below. They were
in degenerative motor disorders, such as
placed in the ‘‘recovery room’’ in which they
Parkinson’s and Huntington’s diseases (e.g.,
studied, slept, played games on the computer,
Mitchell et al., 2008; Nagasaki, Itoh, Mar-
or conversed with others.
uyama, & Hashizume, 1988; Ziv et al., 1999),
in aged populations (e.g., Cousins, Corrow,
Analysis
Finn, & Salamone, 1998; Nagasaki et al., 1988),
Dependent variables included the number
and in alcohol-dependent populations (e.g.,
of responses, reinforcers, and, for the pun-
Parks et al., 2003). Tapping has been used also
ished alternative, punishers delivered on alter-
in assessing the acute effects of CNS depres-
native A or B. Overall response rates on each
sants, such as ethanol and diazepam (e.g.,
alternative (responses per min) were deter-
Lindenschmist, Brown, Cerimele, Walle, &
mined by dividing the number of responses on
Forney, 1983; Palva, 1985). In this task,
each alternative by 8 (since session duration
participants were asked to tap a pencil against
was 8 min long). Local response rates also
paper as quickly as possible for 60 s, and were
were determined by dividing the number of
told that each pencil tap would result in
responses on each alternative by the time
2 cents added to their total earnings at the
spent on the respective alternative (see Ap-
end of the experiment. One pencil-tap test was
pendix for these time intervals), but no strong
given to each participant under each dose of
effects of ethanol or punishment were found
ethanol, and each test was given immediately
on local rates; therefore, overall response rates
before a session block began.
will be reported here. Obtained reinforcers
The participant then was placed in the
(total delivered), as opposed to net reinforcers
experimental conditions for a 1-hr session
(points delivered minus points lost through
block; only one session block was conducted
punishment) were used.
per day. During this time, 8-min sessions of a
Reinforcer and response ratios were con-
particular concurrent schedule were conducted
structed by dividing the number of responses,
until responding stabilized. Since it took more
or reinforcers, on alternative A by that for
than one session block to run all six concurrent
alternative B. The log of the response ratio was
schedules until stability was observed under
then expressed as a linear function of the log
each dose (stability often required more than
reinforcer ratio and this log-transformed ver-
three sessions—see Table 2), multiple session
sion was applied to Equation 1 using linear
blocks were conducted under each dose of
regression. Response and reinforcer ratios
ethanol, usually four to five (though some
from the last three consecutive stable sessions
participants completed a higher number, e.g.,
of each concurrent schedule were used for the
Participant 06), so participants experienced
analyses. The free parameters c (sensitivity)
between 15 to 26 session blocks across the
and log k (bias) were the dependent variables
experiment, with the exception of Participant
of interest. Though the punishment schedule

---

ETHANOL’S ANTIPUNISHMENT EFFECTS
167
Table 2
Number of sessions and session blocks for each condition for each participant.
Number of sessions
Partic.
Schedule
V
0.6
1.2
1.8
Total
02
Conc VI 12-s VI 60-s
5
4
5
6
20
Conc VI 20-s VI 20-s
4
5
3
6
18
Conc VI 60-s VI 12-s
5
3
4
3
15
Conc VI 12-s (PUN) VI 60-s
3
4
5
5
17
Conc VI 20-s (PUN) VI 20-s
4
3
5
4
16
Conc VI 60-s (PUN) VI 12-s
3
4
3
5
15
Total sessions within dose
24
23
25
29
101
Session blocks
4
4
5
5
18
04
Conc VI 12-s VI 60-s
5
6
4
3
18
Conc VI 20-s VI 20-s
8
3
4
4
19
Conc VI 60-s VI 12-s
5
7
8
5
25
Conc VI 12-s (PUN) VI 60-s
4
4
3
3
14
Conc VI 20-s (PUN) VI 20-s
6
3
3
4
16
Conc VI 60-s (PUN) VI 12-s
9
8
4
5
26
Total sessions within dose
37
31
26
24
118
Session blocks
8
5
4
4
21
05
Conc VI 12-s VI 60-s
7
12
9
3
31
Conc VI 20-s VI 20-s
4
3
4
3
14
Conc VI 60-s VI 12-s
5
5
3
6
19
Conc VI 12-s (PUN) VI 60-s
3
5
7
8
23
Conc VI 20-s (PUN) VI 20-s
3
7
3
7
20
Conc VI 60-s (PUN) VI 12-s
6
11
7
8
32
Total sessions within dose
28
43
33
35
139
Session blocks
5
8
6
7
26
06
Conc VI 12-s VI 60-s
8
9
10
16
43
Conc VI 20-s VI 20-s
21
8
12
7
48
Conc VI 60-s VI 12-s
8
12
24
11
55
Conc VI 12-s (PUN) VI 60-s
7
7
9
5
28
Conc VI 20-s (PUN) VI 20-s
10
5
6
4
25
Conc VI 60-s (PUN) VI 12-s
7
10
5
7
29
Total sessions within dose
61
51
66
50
228
Session blocks
11
9
12
8
40
08
Conc VI 12-s VI 60-s
3
4
5
3
15
Conc VI 20-s VI 20-s
3
3
3
3
12
Conc VI 60-s VI 12-s
3
3
3
3
12
Conc VI 12-s (PUN) VI 60-s
3
4
3
5
15
Conc VI 20-s (PUN) VI 20-s
3
3
3
3
12
Conc VI 60-s (PUN) VI 12-s
4
3
3
3
13
Total sessions within dose
19
20
20
20
79
Session blocks
3
4
4
4
15
was superimposed on alternative A for 3
values, thereby limiting potential floor or
participants and on B for 2 participants (see
restriction-of-range effects.
Table 1), all data will be reported as though
the punisher appeared on alternative A for
RESULTS
clarity of presentation.
Bias values for punishment under baseline
The appendix contains the mean number of
conditions were compared using both ob-
responses per session on alternatives A and B,
tained and net reinforcers in Rasmussen and
mean response rate on alternative A (respons-
Newland (2008). The former yielded a range
es/8), and mean obtained reinforcers earned
of bias values that were slightly, though not
on each alternative for each participant under
significantly, lower than the net values. We
each drug dose and concurrent schedule.
used obtained reinforcers in the present study
Means for each represent the data from the
because it slightly enhanced the range of bias
last three stable sessions for each condition. As

---

168
ERIN B. RASMUSSEN and M. CHRISTOPHER NEWLAND
Fig. 1.
Response rate (drug/vehicle) as a function of dose of ethanol for each schedule. The left column represents
the no-punishment condition; the right column, the punishment condition. Top panels: conc VI 12-s VI 60-s; middle
panels: conc VI 20-s VI 20-s, bottom panels: conc VI 60-s VI 12-s. Each participant is represented by a different symbol.
The dotted horizontal line shows no change from placebo.
the appendix shows, under the conc VI 12-s VI
were lower than in the no-punishment condi-
60-s schedule (no-punishment), response rate
tion.
on alternative A during placebo sessions varied
Figure 1 shows mean response rate on
between 32 and 97 responses per min, and
alternative A for each dose as a proportion of
decreased to 2 to 14 under punishment; under
placebo (drug/vehicle) for each subject across
conc VI 20-s VI 20-s schedule (no punish-
all concurrent schedules. Ethanol reduced
ment), rate ranged between 16 and 40
response rates in the no-punishment condi-
responses per min, and decreased to 1 to 22
tion in conc VI 12-s VI 60-s (top), conc VI 60-s
under punishment; under the conc VI 60-s VI
VI 12-s (bottom), and, to some extent, conc VI
12-s schedule (no punishment), rates varied
20-s VI 20-s (middle) for Participants 04, 05,
between 3 and 32 in the no-punishment
and 06. For Participant 02, response rate
condition, and were suppressed to 1 to 7
changed unsystematically, and for Participant
responses per minute in the punishment
08, rate increases were observed in conc VI 20-s
condition. In all placebo conditions, then,
VI 20-s and conc VI 60-s 12-s. In the punish-
response rates in the punishment conditions
ment condition, ethanol-related response rate

---

ETHANOL’S ANTIPUNISHMENT EFFECTS
169
increases were observed for Participants 04, 05,
and 06. Participant 02 showed a dose-depen-
dent increase under the conc VI 60-s VI 12-s
schedule. (Note the scaling for the conc VI 20-
s VI 20-s punishment condition, middle right
panel. Because Participant 04 showed a 16-fold
increase in rate, it reduces the ability to see the
rate increase for Participant 05.) For Partici-
pant 08 there was a rate increase under the VI
60-s VI 12-s schedule (bottom right) equivalent
to that seen in the nonpunishment compo-
nent. Otherwise, there were few rate increases
noted for Participants 02 and 08.
Figure 2 shows mean punishers as a func-
tion of ethanol dose for each participant
across the three concurrent schedules as a
proportion of baseline, i.e, the mean number
of punishers earned under a particular drug
dose divided by those earned under the
placebo. For Participants 04, 05, and 06,
ethanol-related increases were observed in
the number of punishers across the three
concurrent schedules, except for Participant
05, who showed a dose-dependent decrease
under the conc VI 60-s VI 12-s schedule. The
number of punishers earned was not greatly
different, but usually smaller than placebo
conditions, for Participants 02 and 08.
Figure 3 shows the value of log k, the bias
parameter under placebo (P) conditions.
Error bars represent the standard error of
the estimate. Each panel represents a partici-
pant. The mean bias value in the no-punish-
ment placebo condition (closed diamonds)
Fig. 2.
Mean punishers as a function of dose of
was 20.03 (SEM 5 0.03) for the group (see
ethanol. Each participant is represented by a different
Table 3 for individual data), suggesting a very
symbol. Each datum is the mean number of punishers
small amount of bias toward alternative B.
under a drug dose divided by those under the placebo.
The dotted horizontal line represents no change from
Negative values of bias, ranging from 20.46 to
placebo. Top panel: conc VI 12-s VI 60-s; middle panel:
21.03 (M 5 20.68, SEM 5 0.11) were
conc VI 20-s VI 20-s, bottom panel: conc VI 60-s VI 12-s.
obtained under the punishment conditions
(open diamonds) under placebo, suggesting a
04, 05, and 06. Bias was unchanged for
strong bias toward the unpunished alternative.
Participant 02 and became slightly more
A paired samples t-test confirmed a significant
extreme for Participant 08. Bias under the
difference between no-punishment and pun-
schedule without punishment (closed dia-
ishment conditions for log k under placebo
monds) changed little with ethanol for all
conditions, t(4) 5 25.58, p , 0.01.
participants, though Participant 02 showed a
Figure 3 also shows the bias parameter
slight bias toward alternative B at the highest
under each dose of ethanol under no-punish-
dose.
ment and punishment conditions. Table 3
Figure 4 shows the sensitivity parameter
shows corresponding parameter values for
under placebo (P) and each dose of ethanol
each individual. With ethanol, bias under the
(g/kg) under no-punishment (closed dia-
schedule with punishment diminished by at
monds) and punishment conditions (open
least half (a less negative value was obtained)
squares). Table 3 shows corresponding param-
as a function of ethanol dose for Participants
eter values for each individual. Sensitivity values

---

170
ERIN B. RASMUSSEN and M. CHRISTOPHER NEWLAND
Fig. 3.
Bias values for each participant as a function of dose of ethanol for no-punishment (closed diamonds) and
punishment (open squares) conditions. Error bars represent one standard error of the estimate. In some instances, the
error is so small that the data point covers it.
in the no-punishment placebo conditions were
to have no effect on sensitivity to reinforce-
less than 1 (undermatching) for 3 participants
ment for Participant 02 and a slight increase at
(04, 05, 06) and approximated 1 (matching)
one dose for Participant 05.
for 2 participants (02 and 08). Imposing the
Figure 5 shows the number of pencil taps
punishment contingency decreased sensitivity
in a 60-s period as a function of dose of
for Participants 02, 05 and 08, but did not
ethanol for each participant. Overall, ethanol
change sensitivity for the other 2 participants.
decreased pencil tapping in a dose-dependent
Figure 4 also shows little to no ethanol-
fashion, F(4, 24) 5 3.71, p 5 0.02, but
related change in sensitivity under the no-
individual differences were noted. Partici-
punishment condition; see Table 3 for indi-
pants 04, 05, and 06 showed dose-related
vidual values. An ethanol-related decrease in
declines, while Participant 08 showed a decline
the sensitivity parameter was more pro-
only at the highest dose, and Participant 02
nounced under punishment conditions for
appeared insensitive to the doses of ethanol
Participants 04, 06, and 08. Ethanol appeared
used.

---