Content uploaded by Sherry H Stewart
Author content
All content in this area was uploaded by Sherry H Stewart
Content may be subject to copyright.
Psychopharmacology (1995) 119 : 261 267 © Springer-Verlag 1995
Sherry H. Stewart " Peter R. Finn • Robert O. Pihl
A dose-response study of the effects of alcohol on the
perceptions of pain and discomfort due to electric shock
in men at high familial-genetic risk for alcoholism
Received: 11 July 1994/Final version: 22 December 1994
Abstract Alcoholics have previously been found to be
more sensitive to painful stimulation than controls, and
more sensitive to the pain-reducing effects of alcohol.
The present study was designed to examine these effects
in men at high familial-genetic risk for alcoholism and
controls. Subjects were assigned to one of four alcohol
doses [0.135 (active placebo), 0.50, 0.75, or 1.00 m195%
USP alcohol/kg body weight]. Ratings of the amount
of discomfort and pain experienced during an aversive
shock procedure were taken immediately post-shock,
both while subjects were sober and after they had con-
sumed one of the four alcohol doses. High risk men
were found to rate the experience of the shock as more
uncomfortable and painful overall than the tow risk
controls. Pharmacologically significant levels of alco-
hol were found to reduce or eliminate these group
differences, suggesting that alcohol has a "normaliz-
ing" effect on pain and discomfort perceptions in high
risk men. Only the higher doses of alcohol were found
significantly to dampen subjects' shock rating scores.
High risk males' increased sensitivity to pain and dis-
comfort, combined with the negatively reinforcing
effects of reducing these perceptions at moderate to
high alcohol doses, may play a role in predisposing high
risk males for the development of alcoholism.
Key words Alcohol - Pain • Electric shock " Males •
Alcoholism • Familial-genetic risk
S. H. Stewart ([D)
Clinical Psychology Program, Department of Psychology,
Dalhousie University, Life Sciences Centre, 1355 Oxford Street,
Halifax, Nova Scotia, Canada B3H4J1
P. R. Finn
Department of Psychology, Indiana University, Bloomington,
Indiana, USA
R. O. Pihl
Department of Psychology, McGill University, Montreal,
Quebec, Canada
Introduction
The pain-reducing effects of alcohol have long been
known; in fact, alcohol was one of the first anesthet-
ics ever used (Petrie 1978). An early study by Mullin
and Luckhardt (1934) provided an empirical demon-
stration that pain sensitivity decreases after alcohol
consumption. Wolff, Hardy and Goodell (1942) found
that drinking produced increases in pain thresholds of
up to 45 % above baseline levels. There is evidence that
alcohol may have specific analgesic effects at doses com-
monly administered for pharmacological effect (see
review by Pihl and Peterson 1992). Reductions in the
subjective intensity of sensations following alcohol con-
sumption affect not only the magnitude of pain sensa-
tions, but also those associated with discomfort. Petrie
(1978) found that alcohol consumption led to an
increase in the endurance of discomfort, as measured
by the amount of time subjects could hold their legs
in an unsupported position. Given that pain and
discomfort are aversive or punishing states, alcohol's
analgesic effects have been described as one of several
potentially rewarding (i.e., negatively reinforcing)
consequences of drinking (Pihl and Peterson 1992).
It has been suggested, given alcohol's analgesic
properties, that this drug should prove most reinforc-
ing for those most sensitive to pain and discomfort (see
Pihl and Peterson 1992). In fact, research suggests
important individual variability in pain and discomfort
sensitivities and response to the pain- and discomfort-
reducing effects of alcohol. For example, Brown and
Cutter (1977) and Petrie (1978) found sober alcoholics
to be more sensitive to painful stimulation than
controls. Furthermore, problem drinkers and alcoholics
have also been shown to be more sensitive to the
pain-reducing effects of alcohol than social drinkers
and non-alcoholic controls (Cutter et al. 1976; Brown
and Cutter 1977).
Increased sensitivity to painful stimulation has been
correlated with stimulus "augmenting" (Petrie 1978), a
262
perceptual style in which stimuli are subjectively
increased in perceived magnitude; at the opposite end
of this perceptual reactance continuum is the
"reducer", who tends to diminish what is subjectively
perceived (Petrie 1960). Alcoholics tend to be percep-
tual augmenters (Petrie 1978). Consumption of alco-
hol has been found to produce a shift from
augmentation to reduction in alcoholics (Petrie 1978),
and alcohol has been found to lead to significant
increases in augmenters' endurance of discomfort
(Petrie 1952). Furthermore, augmenting alcoholics
have been found to work harder for alcohol than those
who are reducers (Ludwig et al. 1977).
It has been suggested that stimulus augmentation
may be a form of perceptual hyper-reactivity, and the
shift from stimulus augmentation to reduction follow-
ing alcohol consumption in alcoholics, a form of alco-
hol dampening (Pihl et al. 1990). Thus, an increased
tendency to hyper-react to environmental stimuli and
an increased tendency for alcohol to dampen this
hyper-reactivity may predispose certain individuals to
the development of alcoholism (Pihl et al. 1989).
However, studies on the characteristics of seasoned
alcoholics confound the effects of continued alcohol
use with the effects of possible predisposing factors.
Studies of individuals at risk for alcoholism, which
overcome this limitation, have found a number of con-
sistent characteristics in children and young adults
deemed at risk for the development of alcoholism
due to familial-genetic (Cloninger et al. 1981, 1988;
Goodwin 1985) or personality (Jones 1968; Hoffman
et al. 1974; MacAndrew 1979) factors. High risk sub-
jects have been shown to be cardiovascularly hyper-
reactive to aversive (Finn and Pihl, 1987, 1988; Finn
et al. 1990; Stewart et al. 1992) stimulation, and appear
behaviorally hypersensitive to visual and auditory stim-
ulation (Fine et al. 1976). Consistent with this picture,
the incidence of perceptual augmentation has been
found to be higher among children of alcoholics than
among matched controls (Hennecke 1984). High risk
subjects also differ from controls in terms of response
to alcohol: they have been found to be more sensitive
to the cardiovascular stress-response dampening (SRD)
effects of alcohol than low risk controls (Sher and
Levenson 1982, 1983; Sher and Walitzer 1986; Finn
and Pihl 1987, 1988; Levenson et al. 1987; Finn et al.
1990; Stewart et al. 1992). These characteristics have
been hypothesized to place individuals at high risk for
the development of a wide range of behavior problems,
most notably the abuse of alcohol and other depres-
sant drugs (Pihl et al. 1989).
The present study is part of a larger research
project designed to examine alcohol's dose effects on
the cardiovascular stress response in men at high risk
for alcoholism (Stewart et al. 1992). The purposes of
the present study were to investigate the subjective
perceptions of pain and discomfort due to electric
shock in males at high familial-genetic risk for alco-
holism and controls, and to examine the effects of vary-
ing doses of alcohol on these pain and discomfort
ratings. An active placebo dose was included to test for
potential placebo effects. It was hypothesized that high
risk males would rate the experience of electric shock
as more painful and uncomfortable overall than low
risk males. It was also expected that alcohol adminis-
tration would lead to significant reductions in the
ratings of pain and discomfort due to the shock, at
higher alcohol dosage levels. Finally, it was predicted
that pharmacologically significant levels of alcohol
would exert a "normalizing" influence on the pain and
discomfort ratings of high risk men: no risk group
differences in shock ratings were expected in the high
alcohol dosage groups.
Materials and methods
Subjects
Eighty-one non-alcoholic males between the ages of 18 and 30 par-
ticipated as subjects. Alcoholism diagnoses were established based
upon Michigan Alcoholism Screening Test (MAST; Selzer 1971)
(score > 5) and the Diagnostic and Statistical Manual of Mental
Disorders, Third Edition Revised [DSM-III-R; American
Psychiatric Association (APA) 1987] criteria. Subjects were divided
into two groups based on their family histories of alcoholism. The
high-risk multigenerational family history (MFH) group (n = 39)
comprised males with alcoholic biological fathers and paternal
grandfathers, and at least one other paternally related alcoholic
male relative. The low-risk family history negative (FH-) group
(n -- 42) consisted of men with no identifiable alcoholic relatives in
the previous two generations. MFH males were obtained from the
Douglas Hospital - McGill University Alcohol Research Center
where relatives of alcoholics are recruited for a variety of interdis-
ciplinary research projects. At the Center, family histories are taken
with as many family members as possible. Alcoholism diagnoses
are based on Family History Research Diagnostic Criteria (FH-
RDC; Endicott et al. 1975) for unavailable family members, and
MAST score and DSM-III-R criteria for available family members.
FH- males were recruited through advertisements placed in local
newspapers and on the University campus. Respondents were
screened for family histories of alcoholism over the telephone. Only
those who had knowledge of and contact with relatives in the pre-
vious two generations of their family history were interviewed.
Selected subjects were requested to abstain from alcohol for 24 h
and to fast for 4 h prior to testing.
Procedure
Subjects received one of four possible alcohol doses: 0.135 (active
placebo), 0.50, 0.75, or 1.00 ml 95% USP alcohol/kg body weight,
mixed 1 part alcohol to 5 parts orange juice. Upon arrival, subjects
were assigned to one of the four alcohol doses by the principal
experimenter. All subjects were informed that they would receive
alcohol, but both subjects and the experimental assistants interact-
ing with the subjects remained blind as to which dosage group sub-
jects has been assigned. The placebo dose (i.e., 0.135 ml/kg) was
not considered likely to produce significant physiological effects and
was expected to be metabolized before application of the shock
stressor. An active placebo was chosen to provide smell and taste
cues to increase the probability that placebo subjects would remain
blind as to their dosage group membership (Ross and Pihl 1989).
Compliance with the requirements of fasting and abstinence from
alcohol consumption were verified verbally; individuals suspected
of non-compliance were administered a blood alcohol level (BAL)
test to ensure the abstinence criteria had been met.
Subjects completed an informed consent form and provided
demographic (i.e., age; number of years of education; and family
salary range, measured on a 7-point scale), and family drinking his-
tory information. Subjects also completed the MAST and indicated
the average number of occasions per week during which they nor-
mally consumed alcohol, and the average number of alcoholic bev-
erages they normally consumed per drinking occasion. Quantity
and frequency of drinking were multiplied to yield a measure of
average number of drinks per week (see Peterson et al. 1993; Stewart
et al. 1995). A recent review (Sobell and Sobell 1990) concludes that
this type of self-report measure is generally a highly reliable and
valid index of actual alcohol use.
The procedure used to administer aversive stimulation was sim-
ilar to that used in previous studies conducted in our laboratory
(e.g., Finn and Pihl 1987, 1988; Stewart and Pihl 1994). Subjects
were weighed, and Realistic Nova 10 headphones were applied.
Instructions were given about the nature of the shock procedure
and a concentric shock electrode was affixed to the inside of the
right elbow. Three consecutive electric shocks were delivered
through a Farrall Instruments Mark I at an intensity of 1.85 mA
for 0.25 s. Each shock was signalled through the headphones by a
low-frequency tone followed by a countdown from 10 to 1 (actual
time lapse = 14 s) and then a second tone which indicated shock
onset. Individual trials followed one another at 20-s intervals.
Following shock delivery, the subject completed a shock-rating
questionnaire (Finn and Pihl 1987) measuring the degree of
discomfort and pain experienced due to the shock on two
10-point rating scales (1 = pleasurable, 10 = very uncomfortable;
1 = absolutely no pain, 10 = very painful).
The subject was then provided with his assigned dose of alcohol
in two to three equal portions (depending on total volume), and
was instructed to consume the mixture in 15-20 min (i.e., one drink
approximately every 7 min). Subjects then relaxed for 10-15 min
(5 rain per glass) for alcohol absorption. After this waiting period,
subjects completed a measure of subjective intoxication (i.e., the
Sensation Scale; Maisto et al. 1980). The aversive stimulation pro-
cedure was then reintroduced. Subjects were given no instructions
concerning alcohol's effect on shock perceptions. Subjects again
rated the shocks on the two rating scales described previously. BALs
were taken immediately after the shock sequence in the alcohol con-
dition, using an Alco-Sensor III (Intoximeters). Finally, subjects
were debriefed about the alcohol condition to which they had been
assigned, and the purposes of the present experiment, and were paid
$5.00 per hour as compensation for their time.
Shock rating data for the 1.00 ml/kg dosage groups were taken
from a study previously completed in our laboratory (i.e., Finn et al.
1990). Subjects in tile three remaining dosage levels (i.e., 0.135, 0.50
or 0.75 ml/kg body weight) were run shortly after those in the two
1.00 ml/kg dosage group, using subject selection and recruitment
criteria, instructions, and experimental procedures identical to those
used previously. For subjects in the 1.00 ml/kg dosage groups, the
order of presentation of testing conditions (alcohol versus no-alco-
hol) was counterbalanced within groups (i.e., subjects came in for
two 90-min testing sessions). However, since no significant effects of
order of presentation were observed (see Finn et al. 1990), the remain-
ing subjects (i.e., those in the placebo, 0.50 and 0.75 ml/kg dosage
groups) completed both conditions in a single testing session with
the no-alcohol condition always preceding the alcohol condition.
263
aid of the BMDP statistical software package, version
P4V (Dixon et al. 1983). Interpretation of significant
interactions was facilitated with analyses of simple
effects, using the method recommended by Keppel
(1982). Correlations between variables were computed
with the aid of the BMDP program, version P4M
(Dixon et al. 1983).
Demographic Measures
Data for each of the demographic variables (i.e., age,
number of years of education, family salary range, and
drinks per week) were subjected to 2 x 4 (Risk group
by Dose) ANOVAs. These analyses revealed no
significant group differences on any of the demographic
measures: MFH and FH-males were equivalent in
age [Ms (and SDs) = 23.2 (4.5) versus 22.4 (3.5) years,
for the MFH and FH-groups, respectively], educa-
tion [13.4 (2.1) versus 14.3 (2.2) years], and family
salary range code [5.4 (1.7) versus 6.0 (1.3)]. The ten-
dency for MFH men to report consuming more alco-
holic beverages on a weekly basis than FH- men [Ms
and (SDs) = 9.6 (10.6) versus 7.5 (8.1) drinks per week]
did not prove statistically significant [F(1, 73) = 1.28,
n.s.].
BAL and subjective intoxication data
Subjects' total scores on the Sensation Scale were sub-
jected to a 2 x 4 (Risk group by Dose) ANOVA. The
results revealed no significant main effects or interac-
tions involving the Risk Group factor, but a significant
Dose main effect IF(3, 73) = 6.49, P < 0.001] was
revealed. Alcohol increased subjective intoxication in
a
dose-dependent manner [i.e., mean total Sensation
Scale scores (and SDs) = 34.0 (31.7), 43.2 (26.6), 70.8
(51.9), and 77.6 (34.3), for the placebo, 0.50, 0.75, and
1.00 ml/kg dosage groups, respectively].
The post-shock BAL levels were analyzed using a
2 x 4 (Risk group by Dose) ANOVA. This analysis
yielded no significant main effects or interactions
involving the Risk group factor, but a significant main
effect of Dose [F(3,73)= 118.55, P<0.0001] was
revealed. Alcohol increased BAL in a dose-dependent
manner [i.e., mean BAL and (SDs) = 0.001 (0.001),
0.063 (0.019), 0.85, (0.021), and 0.088, (0.017) g
alcohol/100 ml blood, for the placebo, 0.50, 0.75, and
1.00 ml/kg dosage groups, respectively).
Results
Data analyses
Univariate andt multivariate analyses of variance
(ANOVAs and MANOVAs) were performed with the
Shock rating data
Correlations between the two shock rating scores (i.e,
degree of discomfort and degree of pain due to the
shock) were computed, both in the alcohol and in the
no-alcohol Beverage conditions. The highly significant
264
correlation coefficients obtained (r = 0.767, for no-alco-
hol shock ratings, and r = 0.753, for the post-alcohol
ratings; both significant at P < 0.005) suggested that
the degree of discomfort experienced following expo-
sure to shock was not entirely independent of the actual
painful sensation of the shock stimulus. Thus, the two
shock rating scores were analyzed with a 2 x 4 x 2
(Risk group by Dose by Beverage condition)
MANOVA with repeated measures, which revealed
significant multivariate main effects for Risk group
[F(2, 72) = 5.68, P < 0.01], Dose [F(6, 144) = 2.44,
P < 0.05], and Beverage condition [F(2, 72) = 10.40,
P < 0.0005], and significant multivariate interactions
for Risk group by Dose [F(6, 144) = 2.41, P < 0.05],
and Beverage condition by Dose [F(6, 144) = 4.75,
P < 0.0005]. For the pain ratings, the MANOVA
yielded significant univariate main effects for Risk
group [F(1, 73) = 10.75, P < 0.005], and Dose IF(3, 73)
= 4.70, P < 0.005], as well as significant univariate Risk
group by Dose [F(3, 73) = 3.78, P < 0.05] and Beverage
condition by Dose [F(3, 73) = 9.83, P < 0.0001] inter-
actions. For the discomfort ratings, the MANOVA
yielded significant univariate main effects for Risk
group [F(1, 73) = 8.81, P < 0.005], Dose [F(3, 73) =
3.49, P < 0.05] and Beverage condition [F(1, 73)=
13.30, P < 0.001], as well as significant univariate
Risk group by Dose IF(3, 73) = 3.06, P < 0.05], and
Beverage condition by Dose [F(3, 73) = 3.86, P < 0.05]
interactions.
The Risk group main effects were due to MFH men
reporting more pain and discomfort experienced over-
all during the shock than the FH- subjects (see
Table 1). The Beverage condition main effect with the
discomfort measure was due to an overall dampening
in reported experiences of discomfort due to the shock
following alcohol administration (see Table 2).
The significant Risk group by Dose interactions were
further broken down to examine Risk group simple
main effects at each Dose level, in order to determine
whether the Risk group differences in the perception
of pain and discomfort due to shock were evident at
each Dosage level. This set of analyses revealed
significant simple main effects of Risk group at the
active placebo dose, for both the pain [F(1, 73) = 10.42,
P < 0.005] and discomfort IF(l, 73) = 13.26, P < 0.00i]
measures. The simple main effect of Risk group was
also significant at the 0.75 ml/kg dose for the pain mea-
sure only [F(1, 73) = 8.83, P < 0.005], but the magni-
tude of the Risk group difference in pain ratings was
reduced relative to the effect in the active placebo dose
(see Table 1). These simple main effects of Risk group
were due to the MFH men reporting more pain and/or
discomfort due to shock than FH-men (see Table 1).
No other significant Risk group differences in shock
Table 1 Mean pain and
discomfort as a function of
Risk group and Dose.
Standard deviations shown in
parentheses; MFH
multigenerational family
history, FH- family history
negative, Placebo 0.135 ml/kg,
low 0.50 ml/kg, moderate
0.75 ml/kg, high 1.00 ml/kg,
Risk group difference
(MFH
-
FH -) rating.
Asterisk (*) indicates a
significant Risk group
difference (P < 0.01)
Placebo Low Dosage group High Overall mean
Moderate
Pain
MFH group 5.00 (0.96)
(n = 8)
FH group 2.45 (1.67)
(n = 10)
Risk group difference 2.55*
Discomfort
MFH group 6.25 (0.89)
(n = 8)
FH- group 3.35 (1.51)
(n = 10)
Risk group difference 2.90*
2.90 (1.49) 4.72 (1.97) 5.29 (2.09) 4.49 (1.93)
(n = 10) (n = 9) (n = 12) (n = 39)
3.50 (2.19) 2.45 (1.26) 4.63 (1.13) 3.32 (1.80)
(n= 10) (n= 10) (n= 12) (n=42)
- 0.60 2.27* 0.66 1.17*
4.10 (1.78) 5.00 (1.89) 5.88 (1.88) 5.29 (1.83)
(n = 10) (n = 9) (n = 12) (n = 39)
4.30 (2.30) 3.65 (1.43) 5.46 (1.27) 4.25 (1.81)
(n = 10) (n = 10) (n = 12) (n = 42)
- 0.20 1.35 0.42 1.04"
Table 2 Mean pain and
discomfort as a function of
Beverage condition and Dose.
Standard deviations shown in
parentheses; no alcohol prior
to beverage consumption,
alcohol post-beverage
consumption, placebo O. 135
ml/kg, low 0.50 ml/kg,
moderate 0.75 ml/kg, high 1.00
ml/kg, Beverage condition
difference (no-alcohol
-alcohol) rating. Asterisk (*)
indicates a significant Beverage
condition difference (P < 0.01)
Placebo Low Dosage group High Overall mean
Moderate
Pain
No alcohol 2.89 (1.75)
(n =
18)
Alcohol 4.28 (2.16)
(n = 18)
Beverage condition
difference - 1.39"
Discomfort
No alcohol 4.61 (2.09)
(n--
18)
Alcohol 4.67 (1.91)
(n = 18)
Beverage condition
difference - 0.06
3.20 (2.02) 3.53 (2.14) 5.54 (1.69) 3.90 (2.16)
(n = 20) (n = 19) (n = 24) (n = 81)
3.20 (1.99) 3.53 (1.98) 4.38 (2.18) 3.86 (2.1t)
(n - 20) (n = 19) (n = 24) 01 = 81)
0.00 0.00 1.16' 0.04
4.40 (2.35) 4.68 (1.92) 6.50 (2.00) 5.14 (2.24)
(n - 20) (n = 19) (n = 24) (n =
81)
4.00 (2.00) 3.89 (1.79) 4.83 (1.88) 4.37 (1.9l)
(n 20) (n= 19) (n=24) (n=81)
0.40 0.79* 1.67" 0.77*
265
ratings were obtained in the three higher dosage levels
(see Table 1).
The significant Beverage condition by Dose inter-
actions were further broken down into simple main
effects of Beverage condition at each Dose level, in
order to examine Dose-related differences in alcohol's
effects on self-reported pain and discomfort due to
shock. This set of analyses revealed significant simple
main effects of Beverage condition for the discomfort
measure at the 0.75 ml/kg [F(1, 73) = 3.96, P < 0.05],
and the t.00 ml/kg [F(1, 73) = 22.70, P < 0.0001] dose
levels, reflecting dampening of reported discomfort due
to the shock after alcohol consumption in the two
higher dosage groups (see Table 2). These analyses also
revealed significant simple main effects of Beverage
condition for the pain measure at the 1.00 ml/kg
[F(1, 73)= 14.29, P< 0.0005] dose level, reflecting
dampening of reported pain due to the shock follow-
ing alcohol consumption in the highest dosage group
(see Table 2). The tendency for subjects to report expe-
riencing more pain due to shock following ingestion of
the active placebo beverage also proved statistically
significant [F(1, 73) = 15.25, P < 0.0005] (see Table 2).
Relationship between alcohol use and shock ratings
Correlations were computed for the entire sample
(n = 81) between self-reported drinks per week and the
pain and discomfort ratings in the no-alcohol and alco-
hol Beverage conditions. None of these correlations
proved statistically significant (i.e., rs = - 0.056,
-0.087, 0.046, and -0.087, for the no-alcohol dis-
comfort, alcohol discomfort, no-alcohol pain, and
alcohol pain ratings, respectively). Next, two alcohol
dampening indices were computed for each subject (i.e.,
one for the pain and the other for the discomfort rat-
ings), by subtracting the rating obtained in the alcohol
Beverage condition from that obtained in the no-alco-
hol Beverage condition. Correlations were computed
for the entire sample between self-reported drinks per
week and these alcohol-dampening indices. Again,
neither of these correlations proved statistically
significant (i.e., rs = 0.023 and 0.160, for the alcohol
dampening indices for discomfort and pain respec-
tively). The lack of significant correlations between self-
reported drinks per week and the shock ratings
supports the notion that sensitivity of pain and dis-
comfort, and alcohol dampening of pain and discom-
fort, are not effects secondary to a history of heavy
alcohol consumption.
Discussion
In the present study, it was found that men with multi-
generational family histories of alcoholism rated the
experience of electric shock as being more uncomfort-
able and painful than control males, overall. This lower
tolerance for the pain and discomfort caused by elec-
tric shock is consistent with the picture of the high risk
individual as being autonomically hyper-reactive (Sher
and Levenson 1982, 1983; Sher and Walitzer 1986; Finn
and Pihl 1987, 1988; Levenson et al. 1987; Finn et al.
1990; Stewart et al. 1992) and behaviorally hyper-sen-
sitive (Fine et al. 1976) to environmental stimulation,
when compared to low risk controls. Increased sensi-
tivity to painful stimulation may be one factor under-
lying MFH males' physiological hyper-reactivity (Finn
and Pihl 1987, 1988; Finn et al. 1990; Stewart et al.
1992), but it cannot be the only factor, since these men
have also been shown to be autonomically hyper-reac-
tive to non-aversive stimuli presented during an ori-
enting response procedure (Finn et al. 1990).
This increased sensitivity to pain in high risk males
may be part of a more general perceptual reactance
style. Like Hennecke's (1984) sample of children of
alcoholics, the MFH young men in the present study
may be stimulus "augmenters" (Petrie 1960), individ-
uals who tend to increase the subjective magnitude of
perceived stimuli. Since alcohol has been shown to pro-
duce a shift from augmenting to reducing (Petrie 1978),
drinking in males with such stimulus modulation
difficulties could represent an attempt to dampen their
resultant physiological over-reactivity (Pihl et al. 1989).
The relationship between sensitivity to pain and dis-
comfort and more general perceptual reactance could
be evaluated empirically in future research by com-
paring subjective responses to the shock procedure used
in the present study with measures of perceptual
augmenting/reducing (see Petrie 1978). Similarly, the
relationship between increased sensitivity to
painful/uncomfortable sensations and actual drinking
behavior is an area worthy of further research. Given
Ludwig et al.'s (1977) finding that augmenting alco-
holics work significantly harder for alcohol than those
who are reducers, it could be that those high risk sub-
jects who show the greatest sensitivity to pain and dis-
comfort are those who might drink the most heavily in
a laboratory, or in vivo, drinking situation.
The present finding of increased sensitivity to painful
stimulation in males at high familial-genetic risk for
alcoholism is consistent with the results of Brown and
Cutter (1977) and Petrie (1978), who showed that absti-
nent alcoholics are more sensitive to pain than con-
trols. However, unlike these previous studies with
abstinent alcoholics, the group differences revealed in
the present study cannot be attributed to differences in
histories of alcohol use between the groups, as groups
in the present study were found to be equivalent in lev-
els of weekly alcohol consumption, assessed via a self-
report measure (see Peterson et al. 1993 and Stewart
et al. 1995). Moreover, correlational analyses showed
that subjects' ratings of pain and discomfort were unre-
lated to the amount of alcohol they reported consum-
ing on a weekly basis. Thus, it appears safe to conclude
266
that the present finding of greater sensitivity to pain
and discomfort in MFH men is more likely a premor-
bid factor than the result of a history of alcohol abuse.
High risk men may be either more genetically suscep-
tible to be sensitive to painful or uncomfortable stim-
ulation, and/or this increased subjective sensitivity to
aversive stimulation may be the result of growing up
in a chaotic alcoholic family environment.
With the administration of varying doses of alco-
hol, the tendency for MFH males to rate the experi-
ence of the shock as more painful and uncomfortable
than FH- control males proved highly significant in
the active placebo dose. Pharmacologically significant
doses of alcohol were found to much reduce, or more
often eliminate, this Risk group effect in shock ratings.
This elimination of the group differences in shock rat-
ing scores was not merely due to "floor" effects in the
shock rating scores obscuring Risk group differences,
since both Risk groups rated the experience of the
shock as moderately painful and uncomfortable, even
in the higher alcohol dosage groups. Neither was this
reduction of the Risk group differences due merely to
the fact that high doses of alcohol tend to obscure indi-
vidual differences (see Sher and Walitzer t986), since
the tendency for MFH subjects to rate the shocks more
negatively than the FH -controls was dampened even
at a low alcohol dose (i.e., 0.50 ml/kg). Thus, it can
be concluded that alcohol consumption leads to a "nor-
malizing" of the tendency for MFH males to be hyper-
sensitive to subjective experiences of pain and
discomfort. This finding replicates and extends to a
sample of high risk males, the Cutter et al. (1976) and
Brown and Cutter (1977) findings that alcohol con-
sumption reduced the tendency of alcoholics and prob-
lem drinkers to be hyper-sensitive to pain when
compared to controls. The degree of sensitivity to alco-
hol dampening of pain/discomfort ratings was shown
to be unrelated to self-reported drinks per week in the
present study, ruling out the possibility that the MFH
men's sensitivity to alcohol-induced reductions in their
preceptions of pain/discomfort was secondary to their
tendency toward heavier drinking.
The results of the present study are consistent with
previous results (see review by Pihl and Peterson 1992)
in demonstrating that moderate to high doses of alco-
hol serve significantly to reduce subjects' perceptions
of pain and discomfort. Some evidence suggests that
alcohol's negatively reinforcing pain- and discomfort-
reducing effects may be mediated through the opiate
system (e.g., Altshaler et al. 1990). However, in com-
paring the dose-response curves associated with the
pain and discomfort ratings, it is interesting to note
that these two shock ratings follow a somewhat different
pattern in terms of overall alcohol-induced dampening
(see Table 2). While the ratings of discomfort obtained
in the present study were significantly dampened by the
two highest doses of alcohol employed in the present
study (i.e., 0.75 and 1.00 ml/kg), the pain ratings were
significantly dampened only after the administration of
the highest alcohol dose (i.e., 1.00 ml/kg). This finding
suggests that while experiences of discomfort may be
reduced through moderate to high dose of alcohol,
reductions in subjective pain experiences may require
somewhat higher doses of alcohol.
An active placebo dose was included in the present
study in order to test for potential placebo effects. The
alcohol-induced dampening effects observed at the
higher doses should not be attributed to expectancy for
a decrease in perceived pain and discomfort following
alcohol consumption, since the only significant changes
in shock rating scores observed after consumption of
the active placebo were increases, rather than decreases,
in the subjective experience of pain. This oppositional
or anti-placebo effect for the pain ratings in the active
placebo condition may be due to the instructions
employed in the present study. All subjects were
informed that they would be receiving alcohol; how-
ever, they were not informed of the precise alcohol con-
tent of their assigned beverage. Such instructions may
have set up expectations for pain reduction following
beverage ingestion. If subjects in the active placebo con-
dition had therefore underpredicted the subjective
experience of pain due to shock following beverage con-
sumption, the discrepancy between the predicted and
experienced painfulness of the shock may- explain the
oppositional placebo effect observed in the pain rat-
ings of this dosage group. In fact, previous studies have
shown that the underprediction of aversive events is
significantly associated with increases in subjective and
physiological pain responses (see review by Arntz et al.
1991). However, this interpretation of the oppositional
placebo effect remains merely speculative, as we did not
obtain ratings of the expected painfulness of the shock
in the present study.
In summary, our results demonstrate that men at
high risk for alcoholism due to extensive family histo-
ries of alcoholism are significantly more sensitive over-
all to the pain and discomfort experienced during an
aversive shock procedure than men with family histo-
ries negative for alcoholism. Alcohol administration
was found to "normalize" the MFH subjects' hyper-
sensitivity to experiences of pain and discomfort: risk
group differences in the ratings of pain and discomfort
were consistently evident in the active placebo dose.
The higher doses of alcohol were found to lead to
significant reductions in subjects' pain and discomfort
ratings. The characteristic of increased sensitivity to
pain and discomfort in MFH males, and the reduction
in these perceptions (i.e., negative reinforcement)
afforded by consumption of moderate to high doses of
alcohol, may play a role in predisposing these high risk
males to turn to alcohol in order to dampen their
perceptual over-reactance.
Acknowledgements This research was funded by the Medical
Research Council of Canada, and was conducted in the Alcohol
267
Research Laboratory, Department of Psychology, McGill
University. The authors would like to acknowledge the help received
from individuals at the Douglas Hospital-McGill University
Alcohol Research ,Center in subject recruitment, and from Aurelio
Sita and Peter Giancola in helping conduct the study. The authors
would also like to thank three anonymous reviewers for their
helpful comments and suggestions on earlier drafts of this manu-
script.
References
Altshaler HL, Phillips PL, Feinhandler DA (1990) Alteration in
ethanol self-administration by naltrexone. Life Sci 26:679 699
American Psychiatric Association (1987) Diagnostic and statistical
manual of mental disorders (3rd ed. revised). Washington DC
Arntz A, Van den Hout MA, Van den Berg G, Meijboom A (1991)
The effects of incorrect pain expectations on acquired fear and
pain responses. Behav Res Ther 29:547-560
Brown RA, Cutter HSG (1977) Alcohol, customary drinking behav-
ior and pain. J Abnorm Psychol 86:179 188
Cloninger CR, Bohman M, Sigvardsson S (1981) Inheritance of
alcohol abuse:a cross-fostering analysis of adopted men. Arch
Gen Psychiatry 38:861-868
Cloninger CR, Sigvardsson S, Bohman M (1988) Childhood per-
sonality predicts alcohol abuse in young adults. Alcohol Clin
Exp Res 12:494-505
Cutter HSG, Malouf B, Kurtz NR, Jones WC (1976) Feeling no
pain J Stud Alcohol 37:273-277
Dixon W, Brown MB, Engelman L, Frane JW, Hill MA, Jennrich
RI, Toporek JD (eds) (1983) BMDP statistical software: 1983
printing with additions. University of California Press, Berkeley
Calif. 388412
Endicott J, Andreasen N, Spitzer RL (1975) Family history research
diagnostic criteria. Biometrics Research, New York State
Psychiatric Institute, New York
Fine E, Yudin L, Holmes J, Heinemann S (1976) Behavioural dis-
orders in children with parental alcoholism. Ann NY Acad Sci
273:507 517
Finn PR, Pihl RO (1987) Men at high risk for alcoholism: the effect
of alcohol on cardiovascular reactivity to unavoidable shock. J
Abnorm Psychol 96:230-236
Finn PR, Pihl RO (1988) A comparison between two groups of
sons of alcoholics on cardiovascular reactivity and sensitivity to
alcohol. Alcohol Clin Exp Res 12:742 747
Finn PR, Zeitouni N, Pihl RO (1990) The effects of alcohol on psy-
chophysiological hyper-reactivity to non-aversive and aversive
stimuli in men at high risk for alcoholism. J Abnorm Psychol
99 : 79-85
Goodwin DW i1985) Alcoholism and genetics. Arch Gen Psychiatry
42:171-174
Hennecke L (1984.) Stimulus augmenting and field dependence in
children of alcoholic fathers. J Stud Alcohol 45:486492
Hoffman J, Loper R, Kammeier ML (1974) Identifying future alco-
holics with MMPI alcoholism scales. Q J Stud Alcohol
18 : 190-204
Jones MC (1968) Personality correlates and antecedents of drink-
ing patterns in adult males. J Consult Clin Psychol 32:2-12
Keppel, G. (1982) Design and analysis: A researcher's handbook
(2nd ed.). Englewood Cliffs, NJ: Prentice-Hall
Levenson RW, Oyama ON, Meek PS (1987) Greater reinforcement
from alcohol for those at risk: parental risk, personality risk,
and sex. J Abnorm Psychol 96:242-253
Ludwig AM, Cain RB, Wikler A (1977) Stimulus intensity modu-
lation and alcohol consumption. J Stud Alcohol 38:2049 2056
MacAndrew C (I 979) On the possibility of the psychometric detec-
tion of persons who are prone to the use of alcohol and other
substances. Addict Behav 4: 11-20
Maisto SA, Connors G J, Tucker JA, McCollam JB, Adesso VJ
(1980) Validation of the sensation scale, a measure of subjective
physiological responses to alcohol. Behav Res Ther 18:3743
Mullin F, Luckhardt AB (1934) The effect of alcohol on cutaneous
tactile and pain sensitivity. Am J Physiol 109:77 78
Peterson JB, Pihl RO, Seguin JR, Finn PR, Stewart SH, (1993)
Alcohol-induced heart rate change, family history and predic-
tion of weekly alcohol consumption by non-alcoholic males. J
Psychiat Neurosci 18:190-198
Petrie A (1952) Personality and the frontal lobes. Blakiston Press,
New York
Petrie A (1960) Some psychological aspects of pain and the relief
from suffering. Ann NY Acad Sci 86:13 27
Petrie A (1978) Individuality in pain and suffering. University of
Chicago Press, Chicago, Ill.
Pihl RO, Finn PR, Peterson JB (1989) Autonomic hyper-reactivity
and risk for alcoholism. Prog Neuropsychopharmacol Biol
Psychiatry 13 : 489-496
Pihl RO, Peterson JB (1992) Etiology. Ann Rev Addict Res Treat
2:153-175
Pihl RO, Peterson JB, Finn PR (1990) The inherited predisposi-
tion to alcoholism:characteristics of sons of male alcoholics. J
Abnorm Psychol 99:291-301
Ross D, Pihl RO (1989) Modification of the balanced-placebo
design for use at high blood alcohol levels. Addict Behav
14:91-97
Selzer M (1971) The Michigan alcoholism screening test:A quest
for a new diagnostic instrument. Am J Psychiatry 127 : 1653-1658
Sher K J, Levenson RW (1982) Risk for alcoholism and individual
differences in the stress-response dampening effect of alcohol. J
Abnorm Psychol 91:350-367
Sher K J, Levenson RW (1983) Alcohol and tension reduction:the
importance of individual differences. In : Pohorecky LA, Brick J
(eds) Stress and alcohol use. Elsevier, New York, 121-134
Sher K J, Walitzer KS (1986) Individual differences in the stress-
response-dampening effect of alcohol:a dose-response study. J
Abnorm Psychol 95:159-167
Sobell LC, Sobetl MB (1990) Self-reports issues in alcohol
abuse:state of the art and future directions. Behav Assess
12 : 77-90
Stewart SH, Finn PR, Pihl RO (1992) The effects of alcohol on the
cardiovascular stress response in men at high risk for alco-
holism:a dose-response study. J Stud Alcohol 53:499-506
Stewart SH, Peterson JB, Pihl RO (1995) Anxiety sensitivity and
self-reported rates of alcohol consumption in university women.
J Anxiety Disord (in press)
Stewart SH, Pihl RO (1994) The effects of alcohol administration
on psychophysiological and subjective-emotional responses to
aversive stimulation in anxiety sensitive women. Psychology
Addict Behav 8:29-42
Wolff HG, Hardy JD, Goodell H (1942) Studies on pain:mea-
surement of the effect of ethyl alcohol on the pain threshold and
on the "alarm reaction." J Pharmacol 75:38-49