ACTH, corticosterone, and prolactin levels and decreases DA and NE
levels in hypothalamic regions. This stressor attenuates nicotine’s
activation of NE neurons but does not reverse its attenuating effects
on prolactin.

Nicotine appears to be associated with neuroendocrine activity by
NE and DA activation (Fuxe et al. 1987). Immunohistochemical
studies suggest that alterations in NE function are more important
for the control of the pituitary-adrenal-axis, while DA turnover
appears to be crucial for nicotine’s effects on prolactin, LH, and
follicle-stimulating hormone (FSH). Moreover, these studies indicate
that similar nAChRs are located within both DA mesolimbic and
neostriatal systems.

Stimulation of Pituitary Hormones

Nicotine administration and cigarette smoking stimulate the
release of several anterior and posterior pituitary hormones. Seyler
and coworkers (1986) had human subjects smoke two high-nicotine
(2.87 mg) cigarettes in quick succession. Plasma levels of prolactin,
ACTH, B-endorphin/B-lipoprotein, growth hormone (GH), vasopres-
sin, and neurophysin I increased. No change was seen in TSH, LH, or
FSH. The rapid smoking paradigm used by Seyler and coworkers
(1986) may have contributed to the effects of nicotine. Growth
hormone levels exhibited a prolonged increase after subjects smoked
three cigarettes in rapid succession (Sandberg et al. 1973). In
experiments conducted by Winternitz and Quillen (1977) with male
habitual smokers, GH began to rise after two cigarettes, peaked at 1
hr, and then returned to control levels while smoking continued.
Wilkins and colleagues (1982) also found that smoking increases GH
levels and presented evidence that the effect is nicotine mediated.
Coiro and coworkers (1984) reported that the increase in GH
produced by clonidine was greatly enhanced by cigarette smoking,
suggesting that nicotinic cholinergic and adrenergic mechanisms
might interact in the stimulation of GH secretion.

The TSH plasma levels were not affected when nicotine was
administered over a 60-min period to female rats (Blake 1974). In
studies involving exposure to cigarette smoke, Andersen and col-
leagues (1982) reported a lowering of TSH secretion in rats, but as
noted, Seyler and coworkers (1986) found no change in human
subjects. Thus, the data on the effects of nicotine on TSH release are
inconclusive at this time.

ACTH plasma levels increased after i.p. injection of nicotine in the
rat (Conte-Devolx et al. 1981). In similar experiments, Cam and
Bassett (1983b) found that elevated ACTH levels peaked and rapidly
declined to a sustained plateau level. Sharp and Beyer (1986)
reported that the effects of nicotine on ACTH in rats show a rapid
and marked desensitization. Seyler and coworkers (1984) had male

101
subjects smoke cigarettes containing 0.48 or 2.87 mg of nicotine. No
increases in ACTH or cortisol were detected after subjects smoked
0.48-mg-nicotine cigarettes. Cortisol levels rose significantly in 11 of
15 instances after smoking the high-nicotine cigarettes, but ACTH
rose in only 5 of the 11 instances when cortisol increased. Each
ACTH increase occurred in a subject who reported nausea and was
observed to be pale, sweaty, and tachycardic. Seyler and coworkers
(1984) studied smokers and concluded that ACTH release occurs only
in smokers who become nauseated.

LH levels were reduced in male rats exposed to unfiltered
cigarette smoke, while FSH was unchanged (Andersen et al. 1982). In
experiments by Winternitz and Quillen (1977), there were no
differences in LH and FSH among male cigarette smokers while
smoking as compared with not smoking. Seyler and colleagues (1986)
found no change in human LH or FSH levels after smoking. There is
no evidence of gonadotropin release stimulated by nicotine or
smoking.

Prolactin plasma levels were lowered considerably in lactating
rats injected twice daily with nicotine (Terkel et al. 1973). It was
suggested that failure of prolactin release following chronic nicotine
administration was responsible for low milk production and starva-
tion of pups. Blake and Sawyer (1972) found that, in lactating rats,
the rapid suckling-induced release of prolactin into the blood is
inhibited by s.c. injections of nicotine. Ferry, McLean, and Nikitito-
vich-Winer (1974) reported that tobacco smoke inhalation in rats
delays the suckling-induced release of prolactin. Andersen and
coworkers (1982) found that prolactin secretion was reduced in male
rats in a dose-dependent manner by exposure to unfiltered cigarette
smoke. However, Sharp and Beyer (1986) reported that the effects of
nicotine on prolactin in rats shows a biphasic effect, first increasing
and then decreasing. Suppressed prolactin levels were found in
female smokers who were breast feeding (Andersen et al. 1982).
These researchers noted that smokers weaned their babies signifi-
cantly earlier than nonsmokers. However, Wilkins and coworkers
(1982) observed an increased level of prolactin in male chronic
smokers.

Arginine Vasopressin

In addition to its antidiuretic effects, arginine vasopressin acts asa
vasoconstrictor (Munck, Guyre, Holbrook 1984; Waeber et al. 1984).
Arginine vasopressin may also act as a neuromodulator in pathways
that affect behavior. It has been shown to promote memory
consolidation and retrieval in rats (Bohus, Kovacs, de Wied 1978) and
there are reports of memory enhancement following intranasal
administration of a vasopressin analog in both normal and memory-
deficient humans (LeBoeuf, Lodge, Eames 1978; Legros et al. 1978;

102
Weingartner et al. 1981). Nicotinic cholinergic receptors in the
medial basal hypothalamus and muscarinic cholinergic receptors in
the neurohypophysis (posterior pituitary) have been implicated in
the release of vasopressin (Gregg 1985). Nicotine has been found to
stimulate vasopressin release in a dose-related manner in animals
(Reaves et al. 1981; Siegel et al. 1983) and in humans (Dietz et al.
1984; Pomerleau et al. 1983; Seyler et al. 1986). These observations
are consistent. with the effects of nicotine on cognitive performance
(Chapter VD.

The Pro-Opiomelanocorticotropin Group of Hormones

The POMC hormones are released in response to stress and in
response to corticotropin-releasing hormone (Munck, Guyre, Hol-
brook 1984; Krieger and Martin 1981). ACTH has behavioral effects
and stimulates the release of steroids such as cortisol from the
adrenal cortex. ACTH produces rapid cycling between sleeping and
waking as well as sexual stimulation, grooming/scratching, blocking
of opiate effects such as analgesia, and the enhancement of attention
and stimulus discrimination (Bertolini and Gessa 1981). Endogenous
opioids, such as B-endorphin, potentiate vagal reflexes, cause respira-
tory depression, lower blood pressure, block the release of catechol-
amines (Beaumont and Hughes 1979; Schwartz 1981), have antinoci-
ceptive effects (van Ree and de Wied 1981), and modulate neuro-
transmitter systems leading to amnesic effects (Izquierdo et al. 1980;
Introini and Baratti 1984). It has been suggested that the primary
function of the endogenous opioids is metabolic, serving to conserve
body resources and energy (Amir, Brown, Amit 1980; Margules 1979;
Millan and Emrich 1981).

Nicotine appears to stimulate the release of corticotropin-releasing
hormone from the hypothalamus through a nicotinic cholinergic
mechanism (Hillhouse, Burden, Jones 1975; Weidenfeld et al. 1983).
Using an isolated perfused mouse brain preparation, Marty and
coworkers (1985) demonstrated that nicotine stimulates secretion of
B-endorphin and ACTH in a dose-related manner when applied
directly to the hypothalamus but not when applied to the pituitary.
The work of Sharp and Beyer (1986) supports this finding; they
reported that the secretion of ACTH following nicotine was unaffect-
ed by adrenalectomy. Nicotine administration to rats has also been
shown to increase the plasma levels of corticosterone, ACTH, and B-
endorphin in a dose-related manner (Conte-Devolx et al. 1981).
Termination of chronic nicotine administration reduced hypotha-
lamic B-endorphin levels (Rosecrans, Hendry, Hong 1985). Hurlick
and Corrigal (1987) have also observed that the narcotic antagonist
naltrexone inhibits some nicotine-modulated behavior in mice,
providing a possible link between nicotine stimulation of endogenous
opioid activity and behavioral responses. Acute administration of

103
nicotine increases levels of plasma ACTH and corticosterone sharply
(Cam and Bassett 1983b). while chronic exposure results in complete
adaptation (Cam and Bassett 1984). Melanocyte-stimulating hor-
mone was decreased and 3-endorphin was increased by i-p. injections
of nicotine in the rat (Conte-Devolx et al. 1981).

Risch and colleagues (1980, 1982) have accumulated evidence for
cholinergic control of cortisol, prolactin, and B-endorphin release in
humans. Rapid smoking increases circulating cortisol, B-endorphin,
and neurophysin I (Pomerleau et al. 1983; Seyler et al. 1984; Novack
and Allen-Rowlands 1985; Novack, Allen-Rowlands, Gann, in press).
Moreover, in a study that examined the role of endogenous opioid
mechanisms in smoking, Tobin, Jenouri, and Sackner (1982) ob-
served that mean inspiratory flow rate increases during the smoking
of a cigarette but is depressed shortly after smoking. Naloxone had
no effect on the initial stimulation of respiration in response to
smoking but did significantly blunt the subsequent depression of
respiration. The significance of these findings for the control of
cigarette smoking remains equivocal (Karras and Kane 1980;
Nemeth-Coslett and Griffiths 1986; Chapter IV).

Thyroid

Most of the earlier work (1930s through 1950s) assessing the
effects of nicotine on thyroid function involved histological studies of
the thyroid glands from animals treated chronically with nicotine.
The findings are inconsistent in that some studies suggest elevated
thyroid activity and others do not (Cam and Bassett 1983a). In a
more recent study of nicotine’s action on the plasma levels of the
thyroid hormones, thyroxine (T4) and triiodothyronine (T3), Cam
and Bassett (1983a) found that a single i.p. injection of 200 wg/kg did
not alter the level of either hormone, although it did produce an
increase in plasma corticosterone. As mentioned earlier, nicotine
does not consistently affect TSH in animals or humans (Blake 1974;
Seyler et al. 1986).

Adrenal Cortex

Several studies in animals and human subjects have reported that
nicotine and cigarette smoking lead to elevated levels of corticoste-
roids. Kershbaum and colleagues (1968) administered nicotine i.v. to
anesthetized dogs and found a 64 percent rise in plasma corticoste-
roids. In rats, corticosteroid concentrations increased 50 percent
after ip. administration of nicotine. Suzuki and coworkers (1973)
also reported adrenal cortical secretion in response to nicotine in
conscious and anesthetized dogs. The effects of nicotine on plasma
corticosteroids in stressed and unstressed rats were studied by
Balfour, Khuller, and Longden (1975). The administration of nicotine
to unstressed rats caused a rise in corticosterone which persisted for

104
60 min. Nicotine did not affect plasma corticosterone concentration
in rats stressed by being placed on an elevated platform. Other
studies showed increased plasma corticosteroid levels after nicotine
administration (Turner 1975; Cam, Bassett, Cairncross 1979; Cam
and Bassett 1983b). Andersen and colleagues (1982) exposed male
rats to unfiltered cigarette smoke and found a dose-related increase
in corticosterone secretion. Filtered cigarette smoke was inactive.

Seifert and coworkers (1984) found that the chronic administration
of 0.5 or 1.0 mg/kg of nicotine s.c. twice daily for 8 weeks to rats
produced a marked decrease in plasma aldosterone levels. In this
study, nicotine had no effect on plasma corticosterone concentration.

Hokfelt (1961) reported increases in plasma cortisol and urinary
17-hydroxycorticosteroids following cigarette smoking in human
subjects. Kershbaum and coworkers (1968) reported similar results
involving elevations of 11-hydroxycorticostervids. Hill and Wynder
(1974) found that serum corticosteroids were markedly elevated after
high-nicotine (2.73 mg) cigarettes were smoked. No increase was seen
with cigarettes containing less nicotine. Cryer and colleagues (1976)
also found an increase in circulating levels of corticosteroids after
smoking. Winternitz and Quillen (1977) reported a sharp increase in
circulating cortisol after two cigarettes. The levels were maintained
through the smoking period and fell gradually to normal. Wilkins
and coworkers (1982) also observed increased levels of cortisol after
2-mg-nicotine cigarettes were smoked. No increases in cortisol were
detected after smoking 0.48-mg-nicotine cigarettes, but cortisol rose
significantly in 11 of 15 cases smoking 2.87-mg-nicotine cigarettes
(Seyler et al. 1984). Consistent with these results is the observation of
Puddey and colleagues (1984) that cessation of smoking is associated
with a significant fall in cortisol levels.

In contrast to these findings, Tucci and Sode (1972) reported intact
diurnal circadian variations of cortisol and unchanged 24-hr 17-
hydroxycorticosteroids during smoking. Benowitz, Kuyt, and Jacob
(1984) studied 10 subjects who either smoked their usual brand of
cigarettes, some of which contained 2.5 mg nicotine, or abstained.
Plasma cortisol concentrations throughout the day did not differ
during smoking or abstaining. Thus, while the majority of human
and animal data indicates that nicotine or smoking elevates cortico-
steroid levels, the effects appear to be influenced by dose, time, and
perhaps other factors.

Many investigators cited above have proposed that nicotine’s
effects on corticosteroids are mediated by the release of ACTH.
Indeed, hypophysectomy abolished the increase in adrenocortical
secretion following nicotine administration (Suzuki et al. 1973, Cam,
Bassett, Cairncross 1979) and nicotine-induced increase in plasma
ACTH precedes the increase in cortisol (Conte-Devolx et al. 1981).
However, Turner (1975) found that bilateral adrenal! demedullation

105
abolished the rise in corticosterone in response to nicotine and
suggested that the effect of nicotine is mediated via adrenal release
of catecholamines and that centrally mediated stimulation is not
significant. In contrast, the work of Matta and associates (1987)
demonstrates that the effects of nicotine on ACTH secretion are
centrally mediated. Rubin and Warner (1975) have also shown that
nicotine directly stimulates isolated adrenocortical cells of the cat.
The stimulant effect was dose-dependent and required the presence
of calcium. These experiments also indicated that nicotine enhances
the steroidogenic effect of ACTH.

Androgens

In male beagles, chronic smoking of high-nicotine/tar cigarettes
was associated with decreased activity of 7a-hydroxylase active on
testosterone (Mittler, Pogach, Ertel 1983). Testicular 6B- and 16a-
hydroxylases were not altered, while the hepatic androgen 6B-
hydroxylase activity in the testis was stimulated markedly by
smoking. Serum testosterone levels were reduced to 54 percent of
control levels by heavy smoking. It was concluded that chronic
cigarette smoking increased hepatic metabolism of testosterone,
resulting in lowered serum testosterone levels. However, it may be
that total testosterone is lower while free testosterone is not.

Estrogens

Cigarette smoking is associated with antiestrogenic effects in
women, including earlier menopause, lower incidence of breast and
endometrial cancer, _and increased osteoporosis. MacMahon and
colleagues (1982) reported lower urinary estrogen levels in premeno-
pausal smokers than in premenopausal nonsmokers and suggested
that the low estrogen secretion reflected lower estrogen production,
based on decreased estrone, estradiol, and estriol. However, 2-
hydroxyestrogens, the major metabolites of estradiol in women, were
not measured. Jensen, Christiansen, and Rodbro (1985) presented
evidence for increased hepatic metabolism of estrogens as a result of
smoking based on an observation of decreased serum estrogen levels
in postmenopausal smokers receiving exogenous hormone therapy.
This study examined 136 women treated for 1 year with different
doses of estrogen. Reduction of serum estrogen was most pronounced
in the highest estrogen-dose group. There was a significant inverse
correlation between the number of cigarettes smoked daily and
changes in serum estrogen. Michnovicz and colleagues (1986) found a
significant increase in estradiol 2-hydroxylation in premenopausal
women who smoked at least 15 cigarettes/day. They concluded that
smoking exerts a powerful inducing effect on the 2-hydroxylation
pathway of estradiol metabolism, which is likely to lead to decreased
bioavailability of hormone at estrogen target tissues.

106
Pancreas and Carbohydrate Metabolism

The body weight of smokers is consistently lower than that of
nonsmokers, and smokers tend to gain weight after cessation of
smoking (see Chapter VI for a detailed discussion of these relation-
ships). These phenomena are thought to contribute to tobacco use.
Glauser and coworkers (1970) and Hofstetter and coworkers (1986)
suggested that a change in metabolic rate is partially responsible for
these effects. Schechter and Cook (1976) and Grunberg, Bowen, and
Morse (1984) showed that rats which were administered nicotine lost
body weight without reducing food intake, although the body weight
changes were not as great as when eating behavior declined as well
(Grunberg 1982). Grunberg (1986) has pointed out that differences in
body weight between smokers and nonsmokers result from changes
in energy consumption (via changes in specific food consumption)
and changes in energy utilization. Recently, Grunberg and cowork-
ers (1988) have reported reductions of insulin levels accompanying
nicotine administration in rats which could result in an increase in
the utilization of fat, protein, and glycogen. This finding is consistent
with work of Tjdlve and Popov (1973), using rabbit pancreas pieces,
and studies by Florey, Milner, and Miall (1977) of human smokers
versus nonsmokers. Grunberg and coworkers (1988) have suggested
that the effects of nicotine on insulin levels also may be involved in
the nicotine-induced decrease of sweet food preferences.

Electrophysiological Actions of Nicotine
Electrocortical Effects

The brain responds to electrical as well as to chemical stimuli.
Therefore, measurements of the electrophysiological actions of
nicotine complement studies of its chemical effects. In addition,
electrophysiological activity reflects function that may relate to
sensory and cognitive changes observed in humans after smoking
(see Chapter VI). In animals, nicotine produces changes ranging
from subtle latency decreases in the primary auditory pathway to
seizures. The electrophysiological actions of nicotine may help to
relate the anatomical and receptor data (discussed earlier in this
Chapter) with sensory and cognitive data (discussed in greater detail
in Chapter VI).

The human studies on electrocortical effects of nicotine have some
methodological limitations. Most of the human studies had subjects
smoke cigarettes and did not measure blood levels of nicotine. Also,
most studies were performed on smokers whose immediate and long-
term smoking history was determined by questionnaires which may
not accurately reflect tolerance and physical dependence (Chapter
IV). In some studies the subjects were deprived of cigarettes, but no
objective measures such as expired carbon monoxide or blood

107
nicotine levels were collected to verify compliance with the depriva-
tion conditions.

Spontaneous Electroencephalogram

Historically, nicotine and ACh were used in animal experiments to
study the cholinergic mechanisms in the midbrain and thalamus
which produced EEG and behavioral activation (Longo, von Berger,
Bovet 1954; Rinaldi and Himwich 1955a,b). The administration of
nicotine produced EEG activation, consisting of desynchronized low-
voltage, fast activity, and behavioral arousal or alerting. These EEG
and behavioral responses resembled those produced by electrical
stimulation of the midbrain reticulomesencephalic activating system
(Moruzzi and Magoun 1949). With the discovery by Eccles, Eccles,
and Fatt (1956) of nicotinic receptors in the Renshaw cell of the
spinal cord, other investigators began to study the precise pharma-
cology of the EEG and behavioral alerting produced by nicotine and
electrical stimulation of the midbrain. Cigarette smoking in humans
also produced EEG desynchronization (Hauser et al. 1958; Wechsler
1958; Bickford 1960) or EEG desynchronization with an increase in
alpha frequency (Lambiase and Serra 1957). By the late 1950s and
early 1960s it was generally known that nicotine or tobacco smoke
caused EEG and behavioral arousal in animals and humans, but
several important issues were unresolved.

The central effects of nicotine were originally thought to result
from its action on the cardiovascular system (Heymans, Bouckeart,
Dautrebande 1931). Early studies found that EEG desynchronization
occurred when the subjects smoked nicotine cigarettes, nicotine-free
cigarettes, or sucked on glass tubes filled with cotton (Hauser et al.
1958; Wechsler 1958). Schaeppi (1968) injected nicotine into the
vertebral artery, carotid artery, and third and fourth ventricles of a
cat’s brain and was able to dissociate the effects of nicotine on the
EEG from those on the cardiovascular system. Kawamura and
Domino (1969) demonstrated that the EEG changes induced by
nicotine could be obtained in animals whose blood pressure increase
was blocked. Prevention of release of catecholamines in reserpine-
pretreated animals did not interfere with the EEG desynchroniza-
tion produced by nicotine (Knapp and Domino 1962).

Inhaled tobacco smoke (2-mL samples with about 2 pg/kg of
nicotine) and 2 yg of nicotine injected every 30 sec in a cat encephale
isolé preparation produced EEG desynchronization. EEG and behav-
ioral activation after cigarette smoke inhalation was also observed in
unanesthetized cats with implanted electrodes (Hudson 1979). Lukas
and Jasinski (1983) found that i.v. doses (0.75 to 3.0 mg) in human
smokers resulted in dose-dependent decreases in alpha (8 to 12 Hz
EKG activity) power and EEG desynchronization. In an inpatient
study where nicotine deprivation was carefully controlled and

108
monitored by measurement of expired carbon monoxide, the smok-
ing of non-nicotine cigarettes did not change the EEG (Herning,
Jones, Bachman 1983), but EEG changes did occur when subjects
smoked nicotine-containing cigarettes. These studies confirm that
nicotine has a direct action on the CNS separate from the cardiovas-
cular effects and that the effects are produced primarily by the
nicotine in inhaled tobacco smoke.

As experimental physiological manipulations, EEG recording, and
EEG quantification techniques improved, the specific nature of the
nicotine-induced cortical EEG changes and their relationship to
behavior were found to be more complex than originally thought.
The desynchronization produced by nicotine (20 to 100 g/kg) in the
cat was blocked by anterior pontine transections, but not by
midpontine transections (Knapp and Domino 1962). The midbrain
reticular activating system was needed for the cortical EEG desyn-
chronization produced by nicotine. However, larger doses of nicotine
injections also produced synchronous slow high-voltage EEG activity
in the hippocampus (hippocampal theta). Injections of the muscarin-
ic agonist arecoline (20 to 40 mg/kg) in the anteriorly transected
midbrain preparations still produced the hippocampal theta activity
without the cortical desynchronization. Atropine (1 mg/kg) and
mecamylamine (1 mg/kg), but not the ganglionic antagonist trimeth-
idinium (1 mg/kg) block the nicotine induced EEG desynchroniza-
tion in an intact animal. The convulsions observed after nicotine
injections (1 to 5 mg/kg in cats; 0.05 to 0.25 g/g in mice) (Laurence
and Stacey 1952; Stone, Meckelnburg, Torchiana 1958; Stiimpf,
Petsche, Gogolak 1962; Stiimpf and Gogolak 1967) appear to be due
to nicotine’s ability in large doses to stimulate muscarinic choliner-
gic receptors in the hippocampus. Because a high concentration of
labeled nicotine binds to hippocampal cells of the cat (Schmiterléw et
al. 1967) and areas adjacent to the hippocampus in the rat (Clarke,
Pert, Pert 1984), the possibility that nicotine-induced limbic electri-
cal activity contributes to its behavioral effects cannot be discounted.

Nicotine’s alerting effect on the brain may also involve a peripher-
al component. Electrocortical and behavioral arousal occurs in the
cat within 1 to 2 sec after injection of 10 to 15 pg/kg into the right
atrium of the heart, originating in vagal pulmonary C fiber afferents
(Ginzel 1987). The human counterpart to this finding is the
observation by Murphree, Pfeiffer, and Price (1967) that an initial
EEG change occurred within 5 sec after cigarette smoke inhalation,
which is shorter than a chest-to-head circulation time. Another input
from the periphery arises from nicotinic sites in the arterial tree.
Injection of small amounts (2 to 4 yg/kg) of nicotine, even as far
away from the brain as into the lower aorta or femoral artery, causes
instantaneous arousal] from all types of sleep (Ginzel and Lucas
1980).

109
The nicotine-induced release of ACh (MacIntosh and Oborin 1953;
Mitchell 1963) may be responsible for the EEG desynchronization in
animals (Armitage, Hall, Sellers 1969). The effect does not appear to
be due to the direct action of nicotine on the cortex because the
cortical cholinergic receptors are largely muscarinic (Kuhar and
Yamamura 1976; Rotter et al. 1979). Lower doses of nicotine (20
wg/kg/30 sec for 20 min) induced EEG desynchronization and ACh
release in the cat, whereas higher doses (40 pg/kg/30 sec for 20 min)
produced either an increase or decrease in EEG desynchronization
with corresponding increase or decrease in ACh release (Armitage,
Hall, Sellers 1969). The effect of nicotine on the EEG was short lived
relative to the release of ACh. Two separate pathways have been
proposed to explain these results: an ascending cholinergic pathway
mediating the cortical desynchronization and a limbic pathway
mediating the ACh release.

In one strain of mice, C57BL, nicotine increased cortical high-
voltage activity and decreased homovanillic acid (HVA) and 3-
methoxy-4-hydroxyphenthyleneglycol (MHPG) production in a per-
fused brain preparation (Erwin, Cornell, Towell 1986). The decrease
in HVA and MHPG levels reflects an increase in brain DA and NE
levels. In intact C57BL mice, nicotine decreased locomotor activity
(Marks, Burch, Collins 1983a). Thus, at least in one strain of mice,
nicotine induces an increase in cortical EEG synchronization, a
decrease in locomotor activity, and an increase in brain catechol-
amines. Little evidence relates the cortical desynchronization ob-
served in animals and humans to an increase in catecholamine
changes in the brain.

As trends in neuroscience research have shifted away from
spontaneous EKG recording in animals to intracellular recording,
receptor localization, and binding techniques, the precise quantifica-
tion of the nicotine-induced EEG desynchronization and hippocam-
pal synchronization has not been done. This type of quantification
has been done in humans by power spectral analysis. This technique
quantifies the EEG by the distribution and amplitude of brain waves
at different frequencies. Alpha power includes EEG activity in the 8-
to 12-Hz frequency range. Theta power includes EEG activity in the
4- to 7-Hz frequency range. Beta power includes EEG activity in the
frequency range of 13 Hz and higher.

The comparison of nicotine-induced EEG changes in animals and
humans is complicated by an important methodological difference.
Animals usually have not previously been given nicotine, while in
studies of humans, the subjects always are experienced tobacco
smokers. Moreover, in human studies that included a deprivation
period, nicotine abstinence may have produced electrophysiological
changes that are reversed by smoking or nicotine.

110
EEG desynchronization or increased beta power was observed in
smokers after smoking a tobacco cigarette (Hauser et al. 1958;
Wechsler 1958: Bickford 1960; Ulett and Itil 1969). These findings
essentially replicated the animal studies of nicotine. Using power
spectral analysis, Ulett and Itil (1969) also observed a decrease in
theta power and an increase in alpha frequency. The increase in
alpha frequency was previously noted with visual inspection by
Lambriase. However, the increase in theta was not. The subjects in
the study by Ulett and Itil had smoked one pack or more of
cigarettes/day and had been deprived of tobacco cigarettes for 24 hr
when the baseline EEG was recorded. Comparisons of the postsmok-
ing EEG were made with this baseline period. Therefore, the
decrease in alpha frequency and increase in theta power relative to
the data from the postsmoking session may be the result of nicotine
deprivation (Chapter IV).

Knott and Venables (1978) compared the alpha frequencies of
nonsmokers, 12-hr nicotine-deprived smokers, and nondeprived
smokers. They observed a decrease of about 1 Hz in the dominant
alpha frequency of the deprived smokers relative to the nonsmokers
and nondeprived smokers in a passive eyes-closed situation. An
active behavioral task and other frequencies of the EEG were not
studied. Knott and Venables hypothesize that smokers were consti-
tutionally different from nonsmokers. The slower alpha frequency
was interpreted as an arousal deficit, and smoking as compensation
to reduce the arousal deficit. Knott and Venables (1978) and Ulet
and Itil (1969) both found an attentional deficit during tobacco
deprivation.

Herning and coworkers (1983) investigated the EEG changes
related to cigarette smoking in a hospitalized group of healthy
smokers who smoked at least a pack and a half of tobacco cigarettes
with a machine nicotine delivery of 0.8 mg or more. A serial
subtraction task was administered and EEGs were recorded from
subjects in an eyes-open state. Alpha frequency was not affected by
periods of smoking and deprivation. However, theta and alpha power
increased during periods of deprivation and decreased after smoking
tobacco but not placebo cigarettes. The effects were most pronounced
on theta power. Increases in theta power occurred as early as 30 min
after the last cigarette, and were of the same magnitude as those
after 10 to 19 hr of nicotine deprivation. The increase in EEG theta
was interpreted to be a sign of tobacco deprivation (Chapter IV).

An indirect method of observing an increase in cortical activation
was the measurement of alpha power changes after tobacco smoking.
A number of investigators reported a decrease in alpha power or
abundance with cigarette smoking (Murphree, Pfeiffer, Price 1967:
Philips 1971; Caille and Bassano 1974, 1976; Murphree 1979;
Herning, Jones, Bachman 1983; Cinciripini 1986). with nicotine

111
polacrilex gum (Pickworth, Herning, Henningfield 1986, in press),
and with i.v. doses of nicotine (Lukas and Jasinski 1983). In spite of
differences in the number of cigarettes regularly smoked by the
subjects, the length of tobacco deprivation, the type of tobacco
cigarette smoked during the experiment, and the route of adminis-
tration, nicotine reduced alpha power.

Brown (1968) measured the resting EEG for heavy smokers and
nonsmokers. No cigarettes were smoked. The EEG of the heavy
smokers had less alpha and more beta activity. Twelve hours of
nonconfirmed deprivation in the heavy smokers did not change the
EEG patterns.

The EEG of neonates of mothers who smoke is not different from
that of neonates of contro] mothers (Chernick, Childiaeva, Ioffe
1983). Whether acute periods of smoking may affect the EEG of the
child before birth is not known.

In limited animal and human work, individual or species differ-
ences in the effects of nicotine on the EEG have been observed.
Nicotine produced a dose-dependent cortical EEG desynchronization
in C3H mice and an increase in synchronized EEG similar to
hippocampal theta activity in C57BL mice (Erwin, Cornell, Towell
1986). Both effects have been observed at different doses in the same
preparation (Kawamura and Domino 1969). Lower doses produce
EEG desynchronization, and higher doses produce hippocampal
theta. Tobacco cigarette smoking decreased EEG alpha power in
Type A subjects and increased theta power in Type B subjects
deprived of nicotine for about 4 hr (Cinciripini 1986). The relation-
ship between hippocampal theta in animals and cortical theta in
humans is not yet understood. In nondrugged animals cortical
desynchronization and hippocampal theta activity often occur simul-
taneously. Nicotine at low doses produces cortical desynchronization
and at high doses produces both types of EEG activity. Animal data
indicate that nicotine has effects on at least two systems in the brain:
a midbrain area responsible for EEG desynchronization and a limbic
system generating hippocampal theta activity. These findings are
consistent with the observation that some smokers indicate that they
smoke for nicotine’s stimulating effects and others smoke for its
sedating effects.

Sensory Event-Related Potentials

In animals and humans, the brainstem auditory-evoked potential
technique provides a noninvasive method for studying the effects of
nicotine on primary auditory sensory function. In the rat, nicotine
reduced the amplitudes of Waves IIL and IV of the brainstem
auditory-evoked response (BAER) (Bhargava and McKean 1977;
Bhargava, Salamy, McKean 1978; Bhargava, Salamy, Shah 1981),
Serotonergic mechanisms may mediate the nicotine-induced reduc-

112
tion in latency. Lavernhe-Lemaire and Garand (1985) found essen-
tially the opposite. Nicotine increased Waves I-III and did not
decrease Waves IV and V of BAER.

Auditory event-related potentials (AERPs) recorded directly from
the cortex of rat have provided conflicting information about
nicotine’s effects on auditory transmission from the inferior collicu-
lus to the cortical areas. Guha and Pradhan (1976), using pentobarbi-
tal anesthesia, found a dose-dependent increase in P1 (40 ms) and Ni
(110 ms) of the AERP. Bhargava, Salamy, and McKean (1978), using
chloralose anesthesia with atropine pretreatment, reported no
nicotine-related change in P1 (11 ms), N1 (28 ms), P2 (75 ms), and N2
(121 ms) of the AERP.

After smoking, the P1 (50 ms) of the human AERP is increased
during passive tasks at all intensity levels and the N1 (110 ms) is
increased in both passive and active tasks (Knott 1985). The N2
(about 215 ms) to P2 (about 260 ms) component of the AERP recorded
during a passive task was reduced after cigarette smoking when
compared with data from the baseline deprivation test (Friedman
and Meares 1980). P2 was also reduced by nicotine in the study by
Knott (1985). These components also increased in amplitude as the
tobacco deprivation period was lengthened. Any attempt to relate
this finding to results in the anesthetized rat would be speculative
because AERPs recorded from the cortex of unanesthetized animals
and humans are difficult to compare (Wood et al. 1984). Alterations
in AERP components in the 75- to 150-ms latency range have been
attributed to change in attention. The decrease in the later N2-P2
component is more likely to reflect reduced habituation to auditory
stimuli.

The effects of nicotine on visual event-related potentials (VERPs)
are more complicated than those on the AERPs. In unaesthetized
rabbits, i.v. nicotine (0.025 to 0.500 mg/kg) produced a complex
VERP change (Sabelli and Giardini 1972). At 2 min, nicotine
depressed the P1 (100 ms) and the N1 (250 ms). At 5 min, these
components were enhanced. At doses below 0.050 mg/kg, the N1 was
again depressed from 10 to 20 min after the injection. Pretreatment
with catecholamine inhibitors diminished the nicotine-induced
VERP changes. The authors suggested that the effect of nicotine on
VERPs was mediated in part by catecholaminergic mechanisms.

The effects of nicotine on the human VERP using multiple flash
intensities were the focus of four studies. The studies were designed
to test Buchsbaum and Silverman’s (1968) concept of stimulus
intensity control and its modulation by nicotine. According to their
theory, sensory processing in different individuals varies in at least
two ways. Some persons, “augmenters,” are more sensitive to higher
intensities than to lower intensities, and others, “reducers,” are
more sensitive to lower than to higher intensities. Smokers might be

113
one particular type of stimulus processer and may smoke to alter or
normalize stimulus intensity. In all studies the comparison was
between results after 12 hr or more of unconfirmed tobacco
deprivation and those after recent smoking. Components of the
VERP increased after smoking in three studies (Hall et al. 1973;
Friedman and Meares 1980; Woodson et al. 1982) but decreased in
another study (Knott and Venables 1978). The increases and
decreases occurred in components of the same latency range (75 to
250 ms) after flash onset. The fourth study differed only slightly
from the others in that it used a between-subjects and not within-
subject experimental design. Using a single flash intensity, Vasquez
and Toman (1967) also observed a decrease in components IV (‘40 ms)
and V (170 ms) of the VERP when compared with results after 36 hr
of tobacco deprivation. Two studies found a nicotine-induced increase
at earlier components (III-IV and IV-V) for the lower intensities only.
The other study reported an increase in later components (V-VI and
VI-VID at the higher flash intensities. Knott and Venables (1978)
observed the decrease after smoking in the middle components (IV-V
and V-VI) for the lower intensities. Because of these divergent
results, it is premature to conclude that smokers are exclusively
augmenters or reducers who are attempting to optimally adjust
stimulus intensity by smoking.

Cognitive Event-Related Potentials

Cognitive event-related potentials reflect neural events which
appear to be related to different aspects of cognition, such as
attention and stimulus evaluation. They usually follow the sensory
components of event-related potentials when human subjects are
performing active behavioral tasks. They provide information not
normally available from performance measures such as reaction
time. Increases or decreases in these potentials after smoking can aid
in our understanding the effects of nicotine on performance.

When two task-relevant stimuli are separated by a short interval
(1 to 3 sec), a negative slow wave develops between them. In
particular, this contingent negative variation (CNV) develops in
warned or cued reaction times, successive discrimination, and some
language processing tasks. The CNV appears to reflect brain
preparation to process and respond to the second stimulus. Smoked
tobacco and i.v. nicotine either increase or decrease the CNV (Ashton
et al. 1973, 1974, 1980; Minnie and Comer 1978). Extraverted
smokers took longer to smoke and nicotine increased the CNV.
Introverted subjects smoked faster and nicotine decreased the CNV.
Reaction time was inversely correlated with CNV amplitude; that is,
shorter reaction time was associated with larger CNV. With iv.
doses of nicotine (12.5 to 800.0 ug), larger doses produced a decrease
and small doses produced an increase in the CNV in the same

114
subject. O’Connor (1982) studied the effects of smoking on the
orienting (O wave) and expectancy (E wave) components of the CNV
in introverted and extraverted subjects. The O wave was not affected
by smoking. The E wave, recorded in frontal areas, was increased in
extraverted subjects after smoking. The E wave has been interpreted
by some investigators as cortical preparation for a response. Smok-
ing decreased a positive parietal E wave in introverts. Nicotine’s
effect on the E wave suggests the possible enhancement of motor
preparation in the extraverted subjects. The decrease of parietal
positivity indicates a possible enhancement of stimulus-processing
capacities in the introverts.

Poststimulus components P2(00) and P3(00) were affected by
cigarette smoking and nicotine polacrilex gum. P2 is thought to be
an index of habituation (Hillyard and Picton 1979), and P3 an index
of stimulus evaluation (Johnson 1986). Both components were
reduced in deprived smokers after smoking (Knott 1985; Herning
and Jones 1979). Knott (1985) interprets the reduction in P2 as a
more efficient habituation of sensory screening of relevant stimuli.
The reduction in P3 amplitude after smoking indicates a poorer
evaluation of task-relevant stimuli. The P3 latency and reaction time
were reduced only by cigarettes with higher machine-tested nicotine
yields (Edward et al. 1985). Such data indicate faster stimulus and
response processing. These authors did not report any P3 amplitude
changes. If none were present or P3 was reduced, the argument for
enhanced stimulus processing would be weak. Herning and Pick-
worth (1985) reported both dose-dependent increases and decreases
in P3 amplitude as a function of background noise levels when
deprived smokers chewed nicotine polacrilex gum (4 mg and 2 mg
doses). The respective increase or decrease was blocked by mecamy-
lamine pretreatment. Thus, the effect of nicotine on stimulus
evaluation remains unclear and is perhaps confounded by cognitive
deficits after periods of nicotine deprivation.

Motor Potentials

O’Connor (1986) investigated the effect of tobacco smoking on
motor potential and motor performance. Smoking increased the
motor readiness potential in extraverts, but not in introverts. These
results are consistent with his earlier finding of an incréased E wave
in extraverts after smoking. For introverts, smoking improved task
performance, but did not increase the motor readiness potential.

Other Peripheral Effects Relevant to Tobacco Use

In addition to vast central and peripheral effects, cigarette
smoking and nicotine have other peripheral effects that may
contribute to tobacco use. These additional factors have received less

115
research attention, mainly because they involve relatively new
theory or methodological approaches. For example, there is evidence
that direct stimulation of the trachea is important for cigarettes to
satisfy smokers (Rose et al. 1984) (Chapter IV). There is also evidence
that nicotine acts directly on the lung to stimulate afferent neurons
that, in turn, result in skeletal muscle relaxation and electrocortical
arousal (Ginzel 1987). These effects may contribute to the relation-
ship between smoking and stress (Chapter VI). Other research
indicates that smoking affects psychophysiological reactivity, an
integrative mechanism that is different from the classic, physiologi-
cal approach of examining individual systems or pathways. There-
fore, psychophysiological reactivity and its relevance to smoking are
discussed.

Psychophysiological Reactivity and Smoking

Psychophysiological reactivity is emerging as a useful construct in
smoking research, linking basic biological processes (genetic vulnera-
bility, central neurochemical factors) to behavioral coping and other
psychosocial factors. Psychophysiological reactivity refers to a
physiological response to a specific stimulus or as a result of the
absence of stimulation. This response can, in some cases, act as a
stressor. Within the broader conceptual framework of a stress-coping
model of smoking addiction (Shiffman and Wills 1985), smoking
behavior can be viewed both as a potential stimulus and as a coping
response that modulates psychophysiological reactivity.

Studies of psychophysiological reactivity illustrate the value of
controlled laboratory procedures to study person-environment inter-
actions. Psychophysiological reactivity reflects an interaction of the
organism and the environment. It is affected by individual differ-
ences in multiple response modes (physiological, cognitive, behavior-
al) and takes into account the genetic and learning history and
current state of the organism.

This Section reviews two separate but interrelated lines of
psychophysiological reactivity research with humans. The first is the
effect of smoking on psychophysiological reactivity. Related issues
include identification of mechanisms that may help to reveal why
some individuals smoke and the relationship between smoking and
coronary heart disease (CHD). The second research line addresses
the relationship among situational events (general and drug-specif-
ic), psychophysiological reactivity, and relapse.

The effects of smoking on the cardiovascular aspects of physiologi-
cal reactivity have been well documented and appear to be primarily
due to effects of nicotine and carbon monoxide (Suter, Buzzi, Battig
1983; Koch et al. 1980; Rosenberg et al. 1980). In individuals with no
cardiovascular disease, some of the typical effects of smoking and
nicotine are elevated heart rate and blood pressure and a fall in

116
fingertip temperature and capillary blood flow (Richardson 1987;
Ashton et al. 1982; Epstein and Jennings 1986; Henningfield et al.
1983).

Accompanying cardiovascular reactions to smoking are cognitive
reactions, including perceptions of relaxation, and anxiolytic, antino-
ciceptive, euphoric, stimulative, and dysphoric effects (Kozlowski,
Director, Harford 1981). Although there is consistency in the
literature with regard to the self-reported emotional changes experi-
enced as a result of smoking, there are clear differences in response
and direction of effects between individuals and within individuals
over time (Best and Hackstian 1978; Gilbert 1979; Gilbert and
Welser, in press). Smoking can produce physiological changes that
are concurrent with subjective tranquilizing effects (Nesbitt 1973;
Shiffman and Jarvik 1984; Gilbert 1979). This phenomenon has led
investigators to emphasize the importance of incorporating physio-
logical, psychological, and environmental factors into more biobeha-
vioral models to better understand the cognitive and physiological
components of reactivity to smoking (Pomerleau and Pomerleau
1984; Baum, Grunberg, Singer 1982; Abrams et al. 1987; Grunberg
and Baum 1985). For example, nicotine has direct and indirect
actions on central neuroregulatory systems and has biphasic effects
of both stimulation and blockade. These factors can help explain
effects such as the anxiolytic and antinociceptive phenomena
(Pomerleau 1986) at a cognitive and neurochemical level, while at
the same time resulting in increased heart rate and blood pressure
and decreased perception of muscle tension (Epstein et al. 1984).

In addition to dosage, biphasic, and physiological factors, the
influence of setting and expectancy set, the current state of the
individual (smoking, deprived, stressed, not stressed), and individual
differences in dependence, genetic, demographic, and learning
history can all influence psychophysiological reactivity. For exam-
ple, smoking a 1.3-mg-nicotine cigarette under conditions of mild
sensory isolation produced consistent arousal effects (i.e., elevations
in heart rate and skin conductance level with decreases in EEG
alpha waves) in smokers compared with sham smoking or a
situational control group. However, under conditions of stress, as
induced by intermittent noise bursts, a mixed stimulant (heart rate)
and depressant (EEG, skin conductance) response was observed
(Golding and Mangan 1982). Woodson and coworkers (1986) also
reported that during noise, smoking induced cardiovascular stimula-
tion (i.e., heart rate acceleration, peripheral vasoconstriction) but
electrodermal depression (i.e., lowered skin conductance response
amplitude). These findings are consistent with the conclusions of
Gilbert and Welser (in press} that unidimensional models are
inadequate to explain the effects of smoking.

117
In addition to research on the impact of smoking on psychological
and physiological processes, studies have also examined the com-
bined cardiovascular effects of smoking and stress. In this context
the concept of cardiovascular psychophysiological reactivity is used
to help clarify the relationship among stress, smoking, and CHD
(Epstein and Jennings 1986). MacDougall and colleagues (1983)
randomly assigned 51 male smokers to smoking versus sham
smoking and stress versus no stress conditions in a 2 x 2 factorial
design. The stressor was a difficult video game performed under
challenging conditions. Subjects who sham smoked under no stress
showed minimal cardiovascular response. Subjects who smoked
under no stress or who sham smoked under stress evidenced similar
degrees of response of about a 15-bpm increase in heart rate, a 12-
mmHg increase in systolic blood pressure, and a 9-mmHg increase in
diastolic blood pressure. Subjects in the combined smoking and stress
condition had larger increases in all cardiovascular measures. The
combination of mild stress and smoking produced effects that were
twice those of either condition alone. Smoking and stress combined
to increase cardiovascular response in men.

In a followup study of women, using the same 2 x 2 factorial
design, Dembroski and colleagues (1985) found that the combined
effect of stress and smoking produced blood pressure and heart rate
increases that exceeded the sum of the individual effects. However,
because modifications were made in dosage and psychological
challenge, the two studies were not identical. The gender differences
noted could therefore reflect methodological differences, uncon-
trolled factors, or possibly differences between the sexes in response
to the stress and smoking stimuli. Indeed, it has been noted that
females may be more likely than males to smoke to regulate affect
(Ikard and Tomkins 1973), are more likely to relapse after quitting
(Gritz 1986), may differ in biological factors relating to stress
reactivity/sensitivity (Abrams et al. 1987), and show greater changes
in body weight and eating behavior in response to nicotine (Grun-
berg, Bowen, Winders 1986; Grunberg, Winders, Popp 1987). (See
Chapter VII for a discussion of treatment implications of these
possible sex differences.)

In a conceptually related study, the relationship between physio-
logical responses to cognitive (mental arithmetic) and physical (cold
pressor) stressors was examined in female smokers and nonsmokers
who either used or did not use oral contraceptives (Emmons and
Weidner, in press). All subjects showed some physiological response
(heart rate and blood pressure responses) to the stressors, but in
smokers oral contraceptive use significantly enhanced the systolic
blood pressure response to cognitive stress. This finding may be
related to the fact that smokers who use oral contraceptives are 5.6-
times more likely to have a myocardial infarction than are smokers

118
who do not use oral contraceptives, 9.7-times more likely than
nonsmoking users, and 39-times more likely than nonsmokers who
do not use oral contraceptives (Shapiro et al. 1979; Jain 1976; Ory
1977).

In studies of psychophysiological reactivity, it is critical to identify,
measure, and control for factors that might confound or alter the
intended impact of the independent variables. For instance, time
since last drink and beliefs, expectations, and setting are important
variables to consider in the study of alcohol addiction (Abrams and
Wilson 1979; Abrams 1983; Marlatt and Rohsenow 1980). The 2 x 2
balanced placebo design (Marlatt, Demming, Reid 1973), where
expectancy set (told to expect the drug or told to expect no drug) and
actual content (drug versus placebo) are fully controlled, has been
used extensively in the alcohol addiction field to isolate the separate
and interactive elements of cognitive and pharmacologic effects.
With smoking, little is known about the separate and interactive
impacts of expectations of cigarettes’ effects versus their actual
pharmacologic effects. This is partially because it is difficult to find a
method of administration that closely resembles smoking but where
the required manipulations to achieve a credible balanced placebo
design can be accomplished.

Another methodological concern is control over the dosage of
nicotine absorbed by the smoker. Nicotine is thought to be the most
important tobacco constituent responsible for the acute effects of
smoking on reactivity, attention and task performance, mood, and
withdrawal following cessation (Perkins et al., in press; Pomerleau,
Turk, Fertig 1984; Hughes et al. 1984). However, in tobacco smoking,
nicotine is accompanied by more than 4,000 other compounds (Dube
and Green 1982) and smokers are known to smoke in individualized
ways (Epstein et al. 1981) (Chapter IV). The coaching of puff
frequency and other attempts to standardize intake of smoke are
imperfect (Perkins et al., in press). An aerosol nasal spray appears to
be a promising alternative to smoking in studies of behavioral and
physiological effects. It allows for rapid uptake through inhalation,
and a dose-response study indicates patterns of heart rate, blood
pressure, and serum nicotine levels that are very similar to those
obtained by smoking cigarettes of equivalent nicotine content
(Perkins et al., in press).

Perkins and coworkers (in press) studied the separate and interac-
tive effects of nicotine administered by nasal aerosols and stress on
psychophysiological reactivity. The authors note that the previous
studies (MacDougall et al. 1983; Dembroski et al. 1985) could be
confounded because smokers usually smoke more under stress and
therefore they may inhale more nicotine or alter their smoking in
other ways when stressed (Mangan and Golding 1978; Rose, Ananda,
Jarvik 1983) (Chapter VI). In other words, the additive effects of

119
stress and smoking on physiological responses could have resulted
from uncontrolled changes in smoking pattern between the smokers
in the no-stress and stress conditions. Perkins and colleagues (in
press) studied 12 male smokers in a repeated-measures design, where
subjects received all 4 conditions (stress plus nicotine, stress plus
placebo, rest and nicotine, and rest and placebo) on separate days
with the order of condition counterbalanced within subjects. Follow-
ing the methodology of previous studies of psychophysiological
reactivity, the researchers used an active stressor consisting of a
video game under conditions of competitive challenge. Nicotine was
administered in measured 1.0-mg doses by the aerosol nasal method
(Perkins et al., in press). Consistent with observations of MacDougall
and coworkers (1983), results were additive for heart rate reactivity.
However, effects were less than additive for systolic and diastolic
blood pressure.

Taken together, the studies of the effects of smoking cigarettes and
of nicotine aerosol stimuli on the physiological responses of adult
males demonstrate a consistent effect for the stimuli alone, additive
in combination with stress on heart rate, and additive or less than
additive with stress on blood pressure. There is some suggestion that
effects may be more than additive for women, but this finding
requires replication.

Psychophysiological Reactivity, Smoking Cessation, and
Relapse

Psychophysiological reactivity also serves as a conceptual frame-
work to study relapse after cessation from smoking (Shiffman 1986b;
Abrams 1986). Individual differences in psychophysiological reactivi-
ty and associated coping responses, as a function of general and
smoking-specific stressful stimuli, have been hypothesized to medi-
ate relapse. For example, smokers who smoke more when stressed
might be particularly vulnerable to relapse (Pomerleau, Adkins,
Pertschuck 1978). This idea is consistent with the observation that
relapse may be triggered by life stress events and other psychosocial
demands (Ockene et al. 1982) and by high-risk situations including
negative emotions, social conflicts and pressures, and the presence of
alcohol or smoking cues (Marlatt and Gordon 1985; Shiffman 1979,
1982, 1984, 1986a; Abrams et al. 1986). If certain psychophysiological
reactivity responses distinguish potential abstainers from relapsers,
cessation may be better maintained by identifying ‘relapse-prone”
individuals (Chapter VID.

Stressful environmental demands, sensitivity of the individual to
these demands, and the repertoire of coping responses are important
factors in relapse (Shiffman and Wills 1985; Abrams et al. 1987).
These same factors also may contribute to initiation of smoking
among adolescents. Wills (1985) provides evidence for the stress-

120
coping model of smoking in adolescence, relating both stress and
coping patterns to substance use. Results are consistent with other
findings that, in addition to peer pressure to smoke, adolescents
actively seek methods of coping with their perceptions of stress
(Wills 1985; Friedman, Lichtenstein, Biglan 1985; Botvin and
McAlister 1981). Although these survey studies are consistent with
the notion of smoking as a means of coping with psychophysiological
reactivity to environmental demands, research has not yet measured
reactivity in adolescents prior to smoking onset.

Observational and retrospective studies of relapse have identified
other smoking-specific stressful stimuli and cogni-
tive/psychophysiological measures of reactivity that are relevant to
relapse. Situations or stimuli that cue smoking and are associated
with relapse include pharmacologic dependence and withdrawal
symptoms (Jarvik 1977; Pomerleau and Pomerleau, in press; Hughes
et al. 1984), stimuli previously associated with smoking (e.g., coffee
drinking, alcohol) (Shiffman 1984, 1986a; Best and Hakstian 1978),
and urges to smoke (Myrsten, Elgerot, Edgren 1977). Situational
stimuli may or may not have previously been paired with smoking
and may or may not include smoking cues as a trigger for relapse.

Substance use cues themselves (e.g., the sight and smell of
cigarettes) also may precipitate relapse, perhaps in combination with
other stressful] stimuli or in a vulnerable individual (Shiffman 1986b:
Abrams et al. 1987). Models of how substance use cues are related to
relapse have been proposed on the basis of classical, operant, and
social learning principles. Reactions may be conditioned to stimuli
repeatedly paired with smoking, resulting in craving and physiologi-
cal reactivity in their presence and moderated by dependence,
tolerance, and nonpharmacologic withdrawal (Siegel 1983; Cooney,
Baker, Pomerleau 1983; Gritz 1980). Psychophysiological reactivity
to smoking cues could mimic the prior drug response (Wikler 1965),
result in a drug-opposite (compensatory) response (Siegel 1983), or
have other effects on psychological processes such as perceived
anxiety, urges to smoke, and self-efficacy in resisting relapse
according to a social learning model of relapse (Marlatt and Gordon
1985).

Abrams and colleagues (1987) studied the psychophysiological
reactivity and behavioral coping responses of male and female
relapsers and quitters in four simulated situational contexts: general
social situations, smoking-specific negative emotional and interper-
sonal role-plays, high-demand social stress, and relaxation. Com-
pared to abstainers, relapsers had higher heart rates and higher
perceived anxiety and were rated as less skillful at coping in the
smoking-specific intrapersonal (negative affect) situations. There
were no differences on any measures in the high-performance-
demand general-social-stress procedure. There were some differences

121
in heart rate and self-reported anxiety in the general social
situations and in heart rate in the relaxation interval, with relapsers
having higher levels than abstainers. Abstainers and relapsers did
not differ in heart rate, perceived anxiety, or coping skills in the
high-demand social anxiety procedure, but they did differ in the
other situations. The results suggest that selected situational
demands prompt situation-specific psychophysiological changes.

Rickard-Figueroa and Zeichner (1985) used a within-subjects
design to examine the responses of smokers to a confederate of the
experimenter lighting and smoking the subject’s preferred brand of
cigarette behind a glass window. Cigarette paraphernalia were
piaced adjacent to the subject but smoking was not permitted until
after the session. The cue exposure manipulation resulted in higher
urges to smoke, increased systolic and diastolic blood pressure, and
increased heart rate variability compared with a no-cue condition.
Urges were significantly positively correlated with diastolic blood
pressure, the use of active mastery to cope with urges, and the more
rapid smoking of a standard cigarette after the trial.

In a study that shows some evidence for a conditioned response,
Saumet and Dittmar (1985) measured finger-pulse amplitude, a
measure of peripheral vasoconstrictive activity, while subjects
placed an unlit cigarette into their mouths and waited for it to be lit.
Heavy smokers showed an anticipatory vasoconstrictive response to
the cigarette compared with light smokers and nonsmokers.

Abrams and colleagues (in press) used smoking cues and a social
stressor to simulate an interpersonal situation with high risk for
relapse. Relapsers, abstainers, and never smokers were examined for
psychophysiological reactivity. Compared with controls (never smok-
ers), relapsers had significant heart rate reactivity, stronger urges to
smoke, and subjective anxiety. Trained raters, unaware of subject
smoking status, judged relapsers as having significantly less effec-
tive coping skills to resist smoking. In a second study, the same
assessment was used prospectively in a treatment outcome context
to determine whether patterns of psychophysiological reactivity
could discriminate between quitters who maintain abstinence from
those who do not. Both heart rate reactivity and subjective anxiety
were greater in quitters who relapsed at 6-month followup compared
with those who continued to abstain. The groups did not differ with
regard to urges to smoke or behavioral judgments of coping skill.
Thus, the two studies were consistent for heart rate and perceived
anxiety but not for urges or objective ratings of coping effectiveness.

In a reanalysis of the heart rate data from Abrams and coworkers
(in press), Niaura and colleagues (in press) examined beat by beat
event-related heart rate during the period immediately before and
for the 10 sec following the lighting of a cigarette by a confederate
(subjects did not smoke throughout). Prospective relapsers showed a

122
strong decelerative trend at the point of lighting, whereas prospec-
tive abstainers did not. The results may reflect a conditioned
compensatory response (Siegel 1983) or some other information
processing/attentional phenomenon (Sokolov 1963; Knott 1984). In
another treatment study, Emmons (1987) examined smokers’ cardio-
vascular reactivity to mental arithmetic or deep knee bends before
and 6 months after smoking cessation. There was no change in
reactivity (heart rate, systolic and diastolic blood pressure) to either
stressor before and after quitting. Heightened pretreatment heart
rate reactivity significantly discriminated relapse at 6-month follow-
up.

Individual differences in psychophysiological reactivity may influ-
ence the likelihood of relapse. This possibility is discussed in Chapter
VIL.

Summary and Conclusions

1. Nicotine is a powerful pharmacologic agent that acts in the
brain and throughout the body. Actions include electrocortical
activation, skeletal muscle relaxation, and cardiovascular and
endocrine effects. The many biochemical and electrocortical
effects of nicotine may act in concert to reinforce tobacco use.

2. Nicotine acts on specific binding sites or receptors throughout
the nervous system. Nicotine readily crosses the blood-brain
barrier and accumulates in the brain shortly after it enters the
body. Once in the brain, it interacts with specific receptors and
alters brain energy metabolism in a pattern consistent with the
distribution of specific binding sites for the drug.

3. Nicotine and smoking exert effects on nearly all components of
the endocrine and neuroendocrine systems (including catechol-
amines, serotonin, corticosteroids, pituitary hormones). Some
of these endocrine effects are mediated by actions of nicotine
on brain neurotransmitter systems (e.g., hypothalamic-pitu-
itary axis). In addition, nicotine has direct peripherally mediat-
ed effects (e.g., on the adrenal medulla and the adrenal cortex).

123
References

ABOOD., L.G., GRASSI, S. [*H}Methylcarbamylcholine, a new radioligand for studying
brain nicotinic receptors. Biochemical Pharmacology 35(23):4199-4202, December
1, 1986.

ABRAMS, D.B. Psychosocial assessment of alcohol-stress interactions: Bridging the
gap between laboratory and treatment outcome research. In: Pohorecky, L., Brick,
J. teds.) Stress and Alcohol Use. New York: Elsevier, 1983.

ABRAMS, D.B. Roles of psychosocial stress, smoking cues and coping in smoking-
relapse prevention. Health Psychology 5(Supplement).91-92, 1986.

ABRAMS, D.B., MONTI, P.M., CAREY, K.B., PINTO, R.P., JACOBUS, S.I. Reactivity
to smoking cues and relapse: Two studies of discriminant validity. Behaviour
Research and Therapy, in press.

ABRAMS, D.B., MONTI, P.M., PINTO, R.P., ELDER, J.P., BROWN, R.A., JACOBUS,
S.I. Psychosocial stress and coping in smokers who relapse or quit. Health
Psychology 6(4):289-303, 1987.

ABRAMS, D.B., NIAURA, R.S., CAREY, K.B., MONTI, P.M., BINKOFF, J.A.
Understanding relapse and recovery in alcohol abuse. Annals of Behavioral
Medicine 8(2/3):27-32, 1986.

ABRAMS, D.B., WILSON, G.T. Effects of alcohol on social anxiety in women:
Cognitive versus physiological processes. Journal of Abnormal Psychology
88:161-173, 1979.

ACETO, M.D., BENTLEY, H.C., DEMBINSKI, J.R. Effects of ganglion blocking agents
on nicotine extensor convulsions and lethality in mice. British Journal of
Pharmacology 37:104-111, September 1969.

ACETO, M.D., MARTIN, B.R. Central actions of nicotine. Medicinal Research Reviews
2(1):43-62, January-March 1982.

AGHAJANIAN, G.K., ROSECRANS, J.A., SHEARD, M.H. Serotonin release in the
forebrain by stimulation of the midbrain raphé. Science 15:402-404, 1967.

AMIR, S., BROWN, Z.W., AMIT, Z. The role of endorphins in stress: Evidence and
speculations. Neuroscience and Biobehavioral Reviews 4:77-86, 1980.

ANDERSEN, A.N., LUND-ANDERSEN, C., LARSEN, J.F., CHRISTENSEN, N.J.,
LEGROS, J.J., LOUIS, F., ANGELO, H., MOLIN, J. Suppressed prolactin but
normal neurophysin levels in cigarette smoking breast feeding women. Clinical
Endocrinology 17(4):363-368, October 1982.

ANDERSSON, K. Mecamylamine pretreatment counteracts cigarette smoke induced
changes in hypothalamic catecholamine neuron systems and in anterior pituitary
function. Acta Physiologica Scandinavica 125:445—-452, 1985.

APPELGREN, L.-E., HANSSON, E., SCHMITERLOW, C.G. The accumulation and
metabolism of C'*-labelled nicotine in the brain of mice and cats. Acta Physiologica
Scandinavica 56:249-257, 1962.

APPELGREN, L.-E., HANSSON, E., SCHMITERLOW, C.G. Localization of radioactiv-
ity in the superior cervical ganglion of cats following injection of C**-labelled
nicotine. Acta Physiologica Scandinavica 59:330-336, 1963.

ARMITAGE, A.K., HALL, G.H., MORRISON, C.F. Pharmacological basis for the
tobacco smoking habit. Nature 217:331-334, 1968.

ARMITAGE, A.K., HALL, G.H., SELLERS, C.M. Effects of nicotine on electrocortical
activity and actylcholine release from the cat cerebral cortex. British Journal of
Pharmacology 35(1):152-160, 1969.

ASHTON, H., MARSH, V.R., MILLMAN, J.E., RAWLINS, M.D., TELFORD, R.,
THOMPSON, J.W. Biphasic dose-related responses of the CNV (contingent
negative variation) to I.V. nicotine in man. Journal of Clinical Pharmacology
10(6):579-589, 1980.

ASHTON, H., MILLMAN, J.E., TELFORD, R., THOMPSON, J.W. Stimulant and
depressant effects of cigarette smoking on brain activity in man. British Journal of
Pharmacology 48(4):715-717, August 1973.

124
ASHTON, H.. MILLMAN, J.£., TELFORD, R., THOMPSON, J.W. The effect of
caffeine, nitrazepam, and cigarette smoking on the contingent negative variation
in man. Electroencephalugraphy and Clinical Neurophysiology 37(1):59-71, 1974.

ASHTON, H., STEPNEY, R.. TELFORD, R., THOMPSON, J.W. Cardiovascular and
behavioural responses to smoking. In: Ashton, H.. Stepney, R. feds.) Smoking,
Psychology and Pharmacology. London: Tavistock, 1982.

BALFOUR, D.J.K. The effects of nicotine on brain neurotransmitter systems.
Pharmacology and Therapeutics 16(2):269-282, 1982.

BALFOUR, D.J.K. (ed.) Nicotine and the Tobacco Smoking Habit. International
Encyclopedia of Pharmacology and Therapeutics, Section ll4, New York: Pergamon
Press, 1984.

BALFOUR, DuJ.K., GRAHAM, C.A., VALE, A.L. Studies on the possible role of brain
5-HT systems and adrenocortical activity in behavioural responses to nicotine and
diazepam in an elevated x-maze. Psvchopharmacology 90(4):528-532, November
1986.

BALFOUR, D.J.K., KHULLER, A.K.. LONGDEN, A. Effects of nicotine on plasma
corticosterone and brain amines in stress and unstressed rats. Pharmacology
Biochemistry and Behavior 3(2i:179-184, March-April 1975.

BAUM, A., GRUNBERG. N.E.. SINGER, J.E. The use of psychological and neuroendo-
crinological measurements in the study of stress. Health Psychology 1(3):217-236,
Summer 1982.

BEANI, L., BIANCHI, C., NILSSON, L.. NORDBERG, A., ROMANELLI, L., SIVILOT-
TI, L. The effect of nicotine and cytisine on *H-acetylcholine release from cortical
slices of guinea-pig brain. Naunsn-Schmiedeberg'’s Archives of Pharmacology
331(2/3):293-296, November 1985.

BEAUMONT. A., HUGHES, J. Biology of opioid peptides. Annual Review of
Pharmacology and Toxicology 19:245-267, 1979.

BENOWITZ, N.L., KUYT, F., JACOB, P. Influence of nicotine on cardiovascular and
hormonal effects of cigarette smoking. Clinical Pharmacology and Therapeutics
36(1):74-81, 1984.

BENWELL, M.E.M., BALFOUR, D.J.K. Central nicotine binding sites: A study of post-
mortem stability. Neuropharmacology 24(11):1135-1137, 1985a.

BENWELL, M.E.M., BALFOUR, D.J.K. Nicotine binding to brain tissue from drug-
naive and nicotine-treated rats. Journal of Pharmacy and Pharmacology
37(6):405-409, June 198db.

BENWELL, M.E.M., BALFOUR, D.J.K., ANDERSON, J.M. Evidence that tobacco
smoking increases the density of (-)-[*H]-nicotine binding sites in human brain.
Journal of Neurochemistry, probably in press but awaiting verification.

BERTOLINI, A., GESSA, G.L. Behavioral effects of ACTH and MSH peptides. Journal
of Endocrinological Investigation 4(2):241-251, April-June 1981.

BEST, J.A., HACKSTIAN, A.R. A situation-specific model for smoking behavior.
Addictive Behaviors 3(2):79-92, 1978.

BHARGAVA, V.K., McKEAN, C.M. Role of 5-hydroxytryptamine in the modulation of
acoustic brainstem (far-field) potentials. Newropharmacology 16:447-449, 1977.
BHARGAVA, V.K., SALAMY, A., McKEAN, C.M. Effects of cholinergic drugs on
auditory evoked responses (CER) of rat cortex. Neuropharmacology

17(12):1009-1013, December 1978.

BHARGAVA, V., SALAMY, A., SHAH, S. Role of serotonin in the nicotine-induced
depression of the brainstem auditory evoked response. Pharmacology Biochemistry
and Behavtor 15(4):587-589, October 1981.

BICKFORD, R. Physiology and drug action: An electroencephalographic analysis.
Federation Proceedings 19(2):619-625, July 1960.

BISSET, G.W., FELDBERG, W., GUERTZSTEIN. P.G., ROCHA e SILVA, M.
Vasopressin release by nicotine: The site of action. British Journal of Pharmacolo-
gy 54(4):463-474, August 1975.

125