Tobacco Documents Online

°srcnoAn~'sfo uz1. =0 119() , "i. ._ -_=b Camer;Cgr irr.i%ersits Pres, Pnntra r. :nr LS ~ Co^~~~gh: _!y9? Soc:et. tor Pa.cneph~ao~egical Researcn I I I I I I I I I I I I I I I I I I Effects of cigarette smoking on electrodermal orienting reflexes to stimulus change and stimulus significance MICHAEL LYVERS AND YO MIYATA Department of Psychology, Kwansei Gakuin University, Nishinomiya, Japan Abstract Skin conductance responses (SCRs) evoked by novel, signal, and frequent tone stimuli were measured in 20 male heavy smokers and 10 male nonsmokers over two sessions. All smokers abstained from smoking for 12 hr prior to each session. Half of the smokers smoked a cigarette of their preferred brand prior to SCR measurement in the first session, whereas the remaining smokers smoked in the second session. Nonsmokers did not smoke. Results combined across the two sessions indicated that abstinence was associated with selective depression of SCRs to the novel tone. Separate analyses of results from each session revealed that, in the second session, SCRs to both novel and signal tones were depressed in abstinent smokers, partially replicating previous findings. By contrast, first session results showed no significant effects of smoking or abstinence. Results were interpreted in terms of nicotine's effects on nonspecific arousal, with some reservations. Descriptors: Skin conductance response, Orienting reflex, Smoking, Nicotine The specific factors that maintain the remarkably persistent tobacco smoking habit are not well understood. Most recent theories of tobacco smoking focus on the role of the drug nic- otine (Pomerleau & Pomerleau, 1984), regarding smoking as a form of drug addiction. This view is supported by evidence that smokers tend to regulate their nicotine levels in response to changes in the nicotine content of their cigarettes (Robinson, Young, & Rickert, 1982). However, smokers' self-adjustment of nicotine levels through changes in number of cigarettes smoked, puffing rate, depth of inhalation, or other smoking pa- rameters rarely approaches 100% compensation (McMorrow & Foxx, 1983). Moreover, the motive for self-administering nic- otine is unclear. The drug addiction view of the smoking habit assumes that nicotine is self-administered so that the smoker may obtain some intrinsically rewarding drug effect and/or relief from unpleasant abstinence symptoms. But in marked con- trast to other addictive drugs such as cocaine, amphetamine, or heroin, nicotine appears to be a very weak primary reinforcer (Slifer & Balster, 1985). Tolerance to nicotine is weak and short lived (Henningfield, 1984), and termination of regular nicotine administration does not provoke obvious physical withdrawal This research was supported by Ministry of Eduation, Science and Culture Grant 02961005. We thank T. Mino, J. Katayanu, Y. Gondo, and 1. Sugishima for their assistance with the experimental apparatus, the analysis of results, and preparation of the final manuscript. Address requests for reprints to: Professor Yo Miyau, Department of Psychology, Kwansei Gakuin University, Nishinomiya 662, Japan. signs as does termination of opiates or alcohol (Abood, Gra.ssi, Costanzo, & Junig, 1984). Nicotine, a centrally acting cholinergic agonist (Schwartz & Kellar, 1983), appears to affect brain systems that regulate arousal. Smokers typically describe vague symptoms such as restlessness, fatigue, and difficulty concentrating during nico- tine abstinence (Huges et al., 1984: Warburton & Wesnes, 1978). Such self-reporu suggest that abstinent smokers are in a hypo- aroused state that can be relieved by smoking, an idea supported by ample evidence from EEG studies (Herning, Jones, & Bach- man, 1983; Knott, 1979; Knott & Venables, 1977; Szalai, Allon, Doyle, Peng, & Zamel, 1986). Given the evidence for a restor- ative effect of smoking on arousal in abstinent smokers, reports of apparent depressant effects of smoking on certain compo- nents of cortical evoked potentials (Knott, 1985, 1986), ha- bituation of the alpha desynchronization response (Friedman, Horvath, & Meares, 1974), and the skin conductance response (SCR) measure of the orienting reflex (OR) (Mangan & Gold- ing, 1978) are puzzling. Several authors have attempted to ex- plain such contradictory findings by postulating that, in addition to its arousing effecu, nicotine helps smokers "tune out" dis- tracting stimuli (Golding & Mangan, 1982b; Knott, 1986), per- haps by increasing the effectiveness of a hypothetical stimulus barrier mechanism (Friedman et al., 1974). Another interpre- tation is that smoking decreases the arousing effects of high- intensity stimuli, such as those employed in studies by Friedman et al. and Knott, or milder stimuli when arousal is already high because of other factors. According to this view, depressant effects of smoking should be evident only under conditions of 231

I I I I I I I I I I I I I I I I re:attt:. h:grn arousa! (Etisenck. 1985t Consisten: ui;h :hu d; ;~~Id!ne and \lancan t I y82a7 rAr•orted :ha: Smol ine ^ro S:ronc ~u^Ltant ettr.:ts or, a tarlet~ ot eleCtropn)~folo_L'- i,ai measures under reiaxtng senson isolation conditions but had mixed stimulant and depressant effects under conditions of hteh arousal induced by aversive whtte noise. Gilbert, Robin- son, Chamberlain. and Spielberger (1989) similarl% reported that smoktnc had a stimulant effect on electrocortical arousal in a no-stress control condition but reduced electrocortical arousal during presentation of stressful movie scenes. To evaluate the "stimulus barrier" hypothesis, Lyvers. Boyd, and Maltzman (1988) assessed the effects of smoking and ab- stinence on SCRs evoked by randomly presented frequent and infrequent innocuous tones, a typical oddball paradigm, under task and nontask conditions. Smokers who typically smoked at least 20 cigarettes per day abstained from smoking for 12 hr prior to each of two experimental sessions. Half of them smoked a cigarette of their preferred brand during the first session, whereas the remainder smoked during the second session. Non- smoking nonsmokers were also tested. In both sessions, all sub- jects were instructed to simply listen to the tape-recorded series of frequent and infrequent tones, which were easily discrimina- ble by pitch. In the second session, a task condition was added. Subjects were instructed to press a footpedal whenever they heard the infrequent tone. During this pedal-press task in the second session, smoking selectively enhanced SCRs evoked by the signal tone. This result was interpreted as reflecting a spe- cific effect of nicotine on central nervous system (CNS) process- ing of significant stimuli, possibly mediated by frontal cortex (Luria, 1973; Maltzman, 1979). No support for a stimulus bar- rier-enhancing effect of smoking was obtained; rather, smok- ing appeared to restore the CNS response to signals, which was depressed in abstinent smokers, to high nonsmoker levels. One problem with the Lyvers et al. (1988) study was that their oddball paradigm confounded stimulus change with stimulus significance in the task condition. The infrequent tone was not only significant but relatively novel as well. It occurred at ran- dom intervals and thus should have elicited ORs to the unpre- dictable stimulus change (Sokolov, 1960). To resolve this problem, a better oddball design was developed based on an evoked potential study by Roth, Ford, and Kopell (1978). In the improved design, three types of tone were repeatedly presented in randomized order: one frequent tone and two infrequent tones, all of which were readily discriminable by pitch. Task instructions established only one of the two infrequent tones as the signal for a pedal-press response. Normal subjects who are presented with this stimulus design generate significantly larger SCRs to the signal tone than to either the novel tone or the fre- quent tone and also show larger SCRs to the novel tone than to the frequent tone (Lyvers & Maltzman, 1991). The improved design thus allowed simultaneous evaluation of smoking effects on SCR-ORs evoked by stimulus change and stimulus signifi- cance. The present experiment employed the improved oddball de- sign in a replication and extension of the Lyvers et al. (1988)' smoking study. As in the previous experiment, smokers who smoked at least 20 cigarettes per day were tested over two ses- sions in smoking and 12-hr abstinent conditions. In smoking sessions, smokers were allowed to smoke a cigarette of their own preferred brand ad lib, without controlling for nicotine intake. The rationale for this approach was that it allowed habitual smokers to titrate their own nicotine dosage to achieve an opti- mum desired lete!. :1l;iung riiec; ;ha! are rroo::ni_•• .:.: : %k;:h .moktnY re:ntor,:ement unaer natural smoi,inL \compari~, orr group ot norrsmokers uas aiso ttsted. ~Lr rv- pected that abstinence would be associated with low letels of SCR. which would be restored to nonsmoker levels bt smok- ing. consistent with the common finding that smoking has a restorative effect on electrophysiologtcal indices of nonspecific arousal (Herning et al., 1983; I:nott, 1979: Knott & Venables. 1977: Lvvers et al., 1988; Szalai et al., 1986). A more interest- ing question was whether SCRs evoked by the signal tone would be selectively affected by smoking. If the effect of smoking was substantially greater in magnitude for the signal tone than for the novel tone, a selective effect of nicotine on CNS processing of signals would be demonstrated in addition to the anticipated effect on nonspecific arousal. Method Subjects An experimental group of 20 male smokers, defined as nicotine dependent based on Ashton and Stepney's (1982) criterion of smoking at least 20 cigarettes per day for at least I year up to the time of the experiment, was recruited from undergraduate classes at Kwansei Gakuin University, Nishinomiya, Japan. Smokers reported smoking 25 cigarettes per day on average. A comparison group of 10 male nonsmokers who reported that they had never smoked on a regular basis was recruited from the same classes. All subjects were right handed and between 20 and 25 years of age (M = 21.5 years). They were paid Y800 (ap- proximately S6) per hour for a total of 3 hr of participation over two sessions. Data for two smokers had to be discarded because of measurement problems. Skin resistance in these subjects was too high or too low for accurate measurement of tone-evoked SCRs. Materials and Apparatus Nihon Kohden Ag/AgCI skin electrodes were attached to the thenar eminence and center of the left palm of each subject's hand using adhesive electrode collars. The effective skin area contacted by each electrode was 0.95 cm2. Nihon Kohden Gelaid electrode paste (10°ia NaCI) served as the electrolyte. Skin re- sistance was recorded on a Nihon Kohden DZ-603G polygraph using a Nihon Kohden AD-611G high-gain DC amplifier and a constant-current (5 µA) bridge box circuit. Skin resistance baseline was centered on the polygraph manually using a custom computer program on an NEC PC-9801 VX personal computer. A 20-min tape recording of randomly ordered presentations of three computer generated tones was presented to subjects through Audio Technica ATH-20D stereo headphones connected to a Pioneer CT-720 cassette deck and Pioneer A-004 stereo ampli- fier. A 1,000-Hz tone presented 90 times was the frequent tone. A 2,000-Hz tone and a 500-Hz tone, each presented 15 times, were the infrequent tones. All tones had a duration of 50 ms. Rise times were under 50 ns. Interstimulus intervals (ISIs) var- ied randomly between 8 s and 12 s, with an average ISI of 10 s. Sound intensity for all three tones was 60 dB (SPL). Auditory signals were transmitted to an event marker on the polygraph. A footpedal beneath the seated subject's feet, when pressed, activated another event marker on the polygraph. Previous research (Bernstein, Taylor, Weinstein, & Riedel, 1985) and pi- lot work in this laboratory have shown that such simple motor responses do not themselves affect electrodermal ORs.

I I I I I I I I I I I I I I I I I I I E~"iects of sm.oktne on ortenlrne rel7exes i:om%o E:;tacawa -\P-4*(} CO creath tes: 4,;: ~ka~ u~ed to an% moxinc mmed;atei\ prior to :he esp er;men,ai sess;on. Procedure Durtng their usual class period, prospective subjects completed a brief questionnaire assessing self-reported handedness, smok- ing history, gender, and age. Subjects who met all criteria for participation were contacted by telephone. They were instructed to abstain from smoking (if applicable), caffeine, alcohol, or other drug use for 12 hr overnight prior to each of two experi- mental sessions, which would begin at 9:30 a.m. The two sessions were scheduled exactly I week apart. Subjects were telephoned again on the night before each session and reminded of the ab- sttnence requirements. Smokers were instructed to bring ciga- rettes of their usual brand in case they had to smoke during the experiment. By random assignment, half of the smokers smoked during their first session, whereas the remainder smoked dur- ing their second session. Nonsmokers did not smoke in either session. Upon arrival at the laboratory, smokers were given a CO breath test, followed by a short questionnaire (administered to all subjects) asking if the subject had smoked cigarettes or used alcohol, caffeine, or other drugs during the previous 12 hr. The CO breath test had a sensitivity of about 2 ppm and thus could only detect smoking that had occurred immediately prior to the experiment (Nils & Battig, 1989). The CO breath test was em- ployed in spite of this limitation, in part to discourage dishon- est responses on the subsequent self-report questionnaire and to encourage adherence to the nonsmoking requirement. ln the breath test, smokers were asked to exhale forcefully into a plas- tic gas bag, the contents of which were analyzed for expired CO. No evidence of expired CO was indicated by this test for any smoker, nor did questionnaire responses indicate violation of the nonsmoking or other abstinence requirements by any sub- ject. Electrodes were then attached to the subject's left palm. If the subject was in a smoking condition, he was instructed to smoke one of his own cigarettes ad lib using the right hand over a 7-min period while sitting quietly at rest, keeping the left hand resting with palm up. Nonsmoking subjects were instructed to sit quietly for 7 min while keeping the left hand resting with palm up. In this way, electrodes were allowed to stabilize while the experimental treatment (smoking or nonsmoking) was admin- istered. Immediately after the smoking or nonsmoking period, the subject was led into a brightly illuminated, soundproof, electri- cally shielded experimental chamber. He was seated in a chair with both feet resting on the footpedal. Electrodes were con- nected to the recording apparatus via leads that ran to the poly- graph system in an adjacent room. The participant was fitted with headphones and informed that he would hear a 20-min series of three distinct tones, one high, one low, and one medium in pitch, beginning with the medium tone. The subject was told that the medium tone would occur more often than the other two tones and that the high tone (or the low tone, depending on his counterbalancing subgroup) would be a signal to press the foot- pedal forward immediately, using both feet. The participant was then asked if he understood the instructions. If the an- swer was negative, instructions were presented again until fully understood. Following practice of the pedal press, the tones were presented and evoked SCRs were recorded while the subject sat alone in the experimental chamber. Recording was initiated \t(:nt.n, =.'nln )1 Cmok inc. en5llrtnc .',:3f1~Ct~ nICC ^i y.m,l :".i. tine Ie~rk durtr.e the tones :as},. Partm,:ipant• rrrrsnuo:-i% asstgned to ~ounterbalanc;ng sub¢roups such tha; the h;gh tone was the pedal press signal for lialf of the subjects in each group and the low tone was the signal for the remainder. - Following presentation of the tones, electrodes and head- phones were removed. A subgroup of smokers and nonsmok- ers again experienced a 7-min smoking or rest period, after which they performed the Wisconsin Card Sorting Test on an Apple fI computer. These results will be reported elsewhere. Upon completion of both experimental sessions, participants were paid for their participation and released. Results The largest resistance change that began during a 0.5-4-s win- dow following each tone stimulus was measured when the change exceeded 500 ohms. The resistance change values were transformed directly into units of conductance. A square root transformation was then made on the resultant SCR values. Be- cause there were 90 frequent tones. 15 signal tones, and 15 novel tones, for analysis the SCR to every sixth frequent tone was sampled, starting with the first tone of the series. SCRs to non- signal tones that elicited a pedal press response (false alarms) and SCRs to signal tones that failed to elicit a pedal press re- sponse (misses) were not measured because they were not inter- pretable in terms of the signal vs. nonsignal OR dimension. However, such errors were rare on this simple task, and groups did not differ in the number of false alarms or misses. Analy- ses of variance (ANOVAs) incorporating SCR data from the en- tire 20 min of tones presentation, analyzed as five trial blocks, revealed that significant habituation occurred only in the first trial block of 18 frequent tones, 3 novel tones, and 3 signal tones. Because potential effects of smoking on habituation were of some interest. ANOVAs were also conducted on results from within this first trial block only. These analyses revealed essen- tially the same smoking effects as the overall analyses, in addi- tion to showing significant habituation; thus, only the first trial block results are reported here. Greenhouse-Geisser corrected p values are given for all repeated-measures effects. An ANOVA was conducted on all first trial block SCR data for smokers, with smoking treated as a within-subject variable combining results across the two sessions. Other within-subject variables included tone (frequent, novel, signal) and trial (lst. 2nd, or 3rd novel or signal tone or lst, 7th, or 13th frequent tone). This analysis revealed the expected main effect of tone, F(2,34) = 76.41, p < .0001, t: = 0.91, and a Condition x Tone interaction, F(2,34) = 3.35, p < .05. e= 0.95. Multiple t tesu employing the Dunn-Bonferroni correction revealed that SCRs evoked by the signal tone were significantly larger than those evoked by the novel tone or the frequent tone in both smoking and abstinent conditions, p<.001. However, SCRs evoked by the novel tone were significantly larger than those evoked by the frequent tone in the smoking condition only, t(17) = 3.62, p < .01. In the abstinent condition. SCRs evoked by the novel ZZ tone were not different from SCRs evoked by the frequent tone, Q t(17) = 0.32, n.s. (Figure 1, left panel). Only the SCR-OR evoked by stimulus change reflected an effect of smoking ac- Q~ cording to this analysis. W~ib The within-subjects ANOVA also showed a significant effect p of trials, F(2,34) = 21.79, p<.0001, e = 0.79, and a Tone x p C ~

W . L rrPrS anc } 11:1u[u 1 ~ ~ 1 ~ ~ 1 ~ ~ 1 ~ 1 1 ~ ~ Z" 2^ 0 --~- .....~. .. ~ ....... 0 ................ p 1 2 3 Trial signal novel frequent 0 1 2 Trial r 3 Figure 1. Mean square-root transformed skin conductance responses to signal, novel, and frequent tones in smokers (left panel) and nonsmokers (right panel) in the first trial block over two expenmental sessions. In the left panel, open figures denote smoking, solid figures denote abstinence. Trials interaction, F(4,68) = 9.38, p < .0001, s= 0.72. Habitu- ation was greatest for the frequent tone (Figure 1). The comparable ANOVA on nonsmoker data similarly re- vealed significant effects of tone (F[2,16J = 21.72, p<.001), trials (F[2,16] = 7.22, p < .01, s= 0.75), and a Tone x Trials interaction (F[4,321 = 5.82, p < .01, e= 0.55). These nonsmoker resulu reflect the same effects as those obtained in smokers (Fig- ure 1, right panel). - A selective effect of smoking on the OR to stimulus change in the absence of any effect on the signal OR was unexpected given the previous findings of Lyvers et al. (1988), who observed abstinence-related depression of the OR evoked by a signal stim- ulus and restoration of this OR by smoking in the second of two SCR sessions. Thus the possibility that smoking interacted with sessions in the present experiment was examined• A mixed ANOVA with session, tone, and tri.als as within-subject variables and group (smoked in first session, smoked in second session) as a between-subject variable revealed a main effect of session, F(1,16) = 10.67, p < .01, and a Session x Tone x Group inter- action, F(2,32) = 3.3, p < .05, s= 0.96. Smokers generated significantly larger SCRs to all tones in the first session (M = 1.08 ,cS) compared with the second session (M = 0.70 AS). The interaction of session, tone, and smoking was examined by sep- arate analyses of smoking effects within each session, compar- ing smoking smokers, abstinent smokers, and nonsmokers as in Lyvers et al. (1988). In the second session, abstinence was as- sociated with depression of SCRs to both the novel tone and the signal tone, whereas SCRs to the frequent tone were not affected (Figure 2). The mixed ANOVA for the second session revealed a sigttificant effect of tone, F(2,50) = 66.77, p < .0001, e=0.97, and a Tone x Groups interaction, F(4,50) = 2.85, p<.05. The Tukey post test ( p<.05) showed that SCRs evoked by 21 N =L V © !} ~ signal .....~.... novel - 0 frequent i ---1--- (n Q ...~ .......... ~... ...... N M .............~..~ .. . ..A 0 T- r 0 s r--. t 2 3 1 2 3 Trial Trial Fig<tre 2. Mean square-root transformed skin conductance responses to sigztal, novel, and frequent tones in smokers (left panel) and nottsmokers (right panel) in the first trial block of the second session. 1n the left panel, open figures denote smoking, solid figures denote abstinence.

I I I I I I I I I I I I I I I I I I E.''"ec.v o,' sMoh-:nQ on o/lenltnA rej7eXes ,he no\,el tonr Nere vcniii antl~ smaller in abstinent smokers ~an :n )moF.tns: smokers or nonsmokers, SCR~ e%oked b% the stgnai tone were also stgnificantlv smaller in abstinent than in smoking smokers. There were no group differences on SCRs evoked by the frequent tone. The second session ANOVA fur- ther revealed a significant effect of trials, F(2,50) = 9.7'_, p < .001, e= 0.74. and a Tone x Trials interactions, F(4,100) = 8.94, p < .0001, c= 0.76, again reflecting the more pronounced habituation of SCRs to the frequent tone in all groups. An identical analysis was also conducted on results from the first session, in which one nonsmoker was excluded because of an equipment problem that occurred in that session only. The first-session mixed ANOVA indicated the usual significant ef- fects of tone (F(2.48] = 48.1 1, p < .0001, e= 0.83) and trials (F(2,48J = 23.31, p < .0001, e= 0.96) and a Tone x Trials interaction (F(4,96] = 8.52, p < .0001, s= 0.73) similar to that in the second session. The Tone x Groups interaction was not significant in Session 1, F(4,48) = 0.75, n.s., unlike Session 2. However, there was a significant Trials x Groups interaction, F(4.48) = 3.74, p<.01. The Tukey post test ( p<.05) revealed that on the first trial only, abstinent smokers had larger SCRs overall (M = 1.59 µS) than nonsmokers (M = 0.93 ~cS). Smok- ing smokers (M = 1.21 uS) exhibited an intermediate level of SCR that was not significantly different by post test from that of the other groups. Pedal-Press Reaction Time Although the pedal-press task was not presented to subjects as a test of their response speed, pedal-press reaction times (RT) could be recovered by measuring the distance between signal and response event marks on the polygraph. Smoking had no effect on RT nor did RT vary as a function of other experimen- tal variables, according to the appropriate ANOVAs. Smoking smokers (M= 0.58 s), abstinent smokers (M = 0.61 s), and non- smokers (M = 0.65 s) showed similar response times. Spontaneous Responses The number of spontaneous SCRs, defined as changes in resis- tance of more than 50 ohms that began more than 4 s after each tone stimulus but before the following stimulus, was counted for each subject and analyzed as five trial blocks in each session. The appropriate ANOVAs indicated no effect of smoking on spontaneous SCRs in eithet session. There was a significant effect of trial blocks in the first session only, F(4,96) = 7.42, p < .01, t: = 0.46. The Tukey post test (p < .05) revealed that in the first session subjects generated significantly more spon- taneous SCRs in the first trial block than in all subsequent blocks, with no other significant comparisons. Discussion Because this experiment was initially conceived as a within- subjects design, the data analysis initially focused on results combined across sessions. This analysis indicated a selective ef- fect of smoking deprivation on the OR to stimulus change. Whereas smoking smokers displayed the usual nonsmokerlike differentiation between SCRs to signal, novel, and frequent tone stimuli (in descending order of SCR magnitude), during absti- nence they showed no differential response to the novel tone as compared with the frequent tone. This result was surprising given the previous finding of a selective effect of abstinence on the signal OR and its normalization by smoking (Lyvers et al., 19881 Separate anaivs;, o( re~ul:. ;rjm :h,: r- ,ni •_••wn ,•: the present e\periment sho«ed tha; SCR•UR> : the no%ei :on: and the signal tone were both depressed in absttnent as compared with smoking smokers. Second'=sesston results were thus consts- tent with those of the previous study, which had assessed the sig- ?ial OR in only the s.econd of two SCR sesstons. Because the signal tone was novel as well as si¢nificant. and because SCRs to a novel nonsignal tone were similarly affected, the abstinence- related depression of the signal OR observed in the second ses- sion of the present study can be attributed to depression of the component evoked by stimulus change. Present results thus do not support the predicted selective effect of smoking or nico- tine deprivation on CNS processing of signals. These results are equivocal with respect to the hypothesis of a frontal cortical effect of nicotine (Lyvers et al., 1988) but do not support the hypothesis of Friedman et al. (1974) and Golding and Mangan (1982b) that smoking helps smokers to "tune out" or ignore non- task stimuli. Abstinence-related depression of the traditional OR to stim- ulus change, and its restoration by smoking, likely reflects the state of hypoarousal most often associated with nicotine depri- vation in heavy smokers and the normalization of arousal lev- els by smoking or nicotine administration (Herning et al., 1983; Knott, 1979; Knott & Venables, 1977; Szalai et al., 1986). How- ever, SCRs to the frequent tone were not affected by smoking or abstinence in the second session of the present experiment, in contrast to SCRs evoked by novel or signal tones. A general effect of smoking or abstinence on nonspecific arousal should be manifested by corresponding changes on SCR-ORs to all three stimulus types. A floor effect could account for the lack of any effect of smoking on SCRs to the frequent tone, but only after the first trial. SCRs to the frequent tone were of fairly high magnitude on the first trial, with no smoking-related difference (Figure 2, left panel). Thus, it may be significant that the fre- quent tone was the very first stimulus presented. Recent evidence suggests that initial stimuli may evoke a unique electrocortical response that is distinct from those evoked by stimulus change or stimulus significance (Naatanen & Gaillard, 1983). First-session results of the present study did not show clear effects of smoking. Although abstinent smokers showed rela- tively larger SCRs to the first tone of each type as compared with nonsmokers, abstinent smokers did not differ significantly from smoking smokers on any SCR measure. Considering the strong OR-enhancing effects of smoking revealed by the second-session results, the lack of any effect in the first session is surprising. A plausible interpretation is that arousal evoked by the novel, unfamiliar, and potentially threatening experimental environ- ment interacted with or overrode smoking effects in the initial session. The finding that responses to all tones were significantly elevated in the first session as compared with the second session supports this interpretation. Previous research has indicated that variables peculiar to initial sessions, such as anxiety, uncertainty, or interest in the novel experimental situation, may interact with experimental treatments and have various effects on SCR mea- sures (Crider & Lunn, 1971; Lyvers, Boyd, & Maltzrnan, 1987, 1988; Maltzman, Smith, Kantor, & Mandell, 1971). Single- session experiments may give an incomplete or distorted picture of smoking effects on OR measures for this reason. The trend for smoking-related differences in the first session was sugges- tive of a depressant action of smoking, consistent with previ- ous single-session SCR studies of smoking (Golding & Mangan, 1982b; Mangan & Golding, 1978). I

A _6 kf. (-t1Prt OR6' ) ll:,~,, I The oddball de~tcn empio%ed in ,he present ~:ud, ;uli, di,;eren,ta:ee OR., to trequent. no,e!, and ~ionai ct:m- ui; on tne oasts of SCR magnitude. SCRs to the frequent tone habituated rapidly, reaching asymptote within the first trial block. Habituation was less evident for novel and especially stenal tones, signal tones elicited much lar¢er SCRs than did the other tones throughout the experiment. In addition to the ; retrn: 'indrngs per,atntnc to ,mot.inc. thi• de.icn na• aiio~t;,: demon<trarnon of aNriecu,e effect of ai:ohol on ;he •iLaai i)R (Lti%ers &!ttaltzman, 1991) and thus appears to or a u.etui method for assessing specific and-nonspecific effects of drugs. Future experiments will investigate the effects of other drug treatments, as well as trait factors, on the different t,pes of SCR-OR that can be elicited with this design. I I I I I I I I I I I I I I REFERENCES Abood, L. G.. Grassi. S., Costanzo, M., & Junig, J. (1984). Behavioral and biochemical studies in rats after chronic exposure to ntcotine. ,Natronal Institute on Drug A buse: Research Monograph Series, 54, 348-355. Ashton, H., & Stepney, R. (1982). Smoking: Psychology and pharma- cology. London: Tavistock. Bernstein, A. S., Taylor, K. W., Weinstein, E., & Riedel, J. (1985). The effect of stimulus significance on relatively sustained (tonic-like) and relatively transient (phasic-like) aspects of electrodermal, heart rate, and eyeblink response. Biological Psychology, 21, 183-228. Crider. A., & Lunn, R. (1971). Electrodermal ability as a personality dimension. Journal of Experimental Research in Personality, 5, 145-150. Eysenck, H. J. (1985). Stress, personality, and smoking behavior. In C. D. Spielberger, 1. G. Sarason, & P. 8. Defares (Eds.), Stress and an-rrety (Vol. 9, pp. 37-49). New York: Hemisphere. Friedman, J.. Horvath, T., & Meares, R. (1974). Tobacco smoking and a "stimulus barrier." Nature, 248, 455-456. Gilbert, D. G.. Robinson, J. H.. Chamberlain, C. L., & Spielberger, C. D. (1989). Effects of smoking/nicotine on anxiety, heart rate, and lateralization of EEG during a stressful movie. Psychophysiology, 26. 311-320. Golding, J., & Mangan, G. L. (1982a). Arousing and de-arousing effects of cigarette smoking under conditions of stress and mild sensory iso- lation. Ps,vchophysiologv, 19, 449-456. Golding, J., & Mangari, 0. L. (1982b). Effects of cigarette smoking on measures of arousal, response suppression, and excitauon/inhibition balance. International Journal of the Addictions, 17, 793-804. Henningfield, J. E. (1984). Pharmacologic basis and treatment of cig- arette smoking. Journal oJClinical Psychiatry, 45. 24-34. Herning, R. I., Jones, R. T., & Bachman, J. (1983). EEG changes dur- ing tobacco withdrawal. Psychophysiology, 20, 507-512. Hughes, J. R., Hatsukami, D. K.. Pickens, R. W., Krahn, D., Malin, S.. & Luknic, A. (1984). Effect of nicotine on the tobacco withdrawal syndrome. Psychopharmacology, 83, 82-87. Knott, V. J. (1979). Psychophysiological correlates of smokers and non- smokers: Studies on cortical, autonomic and behavioural responsiv- ity. In A. Remond & C. Izard (Eds.), Electrophysiological effects of nicotine (pp. 99-116). Amsterdam: Elsevier. Knott, V. J. (1985). Tobacco effects on cortical evoked potentials to dis- tracting stimuli. Neuropsychobiology, 13, 136-140. Knott, V. J. (1986). Tobacco effects on cortical evoked potentials to task stimuli. Addictive Behaviors. 11, 219-223. Knott, V. J.. & Venables, P. H. (1977). EEG alpha correlates of non- smokers, smokers, smokinj, and smoking deprivation. Psychoph,vs- iology, 14, 150-156. Luria, A. R. (1973). The frontal lobes and the regulation of behavior. In K. H. Pribram & A. R. Luria (Eds.), Psychophysiology of the frontal lobes (pp. 3-26). New York: Academic Press. Lyvers. M., Boyd, G., & Maltzman, 1. (1987). Smoking, personality, and imagery preference in relation to spontaneotu bilateral electro- dermal activity. Pavlovian Journal of Biological Science. 22, 7-15. Lyvers, Vi., Boyd, G., & Maltanan, 1. (1988). Effects of cigarette smok- tng on bilateral electrodermal measures of the orienting reflex. Psv- chophys1ology, 25, 408-417. Lyvers. M., & Maltzman, 1. (1991). Selective effects of alcohol on elec- trodermal indices of orienting reflexes to signal and nonstgnal stim- uli. Psychophysiology, 28, 559-569. Maltzman, 1. (1979). Orienting reflexes and classical conditioning in humans. In H. D. Kimmel, E. H. Van Olst, & J. F. Orlebeke (Eds.), The orienting reJler in humans (pp. 323-351). New York: Erlbaum. Maltzman, I., Smith, M.J., Kantor, W., & Mandell, M. P. (1971). Effect of stress on habituation of the orienting reflex. Journal ojExperi- mental Psychology, 87, 207-214. Mangan, G. L., & Golding, J. (1978). An "enhancement" model of smoking maintenance? In R. E. Thornton (Ed.), Smoking behav- iour: Physiological and psychological influences (pp. 87-1 l4). Lon- don: Churchill Livingstone. McMorrow, M. J., & Foxx, R. M. (1983). Nicotine's role in smoking: An analysis of nicotine regulation. Psychological Bulletin, 93, 302-327. Naatanen. R., & Gaillard, A. W. K. (1983). The orienting reflex and the N2 deflection of the ERP. In A. W. K. Gaillard & W. Ritter (Eds.), Tutorials in event-related potential research: Endogenous compo- nents (pp. 119-141). Amsterdam: Elsevier. Nils, R., & Battig, K. (1989). Smoking behavior: A multivariate pro- cess. In T. Ney & A. Gale (Eds.), Smoking and human behavior (pp. 199-221). Chichester: Wiley, Pomerleau, 0. F., & Pomerleau, C. S. (1984), Neuroregulators and the reinforcement of smoking: Towards a biobehavioral explanation. Neuroscience d Biobehavioral Reviews, 8, 503-513. Robinson, J. C., Young, J. C., & Rickert, W. S. (1982). A comparative study of the amount of smoke absorbed from low yield ("less haz- ardous") cigarettes. 1. Non-invasive measures. British Journal of Addiction, 77, 383-397. Roth, W. T.. Ford, J. M., & Kopell, B. S. (1978). Long latency evoked potentials and reaction time. Psychophysiologv, 15, 17-23. Schwartz, R. D.. & Kellar, K. J. (1983). Nicotinic cholinergic receptor binding sites in the brain: Regulation in vivo. Science, 220, 214-216. Slifer, B. L.. & Balster, R. L. (1985). Intravenous self-administration of nicotine: With and without schedule-induaion. Pharmacology, Biochemistry & Bthavior, 22, 61-69. Sokolov, E. N. (1960). Neuronal models and the orienting reflex. In M. A. B. Brazier (Ed.), The central nervous system and behavior (pp. 187-276). New York: Josiah Macy Jr. Foundation. Szalai, J. P., Allon, R., Doyle, J., Peng, & Zamel, N. (1986). EEG alpha reactivity and self-regulation correlates of smoking and smok- ing deprivation. Psychosomatic Medicine, 48, 67-72. Warburton, D. M., & Wesnes, K. (1978). Individual differences in smok- ing and attentional performance. In R. E. Thornton (Ed.), Smoking behaviour: Physiological and psychological influences (pp. 19-144). London: Churchill Livingstone. (RacsrvEa October 14, 1991; Accarst:n March 20, 1992)