Psychological Reports, 1979, 45, 423-428. @ Psychological Reports 1979
PERSONALITY, AROUSAL A N D INDIVIDUAL DIFFERENCES I N CIGARETTE SMOKING1 VERNER J. KNOTT Royal Otiawa Hospital Summa7y.-This study attempted to examine physiologically based personality correlates of 'low-arousal smokers' and 'high-arousal smokers.' The former were smokers who generally experienced their strongest need to smoke in low-arousal situations, characterized by, e.g., monotony or boredom, while the latter experienced their strongest need to smoke in high-arousal situations, characterized by, e.g., anxiety or excitement. Members of each group completed the Eysenck Personalicy Inventory and the Vando Reduction-Augmentation Scale. High-arousal smokers exhibited significantly greater scores on extraversioon, neuroticism, and reducing dimensions.
McCrae, Costa, and Boss4 ( 1978) have recently questioned the productivity of studies which employ global personality dimensions to differentiate smokers from non-smokers and have forwarded the suggestion that personality research must become more specific in terms of more meaningful classification of smokers and in terms of more specific facets of personality. Although the validity of this approach is questionable (Adesso & Glad, 1978), a number of researchers have attempted classification on the basis of different motives or different occasions for smoking (Tomkins, 1968; McKennell, 1970; Frich, 1971; Russell, Peto, & Patel, 1974; Best & Hakstian, 1978). Typically, these investigations have employed complex multivariate statistical techniques, primarily factor analysis applied to self-report data, for the purpose of isolating factors which represent different smokers or different smoking types. Although these studies have identified a number of situational and motivational factors involved in smoking, the role of personality has been noticeably neglected in these studies. Recently, however, Myrsten, Anderson, Frankenhaeuser, and Elgerot ( 1975) have examined an arousal-based typology which may have particular relevance to physiologically based personalicy theories. Employing a questionnaire concerning interactions between the need to smoke and the external sicuation, Myrsten, et al. classified smokers as either 'low-arousal smokers' or 'higharousal smokers.' The former were smokers who generally experienced their strongest need to smoke in low-arousal situations, characteri~ed'b~, e.g., monbtony or boredom, while the latter experienced their strongest need to smoke in situations of high arousal, characterized by, e.g., anxiety or excitement. Examined under laboratory conditions designed to induce either low or high levels of lResearch was supported by the Canadian Tobacco Manufacturer's Council. Address requests for reprints to: Dr. Verner J. Knon, Specialty Clinics, Royal Ottawa Hospital, 1145 Carling Avenue, Ottawa, Ontario, Canada, K1Z 7K4.
V. J. KNOTT
arousal, the effects of smoking- differed characteristically for low-arousal smokers and high-arousal smokers. For low-arousal smokers, performance and general well-being were favourably affected by smoking in the low-arousal situation only. Conversely, performance and well-being of the high-arousal smokers were enhanced by smoking in the high-arousal situation only. Though Myrsten, et al. did not observe any significant differences between low- and high-arousal groups on physiological and performance variables in nonsmoking conditions, it is reasonable to suggest that examination of physiologically oriented personality measures such as Eysenck's (1967) introversionextraversion, neuroticism-stability and Petrie's ( 1967) augmenting-reducing dimensions would be particularly productive in elucidating differential motives underlying the smoking habits of low- and high-arousal groups. Studies employing this personality approach have been reported by Bartol (1975) and Barnes and Fishlinsky ( 1976). In an attempt to examine the relationship between personality and self-reported desire for cigarettes under stressful and non-stressful conditions, Bartol ( 1975 ) focused directly on Eysenck's hypothesis which stated that ". . . ( a ) introverts should more frequently use nicotine for tranquilizing purposes; and ( b ) people smoking under conditions of 'boredom' should use nicotine for stimulating purposes, people smoking under conditions of 'stress' should use nicotine for tranquilizing purposes" (Eysenck, 1973, pp. 113-146). Employing female smokers, Bartol found, contrary to Eysenck's prediction, that extraverts indicated they prefer to smoke in non-stressful situations. Although neurotics desired cigarettes more often than stable persons in both conditions, neurotic-extraverts tended to report the greatest desire to smoke in stressful situations. Barnes and Fishlinsky ( 1976) examined Petrie's ( 1967) hypothesis that reducers would smoke more cigarettes in low-arousal situations and that augmenters would smoke more cigarettes in high-arousal situations. Employing Vando's ( 1969) reducing-augmenting scale, the findings did not support the hypothesis and in fact, extreme female augmenters tended to crave cigarettes more during low-arousal than high-arousal. Though the results of the two studies run contrary to stated theories, and the relationship of the results to pharmacological actions of tobacco is suspect, i.e., females are less likely to inhale tobacco smoke than males (Todd, 1969), Petrie ( 1967) and Vando's ( 1969) studies have indicated that extraverts tend to be reducers while introverts tend to be augmenters, and as such the findings of Bartol and Barnes and Fishlinsky seem reliable and worthy of further investigation. The purpose of the present study was to attempt to replicate the above findings using the classification method of Myrsten, et al, and a sample of male smokers.
METHOD A questionnaire similar to that by Myrsten, et al. (1975) was employed
INDIVlDUAL DIFFERENCES IN CIGARETTE SMOKlNG
42 5
for selecting subjects with different smoking habits, i.e., subjects whose desire to smoke was greatest in either low-arousal or high-arousal situations. The questionnaire had 10 items referring to high-arousal situations, chosen from the questionnaires constructed by McKennell ( 1970) and Frith ( 1971). Subjects were instructed to imagine themselves in each of the situations and to indicate on a 7-point scale how strong their desire to smoke would be. Classification of smokers on the basis of the smoking questionnaire was accomplished by expressing each subject's score on low-arousal items and higharousal items as a percent of his total score. Subjects whose score on higharousal items was greater than 50% of the total score were classified as higharousal smokers and subjects whose score on low-arousal items was greater than 50% of the total score were classified as low-arousal smokers. One hundred male cigarette smokers were administered the smoking questionnaire along with Form A of the Eysenck Personality Inventory (Eysenck & Eysenck, 1968) and the Vando Reducing Augmenting Scale (Vando, 1969). The subjects were all inhalers with a mean age of 37.3 ( 4 9.9) yr., a mean daily consumption of 22.6 ( -t- 8.6) cigarettes and a mean smoking history of 14.0 ( 2 9.3) y r The mean nicotine and tar yields of the cigarettes consumed by the smokers were 1.15 ( + .29) and 16.49 (f3.88) milligrams, respectively. There were no statistically significant differences between low- and high-arousal groups.
RESULTS The mean raw score for high-arousal items was 22.7 ( + 8.3) and for lowarousal items 22.7 ( f 7.6). Means and standard deviations for extraversionintroversion, neuroticism-stability and reducing-augmenting dimensions were 12.74 ( f 8.03), 9.84 ( f 3.14), and 27.00 ( t 8.03), respectively. Among the 100 subjects, 48 achieved a high-arousal item score greater than 50% of their total score and were classified as high-arousal smokers. An additional 48 subjects achieving a low-arousal item score greater than 50% of their total score were classified as low-arousal smokers. Their mean scores on the relevant items are given in Table 1. There were no significant differences between the lowand high-arousal smokers in smoking consumption, smoking history, or age. Means and standard deviations for extraversion-introversion, neuroticismstability and reduction-augmentation dimensions for each of rhe rwo groups are TABLE 1 MEAN SCORESFORTWOGROUPS OF SMOKERS ON ITEMSREFERRING To HIGH-AROUSAL AND LOW-AROUSAL SITUATIONS(ns = 48) Items
High-arousal smokers
Low-arousal smokers
High-arousal Low-arousal
27.3 19.7
18.2 25.6
TABU 2
MEANSA N D STANDARD DEVIATIONS OF EXTRAVERSION-IN~OVBRSION, NEUROTICISMSTABILITY AND REDUCING-AUGMENTINGBY LOW- AND HIGH-AROUSAL SMOKERS(ns = 48) Dimensions Extraversion-Introversion Neuroticism-Stability Reducing-Augmenting
High-arousal smokers M SD 13.6 4.2 11.3 28.5
5.5 7.0
Low-arousal smokers M SD 11.9 4.2 9.2 5.5 25.7
7.0
shown in Table 2. Independent t tests (one-tailed) indicated that high-arousal smokers exhibited significantly greater scores on extraversion ( t = 1.84, df = 94, p < .05 ) , neuroticism ( t = 1.85, df = 94, p < .05 ), and reduction ( t = 2.16, df = 94, p < .025) than low-arousal smokers. DISCUSSION The present results supported the findings of Bart01 and Barnes and Fishlinsky in that low-arousal smokers exhibited lower extraversion (higher introversion) and lower reducing (higher augmenting) scores than high-arousal smokers. Also, high-arousal smokers exhibited higher neuroticism scores. This latter finding parallels a recent observation by Warburton and Wesnes (1978) who examined the relationship between neuroticism and various factors of the Smoking Motive Questionnaire by Russell, Peto, and Pate1 ( 1974). Here, smokers high on neuroticism were 'sedative' smokers in that they reported smoking when worried or angry, i.e., 'under stress.' This was supported by negative correlation obtained between neuroticism and the 'stimulation' factor, indicating that smokers high on neuroticism are less likely to smoke for stimulation. Of particular interest was Warburton and Wesnes' finding of a negative correlation between neuroticism and the 'pharmacological addiction' dimension. This finding that smokers who score high on neuroticism were less likely to be 'addicted' pharmacologically gives support to Myrsten, et al.'s ( 1975) hypothesis that pharmacological motives are predominant in low-arousal smokers and psychological motives are predominant in high-arousal smokers. As neurotic subjects tend to respond more strongly to stimuli, show greater variability in response, and take longer to recover after the response (Eysenck, 1967), it is logical to expect and observe higher neuroticism scores in high-arousal subjects. However, contrary to the neuroticism-stability dimension, the theoretical basis of extraversion-introversion and reducing-augmenting dimensions conflict with the present findings of higher extraversion and reducing scores in high-arousal smokers. .Relative to extraverts who are hypothesized as exhibiting strong inhibitory tendencies, low cortical arousal and hypo-sensitive responsivicy to stimuli, introverts are hypothesized to exhibit weak inhibitory and strong
INDlVlDUAL DIFFERENCES IN ClGARETI'E SMOKING
42 7
excitatory tendencies resulting in relatively high cortical arousal and hypersensitivity to stimuli. On such a basis, high-arousal smokers would be expected to show lower extraversion (higher introversion) scores. As with extraversionintroversion, reducers are characterized as individuals who, due to a neurophysiological predisposition, decrease or dampen the perceived intensity of stimuli while augmenters are individuals who increase or amplify the perceived intensity of a given stimulus (Petrie, 1967; Barnes, 1976). Again, on this basis high-arousal smokers would be expected to exhibit lower reducing (higher augmenting) scores. The conflict arising between these theoretical predictions and present observations is somewhat tentatively reduced by early statements made by Eysenck (1963) concerning the relationship between drug action and personality. Citing evidence that nicotine acts as a stimulant drug, Eysenck referred to studies which suggested that for people who are high on neuroticism and extraversion, stimulant drugs have a calming effect and improve social behavior. On this argument, Eysenck predicted that high neuroticism should lead to indulgence in smoking only if coupled with extraversion. Though this latter contention is indirectly supported by Cherry and Kiernan's ( 1976) longitudinal study which found that those who became smokers are both more neurotic and more extraverted than those who do not, additional studies which employ direct measures of nervous system activity to examine low-high-arousal group differences and their reactions to smoking are required to clarify the relationship between personality, environment, and desire to smoke. As electroencephalographic measures have been used successfully in differentiating nonsmokers from smokers and smoking from smoking-deprived states (Knott & Venables, 1977, 1978; Knott, 1978), it would be a logical choice of measurement. Additional measurements might include autonomic and electromyographic activity. As nicotine exhibits an apparent biphasic action on the nervous system, i.e., low dose-stimulation, high dose-sedation (Armitage, Hall, & Morrison, 1968), future electrophysiological studies which examine the specific mechanism mediating tranquilization in high-arousal smokers and stimulation in lowarousal smokers would do well to incorporate measures of nicotine intake on the assumption that these smokers will differentially. adjust their doses to suit . the situation. Although indirect measures of nicotine intake are available, e.g., butt analysis and alveolar carbon monoxide levels, direct measures of plasma nicotine together with measures of puff number, duration, interval, strength, and depth would yield the most reliable picture. REFERENCES ADESSO,V., & GLAD, W. A behavioral test of a smoking typology. Addictive Behaviors, 1978, 3, 35-38.
428
V. J. KNOTT
ARMITAGE,A,, ,HALL,G., & MORRISON,C. Pharmacological basis for the tobacco smoklng hablt. Nature, 1968, 217, 331-334. BARNES,G. Individual differences in perceptual reactance: a review of the stimulus intensity modulation individual difference dimension. Canadian Psychological Review, 1976, 1, 29-52. BARNES,G., & FISHLINSKY,M. Stimulus intensity modulation, smoking and craving for cigarettes. Addictive Direares, 1976, 2, 479-484. BARTOL,C. Extraversion and neuroticism and nicotine, caffeine and drug intake. Psychological Reports, 1975, 36, 1007-1010. BEST, J., & HAKSTIAN,A. A situation-specific model for smoking behavior. Addictive Behaviors, 1978, 3, 79-92. CHERRY,N., & KIERNAN,K. Personality scores and smoking behavior. British Journal o f Preventive and Social Medicine, 1976, 30, 123-131. EYSENCK,H. Personality and cigarette smoking. Life Sciences, 1963, 3, 777-792. EYSENCK,H. T h e biological basis o f personality. Springfield, IL: Thomas, 1967. EYSENCK,H. Personality and the maintenance of the smoking habit. In W. Dunn (Ed.), Smoking behavior: motives and incentives. Washington: V. H. Winston, 1973. Pp. 136-147. EYSENCK,H., & EYSENCK,S. Eysenck Personality Inventory manual. San Diego, CA: Educational and Industrial Testing Service, 1968. Pp. 113-146. FRITH, C. Smoking behavior and its relation to the smoker's immediate experience. British Ioufnal o f Social and Clinical Psychology, 1971, 10, 73-78. KNOTT, V. Smoking, EEG and input regulation in smokers and non-smokers. In R. Thornton (Ed. ) , Smoking behavior: physiological and psychological influences. Edinburgh: Churchill Liv~ngsrone,1978. Pp. 115-130. KNOTT. V., & VENABLES,P. EEG alpha correlates of non-smokers, smokers, smoking and smoking-deprivation. Psychophysiology, 1977, 14, 150-156. KNOT, V., & VENABLES,P. Stimulus intens~tycontrol and the cortical evoked response in smokers and non-smokers. Psychophysrology, 1978, 15, 186-192. MCCRAE,R., COSTA,P., & BOSSB,R. Anxiety, extraversion and smoking. British Journal o f Social and Clinical Psychology, 1978, 17, 269-273. MCKENNELL,A. Smoking motivation factors. British Journal o f Social and Clinical Psychology, 1970, 9, 8-22. MYRSTEN,A., ANDERSSON, K., FRANKENHAEUSER, M., & ELGEROT,A. Immediate effects of cigarette smoking as related to different smoking habits. Perceptual and Motor Skills, 1975, 40, 515-523. PETRIE,A. individuality in pain and suffering. Chicago: Chicago Univer. Press, 1967. RUSSELL,M., PETO, J., & PATEL.U. The clmsificarion of smoking by factorial structure of motives. Journal o f the Royal .Statistical Society, 1974, 137, 313-333. TODD,G. Statistics o f smoking i n the United Kingdom. London: Tobacco Research Council, 1969. TOMKINS,S. A modified model of smoking behavior. In E. Borgatta & R. Evans (Eds.), Smoking, health and behavior. Chicago: Aldine, 1968. Pp. 165-186. VANDO, A. A personalicy theory related to pain tolerence. Unpublished doctoral dissertation, Columbia Univer., 1969. WARBURTON, D., & WESNES,K. Individual differences in smoking and attentional performance. In R. Thornton (Ed.). Smoking behavior: physiological and psychological inflrrences. Edinburgh: Churchill Livingstone, 1978. Pp. 19-43.
Accepted August 21, 1979.
PERSONALITY, AROUSAL A N D INDIVIDUAL DIFFERENCES I N CIGARETTE SMOKING1 VERNER J. KNOTT Royal Otiawa Hospital Summa7y.-This study attempted to examine physiologically based personality correlates of 'low-arousal smokers' and 'high-arousal smokers.' The former were smokers who generally experienced their strongest need to smoke in low-arousal situations, characterized by, e.g., monotony or boredom, while the latter experienced their strongest need to smoke in high-arousal situations, characterized by, e.g., anxiety or excitement. Members of each group completed the Eysenck Personalicy Inventory and the Vando Reduction-Augmentation Scale. High-arousal smokers exhibited significantly greater scores on extraversioon, neuroticism, and reducing dimensions.
McCrae, Costa, and Boss4 ( 1978) have recently questioned the productivity of studies which employ global personality dimensions to differentiate smokers from non-smokers and have forwarded the suggestion that personality research must become more specific in terms of more meaningful classification of smokers and in terms of more specific facets of personality. Although the validity of this approach is questionable (Adesso & Glad, 1978), a number of researchers have attempted classification on the basis of different motives or different occasions for smoking (Tomkins, 1968; McKennell, 1970; Frich, 1971; Russell, Peto, & Patel, 1974; Best & Hakstian, 1978). Typically, these investigations have employed complex multivariate statistical techniques, primarily factor analysis applied to self-report data, for the purpose of isolating factors which represent different smokers or different smoking types. Although these studies have identified a number of situational and motivational factors involved in smoking, the role of personality has been noticeably neglected in these studies. Recently, however, Myrsten, Anderson, Frankenhaeuser, and Elgerot ( 1975) have examined an arousal-based typology which may have particular relevance to physiologically based personalicy theories. Employing a questionnaire concerning interactions between the need to smoke and the external sicuation, Myrsten, et al. classified smokers as either 'low-arousal smokers' or 'higharousal smokers.' The former were smokers who generally experienced their strongest need to smoke in low-arousal situations, characteri~ed'b~, e.g., monbtony or boredom, while the latter experienced their strongest need to smoke in situations of high arousal, characterized by, e.g., anxiety or excitement. Examined under laboratory conditions designed to induce either low or high levels of lResearch was supported by the Canadian Tobacco Manufacturer's Council. Address requests for reprints to: Dr. Verner J. Knon, Specialty Clinics, Royal Ottawa Hospital, 1145 Carling Avenue, Ottawa, Ontario, Canada, K1Z 7K4.
V. J. KNOTT
arousal, the effects of smoking- differed characteristically for low-arousal smokers and high-arousal smokers. For low-arousal smokers, performance and general well-being were favourably affected by smoking in the low-arousal situation only. Conversely, performance and well-being of the high-arousal smokers were enhanced by smoking in the high-arousal situation only. Though Myrsten, et al. did not observe any significant differences between low- and high-arousal groups on physiological and performance variables in nonsmoking conditions, it is reasonable to suggest that examination of physiologically oriented personality measures such as Eysenck's (1967) introversionextraversion, neuroticism-stability and Petrie's ( 1967) augmenting-reducing dimensions would be particularly productive in elucidating differential motives underlying the smoking habits of low- and high-arousal groups. Studies employing this personality approach have been reported by Bartol (1975) and Barnes and Fishlinsky ( 1976). In an attempt to examine the relationship between personality and self-reported desire for cigarettes under stressful and non-stressful conditions, Bartol ( 1975 ) focused directly on Eysenck's hypothesis which stated that ". . . ( a ) introverts should more frequently use nicotine for tranquilizing purposes; and ( b ) people smoking under conditions of 'boredom' should use nicotine for stimulating purposes, people smoking under conditions of 'stress' should use nicotine for tranquilizing purposes" (Eysenck, 1973, pp. 113-146). Employing female smokers, Bartol found, contrary to Eysenck's prediction, that extraverts indicated they prefer to smoke in non-stressful situations. Although neurotics desired cigarettes more often than stable persons in both conditions, neurotic-extraverts tended to report the greatest desire to smoke in stressful situations. Barnes and Fishlinsky ( 1976) examined Petrie's ( 1967) hypothesis that reducers would smoke more cigarettes in low-arousal situations and that augmenters would smoke more cigarettes in high-arousal situations. Employing Vando's ( 1969) reducing-augmenting scale, the findings did not support the hypothesis and in fact, extreme female augmenters tended to crave cigarettes more during low-arousal than high-arousal. Though the results of the two studies run contrary to stated theories, and the relationship of the results to pharmacological actions of tobacco is suspect, i.e., females are less likely to inhale tobacco smoke than males (Todd, 1969), Petrie ( 1967) and Vando's ( 1969) studies have indicated that extraverts tend to be reducers while introverts tend to be augmenters, and as such the findings of Bartol and Barnes and Fishlinsky seem reliable and worthy of further investigation. The purpose of the present study was to attempt to replicate the above findings using the classification method of Myrsten, et al, and a sample of male smokers.
METHOD A questionnaire similar to that by Myrsten, et al. (1975) was employed
INDIVlDUAL DIFFERENCES IN CIGARETTE SMOKlNG
42 5
for selecting subjects with different smoking habits, i.e., subjects whose desire to smoke was greatest in either low-arousal or high-arousal situations. The questionnaire had 10 items referring to high-arousal situations, chosen from the questionnaires constructed by McKennell ( 1970) and Frith ( 1971). Subjects were instructed to imagine themselves in each of the situations and to indicate on a 7-point scale how strong their desire to smoke would be. Classification of smokers on the basis of the smoking questionnaire was accomplished by expressing each subject's score on low-arousal items and higharousal items as a percent of his total score. Subjects whose score on higharousal items was greater than 50% of the total score were classified as higharousal smokers and subjects whose score on low-arousal items was greater than 50% of the total score were classified as low-arousal smokers. One hundred male cigarette smokers were administered the smoking questionnaire along with Form A of the Eysenck Personality Inventory (Eysenck & Eysenck, 1968) and the Vando Reducing Augmenting Scale (Vando, 1969). The subjects were all inhalers with a mean age of 37.3 ( 4 9.9) yr., a mean daily consumption of 22.6 ( -t- 8.6) cigarettes and a mean smoking history of 14.0 ( 2 9.3) y r The mean nicotine and tar yields of the cigarettes consumed by the smokers were 1.15 ( + .29) and 16.49 (f3.88) milligrams, respectively. There were no statistically significant differences between low- and high-arousal groups.
RESULTS The mean raw score for high-arousal items was 22.7 ( + 8.3) and for lowarousal items 22.7 ( f 7.6). Means and standard deviations for extraversionintroversion, neuroticism-stability and reducing-augmenting dimensions were 12.74 ( f 8.03), 9.84 ( f 3.14), and 27.00 ( t 8.03), respectively. Among the 100 subjects, 48 achieved a high-arousal item score greater than 50% of their total score and were classified as high-arousal smokers. An additional 48 subjects achieving a low-arousal item score greater than 50% of their total score were classified as low-arousal smokers. Their mean scores on the relevant items are given in Table 1. There were no significant differences between the lowand high-arousal smokers in smoking consumption, smoking history, or age. Means and standard deviations for extraversion-introversion, neuroticismstability and reduction-augmentation dimensions for each of rhe rwo groups are TABLE 1 MEAN SCORESFORTWOGROUPS OF SMOKERS ON ITEMSREFERRING To HIGH-AROUSAL AND LOW-AROUSAL SITUATIONS(ns = 48) Items
High-arousal smokers
Low-arousal smokers
High-arousal Low-arousal
27.3 19.7
18.2 25.6
TABU 2
MEANSA N D STANDARD DEVIATIONS OF EXTRAVERSION-IN~OVBRSION, NEUROTICISMSTABILITY AND REDUCING-AUGMENTINGBY LOW- AND HIGH-AROUSAL SMOKERS(ns = 48) Dimensions Extraversion-Introversion Neuroticism-Stability Reducing-Augmenting
High-arousal smokers M SD 13.6 4.2 11.3 28.5
5.5 7.0
Low-arousal smokers M SD 11.9 4.2 9.2 5.5 25.7
7.0
shown in Table 2. Independent t tests (one-tailed) indicated that high-arousal smokers exhibited significantly greater scores on extraversion ( t = 1.84, df = 94, p < .05 ) , neuroticism ( t = 1.85, df = 94, p < .05 ), and reduction ( t = 2.16, df = 94, p < .025) than low-arousal smokers. DISCUSSION The present results supported the findings of Bart01 and Barnes and Fishlinsky in that low-arousal smokers exhibited lower extraversion (higher introversion) and lower reducing (higher augmenting) scores than high-arousal smokers. Also, high-arousal smokers exhibited higher neuroticism scores. This latter finding parallels a recent observation by Warburton and Wesnes (1978) who examined the relationship between neuroticism and various factors of the Smoking Motive Questionnaire by Russell, Peto, and Pate1 ( 1974). Here, smokers high on neuroticism were 'sedative' smokers in that they reported smoking when worried or angry, i.e., 'under stress.' This was supported by negative correlation obtained between neuroticism and the 'stimulation' factor, indicating that smokers high on neuroticism are less likely to smoke for stimulation. Of particular interest was Warburton and Wesnes' finding of a negative correlation between neuroticism and the 'pharmacological addiction' dimension. This finding that smokers who score high on neuroticism were less likely to be 'addicted' pharmacologically gives support to Myrsten, et al.'s ( 1975) hypothesis that pharmacological motives are predominant in low-arousal smokers and psychological motives are predominant in high-arousal smokers. As neurotic subjects tend to respond more strongly to stimuli, show greater variability in response, and take longer to recover after the response (Eysenck, 1967), it is logical to expect and observe higher neuroticism scores in high-arousal subjects. However, contrary to the neuroticism-stability dimension, the theoretical basis of extraversion-introversion and reducing-augmenting dimensions conflict with the present findings of higher extraversion and reducing scores in high-arousal smokers. .Relative to extraverts who are hypothesized as exhibiting strong inhibitory tendencies, low cortical arousal and hypo-sensitive responsivicy to stimuli, introverts are hypothesized to exhibit weak inhibitory and strong
INDlVlDUAL DIFFERENCES IN ClGARETI'E SMOKING
42 7
excitatory tendencies resulting in relatively high cortical arousal and hypersensitivity to stimuli. On such a basis, high-arousal smokers would be expected to show lower extraversion (higher introversion) scores. As with extraversionintroversion, reducers are characterized as individuals who, due to a neurophysiological predisposition, decrease or dampen the perceived intensity of stimuli while augmenters are individuals who increase or amplify the perceived intensity of a given stimulus (Petrie, 1967; Barnes, 1976). Again, on this basis high-arousal smokers would be expected to exhibit lower reducing (higher augmenting) scores. The conflict arising between these theoretical predictions and present observations is somewhat tentatively reduced by early statements made by Eysenck (1963) concerning the relationship between drug action and personality. Citing evidence that nicotine acts as a stimulant drug, Eysenck referred to studies which suggested that for people who are high on neuroticism and extraversion, stimulant drugs have a calming effect and improve social behavior. On this argument, Eysenck predicted that high neuroticism should lead to indulgence in smoking only if coupled with extraversion. Though this latter contention is indirectly supported by Cherry and Kiernan's ( 1976) longitudinal study which found that those who became smokers are both more neurotic and more extraverted than those who do not, additional studies which employ direct measures of nervous system activity to examine low-high-arousal group differences and their reactions to smoking are required to clarify the relationship between personality, environment, and desire to smoke. As electroencephalographic measures have been used successfully in differentiating nonsmokers from smokers and smoking from smoking-deprived states (Knott & Venables, 1977, 1978; Knott, 1978), it would be a logical choice of measurement. Additional measurements might include autonomic and electromyographic activity. As nicotine exhibits an apparent biphasic action on the nervous system, i.e., low dose-stimulation, high dose-sedation (Armitage, Hall, & Morrison, 1968), future electrophysiological studies which examine the specific mechanism mediating tranquilization in high-arousal smokers and stimulation in lowarousal smokers would do well to incorporate measures of nicotine intake on the assumption that these smokers will differentially. adjust their doses to suit . the situation. Although indirect measures of nicotine intake are available, e.g., butt analysis and alveolar carbon monoxide levels, direct measures of plasma nicotine together with measures of puff number, duration, interval, strength, and depth would yield the most reliable picture. REFERENCES ADESSO,V., & GLAD, W. A behavioral test of a smoking typology. Addictive Behaviors, 1978, 3, 35-38.
428
V. J. KNOTT
ARMITAGE,A,, ,HALL,G., & MORRISON,C. Pharmacological basis for the tobacco smoklng hablt. Nature, 1968, 217, 331-334. BARNES,G. Individual differences in perceptual reactance: a review of the stimulus intensity modulation individual difference dimension. Canadian Psychological Review, 1976, 1, 29-52. BARNES,G., & FISHLINSKY,M. Stimulus intensity modulation, smoking and craving for cigarettes. Addictive Direares, 1976, 2, 479-484. BARTOL,C. Extraversion and neuroticism and nicotine, caffeine and drug intake. Psychological Reports, 1975, 36, 1007-1010. BEST, J., & HAKSTIAN,A. A situation-specific model for smoking behavior. Addictive Behaviors, 1978, 3, 79-92. CHERRY,N., & KIERNAN,K. Personality scores and smoking behavior. British Journal o f Preventive and Social Medicine, 1976, 30, 123-131. EYSENCK,H. Personality and cigarette smoking. Life Sciences, 1963, 3, 777-792. EYSENCK,H. T h e biological basis o f personality. Springfield, IL: Thomas, 1967. EYSENCK,H. Personality and the maintenance of the smoking habit. In W. Dunn (Ed.), Smoking behavior: motives and incentives. Washington: V. H. Winston, 1973. Pp. 136-147. EYSENCK,H., & EYSENCK,S. Eysenck Personality Inventory manual. San Diego, CA: Educational and Industrial Testing Service, 1968. Pp. 113-146. FRITH, C. Smoking behavior and its relation to the smoker's immediate experience. British Ioufnal o f Social and Clinical Psychology, 1971, 10, 73-78. KNOTT, V. Smoking, EEG and input regulation in smokers and non-smokers. In R. Thornton (Ed. ) , Smoking behavior: physiological and psychological influences. Edinburgh: Churchill Liv~ngsrone,1978. Pp. 115-130. KNOTT. V., & VENABLES,P. EEG alpha correlates of non-smokers, smokers, smoking and smoking-deprivation. Psychophysiology, 1977, 14, 150-156. KNOT, V., & VENABLES,P. Stimulus intens~tycontrol and the cortical evoked response in smokers and non-smokers. Psychophysrology, 1978, 15, 186-192. MCCRAE,R., COSTA,P., & BOSSB,R. Anxiety, extraversion and smoking. British Journal o f Social and Clinical Psychology, 1978, 17, 269-273. MCKENNELL,A. Smoking motivation factors. British Journal o f Social and Clinical Psychology, 1970, 9, 8-22. MYRSTEN,A., ANDERSSON, K., FRANKENHAEUSER, M., & ELGEROT,A. Immediate effects of cigarette smoking as related to different smoking habits. Perceptual and Motor Skills, 1975, 40, 515-523. PETRIE,A. individuality in pain and suffering. Chicago: Chicago Univer. Press, 1967. RUSSELL,M., PETO, J., & PATEL.U. The clmsificarion of smoking by factorial structure of motives. Journal o f the Royal .Statistical Society, 1974, 137, 313-333. TODD,G. Statistics o f smoking i n the United Kingdom. London: Tobacco Research Council, 1969. TOMKINS,S. A modified model of smoking behavior. In E. Borgatta & R. Evans (Eds.), Smoking, health and behavior. Chicago: Aldine, 1968. Pp. 165-186. VANDO, A. A personalicy theory related to pain tolerence. Unpublished doctoral dissertation, Columbia Univer., 1969. WARBURTON, D., & WESNES,K. Individual differences in smoking and attentional performance. In R. Thornton (Ed.). Smoking behavior: physiological and psychological inflrrences. Edinburgh: Churchill Livingstone, 1978. Pp. 19-43.
Accepted August 21, 1979.
