- Email: [email protected]
Behavioral inhibition: relation to negative emotion regulation and reactivity
Personality and Individual Differences 36 (2004) 1235–1247 www.elsevier.com/locate/paid
Behavioral inhibition: relation to negative emotion regulation and reactivity Ellen W. Leen-Feldnera,*, Michael J. Zvolenskya, Matthew T. Feldnera, C.W. Lejuezb a
Department of Psychology, The University of Vermont, John Dewey Hall, Burlington, VT 05405-0134, USA b University of Maryland, College Park, USA Received 15 October 2001; received in revised form 12 March 2002; accepted 15 April 2002
Abstract The present experimental psychopathology study sought to address two interrelated theoretical predictions from behavioral inhibition theory and research among young adults. The first was whether individual differences in behavioral inhibition, as indexed by the Behavioral Inhibition Sensitivity (Carver & White, 1994) would relate to negative emotional reactivity elicited by a cognitive stressor. The second aim was to examine how individual differences in behavioral inhibition relate to rumination, a response style associated with prolonged periods of negative affect, particularly depression. Consistent with our hypotheses, behavioral inhibition, relative to other theoretically relevant variables (e.g. basal levels of negative affect), predicted cognitive-affective reactivity as well as a rumination response style. These findings are discussed in relation to understanding how behavioral inhibition is associated with prototypical indices of emotional distress, with implications for forwarding future work with specific types of emotional disorders. # 2002 Elsevier Ltd. All rights reserved. Keywords: Emotional reactivity; Emotion regulation; Behavioral inhibition sensitivity; Laboratory task
Behavioral inhibition is a temperament construct defined by motivational sensitivity to interoceptive and exteroceptive signals of punishment, unfamiliarity, and nonreward (Cloninger, 1987); it is characterized by withdrawal-oriented behavior, bodily agitation, and negatively valenced verbal and non-verbal expression (Davidson, Ekman, Saron, Senulis, & Friesen, 1990; Kagan, 1989). Due to the broad theoretical relevance of behavioral inhibition to negative emotional processes and states, this construct has direct implications for better understanding the * Corresponding author. Tel.: +1-802-656-3831; fax: +1-802-656-8783. E-mail addresses: [email protected] (E.W. Leen-Feldner), [email protected] (M.J. Zvolensky). 0191-8869/03/$ - see front matter# 2002 Elsevier Ltd. All rights reserved. doi:10.1016/S0191-8869(02)00113-7
1236
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
nature of emotional disorders (Arcus, 2001). Specifically, the behavioral inhibition system, comprised of the septohippocampal system, its monoaminergic afferents extending from the brainstem and neocortical projections in the frontal lobe, provides the motivational basis for the inhibition of behavior that may lead to negative outcomes, particularly in aversive or novel contexts (Fowles, 1993; Gray & McNaughton, 1996). As such, the behavioral inhibition system can initiate physiological processes and higher cortical functions, including cognitive-affective reactions to environmental challenges and opportunities. Researchers have studied behavioral inhibition in children (Harmon-Jones & Allen, 1997; Kagan, Reznick, & Snidman, 1988; Reznick et al., 1986; Sobotka, Davidson, & Senulis, 1992). These research studies indicate that a behavioral inhibition response style, as indexed by observable signs of fearfulness and wariness, predict negative emotional states, such as crying in response to novel stimuli, in future circumstances (Davidson & Fox, 1989; Kagan & Snidman, 1991). Behavioral inhibition is also overrepresented in first-order offspring of those with emotion-based psychopathology (Rosenbaum et al., 1990), suggesting these persons may be particularly likely to demonstrate individual susceptibility to negative emotional states. These findings hold promise in regard to forwarding research and theory on how temperamental qualities, an inherited profile of biobehavioral processes (Kagan & Snidman, 1999), relate to individual variation in negative emotional states during adulthood. However, at this juncture, very little experimental psychopathology work (Zvolensky, Lejuez, Stuart, & Curtin, 2001) has attempted to link behavioral inhibition to negative emotional processes among adults. This is unfortunate given that the behavioral inhibition system is theoretically important to vulnerability to both anxiety and depressive states as well as negative affect more generally (Fowles, 1993). Furthermore, although evidence attests to the physiological and behavioral components of the behavioral inhibition construct, markedly less work has been completed at a cognitive level of analysis. The absence of such cognitive-oriented work may impair the translation of behavioral inhibition theory and research to clinical contexts, wherein verbal (self-report) instruments are frequently employed. Carver and White (1994) have developed the Behavioral Inhibition/Activation Scale (BIS/BAS) to assess dispositional sensitivities to Gray’s two general motivational systems at a cognitive level of analysis. Early research using the BIS scale indicates that this verbal report instrument can detect varying sensitivities in the presumed motivational systems. For instance, scores on the BIS scale have been found to predict self-reported anxiety that occurs in response to a cold pressor task (Carver & White, 1994, Study 3). Gable, Reis, and Elliot (2000) similarly found BIS scores moderated affective responding, with higher scores on the scale associated with enhanced negative emotional reactions to naturally occurring stressful life events. Other research indicates that BIS/ BAS scores are predicted by prefrontal brain asymmetry, an electrophysiological index of affective variability, relative to self-reported emotional states (Sutton & Davidson, 1997). Zvolensky, Feldner, Eifert, and Stewart (2001) recently have found that among healthy adults, BIS scores predict self-rated cognitive and affective distress, but not autonomic arousal, during biological challenge. Finally, BIS scores have been found to predict differences in procedural learning under threatening contexts (Corr, Pickering, & Gray, 1997). Additionally, researchers using startle methodology have shown individual differences in trait anxiety affect startle response (Cook, 1999), although these findings have not been consistently replicated (e.g. Grillon, Ameli, Foot, & Davis, 1993). Collectively, the aforementioned findings generally indicate that behavioral inhibition may relate to the differential experience of negative emotional states. Although such studies
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1237
are indeed promising, it remains important and timely to understand if behavioral inhibition, at a cognitive level of analysis, relates to emotional reactivity and regulatory processes characteristic of prototypical aspects of emotion-based psychopathology (Davidson, 1992, 1998). Emotional reactivity reflects variability in the magnitude of an emotional response to evocative events (Tellegen, 1985). In an illustrative example, if two persons with equal levels of pre-experimental emotional distress were exposed to the same event, any observed differences in emotional functioning post-exposure is due to the change from pre- to post-task. Emotional reactivity is important to various anxiety and mood disorders, with greater levels of reactivity indexing greater individual susceptibility to negative emotional states (Larsen & Ketelaar, 1989; Zvolensky & Eifert, 2000). Specifically, persons with heightened emotional reactivity typically experience evocative events as beyond their control, becoming aroused, both physiologically and experientially, to stressful events (Melamed, 1987). If behavioral inhibition is functionally related to negative psychological outcomes, individual variation in BIS should naturally be associated with greater emotional reactivity to particular types of stressful events. Zvolensky, Feldner et al. (2001) found support for this hypothesis in regard to physical stress, yet it is unclear if these effects are content specific (e.g. limited to physical, as opposed to cognitive, stress). Given that many anxiety and mood disorders are characterized by emotional distress that occurs in response to cognitive stressors (e.g. Ruscio, Borkovec, & Ruscio, 2001), an important next research step in this line of inquiry is to test whether BIS relates to ‘‘on line’’ negative emotional reactivity during laboratory induced cognitive stress. Emotional regulation is another affect-relevant construct theoretically related to emotional distress. Gross (1998) has defined emotion regulation as ‘‘the process by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions’’ (p. 275). One clinically important type of emotion regulation is rumination, defined as the tendency to respond to negative emotional states by passively focusing on them (NolenHoeksema, 1991). For example, a person engaging in ruminative coping might attempt to understand the meaning of one’s distressed emotional state (e.g. ‘‘why is this happening?’’) rather than taking constructive action despite such experiential discomfort. Ruminative response styles characterize various depressive and anxiety states (e.g. Cox, Enns, & Taylor, 2001; Nolen-Hoeksema & Morrow, 1992) as well as prolong negative affect symptomatology more generally (Nolen-Hoeksema, Parker, & Larson, 1994). Given that behavioral inhibition is theoretically related to enhanced risk of emotion-based psychopathology (Gray, 1994), it would be expected that this clinically significant variable might be associated with established dysfunctional emotion regulatory styles such as rumination. That is, rather than linking behavioral inhibition simply to basal levels of negative emotional functioning (e.g. Gable et al., 2000), it is important to test whether this variable relates to prototypical emotion regulatory processes pivotal to disorders of emotion (Nolen-Hoeksema, 1991). The present study addressed two interrelated theoretical predictions from behavioral inhibition theory and research. The first was whether individual differences in behavioral inhibition, as indexed by the BIS (Carver & White, 1994), would relate to negative emotional reactivity elicited by a cognitive stressor. Existing research using the BIS has found that behavioral inhibition sensitivity is positively correlated with emotional distress (Sutton & Davidson, 1997; Zvolensky, Feldner et al., 2001), yet no studies have examined its predictive explanatory power in terms of emotional reactivity to a cognitive stressor. It was hypothesized that higher BIS scores would
1238
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
predict cognitive and physiological aspects of negative emotional reactivity. The second focus of the present investigation was to examine how individual differences in behavioral inhibition relate to rumination. It was hypothesized that BIS scores would significantly predict rumination relative to basal levels of negative affect and demographic characteristics. Overall, rather than begin to examine behavioral inhibition in a specific clinical population at this stage of research development, we sought to provide a more conservative test of this construct using a nonclinical population. In this way, we could be more confident that observed differences are not simply second-order consequences of having the disorder itself. The following specific hypotheses guided the present quasi-experimental investigation: 1. BIS scores will predict cognitive reactivity during the experimental task, as indexed by ratings of anxiety, frustration, and control on a visual analog scale, as well as the valence, arousal, and dominance dimensions of the Self Assessment Manikin (Lang, 1980). 2. BIS scores will predict physiological reactivity during the experimental task, as indexed by changes in heart rate and skin conductance from baseline. 3. Scores on the BIS will predict scores from a rumination measure, even after taking into account basal levels of negative affect and demographic characteristics.
1. Method 1.1. Participants Ninety participants (60 females) were recruited from Introductory Psychology courses at a Northeastern University in the United States. Participants were predominantly in their first year of college (M=13.6 years of education) and their mean age was 19.5 years. The ethnic distribution was 87% (n=78) Caucasian, 8% (n=7) African American, 4% (n=4) Asian, and 1% (n=1) Pacific Islander. 1.2. Measures 1.2.1. Pre-experimental psychological assessment The Behavioral Inhibition/Activation scale (BIS/BAS; Carver & White, 1994) is a 22-item selfreport instrument designed to assess BIS and BAS dispositional sensitivities. The scale produces four separate scores, one for BIS sensitivity (seven items), and three BAS subscales (i.e. reward responsiveness—five items, drive—four items, and fun seeking—four items). All scales are derived by summing responses for scale items (reverse scoring when appropriate). Factor analysis indicates the scale taps the hypothesized psychological dimensions (Carver & White, Study 1). Additionally, the BIS/BAS scale has demonstrated adequate internal consistency (e.g. BIS scale alpha: 0.74) and adequate convergent and discriminant validity (Carver & White, Study 2). In the present investigation, only the BIS scale was relevant to the study hypotheses and therefore utilized in the analyses. To assess basal levels of negative affect, we used the Multidimensional Personality Questionnaire (MPQ; Tellegen et al., 1988). The MPQ contains 11 primary scales related to positive
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1239
affect (Well-Being, Social Potency, Social Closeness, and Achievement), negative affect (Stress Reaction, Alienation, and Aggression), behavioral constraint (Constraint, Harm Avoidance, and Traditionalism), and absorption (Absorption). In the present study, we were concerned only with the negative affect super factor. The MPQ has strong internal consistency (alphas range from 0.76 to 0.89; Tellegen, 1985). Additionally, the negative affect super factor is correlated with assessment of current-mood negative affect (r=0.50), indicating adequate convergent validity. The negative affect super factor is essentially unrelated to the positive affect super factor, indicating adequate divergent validity (Tellegen, 1985). To assess rumination, we used the Responses to Situations Questionnaire (RSQ-short form; Nolen-Hoeksema, 1991). This is the shortened version of the original Response Styles Questionnaire developed by Nolen-Hoeksema and colleagues. The RSQ-short form includes items tapping the primary features of a ruminative emotion regulation style: focusing on one’s negative emotional state and thinking repetitively about the causes and consequences of that state (NolenHoeksema, 1991). According to Nolen-Hoeksema (2001, personal communication), the 10 items on the RSQ-short form were chosen from the 22 items on the original questionnaire. The RSQshort form is recommended over the longer version because the longer version includes items addressing ‘‘automatic negative thoughts’’ (Nolen-Hoeksema, 2001, personal communication). The RSQ-short form has high levels of internal consistency (alpha=0.87). Additionally, it is correlated with the Beck Depression Inventory (r=0.54) and Hamilton Rating Scale for Depression (r=0.44), indicating adequate convergent validity. 1.2.2. Experimental cognitive assessment To assess the cognitive response to the experimental task, we employed a series of single-item self-report questions, rated on a visual analog scale (VAS) ranging from 0 (none) to 100 (extreme), to assess moment-to-moment levels of (1) anxiety, (2) frustration, and (3) control. For example, a continuum ranging from 0 to 100 appeared on the screen underneath which was the word ‘anxiety.’ The same exact format was used for the other negative affect scales. Using a mouse to move the cursor along the continuum, participants rated their emotional experience online at three strategic points throughout the task. The distance participants moved the cursor toward 100 was automatically recorded, with larger distances denoting greater anxiety, frustration, and control. Previous research has established the utility of these scales in assessing core features of negative affect during stress-provoking conditions (e.g. Brown, Lejuez, Kahler, & Strong, 2002). A ‘‘global’’ negative affect composite score for the VAS ratings of anxiety, frustration, and control was computed by averaging the total pre- to post- change scores of these scales. 1.2.3. Experimental affective assessment To examine the affective reactivity, we employed the Self-Assessment Manikin (SAM; Lang, 1980). The SAM is a non-verbal pictorial assessment methodology using a human-like stimulus. According to factor analytic research, the SAM permits the rapid assessment of three fundamental dimensions of affective meaning in emotional experience—Valence (scale 1), Arousal (scale 2) and Dominance (scale 3; Lang, 1984). SAM ratings are made along five figures for each scale (Lang, Bradley, & Cuthbert, 1990). Participants select their current level of affective state along the dimensions of valence and arousal on a nine-point scale. Specifically, participants can
1240
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
place an ‘‘X’’ over any of the five figures in each scale, or between any two figures. Total scores range from zero to nine for each of the three dimensions. The SAM has sound psychometric properties and has previously been successfully employed to measure affective responses (Bradley & Lang, 1994; Zvolensky, Eifert, & Lejuez, 2001). Reactivity to the stressor was indexed using pre- to post- change scores along the valence and arousal domains. 1.2.4. Experimental physiological assessment A Coulbourn Modular recording system was used to assess heart rate responding on-line at a sample rate of 10 samples per second (sample error range:5 V). Channels were calibrated online prior to sampling. Heart rate was sampled in beats/per minute (bpm) using a digital Coulbourn tachometer fed through a S75–01 bioamplifier and assessed via Medi-Trace pre-gelled Ag/ AgCl electrodes. Heart rate placement followed standard bilateral positioning on either side of the participant’s rib cage. Skin conductance was sampled in microsiemens (ms) using a Coulbourn S7-23 isolated skin conductance coupler. Skin conductance placement was on the palmer surface of two fingers of the nondominant hand. Physiological reactivity was indexed via pre- to post-task heart rate and skin conductance change scores. 1.2.5. Paced auditory serial addition task (PASAT) As the cognitive stressor, we used a modified computer version of the paced auditory serial addition task (PASAT; see Lejuez, Kahler, & Brown, 2003). The PASAT was originally developed for the assessment of information processing capacity (Gronwall, 1977) and was used in the current study because it has been shown to produce elevated levels of negative affect (Deary et al., 1994; Holdwick & Wingenfeld, 1999; Lezak, 1995; Roman, Edwall, Buchanan, & Patton, 1991). The maximum duration of the PASAT was 20 min. During the task, numbers were sequentially flashed on the screen. Participants summed consecutive numbers in sets of two, adding only two numbers at a time [e.g. 4+3 (correct response=7)+6 (=9)+1 (=7)]. Participants used the computer’s mouse to click on the correct answer displayed on the computer screen in a format resembling a keyboard. After each incorrect response, an aversive auditory stimulus was presented. The first level of the PASAT provided a 3-s latency between number presentations (i.e. low difficulty). The second level provided a 1.5-s latency between number presentations (i.e. medium difficulty). The final level provided a 1-s latency between number presentations (i.e. high difficulty). The first level lasted for 3 min and the second level lasted for 5 min. After a 2-min rest period, during which selfreport forms were completed, the final level lasted up to 10 min. All participants were exposed to all three levels of difficulty (i.e. no between subjects manipulation), unless they elected to quit the task prior to completing level 3. 1.3. Procedure Participants were contacted via telephone and scheduled to come to the psychophysiological laboratory. After completing the consent form, participants completed the BIS, MPQ, and RSQ. Once these measures were completed, participants were seated in the experimental room, a 2-m6-m sound attenuated room. The room contained a chair, a desk with a Pentium microcomputer, SVGA color monitor, mouse, and keyboard. An intercom allowed participants to communicate freely with the experimenter in the adjacent room. Once seated, heart rate and skin
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1241
conductance recording electrodes were attached. The following directions were given to each participant: You are going to be completing a serial math task. You will see some numbers being presented in a series. We are going to ask you to add them in a special way. You will see a number presented at time 1. You will see another number presented at time 2. For the initial answer, add the numbers at time 1 and time 2 together. For all other answers, you will need to recall the number presented at time 2, and add it to a newly presented number (at time 3). Note that you have to REMEMBER the second number in the previous series and add it to the newly presented number. This phase of the task will last about 8 minutes. (This was practiced until the participant reported sufficient understanding of the directions.) After you complete phase 1, you will fill out some forms and proceed with phase 2 of the math task. In this phase, you will have an option to escape. As an incentive to try your best, we will award a $40 gift certificate for the mall to the person who scores the highest on this phase. Of course, we will not be able to award this money until the completion of data collection. We used a small reward to produce some incentive for continuing the task without creating a ceiling effect in the duration of endurance across participants. During the session, participants sat alone in the experimental room. The investigator was able to view participants from a one-way mirror and bi-directional communication was possible via an intercom. After completion of the math task, participants were debriefed.
2. Results 2.1. Descriptive data and zero-order relations between theoretically-relevant variables For comparison purposes, we first computed means and standard deviations for each of the variables (see Table 1).1 Next, to better understand how behavioral inhibition sensitivity relates to other theoretically relevant variables, we computed zero-order relations between each of the primary (psychological) predictor variables. The BIS related positively to the MPQ negative affect super factor (r=0.49, P<0.01) and the RSQ (r=0.40, P< 0.01), and negative affect super factor of the MPQ related positively to the RSQ (r=0.50, P<0.01). The three BAS subscales were negatively correlated (except heart rate change, which was positively correlated) with each of the criterion variables listed in Table 2, but, as would be expected, none of these relations were statistically significant. We then computed zero-order correlations between the predictor variables and each of the dependent variables to explore the relations among these theoretically relevant factors (see Table 2). BIS scores showed a significant positive relation with the valence dimension of the SAM as well as the VAS composite negative affect. As would be expected, the negative affect super 1 Given that gender correlates significantly with BIS, the interaction term for BIS and gender was examined for all multiple regression analyses. Inclusion of this term did not appreciably change the pattern of findings. For this reason, and because we did not have any a priori hypotheses about the role of gender, these analyses are neither presented nor discussed.
1242
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
Table 1 Means and standard deviations for predictor and criterion variables Variable
Mean or n
S.D. or %
Demographic/behavioral variables Gender (female/male) Age Education Level Ethnicity
60 19.53 48 78
66.7 1.94 53.3 86.7
Psychological variables BIS MPQ: negative affect super factor RSQ
20.6 89.7 20.5
3.7 16.2 6.5
Experimental assessments: pre- to post-change scores SAM: Arousal SAM: Valence VAS: Composite Heart Rate Skin Conductance
0.87 1.5 26.86 0.46 0.00
1.8 1.75 19.38 6.83 0.00
n=90; however only 63 participants were included in the heart rate and skin conductance assessments due to equipment failure. Participant gender (1=male, 2=female) and ethnicity (1=Caucasian, 2=African American, 3=Hispanic, 4=Asian, 5=Pacific Islander, 6=Other) were dummy coded. The n for subject gender reflects females, the n for ethnicity reflects Caucasian participants, and the n for education represents college freshmen. BIS: Behavioral Inhibition and Activation Scale (Carver & White, 1994). MPQ: Multidimensional Personality Questionnaire (Tellegen et al., 1988). RSQ: Responses to Situations Questionnaire (Nolen-Hoeksema, 1991).SAM: Self-Assessment Manikin (Lang, 1980); VAS: Visual Analog Scale (composite: average of the total pre- to post- change scores of the anxiety, frustration, and control ratings; Lejuez et al., in press).
Table 2 Zero-order relations between predictor and criterion variables SAM: Arousal
SAM: Valence
VAS: Composite
Heart rate
Skin conductance
RSQ
Demographic/behavioral variables Gender (female/male) 0.07 Age 0.01 Education Level 0.16 Ethnicity 0.10
0.26* 0.17 0.05 0.04
0.24* 0.09 0.09 0.07
0.11 0.18 .12 0.26*
0.05 0.20 0.14 0.08
0.22* 0.16 0.08 0.06
Psychological variables BIS MPQ
0.25* 0.08
0.35* 0.27*
0.06 0.10
0.13 0.02
0.40** 0.50**
0.04 0.19
n=90; however only 63 participants were included in the heart rate and skin conductance assessments due to equipment failure. For all of the criterion variables except the RSQ, pre-to post- change scores were used in all analyses. SAM: SelfAssessment Manikin (Lang, 1980); VAS: Visual Analog Scale (composite: average of the total pre- to post- change scores of the anxiety, frustration, and control ratings; Lejuez et al., in press). RSQ: Responses to Situations Questionnaire (NolenHoeksema, 1991). BIS: Behavioral Inhibition and Activation Scale (Carver & White, 1994). MPQ: Multidimensional Personality Questionnaire—negative affect superfactor (Tellegen et al., 1988). *P<0.05. **P<0.01.
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1243
factor of the MPQ was positively related to the ‘‘global’’ negative affect index during the stressor, as assessed by VAS ratings. None of the pre-experimental psychological variables were significantly related to heart rate or skin conductance, indicating a marked degree of response discordance. 2.2. Multiple regression analyses Hierarchical multiple regression analyses were conducted in order to test each of our primary hypotheses. Analyses were conducted separately for each of the dependent variables. Following the recommendations of Tabachnick and Fidell (1996, see p. 156), we utilized a stepwise approach in the model because it enters the variables into the equation according to the strength of their association with each primary dependent variable. This approach was most consistent with our a priori research objectives of determining the relative predictive power of behavioral inhibition. Each variable that met criteria for inclusion (i.e. P to enter=0.05) and did not meet criteria for removal (i.e. P to remove=0.10) was retained. Semi-partial correlations (sr2) were included to represent effect sizes only for those variables retained in the equation. This index of effect size represents the proportion of total criterion variance for variable Y uniquely accounted for by predictor variable X (Cohen, 1988). 2.2.1. Rumination Predictor variables for rumination included the main effects of gender, age, education, and ethnicity, MPQ ratings, and BIS ratings. Overall, the multiple regression analysis indicated that the predictor variables together explained 26% of the variance in self-reported rumination, F (2, 85)=14.7, P<0.01 (adjusted R2=24%). Both the main effects of negative affect as assessed by the MPQ (b=0.35; sr2=0.10, P<0.01) and BIS (=0.23 sr2=0.04, P<0.05) contributed in a statistically significant manner to the equation. 2.2.2. Emotional reactivity Predictor variables for emotional reactivity included the main effects of age, gender, education level, ethnicity, and BIS scores. Regression analyses predicting the VAS composite indicated that the predictor variables together explained 13% of the variance, F (1, 86)=12.3, P< 0.01 (adjusted R2=12%). As hypothesized, the only variable that met inclusion criteria at any level on the model was the main effect of BIS (b=0.35; sr2=0.12, P< 0.01). For the SAM ratings of emotional valence, the predictor variables together explained 6% of the variance, F (1, 86)=5.8, p < 0.05 (adjusted R2=5%). As hypothesized, the only variable that met inclusion criteria was BIS (=0.25; sr2=0.06, P< 0.05). In terms of the SAM ratings of emotional arousal and dominance, none of the predictors explained a significant amount of variance in the criterion measures. Prior to data analysis, physiological data were screened for outliers due to sampling error (e.g. participant movement). For analyses, the baseline and the mean heart rate during level two were utilized. The baseline reading (i.e. mean of samples taken every 20 s for 2 min) was taken after participants had completed the experimental task in order to control for preexperimental anxiety. With regard to heart rate, the predictor variables together explained 8% of the variance, F (1, 60)=4.9, P<0.05 (adjusted R2=6%). The only variable that met inclusion criteria was ethnicity
1244
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
(=0.28; sr2=0.08, P<0.05). Although we examined skin conductance in the same manner as the other dependent measures, no significant predictors were found.
3. Discussion Contemporary theories of behavioral inhibition suggest that this temperamental variable should relate to a variety of negative emotional outcomes, including high degrees of emotional reactivity and dysfunctional regulatory styles (Gray, 1994). There has been convincing support for this perspective from neuorophysiological (Davidson & Fox, 1989) and behavioral (Kagan et al., 1988) investigations. Yet, there is a relative absence of investigations addressing behavioral inhibition at a cognitive level of analysis. This lack of research stymies efforts to apply behavioral inhibition research and theory to many disorders of emotion. To address the predictive power of behavioral inhibition in terms of negative emotional responding, the present investigation tested two primary hypotheses concerning emotional reactivity and regulation. In terms of emotional reactivity, the present results indicate that, as hypothesized, levels of behavioral inhibition predicted negative affect and negatively valenced ratings of distress during a cognitive stressor. Although these effects were small, it was the only variable in our prediction equations to add a significant amount of variance when predicting post-challenge ratings of affective distress. At the same time, behavioral inhibition did not significantly predict indices of physiological arousal, indexed by self-rated or direct on-line measurement. Such response discordance provides further evidence that behavioral inhibition may relate more to cognitive than physiological aspects of emotional reactivity (see also, Zvolensky, Feldner et al., 2001). Alternatively, it is possible these findings are at least partially a product of the stressor utilized in the current investigation; that is, the PASAT may not have elicited high enough degrees of somatic arousal (i.e. restricted range) to be captured by behavioral inhibition. In all cases, the finding that behavioral inhibition is associated with emotional reactivity at the cognitive level of analysis is important, and specifically illustrates one type of affective variability process linked to this tempermantal construct. Given these findings, psychopathologists are now in a good position to extend this research in an effort to examine how behavioral inhibition confers increased risk for emotional distress. For example, future research could extend the current findings by examining emotional reactivity as it relates to particular cognitive processes (e.g. stroop task). The present study also found that individual variation in behavioral inhibition, along with basal levels of negative affect, significantly predicts a rumination-oriented response style. This finding provides the first evidence that in addition to being associated with negative emotional reactivity, behavioral inhibition is related to dysfunctional styles of regulating emotional distress. Again, these data may highlight some of the specific response styles associated with the negative emotional effects of behavioral inhibition. Although the correlational nature of this particular finding precludes understanding directional or causative roles, considering other research in this domain (Harmon-Jones & Allen, 1997), it may be that behavioral inhibition promotes dysfunctional types of learning experiences. For example, drawing from developmental research (Kagan et al., 1988), it is possible that high levels of behavioral inhibition may increase the likelihood of the rigid application of avoidance-oriented coping strategies for managing emotionally evocative life events. In the absence of corrective learning, such individuals may be less apt to take
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1245
constructive action during negative emotional states, resulting in more frequent, and longer periods of, negative affect (Nolen-Hoeksema et al., 1994). Based upon these preliminary findings, researchers have a sound empirical basis upon which to conduct experimental tests of this and related hypotheses. There are a number of interpretative caveats and directions for future research that warrant consideration. First, all of the psychological variables were assessed using self-report instruments. Accordingly, it is possible the observed findings were, at least in part, due to shared method variance. Although such a methodological strategy was useful at this stage of research development, it likely will be useful to incorporate experimental cognitive science methodologies into future research in this area, particularly those that tap automatic types of processing (e.g. interpretative biases for threat). Second, as with all quasi-experimental designs, causal relations cannot be unambiguously inferred, leaving the results open to a number of interpretations. One way to better address this issue would be to provide a prospective assessment of how behavioral inhibition relates to emotional distress in the natural environment. Finally, although we intentionally designed the study to examine psychologically healthy individuals, it now makes sense to extend these findings to persons with specific types of emotion-based psychopathology. In this way, researchers can determine to what extent behavioral inhibition is related to specific emotional disorders. In summary, the results of the present study help to identify how theoretically relevant variables produce heightened levels of emotional distress. Our results indicate that at a cognitive level of analysis, individual differences in behavioral inhibition predict differential aspects of emotional reactivity to cognitive stress and a rumination response style. These results converge and uniquely extend the existing evidence that behavioral inhibition cognitive processes are important pre-morbid predictors of negative responding.
Acknowledgements This study was submitted in partial fulfillment of the requirements of Ellen W. Leen’s preliminary examination at The University of Vermont. References Arcus, D. (2001). Inhibited and uninhibited children: Biology in the social context. In T. D. Wachs, & G. A. Kohnstamm (Eds.), Temperament in context (pp. 43–60). New Jersey: Laurence Erlbaum Associates. Bradley, M. M., & Lang, P. J. (1994). Measuring emotion: The Self-Assessment Manikin and the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry, 25, 49–59. Brown, R. A., Lejuez, C. W., Kahler, C. W., & Strong, D. R. (2002). Distress tolerance and duration of past smoking cessation attempts. Journal of Abnormal Psychology, 111, 180–185. Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: the BIS/BAS scales. Journal of Personality and Social Psychology, 67, 319–333. Cloninger, C. R. (1987). A systematic method for clinical description and classification of personality variants. Archives of General Psychiatry, 44, 573–588. Cohen, J. (1988). Statistical power analysis (2nd ed.). Hillsdale, NJ: Erlbaum. Cook, Edwin W. III. (1999). Affective individual differences, psychopathology, and startle reflex modification. In
1246
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
M. E. Dawson, & A. M. Schell (Eds.), Startle modification: implications for neuroscience, cognitive science, and clinical science (pp. 187–208). New York, NY: Cambridge University Press. Corr, P. J., Pickering, A. D., & Gray, J. A. (1997). Personality, punishment, and procedural learning: a test of J. A. Gray’s anxiety theory. Journal of Personality and Social Psychology, 73, 337–344. Cox, B. J., Enns, M. W., & Taylor, S. (2001). The effect of rumination as a mediator of elevated anxiety sensitivity in major depression. Cognitive Therapy and Research, 25, 525–534. Davidson, R. J. (1992). Emotion and affective style: hemispheric substrates. Psychological Science, 3, 39–43. Davidson, R. J. (1998). Affective style and affective disorders: Perspectives from affective neuroscience. Cognition and Emotion, 12, 307–320. Davidson, R. J., Ekman, P., Saron, C. D., Senulis, J. A., & Friesen, W. V. (1990). Approach-withdrawal and cerebral asymmetry: emotional expression and brain physiology. Journal of Personality and Social Psychology, 58, 330–341. Davidson, R. J., & Fox, N. A. (1989). Frontal brain asymmetry predicts infants’ response to maternal separation. Journal of Abnormal Psychology, 98, 127–131. Deary, I. J., Ebmeier, K. P., MacLeod, K. M., Dougall, N., Hepburn, Frier, & Goodwin (1994). PASAT performance and the pattern of uptake of super (99m) Tc-exametazime in brain estimated with single photon emission tomography. Biological Psychology, 38, 1–18. Fowles, D. C. (1993). A motivational theory of psychopathology. In W. Spaulding (Ed.), Nebraska symposium on motivation: integrated views of motivation, cognition and emotion (Vol. 41) (pp. 181–238). Lincoln, NE: University of Nebraska Press. Gable, S. L., Reis, H. T., & Elliot, A. J. (2000). Behavioral activation and inhibition in everyday life. Journal of Personality and Social Psychology, 78, 1135–1149. Gray, J. A. (1994). Framework for a taxonomy of psychiatric disorder. In S. H. M. Van Goozen, N. E. Van de Poll, & J. A. Sergeant (Eds.), Emotions: essays on emotion theory (pp. 29–59). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Gray, J. A., & McNaughton, N. (1996). The neuropsychology of anxiety: A reprise. In D. A. Hope (Ed.), Nebraska symposium on motivation: perspectives on anxiety, panic, and fear (Vol. 43) (pp. 61–134). Lincoln: University of Nebraska Press. Grillon, C., Ameli, R., Foot, M., & Davis, M. (1993). Fear-potentiated startle: relationship to the level of state/trait anxiety in healthy subjects. Biological Psychiatry, 33, 566–574. Gronwall, D. M. (1977). Paced auditory serial-addition task: a measure of recovery from concussion. Perceptual Motor Skills, 44, 367–373. Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2, 271–299. Harmon-Jones, E., & Allen, J. J. (1997). Behavioral activation sensitivity and resting frontal EEG asymmetry: covariation of putative indicators related to risk for mood disorders. Journal of Abnormal Psychology, 106, 159–163. Holdwick, D. J., & Wingenfeld, S. A. (1999). The subjective experience of PASAT testing: does the PASAT induce negative mood?. Archives of Clinical Neuropsychology, 14, 273–284. Kagan, J. (1989). Temperamental contributions to social behavior. American Psychologist, 44, 668–674. Kagan, J., Reznick, J. S., & Snidman, N. (1988). Biological bases of childhood shyness. Science, 240, 167–171. Kagan, J., & Snidman, N. (1991). Temperamental factors in human development. American Psychologist, 46, 856–862. Kagan, J., & Snidman, N. (1999). Early childhood predictors of adult anxiety disorders. Biological Psychiatry, 46, 1536–1541. Lang, P. J. (1980). Behavioral treatment and bio-behavioral assessment: computer applications. In J. B. Sidowski, J. H. Johnson, & T. A. Williams (Eds.), Technology in the health care delivery system (pp. 119–137). Norwood, NJ: Ablex. Lang, P. J. (1984). Cognition in emotion: concept and action. In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotions, cognitions, and behavior (pp. 192–228). New York: Cambridge. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex. Psychological Review, 97, 377–395. Larsen, R. J., & Ketelaar, T. (1989). Extraversion neuroticism, and susceptibility to positive and negative mood induction procedures. Personality and Individual Differences, 10, 1221–1228.
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1247
Lejuez, C. W., Kahler, C. W., & Brown, R. A. (2003). A modified version of the Paced Auditory Serial Addition Task (PASAT), as a laboratory-based stressor. Behavior Therapist, 26, 290–293. Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York, NY: Oxford University Press. Melamed, S. (1987). Emotional reactivity and elevated blood pressure. Psychosomatic Medicine, 49, 217–225. Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 100, 569–582. Nolen-Hoeksema, S., & Morrow, J. (1992). A prospective study of depression and posttraumatic stress symptoms after a natural disaster: the 1989 Loma Prieta earthquake. Journal of Personality and Social Psychology, 61, 115–121. Nolen-Hoeksema, S., Parker, L. E., & Larson, J. (1994). Ruminative coping with depressed mood following loss. Journal of Personality and Social Psychology, 67, 92–104. Reznick, J. S., Kagan, J., Snidman, N., & Gersten, M. (1986). Baak, & Rosenberg, A. Inhibited and uninhibited children: a follow-up study. Child Development, 57, 660–680. Roman, D. D., Edwall, G. E., Buchanan, R. J., & Patton, J. H. (1991). Extended norms for the paced auditory serial addition task. Clinical Neuropsychologist, 5, 33–40. Rosenbaum, J. F., Biederman, J., Gersten, M., Hirshfeld, D. R., Meminger, S. R., Herman, J. B., Kagan, J., Reznick, S., & Snidman, N. (1990). Behavioral inhibition in children of parents with panic disorder and agoraphobia: a controlled study. In S. Chess, & M. E. Hertzig (Eds.), Annual progress in child psychiatry and child development (pp. 294– 315). Philadelphia, PA: Brunner/Mazel. Ruscio, A. M., Borkovec, T. D., & Ruscio, J. (2001). A taxometric investigation of the latent structure of worry. Journal of Abnormal Psychology, 110, 413–422. Sobotka, S. S., Davidson, R. J., & Senulis, J. A. (1992). Anterior brain electrical asymmetries in response to reward and punishment. Electroencephalography and Clinical Neurophysiology, 83, 236–247. Sutton, S. K., & Davidson, R. J. (1997). Prefrontal brain asymmetry: a biological substrate of the behavioral approach and inhibition systems. Psychological Science, 8, 204–210. Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). Northridge, CA: Harper Collins. Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In A. H. Tuma, & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Tellegen, A., Lykken, D. T., Bouchard, T. J., Wilcox, K. J., Segal, N. L., & Rich, S. (1988). Personality similarity in twins reared apart and together. Journal of Personality and Social Psychology, 54, 1031–1039. Zvolensky, M. J., & Eifert, G. H. (2000). A review of psychological factors/processes affecting anxious responding during voluntary hyperventilation and inhalations of carbon dioxide-enriched air. Clinical Psychology Review, 21, 375–400. Zvolensky, M. J., Eifert, G. H., & Lejuez, C. W. (2001). Emotional control during recurrent 20% carbon dioxideenriched air induction: relation to individual difference variables. Emotion, 1, 148–165. Zvolensky, M. J., Feldner, M. T., Eifert, G. H., & Stewart, S. H. (2001). Evaluating differential predictions of anxietyrelated reactivity during repeated 20% carbon-dioxide-enriched air challenge. Cognition and Emotion, 15, 767–786. Zvolensky, M. J., Lejuez, C. W., Stuart, G. L., & Curtin, J. J. (2001). Experimental psychopathology in psychological science. Review of General Psychology, 5, 371–381.
Behavioral inhibition: relation to negative emotion regulation and reactivity Ellen W. Leen-Feldnera,*, Michael J. Zvolenskya, Matthew T. Feldnera, C.W. Lejuezb a
Department of Psychology, The University of Vermont, John Dewey Hall, Burlington, VT 05405-0134, USA b University of Maryland, College Park, USA Received 15 October 2001; received in revised form 12 March 2002; accepted 15 April 2002
Abstract The present experimental psychopathology study sought to address two interrelated theoretical predictions from behavioral inhibition theory and research among young adults. The first was whether individual differences in behavioral inhibition, as indexed by the Behavioral Inhibition Sensitivity (Carver & White, 1994) would relate to negative emotional reactivity elicited by a cognitive stressor. The second aim was to examine how individual differences in behavioral inhibition relate to rumination, a response style associated with prolonged periods of negative affect, particularly depression. Consistent with our hypotheses, behavioral inhibition, relative to other theoretically relevant variables (e.g. basal levels of negative affect), predicted cognitive-affective reactivity as well as a rumination response style. These findings are discussed in relation to understanding how behavioral inhibition is associated with prototypical indices of emotional distress, with implications for forwarding future work with specific types of emotional disorders. # 2002 Elsevier Ltd. All rights reserved. Keywords: Emotional reactivity; Emotion regulation; Behavioral inhibition sensitivity; Laboratory task
Behavioral inhibition is a temperament construct defined by motivational sensitivity to interoceptive and exteroceptive signals of punishment, unfamiliarity, and nonreward (Cloninger, 1987); it is characterized by withdrawal-oriented behavior, bodily agitation, and negatively valenced verbal and non-verbal expression (Davidson, Ekman, Saron, Senulis, & Friesen, 1990; Kagan, 1989). Due to the broad theoretical relevance of behavioral inhibition to negative emotional processes and states, this construct has direct implications for better understanding the * Corresponding author. Tel.: +1-802-656-3831; fax: +1-802-656-8783. E-mail addresses: [email protected] (E.W. Leen-Feldner), [email protected] (M.J. Zvolensky). 0191-8869/03/$ - see front matter# 2002 Elsevier Ltd. All rights reserved. doi:10.1016/S0191-8869(02)00113-7
1236
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
nature of emotional disorders (Arcus, 2001). Specifically, the behavioral inhibition system, comprised of the septohippocampal system, its monoaminergic afferents extending from the brainstem and neocortical projections in the frontal lobe, provides the motivational basis for the inhibition of behavior that may lead to negative outcomes, particularly in aversive or novel contexts (Fowles, 1993; Gray & McNaughton, 1996). As such, the behavioral inhibition system can initiate physiological processes and higher cortical functions, including cognitive-affective reactions to environmental challenges and opportunities. Researchers have studied behavioral inhibition in children (Harmon-Jones & Allen, 1997; Kagan, Reznick, & Snidman, 1988; Reznick et al., 1986; Sobotka, Davidson, & Senulis, 1992). These research studies indicate that a behavioral inhibition response style, as indexed by observable signs of fearfulness and wariness, predict negative emotional states, such as crying in response to novel stimuli, in future circumstances (Davidson & Fox, 1989; Kagan & Snidman, 1991). Behavioral inhibition is also overrepresented in first-order offspring of those with emotion-based psychopathology (Rosenbaum et al., 1990), suggesting these persons may be particularly likely to demonstrate individual susceptibility to negative emotional states. These findings hold promise in regard to forwarding research and theory on how temperamental qualities, an inherited profile of biobehavioral processes (Kagan & Snidman, 1999), relate to individual variation in negative emotional states during adulthood. However, at this juncture, very little experimental psychopathology work (Zvolensky, Lejuez, Stuart, & Curtin, 2001) has attempted to link behavioral inhibition to negative emotional processes among adults. This is unfortunate given that the behavioral inhibition system is theoretically important to vulnerability to both anxiety and depressive states as well as negative affect more generally (Fowles, 1993). Furthermore, although evidence attests to the physiological and behavioral components of the behavioral inhibition construct, markedly less work has been completed at a cognitive level of analysis. The absence of such cognitive-oriented work may impair the translation of behavioral inhibition theory and research to clinical contexts, wherein verbal (self-report) instruments are frequently employed. Carver and White (1994) have developed the Behavioral Inhibition/Activation Scale (BIS/BAS) to assess dispositional sensitivities to Gray’s two general motivational systems at a cognitive level of analysis. Early research using the BIS scale indicates that this verbal report instrument can detect varying sensitivities in the presumed motivational systems. For instance, scores on the BIS scale have been found to predict self-reported anxiety that occurs in response to a cold pressor task (Carver & White, 1994, Study 3). Gable, Reis, and Elliot (2000) similarly found BIS scores moderated affective responding, with higher scores on the scale associated with enhanced negative emotional reactions to naturally occurring stressful life events. Other research indicates that BIS/ BAS scores are predicted by prefrontal brain asymmetry, an electrophysiological index of affective variability, relative to self-reported emotional states (Sutton & Davidson, 1997). Zvolensky, Feldner, Eifert, and Stewart (2001) recently have found that among healthy adults, BIS scores predict self-rated cognitive and affective distress, but not autonomic arousal, during biological challenge. Finally, BIS scores have been found to predict differences in procedural learning under threatening contexts (Corr, Pickering, & Gray, 1997). Additionally, researchers using startle methodology have shown individual differences in trait anxiety affect startle response (Cook, 1999), although these findings have not been consistently replicated (e.g. Grillon, Ameli, Foot, & Davis, 1993). Collectively, the aforementioned findings generally indicate that behavioral inhibition may relate to the differential experience of negative emotional states. Although such studies
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1237
are indeed promising, it remains important and timely to understand if behavioral inhibition, at a cognitive level of analysis, relates to emotional reactivity and regulatory processes characteristic of prototypical aspects of emotion-based psychopathology (Davidson, 1992, 1998). Emotional reactivity reflects variability in the magnitude of an emotional response to evocative events (Tellegen, 1985). In an illustrative example, if two persons with equal levels of pre-experimental emotional distress were exposed to the same event, any observed differences in emotional functioning post-exposure is due to the change from pre- to post-task. Emotional reactivity is important to various anxiety and mood disorders, with greater levels of reactivity indexing greater individual susceptibility to negative emotional states (Larsen & Ketelaar, 1989; Zvolensky & Eifert, 2000). Specifically, persons with heightened emotional reactivity typically experience evocative events as beyond their control, becoming aroused, both physiologically and experientially, to stressful events (Melamed, 1987). If behavioral inhibition is functionally related to negative psychological outcomes, individual variation in BIS should naturally be associated with greater emotional reactivity to particular types of stressful events. Zvolensky, Feldner et al. (2001) found support for this hypothesis in regard to physical stress, yet it is unclear if these effects are content specific (e.g. limited to physical, as opposed to cognitive, stress). Given that many anxiety and mood disorders are characterized by emotional distress that occurs in response to cognitive stressors (e.g. Ruscio, Borkovec, & Ruscio, 2001), an important next research step in this line of inquiry is to test whether BIS relates to ‘‘on line’’ negative emotional reactivity during laboratory induced cognitive stress. Emotional regulation is another affect-relevant construct theoretically related to emotional distress. Gross (1998) has defined emotion regulation as ‘‘the process by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions’’ (p. 275). One clinically important type of emotion regulation is rumination, defined as the tendency to respond to negative emotional states by passively focusing on them (NolenHoeksema, 1991). For example, a person engaging in ruminative coping might attempt to understand the meaning of one’s distressed emotional state (e.g. ‘‘why is this happening?’’) rather than taking constructive action despite such experiential discomfort. Ruminative response styles characterize various depressive and anxiety states (e.g. Cox, Enns, & Taylor, 2001; Nolen-Hoeksema & Morrow, 1992) as well as prolong negative affect symptomatology more generally (Nolen-Hoeksema, Parker, & Larson, 1994). Given that behavioral inhibition is theoretically related to enhanced risk of emotion-based psychopathology (Gray, 1994), it would be expected that this clinically significant variable might be associated with established dysfunctional emotion regulatory styles such as rumination. That is, rather than linking behavioral inhibition simply to basal levels of negative emotional functioning (e.g. Gable et al., 2000), it is important to test whether this variable relates to prototypical emotion regulatory processes pivotal to disorders of emotion (Nolen-Hoeksema, 1991). The present study addressed two interrelated theoretical predictions from behavioral inhibition theory and research. The first was whether individual differences in behavioral inhibition, as indexed by the BIS (Carver & White, 1994), would relate to negative emotional reactivity elicited by a cognitive stressor. Existing research using the BIS has found that behavioral inhibition sensitivity is positively correlated with emotional distress (Sutton & Davidson, 1997; Zvolensky, Feldner et al., 2001), yet no studies have examined its predictive explanatory power in terms of emotional reactivity to a cognitive stressor. It was hypothesized that higher BIS scores would
1238
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
predict cognitive and physiological aspects of negative emotional reactivity. The second focus of the present investigation was to examine how individual differences in behavioral inhibition relate to rumination. It was hypothesized that BIS scores would significantly predict rumination relative to basal levels of negative affect and demographic characteristics. Overall, rather than begin to examine behavioral inhibition in a specific clinical population at this stage of research development, we sought to provide a more conservative test of this construct using a nonclinical population. In this way, we could be more confident that observed differences are not simply second-order consequences of having the disorder itself. The following specific hypotheses guided the present quasi-experimental investigation: 1. BIS scores will predict cognitive reactivity during the experimental task, as indexed by ratings of anxiety, frustration, and control on a visual analog scale, as well as the valence, arousal, and dominance dimensions of the Self Assessment Manikin (Lang, 1980). 2. BIS scores will predict physiological reactivity during the experimental task, as indexed by changes in heart rate and skin conductance from baseline. 3. Scores on the BIS will predict scores from a rumination measure, even after taking into account basal levels of negative affect and demographic characteristics.
1. Method 1.1. Participants Ninety participants (60 females) were recruited from Introductory Psychology courses at a Northeastern University in the United States. Participants were predominantly in their first year of college (M=13.6 years of education) and their mean age was 19.5 years. The ethnic distribution was 87% (n=78) Caucasian, 8% (n=7) African American, 4% (n=4) Asian, and 1% (n=1) Pacific Islander. 1.2. Measures 1.2.1. Pre-experimental psychological assessment The Behavioral Inhibition/Activation scale (BIS/BAS; Carver & White, 1994) is a 22-item selfreport instrument designed to assess BIS and BAS dispositional sensitivities. The scale produces four separate scores, one for BIS sensitivity (seven items), and three BAS subscales (i.e. reward responsiveness—five items, drive—four items, and fun seeking—four items). All scales are derived by summing responses for scale items (reverse scoring when appropriate). Factor analysis indicates the scale taps the hypothesized psychological dimensions (Carver & White, Study 1). Additionally, the BIS/BAS scale has demonstrated adequate internal consistency (e.g. BIS scale alpha: 0.74) and adequate convergent and discriminant validity (Carver & White, Study 2). In the present investigation, only the BIS scale was relevant to the study hypotheses and therefore utilized in the analyses. To assess basal levels of negative affect, we used the Multidimensional Personality Questionnaire (MPQ; Tellegen et al., 1988). The MPQ contains 11 primary scales related to positive
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1239
affect (Well-Being, Social Potency, Social Closeness, and Achievement), negative affect (Stress Reaction, Alienation, and Aggression), behavioral constraint (Constraint, Harm Avoidance, and Traditionalism), and absorption (Absorption). In the present study, we were concerned only with the negative affect super factor. The MPQ has strong internal consistency (alphas range from 0.76 to 0.89; Tellegen, 1985). Additionally, the negative affect super factor is correlated with assessment of current-mood negative affect (r=0.50), indicating adequate convergent validity. The negative affect super factor is essentially unrelated to the positive affect super factor, indicating adequate divergent validity (Tellegen, 1985). To assess rumination, we used the Responses to Situations Questionnaire (RSQ-short form; Nolen-Hoeksema, 1991). This is the shortened version of the original Response Styles Questionnaire developed by Nolen-Hoeksema and colleagues. The RSQ-short form includes items tapping the primary features of a ruminative emotion regulation style: focusing on one’s negative emotional state and thinking repetitively about the causes and consequences of that state (NolenHoeksema, 1991). According to Nolen-Hoeksema (2001, personal communication), the 10 items on the RSQ-short form were chosen from the 22 items on the original questionnaire. The RSQshort form is recommended over the longer version because the longer version includes items addressing ‘‘automatic negative thoughts’’ (Nolen-Hoeksema, 2001, personal communication). The RSQ-short form has high levels of internal consistency (alpha=0.87). Additionally, it is correlated with the Beck Depression Inventory (r=0.54) and Hamilton Rating Scale for Depression (r=0.44), indicating adequate convergent validity. 1.2.2. Experimental cognitive assessment To assess the cognitive response to the experimental task, we employed a series of single-item self-report questions, rated on a visual analog scale (VAS) ranging from 0 (none) to 100 (extreme), to assess moment-to-moment levels of (1) anxiety, (2) frustration, and (3) control. For example, a continuum ranging from 0 to 100 appeared on the screen underneath which was the word ‘anxiety.’ The same exact format was used for the other negative affect scales. Using a mouse to move the cursor along the continuum, participants rated their emotional experience online at three strategic points throughout the task. The distance participants moved the cursor toward 100 was automatically recorded, with larger distances denoting greater anxiety, frustration, and control. Previous research has established the utility of these scales in assessing core features of negative affect during stress-provoking conditions (e.g. Brown, Lejuez, Kahler, & Strong, 2002). A ‘‘global’’ negative affect composite score for the VAS ratings of anxiety, frustration, and control was computed by averaging the total pre- to post- change scores of these scales. 1.2.3. Experimental affective assessment To examine the affective reactivity, we employed the Self-Assessment Manikin (SAM; Lang, 1980). The SAM is a non-verbal pictorial assessment methodology using a human-like stimulus. According to factor analytic research, the SAM permits the rapid assessment of three fundamental dimensions of affective meaning in emotional experience—Valence (scale 1), Arousal (scale 2) and Dominance (scale 3; Lang, 1984). SAM ratings are made along five figures for each scale (Lang, Bradley, & Cuthbert, 1990). Participants select their current level of affective state along the dimensions of valence and arousal on a nine-point scale. Specifically, participants can
1240
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
place an ‘‘X’’ over any of the five figures in each scale, or between any two figures. Total scores range from zero to nine for each of the three dimensions. The SAM has sound psychometric properties and has previously been successfully employed to measure affective responses (Bradley & Lang, 1994; Zvolensky, Eifert, & Lejuez, 2001). Reactivity to the stressor was indexed using pre- to post- change scores along the valence and arousal domains. 1.2.4. Experimental physiological assessment A Coulbourn Modular recording system was used to assess heart rate responding on-line at a sample rate of 10 samples per second (sample error range:5 V). Channels were calibrated online prior to sampling. Heart rate was sampled in beats/per minute (bpm) using a digital Coulbourn tachometer fed through a S75–01 bioamplifier and assessed via Medi-Trace pre-gelled Ag/ AgCl electrodes. Heart rate placement followed standard bilateral positioning on either side of the participant’s rib cage. Skin conductance was sampled in microsiemens (ms) using a Coulbourn S7-23 isolated skin conductance coupler. Skin conductance placement was on the palmer surface of two fingers of the nondominant hand. Physiological reactivity was indexed via pre- to post-task heart rate and skin conductance change scores. 1.2.5. Paced auditory serial addition task (PASAT) As the cognitive stressor, we used a modified computer version of the paced auditory serial addition task (PASAT; see Lejuez, Kahler, & Brown, 2003). The PASAT was originally developed for the assessment of information processing capacity (Gronwall, 1977) and was used in the current study because it has been shown to produce elevated levels of negative affect (Deary et al., 1994; Holdwick & Wingenfeld, 1999; Lezak, 1995; Roman, Edwall, Buchanan, & Patton, 1991). The maximum duration of the PASAT was 20 min. During the task, numbers were sequentially flashed on the screen. Participants summed consecutive numbers in sets of two, adding only two numbers at a time [e.g. 4+3 (correct response=7)+6 (=9)+1 (=7)]. Participants used the computer’s mouse to click on the correct answer displayed on the computer screen in a format resembling a keyboard. After each incorrect response, an aversive auditory stimulus was presented. The first level of the PASAT provided a 3-s latency between number presentations (i.e. low difficulty). The second level provided a 1.5-s latency between number presentations (i.e. medium difficulty). The final level provided a 1-s latency between number presentations (i.e. high difficulty). The first level lasted for 3 min and the second level lasted for 5 min. After a 2-min rest period, during which selfreport forms were completed, the final level lasted up to 10 min. All participants were exposed to all three levels of difficulty (i.e. no between subjects manipulation), unless they elected to quit the task prior to completing level 3. 1.3. Procedure Participants were contacted via telephone and scheduled to come to the psychophysiological laboratory. After completing the consent form, participants completed the BIS, MPQ, and RSQ. Once these measures were completed, participants were seated in the experimental room, a 2-m6-m sound attenuated room. The room contained a chair, a desk with a Pentium microcomputer, SVGA color monitor, mouse, and keyboard. An intercom allowed participants to communicate freely with the experimenter in the adjacent room. Once seated, heart rate and skin
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1241
conductance recording electrodes were attached. The following directions were given to each participant: You are going to be completing a serial math task. You will see some numbers being presented in a series. We are going to ask you to add them in a special way. You will see a number presented at time 1. You will see another number presented at time 2. For the initial answer, add the numbers at time 1 and time 2 together. For all other answers, you will need to recall the number presented at time 2, and add it to a newly presented number (at time 3). Note that you have to REMEMBER the second number in the previous series and add it to the newly presented number. This phase of the task will last about 8 minutes. (This was practiced until the participant reported sufficient understanding of the directions.) After you complete phase 1, you will fill out some forms and proceed with phase 2 of the math task. In this phase, you will have an option to escape. As an incentive to try your best, we will award a $40 gift certificate for the mall to the person who scores the highest on this phase. Of course, we will not be able to award this money until the completion of data collection. We used a small reward to produce some incentive for continuing the task without creating a ceiling effect in the duration of endurance across participants. During the session, participants sat alone in the experimental room. The investigator was able to view participants from a one-way mirror and bi-directional communication was possible via an intercom. After completion of the math task, participants were debriefed.
2. Results 2.1. Descriptive data and zero-order relations between theoretically-relevant variables For comparison purposes, we first computed means and standard deviations for each of the variables (see Table 1).1 Next, to better understand how behavioral inhibition sensitivity relates to other theoretically relevant variables, we computed zero-order relations between each of the primary (psychological) predictor variables. The BIS related positively to the MPQ negative affect super factor (r=0.49, P<0.01) and the RSQ (r=0.40, P< 0.01), and negative affect super factor of the MPQ related positively to the RSQ (r=0.50, P<0.01). The three BAS subscales were negatively correlated (except heart rate change, which was positively correlated) with each of the criterion variables listed in Table 2, but, as would be expected, none of these relations were statistically significant. We then computed zero-order correlations between the predictor variables and each of the dependent variables to explore the relations among these theoretically relevant factors (see Table 2). BIS scores showed a significant positive relation with the valence dimension of the SAM as well as the VAS composite negative affect. As would be expected, the negative affect super 1 Given that gender correlates significantly with BIS, the interaction term for BIS and gender was examined for all multiple regression analyses. Inclusion of this term did not appreciably change the pattern of findings. For this reason, and because we did not have any a priori hypotheses about the role of gender, these analyses are neither presented nor discussed.
1242
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
Table 1 Means and standard deviations for predictor and criterion variables Variable
Mean or n
S.D. or %
Demographic/behavioral variables Gender (female/male) Age Education Level Ethnicity
60 19.53 48 78
66.7 1.94 53.3 86.7
Psychological variables BIS MPQ: negative affect super factor RSQ
20.6 89.7 20.5
3.7 16.2 6.5
Experimental assessments: pre- to post-change scores SAM: Arousal SAM: Valence VAS: Composite Heart Rate Skin Conductance
0.87 1.5 26.86 0.46 0.00
1.8 1.75 19.38 6.83 0.00
n=90; however only 63 participants were included in the heart rate and skin conductance assessments due to equipment failure. Participant gender (1=male, 2=female) and ethnicity (1=Caucasian, 2=African American, 3=Hispanic, 4=Asian, 5=Pacific Islander, 6=Other) were dummy coded. The n for subject gender reflects females, the n for ethnicity reflects Caucasian participants, and the n for education represents college freshmen. BIS: Behavioral Inhibition and Activation Scale (Carver & White, 1994). MPQ: Multidimensional Personality Questionnaire (Tellegen et al., 1988). RSQ: Responses to Situations Questionnaire (Nolen-Hoeksema, 1991).SAM: Self-Assessment Manikin (Lang, 1980); VAS: Visual Analog Scale (composite: average of the total pre- to post- change scores of the anxiety, frustration, and control ratings; Lejuez et al., in press).
Table 2 Zero-order relations between predictor and criterion variables SAM: Arousal
SAM: Valence
VAS: Composite
Heart rate
Skin conductance
RSQ
Demographic/behavioral variables Gender (female/male) 0.07 Age 0.01 Education Level 0.16 Ethnicity 0.10
0.26* 0.17 0.05 0.04
0.24* 0.09 0.09 0.07
0.11 0.18 .12 0.26*
0.05 0.20 0.14 0.08
0.22* 0.16 0.08 0.06
Psychological variables BIS MPQ
0.25* 0.08
0.35* 0.27*
0.06 0.10
0.13 0.02
0.40** 0.50**
0.04 0.19
n=90; however only 63 participants were included in the heart rate and skin conductance assessments due to equipment failure. For all of the criterion variables except the RSQ, pre-to post- change scores were used in all analyses. SAM: SelfAssessment Manikin (Lang, 1980); VAS: Visual Analog Scale (composite: average of the total pre- to post- change scores of the anxiety, frustration, and control ratings; Lejuez et al., in press). RSQ: Responses to Situations Questionnaire (NolenHoeksema, 1991). BIS: Behavioral Inhibition and Activation Scale (Carver & White, 1994). MPQ: Multidimensional Personality Questionnaire—negative affect superfactor (Tellegen et al., 1988). *P<0.05. **P<0.01.
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1243
factor of the MPQ was positively related to the ‘‘global’’ negative affect index during the stressor, as assessed by VAS ratings. None of the pre-experimental psychological variables were significantly related to heart rate or skin conductance, indicating a marked degree of response discordance. 2.2. Multiple regression analyses Hierarchical multiple regression analyses were conducted in order to test each of our primary hypotheses. Analyses were conducted separately for each of the dependent variables. Following the recommendations of Tabachnick and Fidell (1996, see p. 156), we utilized a stepwise approach in the model because it enters the variables into the equation according to the strength of their association with each primary dependent variable. This approach was most consistent with our a priori research objectives of determining the relative predictive power of behavioral inhibition. Each variable that met criteria for inclusion (i.e. P to enter=0.05) and did not meet criteria for removal (i.e. P to remove=0.10) was retained. Semi-partial correlations (sr2) were included to represent effect sizes only for those variables retained in the equation. This index of effect size represents the proportion of total criterion variance for variable Y uniquely accounted for by predictor variable X (Cohen, 1988). 2.2.1. Rumination Predictor variables for rumination included the main effects of gender, age, education, and ethnicity, MPQ ratings, and BIS ratings. Overall, the multiple regression analysis indicated that the predictor variables together explained 26% of the variance in self-reported rumination, F (2, 85)=14.7, P<0.01 (adjusted R2=24%). Both the main effects of negative affect as assessed by the MPQ (b=0.35; sr2=0.10, P<0.01) and BIS (=0.23 sr2=0.04, P<0.05) contributed in a statistically significant manner to the equation. 2.2.2. Emotional reactivity Predictor variables for emotional reactivity included the main effects of age, gender, education level, ethnicity, and BIS scores. Regression analyses predicting the VAS composite indicated that the predictor variables together explained 13% of the variance, F (1, 86)=12.3, P< 0.01 (adjusted R2=12%). As hypothesized, the only variable that met inclusion criteria at any level on the model was the main effect of BIS (b=0.35; sr2=0.12, P< 0.01). For the SAM ratings of emotional valence, the predictor variables together explained 6% of the variance, F (1, 86)=5.8, p < 0.05 (adjusted R2=5%). As hypothesized, the only variable that met inclusion criteria was BIS (=0.25; sr2=0.06, P< 0.05). In terms of the SAM ratings of emotional arousal and dominance, none of the predictors explained a significant amount of variance in the criterion measures. Prior to data analysis, physiological data were screened for outliers due to sampling error (e.g. participant movement). For analyses, the baseline and the mean heart rate during level two were utilized. The baseline reading (i.e. mean of samples taken every 20 s for 2 min) was taken after participants had completed the experimental task in order to control for preexperimental anxiety. With regard to heart rate, the predictor variables together explained 8% of the variance, F (1, 60)=4.9, P<0.05 (adjusted R2=6%). The only variable that met inclusion criteria was ethnicity
1244
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
(=0.28; sr2=0.08, P<0.05). Although we examined skin conductance in the same manner as the other dependent measures, no significant predictors were found.
3. Discussion Contemporary theories of behavioral inhibition suggest that this temperamental variable should relate to a variety of negative emotional outcomes, including high degrees of emotional reactivity and dysfunctional regulatory styles (Gray, 1994). There has been convincing support for this perspective from neuorophysiological (Davidson & Fox, 1989) and behavioral (Kagan et al., 1988) investigations. Yet, there is a relative absence of investigations addressing behavioral inhibition at a cognitive level of analysis. This lack of research stymies efforts to apply behavioral inhibition research and theory to many disorders of emotion. To address the predictive power of behavioral inhibition in terms of negative emotional responding, the present investigation tested two primary hypotheses concerning emotional reactivity and regulation. In terms of emotional reactivity, the present results indicate that, as hypothesized, levels of behavioral inhibition predicted negative affect and negatively valenced ratings of distress during a cognitive stressor. Although these effects were small, it was the only variable in our prediction equations to add a significant amount of variance when predicting post-challenge ratings of affective distress. At the same time, behavioral inhibition did not significantly predict indices of physiological arousal, indexed by self-rated or direct on-line measurement. Such response discordance provides further evidence that behavioral inhibition may relate more to cognitive than physiological aspects of emotional reactivity (see also, Zvolensky, Feldner et al., 2001). Alternatively, it is possible these findings are at least partially a product of the stressor utilized in the current investigation; that is, the PASAT may not have elicited high enough degrees of somatic arousal (i.e. restricted range) to be captured by behavioral inhibition. In all cases, the finding that behavioral inhibition is associated with emotional reactivity at the cognitive level of analysis is important, and specifically illustrates one type of affective variability process linked to this tempermantal construct. Given these findings, psychopathologists are now in a good position to extend this research in an effort to examine how behavioral inhibition confers increased risk for emotional distress. For example, future research could extend the current findings by examining emotional reactivity as it relates to particular cognitive processes (e.g. stroop task). The present study also found that individual variation in behavioral inhibition, along with basal levels of negative affect, significantly predicts a rumination-oriented response style. This finding provides the first evidence that in addition to being associated with negative emotional reactivity, behavioral inhibition is related to dysfunctional styles of regulating emotional distress. Again, these data may highlight some of the specific response styles associated with the negative emotional effects of behavioral inhibition. Although the correlational nature of this particular finding precludes understanding directional or causative roles, considering other research in this domain (Harmon-Jones & Allen, 1997), it may be that behavioral inhibition promotes dysfunctional types of learning experiences. For example, drawing from developmental research (Kagan et al., 1988), it is possible that high levels of behavioral inhibition may increase the likelihood of the rigid application of avoidance-oriented coping strategies for managing emotionally evocative life events. In the absence of corrective learning, such individuals may be less apt to take
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1245
constructive action during negative emotional states, resulting in more frequent, and longer periods of, negative affect (Nolen-Hoeksema et al., 1994). Based upon these preliminary findings, researchers have a sound empirical basis upon which to conduct experimental tests of this and related hypotheses. There are a number of interpretative caveats and directions for future research that warrant consideration. First, all of the psychological variables were assessed using self-report instruments. Accordingly, it is possible the observed findings were, at least in part, due to shared method variance. Although such a methodological strategy was useful at this stage of research development, it likely will be useful to incorporate experimental cognitive science methodologies into future research in this area, particularly those that tap automatic types of processing (e.g. interpretative biases for threat). Second, as with all quasi-experimental designs, causal relations cannot be unambiguously inferred, leaving the results open to a number of interpretations. One way to better address this issue would be to provide a prospective assessment of how behavioral inhibition relates to emotional distress in the natural environment. Finally, although we intentionally designed the study to examine psychologically healthy individuals, it now makes sense to extend these findings to persons with specific types of emotion-based psychopathology. In this way, researchers can determine to what extent behavioral inhibition is related to specific emotional disorders. In summary, the results of the present study help to identify how theoretically relevant variables produce heightened levels of emotional distress. Our results indicate that at a cognitive level of analysis, individual differences in behavioral inhibition predict differential aspects of emotional reactivity to cognitive stress and a rumination response style. These results converge and uniquely extend the existing evidence that behavioral inhibition cognitive processes are important pre-morbid predictors of negative responding.
Acknowledgements This study was submitted in partial fulfillment of the requirements of Ellen W. Leen’s preliminary examination at The University of Vermont. References Arcus, D. (2001). Inhibited and uninhibited children: Biology in the social context. In T. D. Wachs, & G. A. Kohnstamm (Eds.), Temperament in context (pp. 43–60). New Jersey: Laurence Erlbaum Associates. Bradley, M. M., & Lang, P. J. (1994). Measuring emotion: The Self-Assessment Manikin and the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry, 25, 49–59. Brown, R. A., Lejuez, C. W., Kahler, C. W., & Strong, D. R. (2002). Distress tolerance and duration of past smoking cessation attempts. Journal of Abnormal Psychology, 111, 180–185. Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: the BIS/BAS scales. Journal of Personality and Social Psychology, 67, 319–333. Cloninger, C. R. (1987). A systematic method for clinical description and classification of personality variants. Archives of General Psychiatry, 44, 573–588. Cohen, J. (1988). Statistical power analysis (2nd ed.). Hillsdale, NJ: Erlbaum. Cook, Edwin W. III. (1999). Affective individual differences, psychopathology, and startle reflex modification. In
1246
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
M. E. Dawson, & A. M. Schell (Eds.), Startle modification: implications for neuroscience, cognitive science, and clinical science (pp. 187–208). New York, NY: Cambridge University Press. Corr, P. J., Pickering, A. D., & Gray, J. A. (1997). Personality, punishment, and procedural learning: a test of J. A. Gray’s anxiety theory. Journal of Personality and Social Psychology, 73, 337–344. Cox, B. J., Enns, M. W., & Taylor, S. (2001). The effect of rumination as a mediator of elevated anxiety sensitivity in major depression. Cognitive Therapy and Research, 25, 525–534. Davidson, R. J. (1992). Emotion and affective style: hemispheric substrates. Psychological Science, 3, 39–43. Davidson, R. J. (1998). Affective style and affective disorders: Perspectives from affective neuroscience. Cognition and Emotion, 12, 307–320. Davidson, R. J., Ekman, P., Saron, C. D., Senulis, J. A., & Friesen, W. V. (1990). Approach-withdrawal and cerebral asymmetry: emotional expression and brain physiology. Journal of Personality and Social Psychology, 58, 330–341. Davidson, R. J., & Fox, N. A. (1989). Frontal brain asymmetry predicts infants’ response to maternal separation. Journal of Abnormal Psychology, 98, 127–131. Deary, I. J., Ebmeier, K. P., MacLeod, K. M., Dougall, N., Hepburn, Frier, & Goodwin (1994). PASAT performance and the pattern of uptake of super (99m) Tc-exametazime in brain estimated with single photon emission tomography. Biological Psychology, 38, 1–18. Fowles, D. C. (1993). A motivational theory of psychopathology. In W. Spaulding (Ed.), Nebraska symposium on motivation: integrated views of motivation, cognition and emotion (Vol. 41) (pp. 181–238). Lincoln, NE: University of Nebraska Press. Gable, S. L., Reis, H. T., & Elliot, A. J. (2000). Behavioral activation and inhibition in everyday life. Journal of Personality and Social Psychology, 78, 1135–1149. Gray, J. A. (1994). Framework for a taxonomy of psychiatric disorder. In S. H. M. Van Goozen, N. E. Van de Poll, & J. A. Sergeant (Eds.), Emotions: essays on emotion theory (pp. 29–59). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Gray, J. A., & McNaughton, N. (1996). The neuropsychology of anxiety: A reprise. In D. A. Hope (Ed.), Nebraska symposium on motivation: perspectives on anxiety, panic, and fear (Vol. 43) (pp. 61–134). Lincoln: University of Nebraska Press. Grillon, C., Ameli, R., Foot, M., & Davis, M. (1993). Fear-potentiated startle: relationship to the level of state/trait anxiety in healthy subjects. Biological Psychiatry, 33, 566–574. Gronwall, D. M. (1977). Paced auditory serial-addition task: a measure of recovery from concussion. Perceptual Motor Skills, 44, 367–373. Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2, 271–299. Harmon-Jones, E., & Allen, J. J. (1997). Behavioral activation sensitivity and resting frontal EEG asymmetry: covariation of putative indicators related to risk for mood disorders. Journal of Abnormal Psychology, 106, 159–163. Holdwick, D. J., & Wingenfeld, S. A. (1999). The subjective experience of PASAT testing: does the PASAT induce negative mood?. Archives of Clinical Neuropsychology, 14, 273–284. Kagan, J. (1989). Temperamental contributions to social behavior. American Psychologist, 44, 668–674. Kagan, J., Reznick, J. S., & Snidman, N. (1988). Biological bases of childhood shyness. Science, 240, 167–171. Kagan, J., & Snidman, N. (1991). Temperamental factors in human development. American Psychologist, 46, 856–862. Kagan, J., & Snidman, N. (1999). Early childhood predictors of adult anxiety disorders. Biological Psychiatry, 46, 1536–1541. Lang, P. J. (1980). Behavioral treatment and bio-behavioral assessment: computer applications. In J. B. Sidowski, J. H. Johnson, & T. A. Williams (Eds.), Technology in the health care delivery system (pp. 119–137). Norwood, NJ: Ablex. Lang, P. J. (1984). Cognition in emotion: concept and action. In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotions, cognitions, and behavior (pp. 192–228). New York: Cambridge. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex. Psychological Review, 97, 377–395. Larsen, R. J., & Ketelaar, T. (1989). Extraversion neuroticism, and susceptibility to positive and negative mood induction procedures. Personality and Individual Differences, 10, 1221–1228.
E.W. Leen-Feldner et al. / Personality and Individual Differences 36 (2004) 1235–1247
1247
Lejuez, C. W., Kahler, C. W., & Brown, R. A. (2003). A modified version of the Paced Auditory Serial Addition Task (PASAT), as a laboratory-based stressor. Behavior Therapist, 26, 290–293. Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). New York, NY: Oxford University Press. Melamed, S. (1987). Emotional reactivity and elevated blood pressure. Psychosomatic Medicine, 49, 217–225. Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 100, 569–582. Nolen-Hoeksema, S., & Morrow, J. (1992). A prospective study of depression and posttraumatic stress symptoms after a natural disaster: the 1989 Loma Prieta earthquake. Journal of Personality and Social Psychology, 61, 115–121. Nolen-Hoeksema, S., Parker, L. E., & Larson, J. (1994). Ruminative coping with depressed mood following loss. Journal of Personality and Social Psychology, 67, 92–104. Reznick, J. S., Kagan, J., Snidman, N., & Gersten, M. (1986). Baak, & Rosenberg, A. Inhibited and uninhibited children: a follow-up study. Child Development, 57, 660–680. Roman, D. D., Edwall, G. E., Buchanan, R. J., & Patton, J. H. (1991). Extended norms for the paced auditory serial addition task. Clinical Neuropsychologist, 5, 33–40. Rosenbaum, J. F., Biederman, J., Gersten, M., Hirshfeld, D. R., Meminger, S. R., Herman, J. B., Kagan, J., Reznick, S., & Snidman, N. (1990). Behavioral inhibition in children of parents with panic disorder and agoraphobia: a controlled study. In S. Chess, & M. E. Hertzig (Eds.), Annual progress in child psychiatry and child development (pp. 294– 315). Philadelphia, PA: Brunner/Mazel. Ruscio, A. M., Borkovec, T. D., & Ruscio, J. (2001). A taxometric investigation of the latent structure of worry. Journal of Abnormal Psychology, 110, 413–422. Sobotka, S. S., Davidson, R. J., & Senulis, J. A. (1992). Anterior brain electrical asymmetries in response to reward and punishment. Electroencephalography and Clinical Neurophysiology, 83, 236–247. Sutton, S. K., & Davidson, R. J. (1997). Prefrontal brain asymmetry: a biological substrate of the behavioral approach and inhibition systems. Psychological Science, 8, 204–210. Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). Northridge, CA: Harper Collins. Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In A. H. Tuma, & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Tellegen, A., Lykken, D. T., Bouchard, T. J., Wilcox, K. J., Segal, N. L., & Rich, S. (1988). Personality similarity in twins reared apart and together. Journal of Personality and Social Psychology, 54, 1031–1039. Zvolensky, M. J., & Eifert, G. H. (2000). A review of psychological factors/processes affecting anxious responding during voluntary hyperventilation and inhalations of carbon dioxide-enriched air. Clinical Psychology Review, 21, 375–400. Zvolensky, M. J., Eifert, G. H., & Lejuez, C. W. (2001). Emotional control during recurrent 20% carbon dioxideenriched air induction: relation to individual difference variables. Emotion, 1, 148–165. Zvolensky, M. J., Feldner, M. T., Eifert, G. H., & Stewart, S. H. (2001). Evaluating differential predictions of anxietyrelated reactivity during repeated 20% carbon-dioxide-enriched air challenge. Cognition and Emotion, 15, 767–786. Zvolensky, M. J., Lejuez, C. W., Stuart, G. L., & Curtin, J. J. (2001). Experimental psychopathology in psychological science. Review of General Psychology, 5, 371–381.