Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Does social support reduce cardiovascular stress reactivity only if you want support: a test of a match/mismatch… Kors, Deborah Joy 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-389170.pdf [ 5.99MB ]
Metadata
JSON: 831-1.0099440.json
JSON-LD: 831-1.0099440-ld.json
RDF/XML (Pretty): 831-1.0099440-rdf.xml
RDF/JSON: 831-1.0099440-rdf.json
Turtle: 831-1.0099440-turtle.txt
N-Triples: 831-1.0099440-rdf-ntriples.txt
Original Record: 831-1.0099440-source.json
Full Text
831-1.0099440-fulltext.txt
Citation
831-1.0099440.ris

Full Text

DOES SOCIAL SUPPORT REDUCE CARDIOVASCULAR STRESS REACTIVITY ONLY IF YOU WANT SUPPORT: A TEST OF A MATCH/MISMATCH HYPOTHESIS by DEBORAH JOY KORS B.S., Emory University, 1984 M.A., The University of British Columbia, 1994 A THESIS SUBMITTED DN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April 1999 © Deborah Joy Kors, 1999  In  presenting  degree  at  this  the  thesis  in  partial  fulfilment  of  University  of  British  Columbia,  I agree  freely available for copying  of  department publication  this or of  reference  thesis by  this  for  his thesis  and study. scholarly  or for  her  Department The University of British Vancouver, Canada  Date  DE-6 (2/88)  ftp*;)  Columbia  purposes  requirements that  agree  may  representatives.  financial  permission.  I further  the  It  gain shall not  be is  that  the  Library  permission  granted  by  understood be  for  an  advanced  shall for  the that  allowed without  make  it  extensive  head  of  my  copying  or  my  written  Abstract Epidemiological studies have suggested that social support may offer a protective role for cardiovascular health. More recently, researchers have begun to examine possible mechanisms by which social support may reduce cardiovascular reactivity to stress. This study was undertaken to determine if the implementation of support (presence or absence) needs to be matched with a person's habitual level of support seeking (high or low) in order to obtain physiological benefits during laboratory stressors. It was hypothesized that high support seekers assigned to a support condition would show decreased reactivity relative to all other matched or mismatched conditions. Following the screening of480 students, 135 high and low support seeking men and women were recruited for the laboratory phase. In this phase, participants performed a math and a speech task, while alone or while receiving support. Participants were randomly assigned to a condition that either matched.(e.g., support provision for support seekers) or mismatched their support seeking style. Heart rate (HR), systolic blood pressure (SBP) and diastolic blood pressure (DBP) were monitored continuously. Self-report measures evaluating the participants' reactions to the task and to support were administered pre- and post-task. Athough the central (match) hypothesis of this study was not supported, several interesting findings emerged: Low support seeking men exhibited larger SBP responses than did high support seeking men during the tasks. Participants receiving support showed greater SBP responses relative to participants who were alone during the tasks. Additionally, supported men showed increased DBP reactivity relative to men who were assigned to an alone condition. Interestingly, high support seeking men and women showed lower resting blood pressures than did low support seeking participants. The findings from the self-report measures did not help to clarify the cardiovascular findings. The findings suggest that future research is needed to better understand how and when support is effective in reducing cardiovascular reactivity to stress.  n  Table of Contents Abstract  ii  Table of Contents  iii  List of Tables  vi  List of Figures  viii  List of Appendices  ix  Acknowledgement  xi  Dedication  xii  Introduction  1  Epidemiological Mortality Studies  2  Limitations of Mortality Studies  3  Coronary Heart Disease Morbidity Studies  4  Limitations of Morbidity Studies  5  A Need for Process Studies  5  General Models Linking Social Support to Health  6  Proposed Mechanisms for the Link between Social Support and Coronary Heart Disease  7  Ambulatory Studies Examining the Relationship Between Social Support and Cardiovascular Indices  8  Laboratory Studies Evaluating the Impact of Social Support on Cardiovascular Stress Reactivity  9  Psychological Mediation of Support  11  Variables Affecting the Support Process  12  Goals of this study  16  Method  20  Overview  :  in  20  Participants  20  Psychological Measures  23  Cardiovascular Measures  28  General Procedure  29  Setting  31  Implementation of Support (for support condition)  31  Specific Procedures  33  Analytic Strategy  41  Cardiovascular Data....  41  Self-report Data  42  Evaluation of Assumptions for Statistical Procedures  42  Alpha Level  43  Missing Data  44  Results Demographic Data  45 ;  45  Validity Checks  45  Cardiovascular Measures  49  Self-Report of Stress/Arousal  56  Reactions to Task  58  Reactions to Provision of Support  61  Impact of Ethnic Grouping on Cardiovascular reactivity, Perception of Support, and Level of Support Seeking  62  Impact of Ethnicity of Confederate on the Cardiovascular Reactivity of Asian and White Participants  64  Discussion  65  Summary of Findings  65  Evaluation of Specific Findings and Hypotheses  65  iv  Potential threats to internal validity  71  General Limitations of the Study  74  Implications of this Study and Future Directions  76  References  80  Appendix 1  •  v  89  List of Tables  Table 1. Men - Demographic Data, Support Seeking Scores, and Perceived Support  22  Table 2. Women - Demographic Data, Support Seeking  Scores, and Perceived Support  23  Table 3. Men - Task-related Stress/Arousal and Task Performance  46  Table 4. Women - Task-related Stress/Arousal and Task Performance  46  Table 5. Number of Support Statements Provided during Math Task, Speech Preparation, and Speech Task  47  Table 6. Men - Means (SD) of Blood Pressure and Heart Rate Baselines and Change Scores for the Math Task, Speech Preparation, and Speech Task  49  Table 7. Women - Means (SD) of Blood Pressure and Heart Rate Baselines and Change Scores for the Math Task, Speech Preparation, and Speech Task  50  Table 8. Men -Visual Analog Scale Data Assessing Perception of Task and Perception of Support/Supporter  vi  5§  Table 9. Women - Visual Analog Scale Data Assessing Perception of Task and Perception of Support/Supporter  vii  / £0  List of Figures  Figure 1. Upper and lower support seeking quartiles for men and women  24  Figure 2. Men - SBP change scores for math task, speech preparation, and speech task  51  Figure 3. Women - SBP change scores for math task, speech preparation, and speech task  51  Figure 4. Men - DBP change scores for math task, speech preparation, and speech task  52  Figure 5. Women - DBP change scores for math task, speech preparation, and speech task  52  Figure 6. Men - H R change scores for math task, speech preparation, and speech task  53  Figure 7. Women - H R change scores for math task, speech preparation, and speech task  53  viii  List of Appendices  Appendix 1. Consent Form for Screening Phase  88  Appendix 2. Consent Form for Laboratory Session (Alone Condition)  89  Appendix 3. Consent Form for Laboratory Session (Support Condition)  91  Appendix 4. Demographic Questionnaire..  93  Appendix 5. Coping Strategy Indicator (screening measure)  94  Appendix 6. Social Support Inventory (pilot support seeking measure)  97  Appendix 7. Interpersonal Support Evaluation List  100  Appendix 8. Stress/Arousal Check List  104  Appendix 9. Controversial Issues Questionnaire  105  Appendix 10. Reactions to Task (Alone Condition)  106  Appendix 11. Reactions to Task/Support (Support Condition)  107  Appendix 12. Evaluations of the Supporter (Support Condition)  108  ix  Appendix 13. Manipulation Check for Support Implementation  109  Appendix 14. Helpful Hints for Preparing the Speech  110  Appendix 15. Experimenter Ratings of Participant's Comfort during the  Speech  Ill  Appendix 16. Data Collection Sheet for Cardiovascular Measures  112  Appendix 17. Experimenter Script for Alone Condition  114  Appendix 18. Experimenter Script for Support Condition  117  Appendix 19. List of Support Statements  121  Appendix 20. Debriefing Form  125  x  Acknowledgement  I want to express my gratitude to the team at the Linden laboratory for their assistance in conducting this research project. In particular, the help receivedfromEva Dehaas and Anuradha Chawla was greatly appreciated. I also would like to thank those who acted as confederates: Leora Stacey-Chung, Susan Sandler, Pam Brown, and Kim Denier. Finally, I want to thank Dr. Wolfgang Linden himself for the guidance and support that he has provided, not only during this project but during my entire career as a graduate student.  xi  Dedication This dissertation is dedicated to those people in my life who have accompanied me "in spirit", during my journey towards the Ph.D. I want to thank all of those close friends who have been both supportive of me and very understanding of my rather "busy schedule" throughout the past few years. In particular, my friend, Karen, has been an incredible source of support throughout my graduate school years. Additionally, her wonderful sense of humor has helped me to laugh, even during times when it was most difficult to do so. The tremendous patience and support of my boyfriend, Chris, has also been greatly appreciated. Finally, I have been blessed with the love and encouragement of two wonderful parents (Ruth and Sandy), throughout my life. Their tremendous belief in me has always given me the strength to take on and conquer new challenges and, has similarly served as a strong source of inspiration in my perseverance towards this doctoral degree.  xii  Introduction  Over the past 20 years, there has been an explosion of research in the area of social support, with a large volume of literature linking social support to both physical and mental health (Berkman, 1985; Bloom, 1990; Cohen, 1988; Ganster & Victor, 1988; House, Landis, & Umberson, 1988). In particular, it has been repeatedly shown that low levels of support are associated with poor health and well-being. In fact, Cassell (1976) and Cobb (1974), in their seminal discussions of thefield,suggested that support might play an etiological role in the development of disease, especially stress-related disorders,fiiehnga tremendous amount of subsequent research in the area. Despite the significant attempts of researchers to study both the impact of support on various health indices, as well as the mechanisms by which people may benefit from support, there remains no agreed upon definition of support over 20 years later. For example, support has been defined as having a large number of social ties, as being well-integrated within one's social network, and as receiving important social resourcesfromnetwork members (Berkman & Syme, 1979; Cohen, 1988; House, Umberson, & Landis, 1988). Additionally, support has sometimes been conceptualized within a more interpersonal scheme and viewed as an exchange process involving interactions between both providers and recipients (Antonucci, 1985; Shumaker & Brownell, 1984). This lack of conceptual clarity has been reflected in the large variety of support measures which have been used - often developed on an ad hoc basis making comparisons between studies difficult at best. A distinction that has proven useful is that of structural versus functional measures (Cohen, 1988; Cohen & McKay, 1984; Cohen & Wills, 1985; Wills, 1985). Structural measures typically assess interconnections between an individual and his/her social network. Thus, the existence of social ties, as well as various characteristics (e.g., density, homogeneity, reciprocity) of the ties are evaluated. The degree of social integration, or how embedded the  1  individual is within the social network, also falls within this category. In general, these measures have been most frequently employed in epidemiological studies evaluating the link between support and mortality. In contrast, functional measures reflect the specific functions that interpersonal relationships are expected to serve. A number of taxonomies of support functions have been proposed (House, 1982; Weiss, 1968; Wills, 1985), in which support is conceived as a multidimensional construct, with different aspects of support provided in different situations. One example is the classification scheme suggested by Wills (1985). Wills (1985) describes a number of support functions which can be measured, including emotional or esteem support, informational support, instrumental support, and social companionship, among several others. It has been speculated that the functional aspects of support are critical to the stress buffering properties of support, and that effective buffering requires a match between support functions and the specific need elicited by the stressful event (Cohen, 1988; Cohen & Wills, 1985).  Epidemiological Mortality Studies Prospective mortality studies have been conducted in an attempt to show a link between social support and mortality (Berkman & Syme, 1979; Blazer, 1982; House, Robbins, & Metzner, 1982; Kaplan et al., 1988; Orth-Gomer & Johnson, 1987; Schoenbach, Kaplan, Friedman, & Kleinbaum, 1986). These studies typically have employed structural measures of support (such as degree of social integration) and have additionally tried to control for risk factors that may contribute to disease and death. Overall, the findings from these studies have shown a strong relationship between social integration and all-cause mortality, such that individuals with minimal social contact have a greater risk of mortality. A number of these studies (Kaplan et al., 1988; Orth-Gomer & Johnson, 1987) have also demonstrated a similar link between social integration and mortality from cardiovascular disease. In both cases, these associations are maintained after controlling for traditional risk factors such as blood pressure, serum cholesterol levels, and smoking  2  (Cohen, 1988). Additionally, the effects of social integration on mortality appear to be relatively non-specific, in that the magnitude of this relationship does not vary significantly with specific cause of death (House, Umberson, & Landis, 1988)  Limitations of Mortality Studies Despite the general similarity in mortality study findings, a number of inconsistencies remain. Specifically, studies tend to vary in the magnitude of the mortality risk, and are inconsistent regarding whether men and women are equally affected by low levels of social integration. It seems likely that some of these inconsistencies may be explained by the variety of populations sampled (House, Landis & Umberson, 1988; Kaplan et al., 1988). Additionally, although the measures tend to be primarily structural measures of support, the specific measures used have varied significantly (Berkman, 1985; Kaplan et al., 1988; OrthGomer & Johnson, 1987). Thus, measurement differences also could account for differences in the magnitude of the risk. Perhaps one of the greatest limitations of these epidemiological studies is that they typically evaluate mortality risks for initially healthy individuals (Cohen, 1988). Thus, it is not possible to determine the endpoint of the disease process (i.e., incidence, severity, course of disease, or recovery) at which social ties may have an impact. Several researchers have shown a strong relationship between social integration and mortality in individuals who are abeady showing signs of serious illness (Berkman, Leo-Sumers, & Horwitz, 1992; Case, Moss, Case, McDermott, & Eberly, 1992; Ruberman, Weinblatt, Goldberg, & Chaudhary, 1984), suggesting that social integration is associated with disease progression and/or recovery. However, suchfindingsdo not rule out the possibility that social ties also may affect the onset of disease (see section below). Hence, the mortality studies point to the necessity of examining the influence of support on specific endpoints in the disease process.  3  Coronary Heart Disease Morbidity Studies The research linking support to coronary heart disease (CHD) morbidity (as well as the link between support and other diseases) does not appear to be as strong as the mortality data. Two studies of Japanese-American men showed that social integration was associated with the prevalence of myocardial infarction (MI), CHD, and angina pectoris (Joseph, 1980; Reed, McGee, Yano, & Feinleib, 1983), although social integration was not associated with CHD incidence (Reed et al., 1983). Another study (Seeman & Syme, 1987) of men and women undergoing angiography in several San Francisco hospitals found that greater instrumental support and one's feeling of being loved were significant predictors of coronary atherosclerosis. Social integration, however, was not related to atherosclerosis development. Moreover, the love and support of one's wife was found to reduce the risk of developing angina pectoris, but this only remained true for men experiencing high levels of anxiety (Medalie & Goldbourt, 1976). Finally, Haynes and Feinleib (1980) reported that female clerical workers with non-supportive bosses were at increasedriskfor the development of coronary heart disease over an 8-year period. It is important to note, however, that this study may not have independently assessed degree of support, as perception of support may have been confounded with job stress (Cohen, 1988). Other researchers have examined the impact of social support after an initial MI. hi one study, emotional support (reported prior to the MI) was independently related to theriskof death during the 6 months following an MI in an elderly sample of men and women (Berkman, Leo-Sumers, & Horwitz, 1992). In another study, evaluation of a sample of 2320 male survivors of an acute MI revealed that high levels of social isolation independently contributed to mortality over a 3 year period following the MI (Ruberman et al., 1984). Additionally, Case et al. (1992) found that following an initial MI, living alone was an independent risk factor for the recurrence of a major cardiac event.  4  Limitations of Morbidity Studies As with the mortality studies, investigations which evaluate the relationship between support and cardiovascular disease do not provide a clear picture as to whether support is most influential in the onset, progression, or recovery from this disease. Thus, future longitudinal studies are needed to try to better evaluate the impact of support on various stages in the disease process and to assess more clearly the type of support (e.g., integration, network size, or specific functional aspect of support) that is most useful at each stage. Further investigation of gender similarities and differences regarding the kind of support that is most likely to be protective is also of particular importance. In addition, the lack of association between social integration and incidence of disease (combined with the link between integration and disease prevalence) may suggest an alternative explanation, namely that people who are ill become increasingly socially isolated. Several authors, however, have provided evidence against this explanation, and in support of the argument that decreased support leads to poorer health, by controlling for preexisting health status (Blazer, 1982; Kaplan et al., 1988). Additionally, Fontana, Kerns, Rosenberg, and Colonese (1989) have tested a number of these hypotheses by using structural equation modeling over three time-lags in a longitudinal, 12-month follow-up of post-myocardial infarction patients. The results supported the hypothesis that social support attenuated cardiac symptomatology during thefirst6 months following MI, while alternative causal hypotheses were supported only at the level of chance.  A Need for Process Studies Although the epidemiological studies have been extremely influential in demonstrating a link between social support and CHD morbidity, as well as between support and CFfD-related mortality, they do not provide a clear understanding of the mechanism by which support may be protective against this disease. In response to this limitation, there has been a move away  5  from epidemiological studies, with a shift towards more process-oriented studies. Within the last 10 years, laboratory and ambulatory monitoring studies have been conducted to attempt to delineate more clearly the process by which support may be beneficial (see later sections). However, prior to a detailed description of these studies, it is necessary to address the theories which have guided them.  General Models Linking Social Support to Health Two general models have been suggested to explain the general protective influence of social support on health. The main effects model, which primarily has been tested epidemiologically, suggests that social support has a direct beneficial impact on health, regardless of level of stress. This is typically described as an additive model, with increased support leading to enhanced well-being, although some investigators have found a ceiling effect, in which there is no added benefit of support beyond a certain minimal level (Berkman & Syme, 1979; House, 1982). There is evidence that social integration may account for this model of support (Cohen & Wills, 1985). In contrast, the buffering model hypothesizes that there is an interaction between social support and stress, such that support is only beneficial for individuals during stressful situations. Specifically, social support protects (or buffers) the individual from the adverse effect of stress, thereby preventing the onset of disease. The individual's perception of the adequacy of the support function appears to be most important for buffering to take place (Cassel, 1976; Cohen, 1988; Cohen & McKay, 1984; Cohen & Wills, 1985). This model has been tested with a variety of laboratory experiments (see below). It has been suggested that the main effects model of support could entail a perception of availabihty of support during episodes of stress (as well as during periods of no stress), which, in turn, would lead one to question whether the two models of support are conceptually distinct. However, the use of social integration measures provides evidence for main effects but not for buffering effects, suggesting that embeddedness in a social network is useful to  6  promote a sense of well being but not particularly helpful during periods of stress (Cohen & Wills, 1985). Jn explaining this finding, Cohen and Wills (1985) indicate that having a certain number of social contacts does not imply that the relationships are close ones or are available to provide support during stress. Along these lines, low correlations between measures of social integration and measures of functional support (i.e., availabihty of support functions) have consistently been observed. With both the main effects and buffering models, there are several pathways by which support could decrease the likelihood of the development of disease. Individuals with adequate support may be more likely to engage in health-promoting behaviors, such as adhering to a healthy diet, engaging in regular exercise, and going for regular medical checkups, while avoiding harmful behaviors, such as smoking and excess alcohol use (Bloom, 1990; Cohen, 1988; Shumaker & Hill, 1991). A second pathway suggests that support may provide direct physiological benefits to the individual via changes in neuroendocrine, hemodynamic, and immune responses which, in turn, would have a direct effect on health status (Cohen, 1988; Davidson & Shumaker, 1987). This mechanism is hypothesized to be particularly important during stressful situations (see below). Finally, it is also possible that certain psychological states may mediate the link between social support and health. Along these lines, social support may promote positive affect by increasing feelings of self-esteem, a sense of belonging, and a sense of control. Such positive psychological states would then be beneficial to health by facilitating more health-promoting behaviors or by their strong impact on biological processes, directly implicated in the development of disease (House, Umberson, & Landis, 1988; Shumaker & Hill, 1991).  Proposed Mechanisms for the Link between Social Support and Coronary Heart Disease In applying the above pathways to describe the link between social support and coronary heart disease, it has been proposed that the influence of support on health-related lifestyle factors may play a role in the development and progression of this disease (Cohen, 1988;  7  Umberson, 1987). An even greater emphasis over the past 10 years, however, has been placed on the possibility that social support may exert a strong influence on physiological processes, such as neuroendocrine or hemodynamic functioning, which would directly affect cardiovascular health. Along these lines, the reactivity hypothesis suggests that individuals who experience excessive cardiovascular responses to stress may be at greater risk for coronary heart disease and hypertension (Krantz & Manuck, 1984; Manuck, 1994; Manuck, Kasprowicz & Muldoon, 1990). Greater stress reactivity may also be associated with more severe coronary events in diseased patients (Krantz et al., 1991; Rozanski et al., 1988). Presumably, then, social support might buffer the individual from the impact of these stressors, thereby preventing or decreasing the physiological responses to the stressful situations. The effects of support could be mediated at two time points. In thefirstcase, support may prevent an individualfromappraising a situation as stressful (or allow the event to be perceived as less stressful), thereby preventing the stress response (Cohen & McKay, 1984; House, 1981). According to Lazarus's model, support would decrease the primary appraisal of the threat (Lazarus & Folkman, 1984). The other possibility is that support may intervene after the situation has been appraised as stressful, by accelerating recovery from the physiological stress response. In the latter case, support would have a direct psychophysiological influence, dampening the neuroendocrine activation so that the individual is less reactive to stress (House, 1982; Uchino, Cacioppo, & Kiecolt-Glaser, 1996). This decrease in cardiovascular reactivity might also be mediated by the facilitation of coping behaviors and decreased affective response that occur in response to support. Ambulatory Studies Examining the Relationship Between Social Support and Cardiovascular Indices In order to better examine the impact of social support on cardiovascular functioning during one's daily life, ambulatory studies have been conducted. Typically, these studies involve 8-24 hour ambulatory recordings of blood pressure and heart rate as the individual  8  goes about his/her daily routine. The perception of support, among other measures, is also assessed. Using this type of protocol, Linden, Chambers, Maurice and Lenz (1993) found that high levels of social support were associated with low systolic blood pressure (SBP) in women but not in men. Additionally, in a study of cardiovascular effects in the work place, low social support was associated with elevated heart rate (HR) and SBP during work, and these effects persisted during leisure time and rest (Unden, Orth-Gomer, & Elofsson, 1991). Finally, Spitzer et al. (1992) used 12-hour ambulatory recordings to compare the cardiovascular reactions of participants while in the presence of strangers, friends, or family members. Blood pressures were lowest when spending time with family members and highest while in the presence of strangers. Although this study was not specifically designed to evaluate support, one might argue that greater support would be expected from family members than from strangers, and this might, in part, account for these physiological differences. It is worth noting that although ambulatory studies (Linden et al., 1993; Spitzer et al., 1992; Unden et al., 1991) have demonstrated a strong relationship between social support and a number of cardiovascular parameters, they have not provided clear evidence of a buffering effect of support. One possibility is that ambulatory studies may be limited in their ability to detect such effects because they lack a clear baseline and do not directly manipulate support. Experimental laboratory studies do not suffer from these limitations, and are therefore in a better position to examine buffering effects of support on cardiovascular reactivity. Laboratory Studies Evaluating the Impact of Social Support on Cardiovascular Stress Reactivity To further elucidate the mechanism by which social support may protect cardiovascular health, experimental laboratory studies have been conducted (Allen, Blascovich, Tomaka, & Kelsey, 1991; Christenfeld et al., 1997; Edens, Larkin & Abel, 1992; Gerin, Milner, Chawla, & Pickering, 1995; Gerin, Pieper, Levy, & Pickering, 1992; Glynn, Christenfeld, Sanders, &  9  Gerin, 1996; Kamarck, Annuziato, & Amateau, 1995; Kamarck, Manuck, & Jennings, 1990; Kors, Linden, & Gerin, 1997; Lenz, 1995; Lepore, 1995; Lepore, Allen, & Evans, 1993; Sheffield & Carroll, 1994, 1996; Snydersmith & Cacioppo, 1992). In accordance with the reactivity hypothesis, these studies have examined whether laboratory manipulations of support buffer against cardiovascular reactivity that occurs in response to a psychological challenge. The typical paradigm of these studies is one in which participants perform a stressful task while receiving social support - usually operationalized as a form of emotional support. Participants in one or more support conditions are compared with an alone condition or with some other condition that is designed to be non-supportive. The laboratory challenges have varied across studies but can nevertheless be subdivided into two major categories: mental challenges, such as a mental arithmetic task (Allen et al., 1991; Edens et al., 1992; Kamarck et al., 1990, 1995; Kors et al., 1997; Sheffield & Carroll, 1994), stroop task (Kamarck et al., 1995), or mirror tracing task (Edens et al., 1992), and psychosocial challenges which require the individual to give a speech (Christenfeld et al., 1997; Glynn et al., 1996; Lenz, 1995; Lepore, 1995; Lepore et al., 1993) or engage in a discussion (Gerin et al., 1992), in which they must defend their opinions on a controversial issue (e.g., abortion). While both types of tasks lead to significant elevations in blood pressure and heart rate, the psychosocial tasks have been shown to elicit the strongest cardiovascular responses. Because of the nature of the tasks and the importance of matching supportive functions with the specific coping requirements of the stressor (Cohen & McKay, 1984; Uchino et al., 1996), the implementation of support has been somewhat different for these tasks. With the speech tasks, the support provision has usually included both verbal and nonverbal behaviors designed to provide validation of the participant's opinions. In contrast, researchers using arithmetical tasks have typically employed silent support, with few or no non-verbal gestures provided. In addition, the supporter is frequently prevented from being able to evaluate the participant's performance.  10  In thefirstlaboratory experiment evaluating the impact of support on cardiovascular reactivity, Kamarck et al. (1990) compared females who performed a stressful arithmetical task alone with those who performed the task in the presence of a close friend who was instructed to provide silent support. In addition, the situation was designed to ensure that the friend would be perceived as non-evaluative by the participant. Findings demonstrated a decrease in SBP and FfR in response to the provision of support and thus provided clear evidence of a buffering effect on cardiovascular stress reactivity in the laboratory. As the buffering effect could not be explained by either distraction or a general decrease in emotional arousal, the authors suggested that a number of cuesfromthe support situation directly affected physiological processes, dampening autonomic activity.  Psychological Mediation of Support It is worth noting that a number of studies which have followed the Kamarck et al. (1990) study have continued to demonstrate the physiological benefits of support while failing to detect changes in perceived stress or other psychological variables that might mediate the impact of support on cardiovascular reactivity (Edens et al., 1992; Gerin et al., 1992; Kamarck et al., 1990,1995; Kors et al., 1997; Lepore et al., 1993). Hence, it is not entirely clear whether support exerts a direct influence on reactivity or whether psychological processes may mediate the impact of support on physiology. One reason for the discrepancy between self-report and physiological measures may be that the self-report measures are relatively insensitive to such changes because of their retrospective nature (Gerin et al., 1992; Lepore et al., 1993). It is also possible that these measures are not assessing the psychological changes most valuable to the support process. Thus, at present, there remains a lack of clarity regarding psychological processes involved in the support process as well as how most effectively to measure them.  11  Variables Affecting the Support Process Although the Kamarck et al. (1990) study provides some evidence for one pathway through which support may exert a protective influence on cardiovascular measures, the results do not clarify what is necessary for a supportive interaction to be protective. Thus, one can speculate that the familiarity of the friend, the physical proximity of the friend, the meaning that the interaction has for the participant (Kamarck et al., 1990), and the fact that the situation was a non-evaluative one (Allen et al., 1991; Kors et al., 1997) may each have contributed to the buffering effect. Hence, the relative importance of each of these (and other) factors in supportive transactions remains unknown. The support experiments which have followed the Kamarck et al (1990) study have continued to examine the buffering process in the laboratory, in an attempt to determine whether there are certain characteristics of the supporter, the participant, or the situation that either facilitate or hinder the protective physiological effects of support. Thus, while similar in prhnarily using university student populations, these protocols have differed in a number of components, which have included: the supporter's relationship to the participant (i.e., friend or stranger), gender of the supporter, the degree to which the supporter could be perceived as "evaluative" by the participant, the degree of threat or stress present during the psychological tasks, and certain participant characteristics, such as cynical hostility. Asfindingsamong these studies have been inconsistent, with nine studiesfindingpositive effects of support (Christenfeld et al., 1997; Edens et al., 1992; Gerin et al., 1992, 1995; Kamarck et al., 1990, 1995; Kors et al., 1997; Lepore, 1995; Lepore et al., 1993) and six studies failing to obtain these effects (Allen et al., 1991; Edens et al., 1992; Lenz, 1995; Sheffield & Carroll, 1994, 1996; Snydersrnith & Cacioppo, 1992), it seems likely that a number of these factors may account for the differences. Current conclusions regarding a number of these components will be summarized below.  12  Relationship to supporter: In women, the nature of the supporter's relationship to the participant does appear to matter. Thus, it has been shown that one's relationship to a supporter (i.e., friendship) provides an additional physiological benefit beyond the specific supportive behaviors that are provided during a supportive interaction (Christenfeld et al., 1997). Similarly, when support is provided by a female friend, perceived closeness with the friend is associated with decreased SBP reactivity during a stressful mental arithmetic task (Kors et al., 1997). Not surprisingly, when participants are assigned to a support condition with either a same-sex friend or samesex stranger (i.e., laboratory confederates) condition, the friend condition has usually led to decreased cardiovascular reactivity relative to the stranger condition (Christenfeld et al., 1997; Edens et al., 1992; Snydersmith & Cacioppo, 1992). It should be noted, however, that strangers can also be effective "supporters" for women. In fact, Christenfeld et al. (1997) directly compared support conditions which included a supportive female friend, a supportive female stranger, and a non-supportive (i.e., neutral) female stranger, while the participant performed a 6-minute speech task. They found that although the supportive friend condition was associated with decreased SBP and diastolic blood pressure (DBP) response when compared with the supportive stranger condition, the supportive stranger condition showed lower reactivity on these measures when compared with the non-supportive condition. These results are consistent with the findings of other studies which also have found supportive strangers to reduce reactivity relative to an alone or nonsupportive stranger condition (Gerin et al., 1992; Lepore, 1995; Lepore et al., 1993). Unfortunately, there have been very few laboratory studies which have examined the effect of support on cardiovascular reactivity in men. Thus, it is not clear whether the nature of the relationship to the supporter has a differential impact on cardiovascular reactivity in men.  13  Gender of Supporter: Several lines of evidence suggest that participants may show greater physiological benefits when receiving support from female supporters than from male supporters. For example, Glynn et al. (1996) found that both males and females showed a decreased cardiovascular response when paired with a female support person (lab confederate), but neither benefited when paired with a male supporter (also a confederate). Similarly, Kirshbaum, Klauer, Filipp, and Hellhammer (1995) found that males who had their live-in gklfriends provide support during the anticipation of a stressor showed attenuation of a Cortisol response compared with unsupported men; women were not benefited by the presence of their boyfriends, and, in fact, displayed larger adrenocortical stress responses after receiving support from their boyfriends. These laboratory findings are consistent with the epidemiological literature which suggests : that compared to those who are unmarried, marriage affords a greater protective influence against mortality for men than for women (Berkman & Syme, 1979; House et al., 1982; Shumaker & Hill, 1991). Additionally, becoming widowed has been associated with increased risk of mortality for men, although there has been no effect for women. In accordance with these findings, the present study used a female supporter for participants assigned to the support condition.  Perceived Threat: It is worth noting that studies that have compared a supportive friend condition with an alone condition during mental challenges have shown an inconsistent pattern of results. While a number of studies have found a supportive friend condition to lead to decreased reactivity relative to the alone condition (Edens et al., 1992; Gerin et al., 1995; Kamarck et al, 1990, 1995; Kors et al., 1997), other studies have found no differences between the two conditions (Sheffield & Carroll 1994,1996; Snydersmith & Cacioppo, 1992), and one study actually found greater reactivity in thefriendcondition than in the alone condition (Allen et al., 1991).  14  One explanation for these differences may be that support effects during a mental challenge, such as a mental arithmetic task, are most likely to occur if the evaluation potential of the task is minimized (Allen et al., 1991; Kors et al., 1997; Uchino et al., 1996). This is most likely to occur if thefriendsare preventedfromseeing/hearing the participants' questions and/or answers, and, hence, are perceived as non-evaluative by the participants. In accordance with this hypothesis, most of the studies which found the friend to be a beneficial supporter designed the friend condition as one in which thefriendwould be clearly perceived as nonevaluative (Edens et al., 1992; Kamarck et al., 1990,1995; Kors et al., 1997). In contrast, protocols which have found no physiological benefits or even a negative effect of support (Allen et al., 1991; Sheffield & Carroll, 1994,1996) have employed protocols that fail to minimize the participants' perceptions of evaluation. Kors et al. (1997) directly tested this hypothesis by comparing an alone condition, a supportive "non-evaluative" friend condition, and a supportive "evaluative"friendcondition. Results were in accordance with these hypotheses and demonstrated that only participants in the non-evaluative condition were physiologically bufferedfromthe stressfulness of the task. Furthermore, several studies have shown that the stressfulness of the task, as well as the social threat due to the experimenter's perceived status and behavior, may influence the extent to which social support affects cardiovascular response (Gerin et al, 1995; Kamarck et al., 1995). It has been speculated that studies which have failed to find differences between the friend and alone conditions (Sheffield & Carroll, 1994; Snydersrnith & Cacioppo, 1992) did not use an explicit manipulation of social threat within the experiment.  Participant Characteristics: Although participant characteristics have been largely ignored in laboratory studies of support, one study has examined whether individual differences in cynical hostility (i.e., hostility involving a pervasive mistrust of others) moderate the impact of social support on cardiovascular response (Lepore, 1995). Participants who were high or low in cynical hostility 15  were asked to give a speech, in either a support or an alone condition. Results showed that individuals characterized by high levels of cynical hostility did not benefit from social support to the same extent as those with lower levels of this trait. Findings from this study suggest that individual differences may indeed influence the impact of support on cardiovascular reactivity. Hence, rather than exclusively focusing on situational aspects of the support paradigm, a trait X situation approach appears to be warranted (Mischel, 1968, 1973). In fact, such an approach may help to reconcile the inconsistent findings within this body of research. It is important to mention that a trait X situation approach has been frequently employed in the literature examining the relationship between anger/hostility and cardiovascular response. Along these lines, a recent meta-analysis found that reactivity differences between persons varying in antagonism were only evident in response to anger-provoking stressors. No differences emerged when participants were exposed to a non-provocative manipulation (Suls & Wan, 1993). Furthermore, a match-mismatch hypothesis has been proposed which suggests that following an anger-inducing situation, a match between the behavioral response of anger expression/suppression with one's actual anger-expression/suppression style will decrease reactivity relative to a mismatch between the behavior and the person's expression style (Engebretson, Mathews, & Scheier, 1989). In support of these hypotheses, these authors found that anger- expressors and anger-suppressors who were given the opportunity to express/suppress anger consistent with their expression style showed more rapid recovery following a harassment task than those who were expected to behave in a way that contradicted their natural style of coping with anger (Engebretson et al., 1989).  Goals of this study The construct of social support seeking has been examined by researchers wanting to gain a better understanding of the process by which social support is provided. It has been hypothesized that individuals are not just passive recipients of support, but, rather, an  16  individual's desire for support may influence whether support is actually provided (e.g., Conn & Peterson, 1989; Gottlieb, 1983; Thoits, 1986). It is clear that individuals differ in their need for support; some individuals may desire high levels of support, whereas others may not want support and may avoid it, when offered (e.g., Butler, Giordano, & Neren, 1985; Conn & Peterson, 1989; Dunkel-Schetter, Folkman, & Lazarus, 1987). Thus, researchers have tried to examine what personality variables or person X situation factors may influence the use of support seeking as a strategy to manage stressful situations (e.g., Amirkhan, 1995; Barbee et al, 1993). Support seeking is a construct that has been overlooked in the support-cardiovascular reactivity literature. Since individuals differ in their desire for support (as described above), it seems likely that level of support seeking may play an important role in the relationship between social support and cardiovascular stress reactivity. Thus, whether support is of physiological benefit may depend on whether one actually wants it or not. Along these lines, this study was designed to test a matching hypothesis within a social support-reactivity laboratory paradigm. This hypothesis specifies that a match must occur between one's support seeking tendencies (high or low support seeking) and the actual provision of support (i.e., presence or absence of support) in order for physiological benefits to be obtained. If a mismatch occurs, benefits are expected to be less likely to occur. In order to test this hypothesis, men and women who were high or low in support seeking (using gender-specific quartiles) were assigned to either a support condition or a non-support condition (i.e., alone condition) as they engaged in two psychological laboratory challenges. It was predicted that high support seekers assigned to a support condition would show decreased reactivity relative to all other matched or mismatched conditions. This would suggest that people who are receptive to support in their daily lives will show attenuated stress reactivity when provided with support in the laboratory. This study also explored possible gender differences. Although previous studies have not found clear gender differences in response to the provision of support, conclusions are 17  difficult to draw because of the small number of studies that have included both male and female participants. Most previous studies have recruited only women. Witbin the support seeking literature, however, significant gender differences have been reported, with females tending to show greater support seeking than males (Ashton & Fuehrer, 1993; Barbee et al, 1993; Butler, Giordano, &Neren, 1985), particularly in the seeking of emotional support (Ashton & Fuehrer, 1983). Consequently, such differences could lead to gender differences in the match-mismatch paradigm of the present study. It is worth noting that this study was intentionally designed with female confederates as the supporters for both male and female participants. Although support provided by a stranger may be viewed as less ecologically valid than support provided by a friend, the decision to use female confederates was made because of the need to find an appropriate supporter for the male participants. Along these lines, it was not clear if men would show a reduction in cardiovascular reactivity when receiving support from friends, because only one previous study had examined support provided by male friends, and this study demonstrated no physiological benefits for men in this condition (Sheffield & Carroll, 1994). Additionally, no prior studies had tested whether men physiologically benefitfromthe support of a female friend. While several studies had shown physiological benefits for men receiving support from strangers (Glynn et al., 1996; Lepore, 1995; Lepore et al, 1993), one study showed that men (and women) benefited only when supported by a female confederate, and not by a male confederate (Glynn et al., 1996). Taken together, thesefindingssuggested that supportfroma female confederate would be associated with reductions in cardiovascular reactivity in male participants. Because a strong test of the match-mismatch hypothesis requires the implementation of support to be maximized, the use of a female confederate was viewed as the best option for this study. Finally, it is important to note that previous laboratory studies exaniining the relationship between support and cardiovascular response have employed either a psychosocial stressor or a mental challenge task. This study added to the literature by incorporating both kinds of 18  tasks, in an attempt to generalizefindingsto a broader range of psychological stressors. In addition, the research was novel in its examination of the impact of verbal support on cardiovascular measures during a math task.  c  19  Method  Overview This study involved two phases: a screening phase and a laboratory phase. During the screening phase, participants completed a questionnaire that assesses social support seeking. Participants who were either high or low support seeking (i.e., in the top and bottom quartiles, using gender-specific cuts) were recruited for the laboratory phase. In this phase, participants performed two stressful tasks (math task and speech task), while randomly assigned to a condition that either matched (e.g., support provision for support seekers) or mismatched their support seeking style. In the alone condition, participants did not receive support and performed both tasks without anyone else in the room. In the support condition, participants performed the tasks in the presence of a supportive female laboratory confederate, who was introduced as another participant. The confederate was trained to provide both verbal and non-verbal support, and support was equated in terms of specific supportive comments provided, as well as non-verbal behaviors such as eye contact, posture, proximity, and facial expression. Assignment to an alone or support condition remained consistent across both tasks. In addition, participants assigned to the support condition received support during the speech preparation phase. Prior to and following the math and speech tasks, participants (and confederates, when present) completed a number of self-report measures. Blood pressure and heart rate were monitored continuously during all phases of the study.  Participants Four hundred and eighty participants were included in the screening phase of this study. Since only the upper and lower quartiles for support seeking were used, a total of 240 of the screened participants were eligible for the study. From these participants, 140 agreed to participate in the laboratory phase. They were recruitedfromthe University of British  20  Columbia undergraduate psychology participant pool and were awarded class credit for their participation. The data from five participants who completed the laboratory phase of the study had to be excluded from the analyses. In two cases, no cardiovascular measures could be obtained during the speech task due to instrument failure of the blood pressure monitor. Additionally, three participants were provided with less than the minimum number of required support statements for the speech task (see section on Support Implementation). Thefinaldata analyses were based on 135 participants (68 women; 67 men). From this data set, the number of participants in each condition was as follows: 34 in the low support seeking/alone condition, 35 in the low support seeking/ support condition, 30 participants in the high support seeking/alone condition, and 36 participants in the high support seeking/support condition. Inclusion and exclusion criteria for laboratory phase. In order to qualify for the laboratory phase, participants were either high or low in support seeking (falling within the top or bottom quartiles), as measured by a support seeking measure (the Coping Strategy Indicator). Prior to scheduling the laboratory phase, participants were asked about thenphysical health and the use of various medications. Participants with essential hypertension, cardiac disease, kidney disease, and diabetes were excluded, as well as those taking medications with cardiovascular effects. These criteria resulted in the exclusion of two participantsfromthe laboratory phase of the study. Demographic composition of sample. The demographic characteristics of the sample are described in Tables 1 and 2. Participants ranged in age from 17 to 43 years (M = 20 yrs., SD = 3.1). Each participant was asked to specify his/her ethnic origin on a demographic questionnaire, and the typical answers were "Chinese", "Canadian", and "Korean." Based on this information, participants were coarsely subdivided into two groups, with a group called  21  Table 1. Men: Demographic Data, Support Seeking Scores, and Perceived Support. low ss, alone Age Generations in North America  low ss, support  high ss, alone  high ss, support  19.4  (2.1)  20.6  (3.1)  19.1  (1.3)  19.4  (2.2)  2.2  (2.2)  1.8  (1.1)  1.9  (1.3)  2.1  (1.4)  57.9 % 31.6% 5.3 % 5.2 %  A: 64.7% W: 35.3%  A: 62.5 % W: 31.3 % M: 6.2%  CSI Score  16.9  (1.9)  14.4  (3.0)  29.4  (2.4)  29.7  (2.0)  SSI Score  42.5  (7.0)  39.9  (9.2)  50.6  (6.3)  47.4  (7.5)  ISEL Score  31.2  (8.2)  35.2  (6.2)  35.7  (7.0)  35.1  (9.2)  ss = support seeking support = support condition alone = alone condition  A: 60 % W: 40 %  A: W: M: U:  Ethnic Composition (%age of group)  A = Asian M = Mixed W= White U = Unknown CSI = Coping Strategy Indicator SSI = Social Support Inventory ISEL = Interpersonal Support Evaluation List  "White" which included participants of European origin, and another group, called "Asian", which included those who had identified with Asian origin (i.e., South Asian, Japanese, Korean, Chinese, etc.). The ethnic composition of the sample was predominantly Asian, with Asian students comprising a larger portion of each group and of the overall sample. The demographic composition of this sample, and in particular the large Asian contingent, is typical of the demographic composition of students attending this university. On average, participants reported that their families had lived in North America for a total of 1.8 generations (SD = 1.3.) (see Appendix 4 for Demographic Questionnaire.)  22  Table 2. Women: Demographic Data, Support Seeking Scores, and Perceived Support. low ss, alone  low ss, support  high ss, alone  high ss, support  20.4 (6.1)  21.3  (3.9)  19.5  (1.4)  20.4  (4.6)  Generations in North America  1.4 (.72)  2.1  (1.6)  1.7  (.88)  1.3  (1.0)  Ethnic Composition (%age of group)  A: 58.8 % W: 35.3 % N: 5.9 %  A: 52.6 % W: 42.1 % N: 5.3 %  A: 66.7% W: 26.7% U: 6.6 %  A: 64.7% W:23.5% M: 11.8 %  CSI Score  19.7 (2.9)  18.7  (2.9)  31.7 (.80)  32.2 (.81)  SSI Score  48.3  (9.5)  48.2  (9.8)  55.3  (7.4)  54.7  (6.5)  ISEL Score  33.5 (8.7)  32.3  (8.2)  39.4  (6.3)  34.6  (8.9)  Age  ss = support seeking support = support condition alone = alone condition  A M W U N  = Asian = Mixed == White = Unknown = Native American  CSI = Coping Strategy Indicator SSI = Social Support Inventory I S E L = Interpersonal Support Evaluation List  Psychological Measures Support Seeking. The screening measure was the Coping Strategy Indicator (CSI), which is a measure of coping strategies that individuals employ when experiencing stress (Amirkhan, 1990). The CSI requires participants to briefly describe a stressful event that has occurred within the past 6 months. Respondents then indicate the degree (on a 3-point scale) to which each of 33 specific coping behaviors was used to deal with the event. In addition to providing an overall coping profile, this measure provides three partially independent coping strategies, each forming a separate scale. These scales include seeking support, problem solving, and avoidance (see Appendix 5 for CSI).  23  The seeking support subscale was of primary interest and was used as the basis for recruitment. Norms for high and low support seeking in the student sample were obtained from an initial screening of 480 students. Thereafter, high and low support seekers were selected based on their scores falling within the upper and lower quartiles relative to these norms. As gender differences on this subscale were observed, gender specific cuts were used to determine separately the upper and lower quartiles for men and women. The upper and lower quartilesfromthis screening were as follows: high and low support seeking men had scores of > 27 and < 19, respectively; high and low support seeking women had scores of > 31 and < 22, respectively. Figure 1 shows the range of scores that fall within the upper and lower quartiles (which are shaded) and the range of scores that fall within the two middle quartiles (which are not shaded). The CSI has been shown to be unaffected by social desirabihty andfreeof demographic influence, with the exception that gender was related to scores on the seeking support scale (Aniirkhan, 1990). Rehability of this measure has been shown to be good, with mean testretest correlations of .82 for the entire CSI and .80 for the seeking support scale (with a student sample). Internal consistency is reported as equivalent or superior to that found for other coping measures (Cronbach's a coefficient = .89 for the entire scale and .93 for the seeking support scale). The vahdity of this measure has been demonstrated in tests of criterion vahdity. For example, one test showed that choice of a coping strategy during stressful conditions was highly associated with the coping preference indicated by the CSI (Amirkhan, 1994). Additionally, the seeking support scale was correlated with total number of supports, providing some evidence of convergent vahdity for this subscale (Amirkhan, 1990).  24  Men F i g u r e 1.  Women  The range of s c o r e s falling within  the upper and lower support seeking quartiles for men and  women.  In order to provide further validation for the measurement of support seeking with the CSI, the Social Support Inventory (SSI), a pilot measure of support seeking newly devised for this study, was also achmnistered. Specifically, it was important to determine if the high and low support seeking participants classified by the CSI were similarly categorized by the Social Support Inventory (SSI). In order to prevent responses on the seeking support scale of the CSI from influencing responses to the SSI, the SSI was completed by participants at a later time (i.e., during the initial baseline period of the laboratory phase). The SSI is comprised of seven vignettes, with each vignette describing a different type of stressor (e.g., performance, relationship, daily hassle, etc.). For each vignette, the participant is asked to rate the degree to which he/she would actively seek support and the degree to which he/she would accept support that is provided. Both questions are answered with a 5-point rating scale, with higher numbers reflecting higher levels of support seeking. From the SSI, a total overall support seeking score is obtained, as well as separate scores for the actively seek and accept subscales.  25  The mean CSI and SSI scores for each group (for male and female samples) are presented in Tables 1 and 2 (see Appendix 6 for the SSL). Social Support. During the initial baseline period of the laboratory phase, participants completed the Interpersonal Support Evaluation List (ISEL), (Cohen, Mermelstein, Kamarck, & Hoberman, 1985). The ISEL is a 48-item scale, presented in a true/false format, that assesses perceived availabihty of support throughout the individual's support network. The scale is subdivided into four 12-item subscales which measure appraisal, belonging, tangible, and self-esteem support, in addition to providing an overall support score. Only the overall support score was included in the data analyses for this study. An evaluation of the measure's reliability shows that test-retest reUabihty (for the entire scale) is good (r = .87) and internal consistency is strong, with coefficient a rangingfrom.88 to .90. Internal consistency of the subscales has typically rangedfrom.62 to .82. As a measure of construct vahdity, the ISEL shows a correlation of .46 with the Inventory of Socially Supportive Behaviours (ISSB). As a measure of discriminant vahdity, the ISEL shows correlations of-.52 to -.64 with various measures of social anxiety. Additionally, this measure has been found to befreeof social desirabihty bias (see Appendix 7 for the ISEL).  Affective State. The Stress/Arousal Adjective Check List (SACL; King, Burrows, & Stanley, 1983) was completed during the baseline and recovery period for each task, in order to examine task-related changes in stress and arousal. This measure consists of a list of 20 adjectives rated on a categorical scale (++ = definitely yes; + = slightly yes; ? = not sure or don't understand; - = definitely not). In scoring this questionnaire, a (++) or (+) associated with a positive adjective is scored as 1 and scored as 0 when associated with a negative adjective. A (?) or (-) circled for a negative adjective is scored as 1 but scored as 0 for a positive adjective. This measure comprises two orthogonal subscales: stress and arousal, with higher scores indicating greater perceived stress and greater perceived arousal. The arousal  26  scale is indicative of a coping response to perceived demand, whereas the stress scale appears to reflect one's response to perceived threat combined with a dnninished belief in one's ability to cope. Internal consistency of this measure is adequate, with a = .86 for the stress subscale and a = .74 for the arousal subscale. VaHdity has been demonstrated by showing that this measure was able to differentiate between groups of people expected to differ in stress and arousal during different stressful situations (King et al., 1983). This measure was used in this study because it appears to be somewhat more sensitive to changes in perceived stress than are other measures reported in the literature. For example, a previous study (Lepore, 1995) found that changes in perceived stress on the SACL differentiated between supported and non-supported participants (see Appendix-8 for the SACL). Controversial Issues Questionnaire. In order to assign a topic to be discussed during the speech task, participants were asked to complete the Controversial Issues Questionnaire. This questionnaire, designed for this study, lists eight controversial issues (e.g., abortion, gun control) and requires participants to rank the three issues which are of greatest concern to them. Participants were informed that these rankings should reflect both the strength of their opinions and their level of knowledge about the issue. Participants were then instructed to give a speech about the issue ranked as their top choice (see Appendix 9 for the Controversial Issues Questionnaire). Additional state measures. Immediately following the math and speech task, participants completed a variety of 10 cm visual analog scales which assessed the participant's reaction to the task, and when appropriate, the participant's response to the support and to the laboratory supporter. For participants in both conditions (i.e., support and alone conditions), measures included degree of stress during the task and the participants' perceptions of their task performance. Participants assigned to the support condition completed several additional scales which assessed their reactions to the provision of support. These scales included the  27  perception of support, perception of evaluation, and degree of happiness in having received support. Additionally, participants in the support conditions were asked to evaluate the supporter by indicating how much they liked the supporter, how much they would want to become friends with the supporter, and the degree to which they believed that the supporter is supportive towards her own friends. The participants' evaluation of the supporter was also assessed by having them rate the supporter on a number of general dimensions, including Mendliness, warmth, and acceptance (see Appendices 10, 11, and 12 for the visual analog scales). Task Performance. In order to examine whether potential differences in self-report and cardiovascular reactivity were influenced by differences in task performance, the math task performance, math task effort, and speech task performance (i.e., comfort during the speech) of the four groups were examined. Math task performance was evaluated by the total number of correct responses, and math task effort was measured by the total number of responses. Speech performance was assessed by the experimenter on a 1-5 scale (with a score of 5 indicating the highest rating), with the rating reflecting the participant's degree of apparent comfort in speaking for the 5-minute period (see Appendix 15 for the speech performance evaluation).  Cardiovascular Measures Systolic blood pressure (SBP), diastolic blood pressure (DBP) and heart rate (HR) were measured with a Dinamap 845 automated blood pressure monitor (Critikon Corp., Tampa, FI.). Standard occluding cuffs were attached to the participant's non-dominant arm, following the signing of consent forms. Participants were asked to refrain from moving their nondominant arm during cuff inflations, as any movement could interfere with the reading. To  28  facilitate the participant's perception of the confederate as another participant in the support condition, the confederate was also hooked up to a blood pressure monitor, but readings were taken only from the participant. All cardiovascular measures were monitored during baseline, task, and recovery phases of the study. Specifically, cardiovascular readings included four readings during the adaptation period, five readings during the math task, two readings during the speech preparation, three readings during the speech task, and three readings taken during each of the recovery periods which followed the math and speech tasks. The last two readings of the 20-minute adaptation period (minutes 14 and 19) served as a baseline for the math task, and the last two readings of the math task recovery period served as a baseline for both the speech preparation phase and the speech task (see Appendix 16 for data collection sheet).  General Procedure The sequence of phases in this experiment was as follows: Initially, there was a 20-minute adaptation period, in which the last 10 minutes constituted the pre-task baseline. The 10minute math task then took place, followed by a 10-minute recovery period. Thereafter, a 5minute speech preparation period and 5-minute speech task occurred, followed by afinal5minute recovery period. Debriefing occurred at the termination of the experiment (see time line for order of study phases). The math task was givenfirstbecause it produces less cardiovascular reactivity than the speech task. Thus, the novelty associated with it being thefirsttask was expected to increase the reactivity of the task. Specific instructions regarding each task were provided immediately prior to the task. To legitimize the confederate's role as another student (in the support condition), task instructions to the confederate were provided infrontof the participant.  29  STUDY PHASES  SCREENING PHASE 480 participants  LAB PHASE (135 participants)  low SS alone  low ss support  high ss alone  high ss support  Baseline 10 minutes Math Task - 10 minutes Recovery Period/Baseline for Speech preparation Phase and Speech Task - 10 minutes Speech Prep. Phase - 5 minutes Speech Task - 5 minutes Recovery Period - 5 minutes  30  Setting All phases of the study took place in a sound-proofed room, measuring 3m X 4m. Against one wall, there was a long table which held the television monitor. On a wall perpendicular to this wall, there was a one-way mirror which measured l m X l m Participants were told that the experimenter was seated behind the mirror and would be able to observe the participant during the entire experiment. Additionally, participants were told that an intercom allowed the experimenter to hear them speak, as they read out loud the math equations and while they gave a speech. Participants were seated about 2.2 m from the oneway mirror, on the same chair throughout the study, although the positioning of the chair (i.e., angle of the chair in relation to the mirror) was somewhat different in the two tasks (see below). In addition, a video camera, situated on a tripod, was located in the far corner of the room, 2.2 m from the participant. In the mental arithmetic task, the participant directly faced a television monitor displaying the task, and was seated approximately 1.2 m awayfromthe monitor. Because of the necessity of facing the monitor, the participant was facing the one way mirror at a 45 degree angle. In the support condition for this task, the confederate was seated facing the participant, within the participant's field of view at a 45 degree anglefromthe participant. Chairs of the participant and confederate were approximately 1.5 m apart. During the speech preparation and speech task, the participant's chair was moved so that the participant directly faced the one way mirror. In the support condition, the confederate remained at a 45 degree anglefromthe participant. All other distances were the same as in the math task.  Implementation of Support (for support condition) In the support condition, the implementation of support was provided by a female confederate, with a total of four female confederates used in the study. The confederates  31  were upper-level undergraduate students or women who had recently completed an undergraduate degree. Three of the confederates were White, and one confederate was Asian, although the Asian confederate acted as the supporter for approximately 40% of the participants assigned to a support condition. Each confederate served in both support seeking conditions in order to minimize the risk of obtaining support effects that were associated with a particular confederate. Additionally, in order to decrease any potential impact on study results, confederates were provided with minimal knowledge of the study hypotheses and, during each experimental session, were not informed as to whether they had been assigned to a high or low support seeking participant. During debriefing (see section below), participants were encouraged to not discuss the role of the confederate with other students, in order to reduce the likelihood that subsequent participants would have knowledge of the confederate. All confederates were trained to provide support in terms of specific supportive verbal statements, as well as supportive nonverbal behavior. This supportive input was designed to provide emotional support, thereby providing validation and enhancing self-esteem. A list of supportive statements and nonverbal gestures for each task (and for the speech preparation phase) was provided for the confederate to choose from. Specifically, the supportive statements (for the math and speech tasks) were listed within a number of different categories, which were designed to represent somewhat different ways of providing emotional support (e.g., validating the participant's abilities, validating the participant's effort, and encouraging the participant's perseverance during the task). There were 2-6 statements listed within each category, and the confederate was expected to choose one statementfromeach category (i.e., five statements in total) (see Appendix 19 for list of support statements). The list of statements was designed in this way so that the provision of support would be consistent for participants within this condition, while also allowing the statements to be tailored to the individual participant. During the math and speech tasks, the confederate was expected to say a total offivesupportive statements in addition to employing two nonverbal gesturesfromthis list. During the speech preparation phase, the confederate was expected to say three 32  supportive statements and to use two nonverbal gestures. As a manipulation check, the number of statements/non-verbal behaviors that were actually provided were monitored by the experimenterfrombehind the one-way rnirror (see Appendix 13). Only participants who had received a minimum of four supportive statements and two non-verbal behaviors (for the math and speech tasks) were included in the data analyses. In providing support for both tasks, every effort was made to keep the provision of support similar across the two tasks. Along these lines, the non-verbal gestures for both tasks included the following: demonstrating an open body posture, while conveying interest in the participant's task performance; maintaining adequate eye contact, without appearing to stare at the participant; smiling and nodding periodically throughout the math task and at appropriate times during the speech task. In addition, there were a number of general supportive statements which were focused on helping the participant to cope with the stressfulness of the task (and complete the task) and could therefore be used in both tasks. Examples of such statements are "Hang in there", and "You only have 2 minutes to go." As the tasks were quite different, however, several statements were specific to each task (see sections on Math Task Support and Speech Preparation/Speech Task Support).  Specific Procedures Pre-study procedures. Prior to the study, participants were asked to refrain from smoking, exercising, chinking alcohol, or ingesting caffeine for at least 2 hours prior to their laboratory session. This was necessary as these activities may increase reactivity to the laboratory tasks. Participants were contacted by telephone to remind them of these instructions on the evening prior to their study visit. Participants in the support condition were additionally informed that there would be another student also participating in the same laboratory session.  33  Waiting period. Upon arrival at the laboratory, participants in all conditions were informed that the experimenter was nrnning several minutes behind schedule and were asked to wait in a waiting area down the hall from the lab.  The waiting period was 5 minutes and  was included because of the need to have a rapport building phase for the support condition (see below). Participants in the alone condition waited alone, and after 5 minutes, were greeted by the experimenter and taken to the laboratory. In the support condition, the confederate, acting as the other participant, was also instructed to wait in the waiting area after arriving at the laboratory. The confederate asked the participant if he/she was waiting for the same study and then introduced herself as the other participant. Following this introduction, the confederate used the 5-minute waiting period to make friendly conversation with the participant. This period of interaction was viewed as important to help build rapport. As no previous studies have had strangers provide verbal support during a math task, it seemed that the rapport building phase could help to facilitate the participant's perception of support during this task.  Study description. Following the waiting period, participants were taken to the lab, given a general description of the study, and asked to sign consent forms. Participants (and confederates, when present) were also asked to complete the Controversial Issues 1  Questionnaire. In the alone condition, participants were told that the purpose of the study was to examine cardiovascular responses as they perform a stressful mental arithmetic task, followed by a speech task. In the support condition, participants (in the presence of confederates) were given the same description but were informed that one student would perform the task while receiving support from the other student. They were additionally told that cardiovascular measures would be takenfromboth task performer and task supporter, and that their specific roles in the task would now be assigned by a random draw.  The confederate was asked to complete this form to make her role more legitimate. 34  Role assignment (for support condition). Immediately foUowing the study description, the experimenter assigned the roles of task performer and task supporter. In order to assign these roles, the experimenter presented the participant and confederate with a choice of two envelopes to select from. One envelope had the word "performer" inside; the other had the word "supporter" in it. Although the experimenter appeared to shuffle the envelopes in front of the two participants, the position of the envelopes was rigged so that the "supporter" envelope was always placed in a prearranged position. The confederate, aware of this positioning, was asked to choosefirstand selected the envelope containing the supporter role, leaving the performer assignment to the participant. Adaptation period. After the role assignment, a 20-minute adaptation period took place. This period was necessary to allow participants to adjust to the laboratory environment. During this time, participants (and confederates) completed several self-report measures. If questionnaires were completed before the end of the adaptation period, participants were encouraged to relax or to look through magazines which were placed near their chairs. r^uring the first 10 minutes of adaptation, all participants remained alone in the room. In the alone condition, participants continued to remain alone for the duration of the adaptation period, whereas in the support condition, the confederate was present for the last 10 minutes of this period. This 10-minute baseline with the confederate was necessary to ensure that resting cardiovascular measures were not affected by the presence of the confederate, in the absence of a task challenge. In order to make the confederate's absence during thefirst10 minutes appear legitimate, participants and confederates were told that they needed to be monitored separately for the first 10 minutes of the task.  Role instructions. Immediately prior to each task, the experimenter entered the room and provided some general instructions to the participant and confederate regarding their roles during the tasks; The participant was informed that he/she would be performing difficult  35  tasks, during which time he/she would be supported by the "other student." The confederate, in the presence of the participant, was instructed to act as supportive as possible, in whatever manner seemed natural. Thus, it was stressed that she should try to incorporate both verbal statements of support, as well as non-verbal supportive gestures. For the speech task and speech preparation, she was instructed to plan her support statements for times that appeared appropriate. For the math task, she was told that non-verbal gestures could be provided at any time during the task but supportive statements should be provided strictly during the rest intervals (see description of task below), as statements during the actual mental calculations could distract the participant. Mental arithmetic task. The 10-rninute math task consisted of 50 equations delivered by video to a television screen. Questions were delivered at a rate of one question every 12 seconds and were a mixture of addition, subtraction, multiplication, and division. Participants were instructed to read the question out loud and then write their answer on a designated answer sheet. They were expected to calculate the answer in their head, as no written calculations were allowed. Participants were instructed to work as quickly as possible and were told that their performance would be evaluated based on the accuracy of their answers. They were expected to proceed to the next question if they were not able to answer the previous question within the allotted 12 seconds. Approximately every 2 minutes (or after every 10 problems) there was a 6-second rest interval, in which no problems were presented on the screen. These intervals were incorporated so that the support statements (in support condition) could be provided in a way that was not distracting to the participant. In order to maintain the reactivity of the task for the entire 10-minute duration, questions became increasingly difficult in the last 5 minutes of the task. Additionally, to increase the perceived stressfulness of the task, participants were told that there would be a prize for the individual who demonstrated the best performance on this task. (The prize was a coffee mug, although participants were not given a description of the prize at the time of the experimental session.)  36  It is worth noting that the participant was instructed to say only the equations aloud (instead of both equations and answers) in order to minimize the evaluation potential of the task and hence, maximize support. The participant was expected to say the questions, as opposed to the answers, because it was important to maximize the speaking time during the math task so that it was roughly equivalent to the speaking time of the 5-minute speech task. Furthermore, the math task in this study was longer than the 5-minute math task which has frequently been used in the stress-reactivity literature. A longer task was viewed as essential, because previous research suggested that supportive effects may not become evident until the last minute of a 5-minute mental arithmetic task (Kors, Linden, & Gerin, 1997). To provide a strong manipulation of support, the length of the math task was extended to 10 minutes. This task was a combination of a 5-minute math task that our lab had used previously (and is known to reliably trigger increases on all cardiovascular measures) and an additional 5-minute task, which functioned as the second half of the task and consisted of more difficult problems. Piloting of this new 5-minute task demonstrated that the task led to an increased cardiovascular response which was maintained for the duration of the task and was not attenuated with the addition of the 6-second rest periods. The 10-minute recovery period following the math task was expected to be sufficient to allow blood pressure and heart readings to return to baseline, as cardiovascular recovery from a math task typically occurs within 1-2 minutes following the task. In the event that thefinalrecovery reading was not within 10% of the pre-task baseline reading, the participant's data for this task were not used. However, this criterion did not result in the exclusion of datafromany participant.  Math task support. Louring the math task, the specific supportive statements were expressed during the 6-second rest intervals which occurred approximately every 2 minutes during the task. This appeared to be essential, because pre-study piloting indicated that supportive statements provided in a math task that had no rest intervals were perceived as extremely distracting and viewed as more of an annoyance than a source of support.  37  Because of the importance of m i n i m i z i n g evaluation potential in the math task, statements were chosen that were less likely to be perceived as evaluative and that would not be expected to lead to a negative reaction, if the participant was experiencing difficulty with the task. Additionally, the confederate was asked to complete several brief questionnaires during the task, so that the participant would not feel that he/she was being stared at. This was necessary because pre-study pilot work suggested that it is much more natural for the confederate to continuously look at the participant while the participant is speaking than while the participant is calculating mental arithmetic problems. Speech task. The speech task involved two phases: the speech preparation and the actual speech task. For the support condition, the confederate was present during both phases of the speech task. I)uring the preparation period, participants had 5 minutes to prepare a speech on a controversial topic. Specifically, they were instructed to prepare a speech on the controversial topic about which they indicated their strongest opinions, as determined by the Controversial Issues Questionnaire. They were encouraged to use this period to think of ideas and to plan the speech. They were given a notepad to write down some notes and to organize their speech. They were also given a sheet of helpful hints, which provided some general strategies for preparing a speech, (see Appendix 14) During the speech task, participants were required to give a 5-minute speech on the controversial topic in either the alone or support condition. Regardless of condition, they were told that the experimenter was observing the speech from behind the one-way mirror and would be evaluating their speech. Participants were informed that the evaluation would depend on the quality of their arguments and the evidence they used to defend their position, rather than on the actual stand that they took. Participants were instructed to speak as continuously as possible for the full 5-minute period. To further increase reactivity,  38  participants were positioned facing a video camera and were told that the speech was being taped. Similar types of speech tasks have been used previously in studies exarnining the impact of support on cardiovascular reactivity (Christenfeld et al., 1997; Lepore, 1995; Lepore, 1995), and our laboratory has recently used this task in a similar manner (Lenz, 1995). Findings consistently show this type of task to lead to substantial increases in blood pressure and heart rate (Christenfeld et al., 1997; Lenz, 1995; Lepore, 1995; Lepore etal., 1993). Speech preparation/speech task support. Support provided during the speech preparation phase was designed to help the participant to appraise the impending speech task as less stressful and to help the participant feel more confident in his/her ability to prepare and give a speech. Comments, such as "Try to not get too stressed about this" or "Don't worry - it will be over before you know it" are examples of statements that were used during this phase. During the actual speech task, supportive statements were provided at regular intervals throughout the task. (As it was only a 5-minute task, there was approximately one comment every minute.) The focus of support was on validating the participant's abilities and on enhancing the participant's ability to cope with the stressfulness of the situation (and to complete the task), rather than on agreement, per se, with the participant's opinions. Thus, the confederate verbalized statements such as: "Good point," "Wow, I don't know if I could think off the top of my head like that" and "You're almostfinished."If the participant requested assistance in any phase of this task or during the speech preparation phase, the confederate made it clear that she could not help the participant by saying "I think you can probably do a lot better without my help." Debriefing. After completion of thefinalrecovery period, the participant and confederate were unhooked from their respective blood pressure cuffs, and debriefing took place. The experimenter indicated that it was necessary to explain the study to each of them individually,  39  and therefore took the participant into a separate room to do this. Immediately prior to the debriefing, participants were asked to indicate their understanding o f the study and its hypotheses. Thereafter, during the debriefing, the participant was given a detailed explanation of the study. Specifically, the participant was informed that the purpose o f the study was to determine i f a match between support seeking style and the provision or absence of support was necessary in order to obtain the greatest cardiovascular benefits. Participants were informed that the other "participant" in the study was actually a research assistant o f the laboratory who had been trained to act in a particular way. Participants were asked whether they were aware of the deception or had any idea of the study's purposes. Towards the end o f the debriefing, the participants were reintroduced to the confederate and had the opportunity to ask her any questions that they might have had about her role or behavior during the study.  Participants were additionally given the  laboratory phone number to contact i f they wanted to receive a copy of the study results following completion o f the study. A l l participants were strongly encouraged to not discuss the study with classmates, as this could affect the results o f the study. In addition to this oral debriefing, participants received a written debriefing form to take with them.  40  Analytic Strategy Cardiovascular Data All cardiovascular data were analyzed using residualized change scores in a betweenwithin analysis of variance (ANOVA). Support seeking group (high vs. low support seeking), support condition (support vs. alone) and gender (men vs. women) constituted the between factors, and task (math task vs. speech preparation phase vs. speech task) was the within factor. The use of residualized change scores was viewed as advantageous for controlling potential baseline differences. Additionally, a weak correlation between baseline values and change scores has frequently been observed in the cardiovascular psychophysiological evaluations of normotensive participants (Manuck, Kamarck, Kasprowicz, & Waldstein, 1993) and was similarly observed in this study. The data thus failed to meet the linearity requirement of ANCOVA, and an analysis of residualized change scores was viewed as the best option. For the analyses, separate SBP, DBP, and HR residualized change scores were calculated for the math task, speech preparation phase, and speech task, with the mean of the task period and the mean of the preceding baseline period entered into the calculations. Residualized change scores represented the difference between the obtained task value and the task value predicted by linear regression from the respective baseline period. Consistent with much of the cardiovascular psychophysiological research, the cardiovascular data in this study were analyzed with separate univariate analyses for SBP, DBP, and HR Univariate, rather than multivariate, analyses were employed, because these cardiovascular measures are viewed as representing separate systems of autonomic activity (Jenkins, Miller, Hewitt, Wallace, & Pool, 1998). The use of a multivariate analysis would falsely treat the change in cardiovascular response as a unitary phenomenon and would overlook the frequently observed desynchrony between stress-related changes in blood pressure and heart rate (Jenkins et al., 1998).  41  Self-report Data One-way ANOVAs were used to analyze the demographic data (using the experimental group as the independent variable) and to compare the two support conditions with respect to the number of support statements that were administered. Factorial ANOVAs (support seeking X support X gender) were used to examine measures of task performance and the participants' reactions to the tasks. Factorial ANOVAs (support seeking X gender) were also employed to compare the two support conditions with respect to reactions to support and evaluations of the supporter. The change in stress/arousal from baseline to task was analyzed in a between-within design (support seeking X support X gender X task), using residualized change scores.  Evaluation of Assumptions for Statistical Procedures All statistical analyses were preceded by tests examining conformity with various statistical assumptions, with each test (i.e., Shapiro-Wilk test, Box Test, Levene test, and Mauchley test) evaluated at a .05 alpha level. Only violations of assumptions are reported along with the specific analyses in which they occur. For all analyses, independence of observations was assumed from the controlled implementation of the experimental design, with each student participating individually in the study. The normality of distributions of the dependent variables across each cell was evaluated by a visual inspection of stem and leaf plots and by computing the Shapiro-Wilk statistic. In general, however, ANOVAs tend to be fairly robust to non-normality. For all between-within univariate analyses, homogeneity of covariance matrices across all cells was tested with the F approximation to the Box Test. Following a significant Box test, the correlation between determinants of the Covariance matrices and cell size was calculated to determine whether a liberal or conservative bias was evident. The homogeneity of variance assumption for the one-way ANOVAs was tested with the Levene test, with correlations  42  between variances and cell size computed to determine the direction of potential bias. All violations of the homogeneity of covariance matrices and homogeneity of variance assumptions were reported, but the anticipated impact of these violations on ANOVA results in this study is quite low due to the near equal cell sizes for all groups (Stevens, 1996). All repeated measures analyses involving three or more levels were tested for violations of sphericity with the Mauchley test. All non-spherical data was adjusted with the GreenhouseGeisser correction.  Alpha Level All demographic data and manipulation checks (i.e., comparing task performance measures and comparing number of administered support statements) were each evaluated at a .05 alpha level. The stress/arousal data were analyzed with separate univariate analyses, 2  with each test evaluated at a .05 alpha level. Following the completion of the study, a power analysis conducted on the cardiovascular data indicated that there was sufficient power to detect main effects, although the power to detect interactions was reduced. (The power to detect interactions ranged from .04 to .46.) As a result, all main effects (in the analysis of cardiovascular data) were evaluated at a .05 alpha, although a more liberal alpha of. 1 was used to evaluate possible interactions. The three sets of visual analog scale data (i.e., reaction to task, reaction to support, and evaluation of supporter) were analyzed with the family-wise alpha level for each group of analyses set at .2. The alpha levels for individual tests (vsdthin each family of analyses) were obtained by applying a Bonferroni adjustment, which involved dividing .2 by the total number of tests in the set of analyses. As each group of analyses included 4-6 univariate tests, the alpha levels of the individual analyses ranged from .033 (i.e., with 6 tests) to .05 (i.e., with 4 tests).  2  T h e .05 alpha level was used for each of these tests because it was important not to miss potentially important  group differences on these measures, as such differences could pose threats to internal validity.  43  Although the alphas employed in the analyses of the cardiovascular and visual analog scale data may result in inflated family-wise and experiment-wise Type I error, the exploratory nature of the research suggested the need to use a more liberal criterion for evaluating the data. Thus, in accordance with Stevens (1996), the possibility of not detecting potentially meaningful effects (i.e., Type n error) was viewed as having greater negative consequences than the inflated risk of Type I error.  Missing Data It is worth noting that there were occasional missing self-report data points (i.e., approximately 1.5% in total), randomly scattered throughout the data matrix. This resulted in a slight variation in the number of participants per cell used in the different analyses of selfreport data.  Support for Central Hypothesis In the present study, the match-mismatch hypothesis predicts that high support seeking individuals who receive support would derive the greatest physiological benefits compared with other combinations of support seeking and support provision. Support for this central hypothesis is provided if there an interaction between the support condition and support seeking group, such that participants in the high support seeking/receiving support condition show decreased cardiovascular reactivity relative to the other three conditions. Alternatively, support for this hypothesis is also provided if the low support seeking participants show the largest cardiovascular response relative to the other three conditions.  44  Results  Demographic Data One-way ANOVAs indicated that there were no significant differences between the groups in terms of demographic composition. The average age of the four groups was not different, £(3^130) = 1.42, p_ = .231, and the number of generations that the participants' families had resided in North America did not differ between the groups, F p 124) = . 120, p = .948. Perceived Support The perceived availabihty of support from the participants' respective support networks was assessed with the Interpersonal Support Evaluation List (ISEL) (see Tables 1 and 2 in Methods Section). Specifically, a factorial (support seeking X gender) ANOVA was conducted to compare high and low support seeking participants in terms of their perceptions of available support. A significant difference was observed between high and low support seeking participants, F(j \30) = 4.93, p = .028, such that high support seekers showed greater levels of perceived support than did low support seekers. There were no differences between men and women in perceived support, F(i 130) = .21, p = .65.  Vahdity Checks Support seeking classification. Validation of the Coping Strategy Indicator (i.e., the screening measure) for this study was demonstrated by the moderate correlations obtained between support seeking scoresfromthis measure and support seeking scoresfromthe Social Support Inventory, a pilot measure designed to evaluate support seeking (for men, r = .47, p = .001; for women, r = .44, p = .001) This finding provides additional evidence that the CSI was able to discriniinate between high and low support seeking participants.  45  Mean CSI and SSI scores for each group (for male and female samples) are presented in Tables 1 and 2. (see Methods section.) T-tests were performed to compare the CSI scores of i the two high support seeking groups and to compare the CSI scores of the two low support seeking groups. The same comparisons were also made using the SSI scores. For male participants, the low support seeking/support group had a lower CSI score than did the low support seeking/alone group, t p i ) = 2.87, p = .007, but these groups did not significantly differ on SSI scores. Findings from the other t-tests indicated no further group differences in CSI or SSI scores for either men or women.  3  Task performance. Means and standard deviations of the math and speech task performance indices are presented in Tables 3 and 4. In order to examine whether the selfreport and cardiovascular data could have been influenced by differences in task performance, a factorial (support seeking X support X gender) MANOVA was conducted for math task performance and math task effort, and a factorial (support seeking X support X gender) ANOVA was performed to evaluate the speech performance (i.e., comfort during the speech). Results indicated no main or interactive effects of support seeking, support, or gender on these measures, with the exception that women appeared more comfortable during the speech task than men, E(ii25)  =  8.69, p_ = .004.  Support implementation. In order to detennine if the implementation of support was consistent across both support conditions, the two support conditions were compared with respect to the total number of support statements that were provided during the math task, the speech preparation phase, and the speech task. The support conditions were also compared in terms of the minimum number of support statements (for each task) that were provided to any participant within each condition, (see Table 5). One-way ANOVAs indicated no differences  T h e comparisons were made separately for the men's and women's data, because of the separate support seeking quartiles used to recruit men and women.  3  46  Table 3. Men - Task-related Stress/Arousal and Task Performance. low ss, low ss, high ss, alone support alone  high ss, support  SACL-A (BL) SACL-A (MT change) SACL-A (SP change)  4.2 (2.5) 0.8 (2.3) 0.9 (3.0)  5.8 (2.4) 1.1 (2.2) 1.2 (2.0)  4.7 (2.4) 1.9 (1.7) 1.3 (2.2)  4.2 2.1 1.5  (2.1) (2.2) (2.5)  SACL-S (BL) SACL-S (MT change) SACL-S (SP change)  2.8 (2.8) 0.8 (2.6) 2.6 (3.0)  2.1 (2.3) 1.1 (2.3) 0.0 (2.6)  0.9 (.92) 2.5 (3.2) 2.8 (3.3)  2.7 0.3 -0.5  (2.3) (3.1) (3.4)  MT - total correct MT - total # answered SP - degree of comfort  24.1 (12.3) 39.1 (11.9) 2.7 (.86)  ss = support seeking support = support condition alone = alone condition  M T = math task SP = speech task B L = baseline  27.8(10.2) 43.9 (5.6) . 2.8 (0.93)  24.1 (9.6) 43.2 (6.9) 2.9 (.96)  28.6(10.2) 42.3 (10.2) 3.1 (1.1)  S A C L - A = Arousal scale of Stress/Arousal Checklist S A C L - S = Stress scale of Stress/Arousal Checklist  Table 4. Women - Task-related Stress/Arousal and Task Performance. low ss, low ss, high ss, alone support alone  high ss, support  SACL-A (BL) SACL-A (MT change) SACL-A (SP change)  4.4 (2.4) 2.1 (2-1) 2.4 (3.0)  3.8 (2.5) 2.2 (3.0) 2.8 (3.1)  4.9 (2.3) 0.8 (2.1) 1.1 (2-0)  3.3 2.1 2.5  (2.0) (2.1) (3.0)  SACL-S (BL) SACL-S (MT change) SACL-S (SP change)  2.0 (2-5) 3.1 (3.2) 3.8 (3-2)  3.0 (2.4) 2.4 (3.4) 0.1 (3.3)  1.5 (1.8) 1.3 (2.3) 1.3 (2.8)  2.5 2.0 2.0  (2.4) (2.5) (3.5)  MT - total correct MT - total # answered SP - degree of comfort  20.6 (11.1) 40.7 (9.3) 3.4 (.94)  22.1 (12.7) 40.4 (8.5) 3.2 (1.0)  24.0 (14.7) 38.3 (13.7) 3.5 (.92)  ss = support seeking support = support condition alone = alone condition  M T = math task SP = speech task B L = baseline  25.0(12.3) 41.1 (10.3) 3.3 (1.1)  S A C L - A = Arousal scale of Stress/Arousal Checklist S A C L - S = Stress scale of Stress/Arousal Checklist  47  Table 5. Number of Support Statements Provided during Math Task, Speech Preparation, and Speech Task for Participants Assigned to a Support Condition. low ss, support  high ss, support  4.8 4.0  (.42)  4.9 4.0  (.24)  # Speech Prep. Support Statements 2.7 minimum provided 2.0  (.48)  2.8 2.0  (.39)  # Speech Task Support Statements minimum provided  4.4 4.0  (.60)  4.1 4.0  (.97)  Men # Math Task Support Statements minimum provided  4.9 4.0  (.34)  5.0 4.0  (.23)  # Speech Prep. Support Statements 2.8 minimum provided 2.0  (.40)  2.9 2.0  (.32)  # Speech Task Support Statements niinimum provided  (.72)  4.2 4.0  (.90)  Women # Math Task Support Statements minimum provided  4.4 4.0  ss = support seeking support = support condition alone = alone condition speech prep. = speech preparation phase  between the support conditions in terms of the total number of support statements that were administered during each task. Additionally, a visual inspection of the data indicates that participants in both support conditions received a minimum of four support statements during both the math and speech tasks and a minimum of two support statements during the speech preparation phase. These results suggest that the provision of support was effectively (i.e., according to design requirements) manipulated and similarly implemented in both support conditions. 48  Cardiovascular Measures Baseline cardiovascular measures. Baseline cardiovascular measures were compared in order to deteraiine if there were any preexisting group differences in cardiovascular levels prior to the onset of the tasks (see Tables 6 and 7 for means and standard deviations of all baseline measures). Separate 2 (support seeking) X 2 (support) X 2 (gender) X 2 (task baseline) ANOVAs were conducted to examine potential baseline differences in SBP, DBP, and HR. Findings indicated that high support seeking participants showed lower SBP, E(l,127)  =  8  "> E  =  003  >  a  n  d  DBP baselines, F(i  12  7) = 11.03, p = .001, than did low  support seeking participants. Gender differences were also observed, with men showing higher SBP. F (  1 1 2  7 ) = 44.49, p < .001, and DBP baselines, F ( i 7 ) = 5.37, p = .022, than 1;  2  women. No other baseline differences were observed. Task-related cardiovascular response. In order to display the cardiovascular measures for each task in the clearest manner, raw, rather than residualized, change scores for SBP, DBP, and HR are presented in Tables 6 and 7. The change scores for each task are also displayed in Figures 2-7. As is evident from the tables, participants showed considerable baseline to task change in all cardiovascular measures during the math task, the speech preparation phase, and the speech task. A visual inspection of Tables 6 and 7 also indicates that speech task change scores were substantially larger than those obtained during the math task and during the speech preparation phase. In general, the math and speech task change scores obtained in this study are consistent with the magnitude of change scores obtained with similar tasks in previous studies (Kamarck et al., 1995; Kors et al., 1997; Lepore, 1995).  4  It is not clear if the change scores observed during the speech preparation phase in this study are consistent with change scores obtained in earlier studies, because few studies have presented the magnitude of cardiovascular change for a speech preparation phase. 4  49  Table 6. Men - Means (SD) of Blood Pressure and Heart rate Baselines and Change Scores for the Math Task, Speech Preparation, and Speech Task. low ss, alone (N=17)  low ss, support (N=16)  high ss, alone (N=15)  high ss, support (N=19)  Baseline - Math Task  119.9 (10.9)  124.5 (11.1)  118.9  (9.0)  119.9 (6.1)  Math Task Change  +11.4  (8.7)  +17.9 (9.8)  +10.8 (11.6)  +12.5 (8.7)  Baseline - Speech Prep./Speech Task  118.5  (9.1)  123.6  (8.9)  119.3  (6.5)  120.1 (5.6)  Speech Prep. Change  + 9.9  (7.7)  +14.8  (7.1)  + 7.9  (6.8)  +12.5 10.6)  Speech Task Change  +21.9  (9.1)  +24.7  (7.4)  +17.5 (10.0)  +19.2 (9.8)  76.1 (10.0)  70.0 (6.4)  69.3 (4.8)  SBP  DBP Baseline - Math Task  71.7 (8.2)  Math Task Change  +11.3 (6.3)  +15.2  (8.3)  +9.9 (6.6)  +14.2 (6.1)  Baseline - Speech Prep./Speech Task  73.1 (7.7)  79.0  (9.0)  68.7 (8.1)  72.3 (9.0)  Speech Prep. Change  +7.5 (7.8)  +9.1  (6.0)  +8.4 (5.9)  +9.6 (10.8)  Speech Task Change  +16.2(10.4)  +20.9 (9.2)  +18.7 (7.6)  +19.0 (11.9)  Baseline - Math Task  74.6(13.5)  70.5 (11.7)  69.9 (11.1)  67.5 (10.2)  Math Task Change  +7.8 (6.8)  +7.3  (4.5)  +8.8 (7.1)  Baseline - Speech Prep./Speech Task  71.9(12.6)  69.1 (9.6)  67.2 (8.6)  Speech Prep. Change  +8.8 (7.0)  +8.3  (6.4)  +5.5 (7.5)  +8.0 (6.7)  Speech Task Change  +12.3 (8.1)  +14.3  (9.4)  +13.5 (7.8)  +11.8 (9.2)  HR  +12.7  (8.4)  70.0 (10.3)  alone = alone condition support = support condition ss = support seeking speech prep. = speech preparation phase  50  Table 7. Women - Means (SD) of Blood Pressure and Heart rate Baselines and Change Scores for the Math Task, Speech Preparation, and Speech Task. low ss, alone (N = 17)  low ss,support (N = 19)  high ss, alone (N=15)  high ss, support (N=17)  SBP Baseline - Math Task  116.0 (9.6)  112.2 (9.0)  108.5 (6.6)  107.6 (8.6)  Math Task Change  +9.8 (7.0)  +10.9 (8.8)  +6.6  (9.8)  +12.5 (4.7)  Baseline - Speech Prep./Speech Task  116.9 (8.0)  112.7 (9.5)  106.6 (6.6)  110.2(10.7)  Speech Prep. Change  +8.5 (7.5)  +10.1 (4-2)  +12.6 (4.5)  + 10.7 (6.7)  Speech Task Change  +19.3 (8.3)  +18.1 (6.2)  +18.9(10.3)  +20.1 (9.8)  Baseline - Math Task  '72.6 (7.1)  71.0 (8.2)  67.5 (7.8)  68.8 (7.8)  Math Task Change  +8.1 (4.0)  +8.5 (6.5)  +5.3 (5.8)  +8.3 (4.4)  Baseline - Speech Prep./Speech Task  71.8 (6.7)  70.1 (9.1)  67.2 (7.0)  67.4 (9.1)  Speech Prep. Change  +9.7 (3.1)  +8.5 (6-2)  +9.7 (9.8)  +7.7 (6.1)  Speech Task Change  +17.4 (7.5)  +14.9 (7.2)  +13.7 (8.5)  +17.0 (8.2)  Baseline - Math Task  71.8(13.1)  73.3 (14.2)  71.8 (7.0)  70.5 (11.8)  Math Task Change  +7.1 (5.2)  +7.8 (7.5)  +7.8 (6.6)  +6.2 (5.7)  Baseline - Speech Prep./Speech Task  70.4(13.3)  70.2 (12.3)  71.4 (6.7)  68.7(11.5)  Speech Prep. Change  +7.2 (3.3)  +9.6 (4.6)  +10.7 (7.9)  +12.1 (8.2)  Speech Task Change  +11.5 (8.5)  +14.7 (8.1)  +13.6(13.4)  +13.0 (9.1)  DBP  HR  alone = alone condition support = support condition ss = support seeking speech prep. = speech preparation phase  51  25 H 20  a  |  o  n  e  [TJ support  co  O  o to 15 CD  10 1  low  high Math Task  low Speech  high Preparation  low Speech  high Task  Figure 2. M e n : SBP c h a n g e s c o r e s f o r t h e m a t h t a s k , speech p r e p a r a t i o n , and s p e e c h t a s k .  25 ^ 20 cn CO k_  alone  [TJ support  O CJ 15  to cu c  id  •C  O Q.  10  CD CO  low  high Math  Task  low Speech  high Preparation  low Speech  high Task  Figure 3. W o m e n : S B P c h a n g e s c o r e s f o r t h e m a t h t a s k , speech p r e p a r a t i o n , and s p e e c h t a s k .  52  25  20  M  a l o n e  •  support  • u  w 15 CD OS  •c 10 1 O 0.  m G  low  high  Math Task  low  high  Speech Preparation  low  high  Speech Task  Figure 4. Men: DBP change scores for the math task, speech preparation, and speech task.  18 16 11 alone  14 tn  CU  [TJ support  • 12 u CO Q)  10  o> a  •c O 0.  CD  8 6 4 1 2  1 + low  high  Math Task  low  high  Speech Preparation  low  high  Speech Task  Figure 5. Women: DBP change scores for the math task, speech preparation, and speech task.  53  M alone  16  •  support  14 12  S  CO o> DI c  10 ~ 8  09  5  6  2 0 low  high  Math T a s k  low  high  Speech Preparation  low  high  Speech Task  F i g u r e 6. M e n : H R c h a n g e s c o r e s for t h e m a t h t a s k , s p e e c h p r e p a r a t i o n , a n d s p e e c h t a s k .  16  T  low Math T a s k  F i g u r e 7.  high  low  high  Speech Preparation  low  high  Speech Task  W o m e n : H R c h a n g e s c o r e s f o r the m a t h t a s k , s p e e c h p r e p a r a t i o n , a n d s p e e c h t a s k .  54  Summary of cardiovascular task response. The overall findings indicated that men and women receiving support showed heightened SBP responses, relative to those who performed the same tasks alone. Men who were supported also showed increased DBP reactivity compared to men in an alone condition. Additionally, low support seeking men showed greater SBP reactivity to the tasks than did high support seeking men. These findings were equally true in both types of laboratory stressors. Gender differences also were observed, with men showing larger SBP and DBP responses during the tasks than did women. No differences in HR reactivity were detected. Finally, the predicted support by support seeking interaction was not observed for any cardiovascular measure. Specific cardiovascular findings. For SBP. a 2 (support seeking) X 2 (support) X 2 (gender) X 3 (task) between-within analysis of variance, using residualized change scores, indicated a main effect of support, F^i^J)  =  5.60, p_ = .019, with participants who had  received support showing significantly greater SBP reactivity than participants who had not received support during the tasks. Additionally, a trend was observed for the interaction between support seeking and gender, £ ( 1 1 2 7 ) = 3.51, p_ = .063. Follow-up simple effects analyses were conducted and revealed a simple main effect of support seeking for men, F_(l 127) = 5.10, p < .05 , but no simple main effect of support seeking for women, £ ( 1 1 2 7 ) 5  5  = .07, p_ > .25. Specifically, this simple effects analysis indicated that low support seeking men showed greater SBP reactivity than high support seeking men, although there were no differences between high and low support seeking women. There also was a main effect of gender, F^i 127) = 6.12, p_ = .015, with men showing larger SBP responses during the tasks than women. The predicted interaction between support seeking and support on SBP reactivity was not significant, F(q 127) 3  =  04, j> = .846.  Exact p values are not presented due to the fact that the calculations were performed without the assistance of a computer. 5  55  For DBP, there was a support by gender interaction, ¥ / \ 1 2 7 ) 2.94, rj - .089. Follow-up =  ;  simple effects analyses indicated a simple main effect of support for men, E(ii27) = 5.78, p_ < .025, but no simple main effect of support for women, E(l,127)  =  008, p_ > .25. This finding  indicated that men receiving support showed greater DBP reactivity than men who were assigned to an alone condition. Additionally, there was a main effect for gender, E(l \ 27) = 11.16, p_ = .001, although these differences were moderated by a gender by task interaction, E(2,254) 5.11, p. = .007. Follow-up simple effects analyses indicated that men showed larger =  DBP responses than women during the math task, V(\ 254) ;  speech task, E(i,254)  =  25.30, p_ < .001, and during the  12.53, p_ < .001, although gender differences in DBP response were  =  not observed during the speech preparation phase F(i 254) = 42, p_ > .25. The hypothesized interaction between support seeking and support on DBP response was not significant, E(l,127) = -07,p_ = .799. For 1 IR. no significant main or interactive effects were detected for support seeking, support, gender, or task. Thus, the predicted interaction between support seeking and support was not significant, F(i 127) ~ 1-40, p_ = .239. ?  It is worth noting that the assumption of homogeneity of covariance matrices was violated in the analysis of SBP (Box M approximate F = 1.50, p = .019), with a mild conservative bias noted in the analysis. However, the expected degree of impact on this analysis is assumed to he negligible, due to the close to equivalent cell sizes that were used.  Self-Report of Stress/Arousal Summary of findings. The overall findings from the Stress-Arousal Checklist (SACL) showed that participants who received support reported less stress than participants who performed the tasks alone. During the speech task, low support seeking participants who were assigned to an alone condition indicated feeling more stressed than high support seeking participants assigned to an alone condition. Additionally, low support seeking women who  56  performed the tasks alone reported more stress than high support seeking women who were in an alone condition. There were no group differences in self-reported arousal. Specific findings. Baseline levels of stress/arousal and mean change scores, reflecting taskrelated changes in stress and arousal, are presented in Tables 3 and 4. A visual inspection of these tables suggests that stress and arousal levels tended to increase from baseline to task during the math and speech tasks. T-tests confirmed that self-reported stress was significantly elevated during both the math task, t(i 32) = -6.55, p < .001 and the speech task, t(i 33) = 5.18, p < .001. Similarly, T-tests indicated that self-reported arousal was significantly greater than baseline levels during the math task, t(i 32) = -8.38, p < .001, and the speech task, t(i33) =-7.51, p<.001. Prior to comparing the stress/arousal change scores, separate factorial (support seeking X support X gender) ANOVAs for stress and arousal were conducted to examine potential baseline differences between conditions in these measures. Results indicated a main effect of support on baseline levels of stress, F ( i i 2 6 )  =  3.91, p = .050, and a support by gender  interaction on baseline levels of arousal, £(1^26) = 4.33, p = .039. As a result of these baseline differences, residualized (rather than, raw) change scores were used in the subsequent repeated measures (support seeking X support X gender X task) analyses of stress and arousal levels. The results of these analyses indicated a significant main effect of support on self-reported stress, F^ii25)  =  5.14, p = .025, with support condition-  participants reporting lower levels of stress than those in the alone condition. Additionally, there was a significant support seeking X support X task interaction on reported stress levels, E(l,125)  =  7-99, P  =  005. Follow-up simple interaction effects analyses were conducted to  separately examine the interaction of support seeking and support at the math task and at the speech task. These analyses indicated a significant interaction for the speech task only, E(l,125)  =  11-52, p < .001. This support seeking by support interaction (for the speech task)  was examined with a simple effects analysis which revealed a simple main effect for support seeking within the alone condition, F(i 125) = 9.98, p < .01. Specifically, the results of this 57  simple effects analysis indicated that low support seeking participants in the alone condition felt more stressed during the speech task than high support seeking participants assigned to this condition. Finally, there was a significant support seeking X support X gender interaction on self-reported stress, F(ii25) = 4.26, p = .041. Follow-up analyses revealed a support seeking by support interaction for women only, F^j 125) = 6.99, p < .01. Follow-up simple effects analyses indicated a simple main effect of support seeking for women in the alone condition, £^1^25) = 8.87, p < .01. That is, low support seeking women in the alone condition reported feeling more stressed during the speech task than high support seeking women assigned to an alone condition. There were no significant main effects or interactions on self-reported levels of arousal. In analyzing the self-reported arousal data, a violation of the assumption of homogeneous covariance matrices was noted (Box M approximate F = 41.25, p < .001), with a mild liberal bias. However, the analysis is presumed to be robust to heterogeneity of covariance matrices, because the cell sizes in the analysis were close to equivalent.  Reactions to Task Summary offindings.The means and standard deviations of the visual analog scale data are presented in Tables 8 & 9. Separate univariate factorial (support seeking X support X gender) ANOVAs were conducted to evaluate the participants' perceptions of their performance, as well as their perceived stress, during the math and speech tasks. Using the Bonferroni adjustment, each of these four univariate tests were evaluated with alpha set at .05. The main findings were that low support seeking participants perceived a greater degree of stress during the speech task than did high support seeking participants. Low support seeking men felt worse about their performance during the speech task than did high support seeking men. Additionally, women reported more stress during the math task than did men.  58  Table 8. Men: Visual Analog Scale Data Assessing Perception of Task and Perception of Support/Supporter. low ss, alone  low ss,support  high ss, alone  high ss, support  4.9 (2.1) 6.0 (2.5)  4.4 (2.7) 6.5 (1.8)  5.2 (2.3) 5.2 (2.4)  4.2 (2.4) 5.5 (2.7)  3.8 (2.8) 2.6 (1.7)  4.4 (2.3) 4.3 (2.4)  3.2 (2.2) 4.4 (2.9)  4.5 (2.9) 4.9 (2.1)  Task Perceived Stress -MT - SP Perceived Performance -MT - SP Support Supportive -MT - SP  5.0 (2.3) 5.9 (2.1)  4.7 (3.0) 5.6 (2.7)  Happy with -MT - SP  4.8 (2.3) 5.5 (2.5)  5.4 (2.6) 6.0 (2.8)  Evaluative -MT - SP  3.5 (2.7) 4.3 (2.1)  2.9 (2.6) 3.8 (2.4)  7.3 6.7 7.4 7.5 8.1 8.3  5.9 5.0 6.1 7.3 7.7 8.3  Supporter Liked Friends Supportive Warm Accepting Friendly  ss = support seeking  M T = math task  support = support condition  SP = speech task  alone = alone condition  59  (1.3) (1.5) (1.0) (1.1) (.86) (1.1)  (2.1) (2.9) (2.1) (1.4) (1.7) (.77)  Table 9. Women; Visual Analog Scale Data Assessing Perception of Task and Perception of Support/Supporter. low ss, alone  low ss,support  high ss, alone  high ss, support  6.7 (1.8) 6.5 (2.2)  6.2 (2.4) 7.0 (2.3)  4.9 (2.7) 5.5 (2.5)  5.3 (2.3) 6.4 (2.5)  2.9 (2.3) 3.7 (2.6)  2.6 (2.2) 4.1 (2.5)  3.3 (2.8) 3.4 (1.9)  3.8 (2.4) 3.6 (2.1)  Task Perceived Stress -MT - SP Perceived Performance -MT - SP Support Supportive -MT - SP  5.4 (2.6) 6.8 (1.8)  6.0 (2.5) 6.8 (2.7)  Happy with -MT - SP  5.7 (2.9) 6.4 (2.2)  5.9 (2.1) 7.7 (1.8)  Evaluative -MT - SP  3.3 (2.7) 4.6 (2.8)  4.4 (2.5) 4.7 (2.3)  7.4 7.0 7.6 7.8 8.2 8.8  7.5 7.4 7.5 7.8 8.5 8.7  Supporter Liked Friends Supportive Warm Accepting Friendly  ss = support seeking  M T = math task  support = support condition  SP = speech task  alone = alone condition  60  (1.6) (1.8) (1.3) (.98) (1.1) (.86)  (1.3) (1.3) (1.4) (1.2) (1.1) (.92)  Specific findings. For participants performing the speech task, the factorial ANOVAs indicated a main effect of support seeking on perceived stress, F ^ 126) = 4.18, p_ = .043, with low support seeking participants reporting more stress during this task than high support seeking participants. There was also a support seeking X gender interaction on participants' perceptions of their performance during the speech task, F(i i26) ?  =  4.27, p_ = .041. Simple  effects analyses revealed a simple main effect of support seeking for men's perceptions of their performance during this task, F^ii26) = 5.25, p_ < .025. Specifically, low support seeking men felt worse about their speech performance than did high support seeking men. For participants performing the math task, gender differences were observed in ratings of perceived stress, F(i 126) 5  =  7.09, p_ = .009, with women reporting more stress than men.  There were no differences in participants' perceptions of their performance during the math task.  Reactions to Provision of Support Summary offindings.Univariate ANOVAs were conducted to compare the two support conditions in terms of the degree to which participants felt supported and evaluated by the supporter, as well as how happy the participants felt to have received support during the tasks. Using the Bonferroni adjustment, each of the univariate analyses was tested at a .033 alpha level. Although participants' did not differ in their perceptions of support or evaluation, women were happier to have received support during the speech task than were men. A separate set of ANOVAs (with a .033 alpha level set for each test) examined participants' ratings of the confederate. Findings indicated that women liked the confederate more and wanted to be friends with the confederate more than did men. Additionally, low support seeking men receiving support wanted to be friends with the confederate more than did high support seeking men receiving support (see Tables 8 and 9 for the visual analog scale ratings).  61  Specific findings. For participants performing the speech task, a gender main effect was observed for degree of happiness in having received support, F(i 57) = 5.52, p_ = .022, with women reporting that they felt happier to have received support during this task than did men. There were no main or interactive effects on participants' perceptions of support or evaluation during either the speech or the math task. An evaluation of the participants' ratings of the confederate indicated several gender differences. Women liked the confederate more than did men, F(igg) = 5.17, p_ = .026 and women wanted to be friends with the confederate more than did.men, F(i gg) = 7.29, p_ = .009. Additionally, a support seeking by gender interaction was observed in terms of the participants' desire to be friends with the confederate, F(igg) = 5.01, j> = .029. Follow-up simple effects analyses indicated that low support seeking men receiving support wanted to be friends with the confederate more than did high support seeking men receiving support, F gg) = 6.15,p_<.025. (L  Impact of Ethnic Grouping on Cardiovascular reactivity. Perception of Support, and Level of Support Seeking In order to determine if ethnic origin could have affected the results, the Asian and White participants were compared in terms of their cardiovascular reactivity, perceptions of support during the tasks, and level of support seeking (as measured by the screening measure). This was particularly important, because Asians comprised more than 50% of the sample, and previous support/reactivity laboratory studies have not compared Asian and White participants on these measures.  6  6  Previous studies (in other laboratories) have not compared the cardiovascular reactivity of Asian and White  participants, because of the much smaller number of Asian students that comprised the samples of these studies. The only previous study in this laboratory to examine potential ethnic differences did not find any reactivity differences between Asian and White students on a speech task.  62  With respect to the cardiovascular data, comparisons were made in terms of overall cardiovascular reactivity (i.e., with all tasks combined) and task-related cardiovascular response (i.e., math task, speech preparation phase, and speech task). T-tests revealed no differences between Asian and White students in overall SBP, DBP, or HR reactivity. T-tests conducted separately for each task indicated a trend for SBP during the math task , J_(i 13 73) = -1.97, p = .052, with a tendency for Asian students to show greater SBP reactivity during this task than White students (raw change scores are 12.9 and 9.7, for Asian and White students, respectively). No ethnic differences were observed for SBP during the speech preparation phase or the speech task. Additionally, there were no AsianAVhite differences in task-related DBP or HR response. Potential ethnic differences in the support seeking scores from the screening measure (CSI) were also evaluated. Specifically, separate t-tests were used to compare the CSI scores of high support seeking Asian and White participants and to compare the CSI scores of low support seeking Asian and White students. Because of the gender-specific cuts (on level of support seeking) that were used in recruitment, these comparisons were conducted separately for men and women. For the men's sample, the CSI scores of low support seeking Asian participants were not differentfromthe CSI scores of low support seeking White participants. However, the CSI scores of high support seeking Asian and high support seeking White men were different, 1(23.78) -2.34, p = .028, with Asian men having higher CSI scores than =  White men (CSI scores are 30.2 and 28.5 for Asian and White men, respectively). No ethnic differences were observed in the CSI scores of either low support seeking or high support seeking women. Asian and White participants were also compared in terms of their perception of support (as measured by visual analog scales) during the math and speech tasks. No differences were found between Asian and White students in their perception of support for either task. T-tests were then performed to compare high support seeking AsianAVhite students and to compare low support seeking AsianAVhite students in their perception of support. The comparison of 63  AsianAVhite high support seeking students revealed a trend, T(i 5 70) = -2.08, p. = .054, with high support seeking Asian students tending to feel more supported during the math task than high support seeking White students (scores are 6.0 and 3.8 for Asian and white students, respectively). No other differences were observed. 7  Impact of Ethnicity of Confederate on the Cardiovascular Reactivity of Asian and White Participants It was important to determine if the cardiovascular reactivity of participants differed according to whether they were supported by an Asian or White confederate. This was particularly important because 60% of the participants were supported by White confederates, and 40% of the participants were supported by an Asian confederate. In order to test this, ttests were used to examine the participants' overall and task-related cardiovascular reactivity when supported by either an Asian or White confederate, and analyses were conducted separately for the Asian and White sub-samples. Findings indicated that Asian participants supported by a White confederate did not differ in overall or task-related SBP, DBP, or HR from Asian participants supported by the Asian confederate. Additionally, White participants supported by a White confederate did not show differences on any measure of reactivity when compared with White participants supported by the Asian confederate.  7  As perception of support was measured by a 10 cm. visual analog scale, the maximum score for perceived  support is 10.  64  Discussion  Summary of Findings This study was designed to examine whether the acute implementation of support needs to be matched with a person's habitual level of support seeking in order to derive cardiovascular benefits during stressful laboratory tasks. The results of the study did not lend support for the prediction that high support seekers who received support would show decreased reactivity during stressful tasks relative to the other groups. However, several unexpected and interesting findings emerged from the study: Participants receiving support showed greater SBP responses during both tasks than did those who performed the tasks alone. Men who received support also showed heightened DBP reactivity during the tasks compared with men assigned to an alone condition. Additionally, low support seeking men showed elevated SBP responses to the tasks relative to men in the alone condition. Interestingly, high support seeking men and women showed lower SBP and DBP baselines than low support seeking participants. In general, the findingsfromthe self-report measures did not provide much additional clarification to the cardiovascular findings except that support was associated with less reported distress, thus providing some support for the effectiveness of the support manipulation.  Evaluation of Specific Findings and Hypotheses An evaluation of the central hypotheses of this study, as well as an exainination of each of the findings, is presented below. As several of the findings were not predicted, an exploration of potential reasons for thesefindingsis also included. The effect of support seeking on cardiovascular measures. Low support seeking men showed greater SBP response than did high support seeking men, regardless of whether they were supported or alone during the tasks. Thefindingthat a  65  trait was associated with differences in blood pressure response was unexpected and contrary to the prediction of a trait by situation (i.e., support) interaction in this study. The reason for this cardiovascular difference is not entirely clear. One hypothesis is that support seeking is linked to a second personality trait, which, in turn, was associated with blood pressure response in this study. In accordance with this view, it has been shown that individuals who are high in cynical hostility tend to seek low levels of support (Houston & Vavac, 1991). The heightened SBP response of low support seeking men in this study could have been linked to a greater level of cynical hostility on the part of these participants (and not to the low level of support seeking). Along these lines, cynical hostility has been positively associated with blood pressure responses during a laboratory stressor (Hardy & Smith, 1988; Lepore, 1995). The hypothesis that cynical hostihty might be linked to the SBP differences observed between low and high support seeking men in this study can not be directly tested, because cynical hostility was not measured. A second, related explanation is that support seeking is associated with another variable which itself is linked to cardiovascular response. In accordance with this view, support seeking has been correlated with both perceived availabihty of support (Conn & Peterson, 1989) and with the actual receipt of support (Dunkel-Schetter, Folkman, & Lazarus, 1987). Thus, low support seekers may have less support available (and perceive less as available), and high support seekers may have a greater amount of support available in their daily lives. In this study, the elevated cardiovascular task response in low support seekers could be the indirect result of participants in this group perceiving less support in their daily lives. This would be consistent with a previous study which found that men characterized by low perceived instrumental and emotional support showed greater DBP response during a mental arithmetic task and cold pressor task (Knox, 1993). Additionally, in an ambulatory monitoring study of cardiovascular activity, Linden et al. (1993) found a higher ambulatory SBP in female students who perceived less available support in their lives. Along these lines, high and low support seeking participants in this study differed in their perception of support from their 66  respective support networks. However, the lack of a correlation (r = .02) between perception of support and SBP reactivity in this study substantially weakens the argument that the support seeking main effect was accounted for by differing perceptions of support. Interestingly, high and low support seeking participants exhibited different resting blood pressures, with high support seekers showing lower SBP and DBP baselines than did low support seekers. As with the task differences (described above), it is possible that this difference in resting blood pressure was obtained because high and low support seekers differ on a second personality trait or variable which itself is linked to blood pressure response. However, as with the SBP task differences, there does not appear to be a personality construct or variable which can readily explain these findings at the present time. The effect of support on cardiovascular measures. In this study, participants who received support showed increased SBP reactivity relative to those who performed the tasks alone. In addition, supported men showed elevated DBP responses relative to alone-condition men. These findings are contrary to prior studies which frequently have demonstrated a buffering effect of support, in the comparisons of a support condition with an alone condition (Gerin, 1995; Kamarck et al., 1990, 1995; Kors et al., 1997; Lepore, 1995; Lepore et al., 1993). One reason for these differences may be that the samples of the present and previous studies were different, in that previous studies did not recruit participants who were at the upper and lower quartiles for support seeking. It is possible that certain (unknown) characteristics of these extreme groups may interfere with the ability to obtain a beneficial effect of support. It is worth noting, however, that a positive main effect of support was not predicted in this study, because it was hypothesized that only high support seeking participants would have benefited from support. Although the sample differences between the current and previous studies may explain the lack of support benefits for both women and men, it is not clear why supported participants actually showed increases in blood pressure response during both tasks. One possibility is that  67  low and high support seeking participants physiologically react to support, although for different reasons. While speculative, it is conceivable that low support seekers show increased blood pressure response to support provided in the laboratory because they are not accustomed to receiving support in their daily lives. In contrast, high support seekers may be particular about the type of support they receive (i.e., meaning of support, how it is delivered, and who the supporter is) and may have shown heightened blood pressure response because the laboratory support was artificial and "inferior" to the support that they are accustomed to receiving. The heightened blood pressure response to support in this study may be linked to the fact that support was provided by strangers (i.e., confederates). Although support provided by strangers is typically beneficial when compared with a non-support stranger condition (Christenfeld et al., 1997; Gerin et al., 1992; Glynn et al., 1996; Lepore et al., 1993), the studies that compare a stranger-support condition with an alone condition are wholly inconsistent, with studies showing decreased reactivity in response to support (Lepore et al., 1993; Lepore, 1995), no change in response to support (Edens et al., 1992; Sheffield & Carroll, 1994,1996), and one study showing increased reactivity to support (Snydersrnith & Cacioppo, 1992). The supportfindingsfor participants in this study are not entirely inconsistent with other studies comparing stranger-support and alone conditions. It is important to note that participants assigned to alone and support conditions did not differ on any baseline measure. This indicates that the mere presence of a supporter did not affect resting blood pressure and heart rate. Is there any support for a mismatch hypothesis? The mismatch hypothesis was not supported in this study, as the use of inferential statistics failed to detect an interaction between support seeking group and support condition. However, the existence of a significant main effect for men in the support condition and a second main effect for men in the low support seeking group adds up to a pattern in which men in the low support seeking/support  68  receiving condition (which represents a mismatch) had the largest absolute SBP response. This particular pattern of results with additive main effects suggests that it would be premature to conclude that a mismatch hypothesis is unwarranted; it needs to be further tested in future studies. No evidence for the match hypothesis. The central hypothesis of this study was that people who habitually seek support in their daily lives would physiologically benefit when actually provided with support in the laboratory. Thefindingsdo not support this hypothesis. One obvious explanation of thesefindingsis that one's desire for support simply does not moderate the impact of support on cardiovascular reactivity. However, it is also conceivable that high support seeking men and women are particular about the kind of support that they want and may be selective about who they get it from. It is possible that if someone is used to receiving support from close fnends/family, they may not be as receptive to and may not physiologically benefit as muchfromsupport that is providedfroma stranger. This would be in accordance with previous studies which have shown that greater physiological benefits are associated with the presence of a supportive friend than with a supportive stranger (Christenfeld et al., 1997; Edens et al., 1992; Snydersmith & Cacioppo, 1992), although these studies did not include participants who were differentiated by level of support seeking. Thus, it is not clear if the matching hypothesis would have been supported if friends had been used as supporters in this study.  Self-report data. The self-report measures in this study were exploratory in nature, and findingsfromthese measures were limited in their ability to clarify the observed cardiovascular findings. The fact that low support seeking men felt worse about their speech performance than did high support seeking men is consistent with the main effect of support seeking on blood pressure response in men. Thefindingthat low support seekers felt more stressed during the speech task than did high support seekers is also consistent with the blood pressure differences observed between high and low support seeking men. However, thefindingthat  69  low support seeking participants assigned to an alone condition were more stressed than high support seeking participants assigned to an alone condition does not fit with or clarify the observed cardiovascular findings. It is interesting to note that supported participants felt less stressed than did alonecondition participants. This finding is in direct contrast with the increased cardiovascular response to support that was observed in men and women and suggests that support may work at a perceptual level, even when it is not associated with physiological benefits. It is not clear why low support seeking men wanted to be friends with the confederate more than did the high support seeking men. One possibility is that high support seeking men are very selective in what they look for in a supporter and consequently did not rate the confederate as highly as did the low support seeking men. A second possibility is that low support seeking men engaged in a form of self-deception which affected their ratings. It is not possible to evaluate this latter hypothesis, as measures of self-deception were not actannistered in this study. There were a number of gender differences observed with the self-report measures. Women felt more stressed during the math task and were happier to have received support during the speech task than men. Women also liked the confederate more and had a greater desire to be friends with the confederate than did men. Such gender differences in subjective measures are not uncommon and also do not help to clarify the pattern of cardiovascular data in this study. In general, many of the self-report measures did not show differences between groups, and the self-report findings that were significant were not always consistent with the observed pattern of cardiovascular data. Thus, it is not clear if certain psychological variables mediated the increases in cardiovascular reactivity shown by supported participants and by low support seeking men. In particular, the lack of concordance between participants' perceptions of support and their cardiovascular reactions to support is consistent with the fact that a striking desynchrony between self-report and cardiovascular measures is frequently observed in studies 70  with support paradigms (e.g., Christenfeld et al., 1997; Gerin et al., 1992; Kamarck et al., 1990). Findings from these studies typically have revealed that the perception of subjective indices is not different for participants assigned to support/non-support conditions, despite powerful physiological benefits that are obtained following the provision of support. This could result from the retrospective nature of the questioning in some studies (Gerin et al., 1992), or may be due to the fact that the psychological variables which are most critical to the support process are not being examined. It is also possible that social support may directly affect physiological processes, without conscious psychological variables mediating this relationship at all (Kamarck et al., 1990), although this appears to be less likely in this study.  Potential threats to internal vahdity In order to evaluate the observed cardiovascular findings, as well as to evaluate the lack of evidence for a match hypothesis, it is necessary to examine whether the experimental design was implemented as planned, and/or whether other factors (e.g., a weakened or inconsistent support manipulation) may have influenced thefindings.Numerous vahdity checks are available to draw on. Cardiovascular response to the task. To evaluate the impact of support on cardiovascular response, it is necessary to show that the two laboratory tasks evoked substantial cardiovascular reactivity in the alone conditions. In this study, the reactivity of the alone conditions was found to be similar to the reactivity of alone conditions in previous studies (e.g., Kamarck et al., 1995; Lepore, 1995), indicating that the laboratory tasks in this study were effective stressorsfromwhich beneficial support effects could be detected. Implementation of support. The implementation of support was in accordance with design requirements. That is, the total number of support statements and non verbal gestures was equivalent in both support conditions, and all participants received the ininimum number of  71  required support statements for each task. Additionally, the confederate's role appeared to be properly implemented and credible. Thus, during debriefing, few participants indicated that they had been suspicious of the confederate's role; rather, most expressed genuine surprise when informed that the other student had been a confederate working for the lab. In examining the participants' reactions to support (as a manipulation check), it appears that support-condition participants gave only moderate ratings of perceived support (i.e., scores of 4.7 - 6.8 out of a total of 10) during the tasks (see Tables 8 and 9). Although this normally would be of some concern, the frequent desynchrony observed between self-report and cardiovascular measures in support paradigms suggests that perception of support (as measured in this study) may not relate to whether an individual shows reduced cardiovascular reactivity when receiving support. Thus, high levels of perceived support are not necessary to obtain physiological benefits of support. Additionally, the decreased feelings of stress reported by supported participants in this study suggest that support may have indeed influenced participants' perceptions of the tasks. However, it is conceivable that some of the findings may have been influenced by the fact that support was not provided spontaneously in response to the participant's behavior, but was provided because the confederate was asked to do so. That is, the confederate was provided with instructions to provide support to the participant while both the participant and confederate were together in the room. Although this was deliberately done to make the confederate's role more credible (particularly in the math task), it is possible that support does not feel natural and lacks meaning when it does not occur spontaneously in response to one's behavior. In accordance with this view, prior studies that have obtained cardiovascular benefits associated with supportfroma stranger during a speech task (e.g., Christenfeld et al., 1997; Gerin, 1992; Glynn etal., 1996; Lepore, 1995; Lepore etal., 1993) have not had instructions informing the participant that the supporter's role is to provide support; rather, the comments have been provided spontaneously during the task. (The confederate was described 72  as an audience member in these studies.) The issue of non-spontaneous support has not been a problem in math tasks, as verbal support has not been provided. Glynn et al. (1996) tested and found support for the hypothesis that support during a speech task had to occur spontaneously in reaction to the participant's performance, in order to provide cardiovascular benefits. Specifically, they showed that support effects were only obtained when the participant was informed that the confederate had been asked "to behave as a typical audience member" in which he/she would be reacting naturally to the participant's speech. However, support effects were not obtained when the participant was told that the confederate had been asked to behave in a supportive manner and that any supportive behaviors on the part of the confederate (audience) were in response to these instructions. In accordance with the Glynn et al. (1996) findings, it is possible that the participants' knowledge of the support instructions in this study may have resulted in a weaker manipulation of support, which, in turn, may have influenced the findings. Unfortunately, the Glynn et al.(1996) findings were unknown when this study's protocol was developed. Task performance differences. No group differences were observed in math task performance (i.e., total number of correct responses), which suggests that the observed findings are not attributable to differences in level of distraction during this task. Additionally, the number of math task questions answered was similar for each group, indicating that the findings were not due to the differential impact of effort on cardiovascular reactivity. Finally, the apparent comfort during the speech task (as rated by the experimenter) was equivalent across all groups, which suggests that thefindingswere not affected by degree of comfort in giving a speech. Although gender differences were observed on this speech performance measure, these differences are assumed to have had negligible impact on the results.  Demographic composition. All groups showed similar demographic composition, in terms of age, ethnicity, and number of generations that participants' families had resided in North  73  America. Thus, these variables are not likely to have differentially influenced the findings. Additionally, results indicated that the ethnicity of the confederate (i.e., Asian or White) did not differentially affect the cardiovascular reactivity of either Asian or White participants. Screening measure. The moderate correlation that was observed between participants' scores on the support seeking screening measure (Coping Strategy Indicator, (CSI)) and participants' scores on the pilot measure of support seeking suggests that the CSI adequately discriminated among high and low support seeking participants. Thus, it can be assumed that participants were accurately assigned to groups, based on their level of support seeking. The reasons for the difference on mean CSI scores between the two low support seeking groups of men is not clear but is assumed to be the result of sampling error. This explanation is consistent with the finding that these two groups of men did not differ significantly on support seeking scores obtained from the pilot measure.  General Limitations of the Study This study had a number of general limitations. The use of confederates as support providers means that results can only be interpreted in terms of what happens when participants receive support from a stranger. Thus, it is not clear whether the increased blood pressure reactivity on the part of support-condition participants is attributed to a reaction to the support, to the fact that support was being provided by a stranger, or attributable to both of these factors. Hence, it is not clear whether similarfindingswould have been obtained if friends had instead been used as supporters. Moreover, one might question whether thefindingsof this study could reflect the fact that the sample was more than 50% Asian, which is a much larger percentage of Asians than has been reported in previous social support/reactivity studies. Since there has been little or no previous research exarnining the cardiovascular reactivity of Asians, it was not clear whether similarfindingswould have been observed in samples comprised of primarily White students.  74  Because of these concerns, comparisons were made between Asian and White students on a number of different measures of cardiovascular reactivity. The comparisons revealed no differences between these groups in overall (i.e., not related to a specific task) blood pressure and heart rate reactivity to the tasks. In exarnining task-related reactivity, the only observed difference was that Asian students tended to show increased SBP reactivity during the math task, relative to the White students. While this difference is informative for reactivity studies, it is unlikely to have had a significant impact on the results of this study. Further comparisons suggested that Asian and White participants had similar support seeking screening scores and were not different in terms of their perception of support during the tasks. The only observed differences were that high support seeking Asian men had somewhat higher support seeking scores than did high support seeking White men, and high support seeking Asians felt more supported during the math task than did high support seeking White participants. These findings argue against the idea that the increased reactivity to support reflected Asian students' feeling less supported during the tasks or was due to a heightened cardiovascular response on the part of the Asian students. Thus, it seems unlikely that the differing ethnic composition of this study, relative to that of previous studies, can account for the findings. Finally, there has been considerable disagreement in the literature with respect to the issue of trait vs. situationally dependent coping styles. Some (Folkman, Lazarus, Gruen, & Delongis, 1986) view support seeking and general coping styles as situationally dependent, but it has also been shown (Amirkhan, Risinger, & Swickert, 1995) that individuals are fairly consistent in their preferred mode of coping (e.g., support seeking). This raises the question of whether participants who were classified as high/low support seeking based on the particular stressor that they select when completing the screening measure (CSI) would be similarly classified when coping with the type of performance stressors that were employed in this study. This concern was addressed, in part, by thefindingthat scores on the support seeking screening measure were moderately correlated with scores on the pilot measure of support seeking. The pilot measure included a broad range of stressors (in the form of vignettes) as 75  opposed to the single stressor that was listed on the CSI. This suggests that participants who were classified as high or low support seeking by the CSI are likely to exhibit similar levels of support seeking when coping with other stressors, including the type of performance stressors that were used in this study.  Implications of this Study and Future Directions This study was conducted because previous studies examining the impact of social support on cardiovascular stress reactivity yielded inconsistentfindings,which, in turn, suggested a need for "microanalyses" to explain how and when support is effective in buffering cardiovascular stress reactivity. This study adds to the existing literature by exarniuing whether the personality construct of support seeking moderates the relationship between provision of support and reduced cardiovascular reactivity to laboratory stressors. The findings indicated that high support seeking participants assigned to a support condition did not show reduced reactivity, relative to the other three groups. Thus, the hypothesis that a match must occur between one's support seeking tendencies and the actual provision of support in order for physiological benefits to be obtained was not supported in this study. However, failure to reject the null hypothesis does not imply either that the null hypothesis is correct or that the study hypothesis is wrong. Rather, it indicates that the null hypothesis could not be rejected based on how it was tested in this study. It is possible that the implementation of support that did not occur spontaneously (in response to the participant's behavior) may have interfered with a stronger test of the match hypothesis in this study. Additionally, the provision of support by strangers may have weakened the test of the match hypothesis in this study. Thus, it is possible that high support seeking/support condition participants are selective regarding their preferance for source of support. Hence, they may not physiologically benefit from support that is provided by strangers, as was done in this study. Thus, it is conceivable that the match hypothesis of this study could have been  76  supported if friends or family members had provided support during the laboratory stressors, and this needs to be tested in future studies. With respect to the central findings, the significant main effect of support seeking on blood pressure response in men and the significant main effect of support provision on blood pressure response in both men and women occurred across both a cognitive and interpersonal stressor. The main effect of support seeking on cardiovascular reactivity was not predicted and is not easily explained. Additionally, the fact that supported participants showed higher SBP and DBP readings is contrary to what is typically reported in the literature. This may have resulted because of the extreme groups of participants used in this study or because the participants reacted to support provided by strangers. Although it is frequently overlooked, only two studies to date (Lepore, 1995; Lepore et al., 1993) have shown buffering effects of stranger-provided support, when a support condition is compared with an alone condition. The other studies that report buffering effects of support have compared a stranger-support condition with a stranger-non-support condition (Christenfeld et al., 1997; Gerin et al., 1992; Glynn et al., 1996), a comparison which would be expected to yield significant differences in reactivity. It is worth noting that the heightened cardiovascular response to support observed in this study is consistent with thefindingsof two previous studies which also showed an increase in blood pressure in response to the provision of support (Glynn et al., 1996; Syndersmith & Cacioppo, 1992). Bothfindingslead one to question the implicit assumption that support is always beneficial for one's cardiovascular health. Thus, perhaps social support should not be conceptualized as buffering cardiovascular reactivity to stress in every situation and with every supporter. It may be worthwhile for future laboratory studies to focus not only on examining beneficial effects of support on cardiovascular measures but also to delineate situations in which social support may lead to heightened cardiovascular stress reactivity. It is interesting that the negative physiological response to support was observed for both genders in this study. Unfortunately, the small number of laboratory studies of social support 77  which have included both male and female participants makes it difficult to make gender comparisons across studies. However, findings from these studies (Glynn et al., 1996; Lenz, 1995; Lepore, 1995; Lepore et al., 1993; Sheffield & Carroll, 1994, 1996) tend to demonstrate gender similarities, in that either men and women both benefit from a specific support manipulation or neither gender shows any physiological benefits. Additional studies comparing the cardiovascular responses of men and women within various support protocols need to be conducted in order to detennine if these gender similarities will remain. Thefindingsand limitations of this study suggest that further testing of the matchmismatch hypothesis is warranted before any definitive conclusions can be drawn. In order to provide the strongest test of this hypothesis, future studies should use friends as support providers, and the support that is provided should occur spontaneously in response to the participants' behavior. For this reason, it would be preferable to use interpersonal stressors (such as speech tasks) and not math tasks, because it is difficult to provide verbal support which appears spontaneous during a math task. That is, the nature of this type of task necessitates the provision of support instructions in order to justify the implementation of verbal support. A second reason for excluding math tasksfromfurther studies is that the lower reactivity associated with math tasks (compared with that of interpersonal stressors) may make it more difficult to detect support effects (e.g., Edens et al., 1992). In conclusion, the results of this study, in combination with previousfindings,suggest that there is much still to be learned about the process by which support buffers individuals from cardiovascular stress reactivity, as well as about situations in which support may not be beneficial or might even adversely affect the cardiovascular health of the individual. Thus, it seems essential for future researchers to continue to employ a "microanalysis" of the support process in order to examine whether there are characteristics of the participant, the supporter, or the support situation that may be necessary for physiological benefits to occur (or not occur). The inconsistencies infindingsacross studies (including those between this study and a number of previous studies) also point to the need for researchers to provide a more detailed 78  description of all aspects of the support manipulation. This would include the types of verbal and non-verbal gestures used, the number and timing of statements employed, and other situational elements that may or may not be hypothesized to affect the support process. Finally, the differences between high and low support seeking men in SBP reactivity suggest a need for future reactivity studies to better understand why and in what situations a low level of support seeking may be linked to heightened cardiovascular reactivity in men.  79  References  Allen, K M . , Blascovich, J., Tomaka, J., & Kelsey, K M . (1991). Presence of human friends and pet dogs as moderators of autonomic responses to stress in women. Journal of Personality and Social Psychology. 61. 582-589. Amirkhan, J. (1994). Criterion vahdity of a coping measure. Journal of Personality Assessment. 62. 242-261. Amirkhan, J. (1990). A factor analytically derived measure of coping: the Coping Strategy Indicator. Journal of Per sonahty and Social Psychology, 59, 1066-1074. Amirkhan, J., Risinger, K , & Swickert, J. (1995). Extroversion: A "hidden" personality factor in coping? Journal of Personality. 63. 189-212. Antonucci, T.C. (1985). Personal characteristics, social support, and social behavior. In K H . Binstock & E. Shanas (Eds.), Handbook of Aging and the Social Sciences, pp. 94-128. Ashton, W.A. & Fuehrer, A. (1993). Effects of gender and gender role identification of participant and type of social support resource on support seeking. Sex Roles. 28. 461-476. Barbee, A., Cunningham, M., Winstead, B., Derlega, V., Gulley, M., Yankeelov, P., & Druen, P. (1993). Effects of gender role expectations on the social support process. Journal of Social Issues. 49. 175-190. Berkman, L. (1985). The relationship of social networks and social support to morbidity and mortality. In S. Cohen & S.L. Syme (Eds.), Social Support and Health (pp. 241-262). Orlando, FL: Academic Press. Berkman, L., Leo-Summers, L., & Horwitz, K (1992). Emotional support and survival after myocardial infarction. Annals of Internal Medicine. 117, 1003-1009. Berkman, L. & Syme, S.L. (1979). Social networks, host resistance and mortality: a nine year follow-up study of Alameda County residents. American Journal of Epidemiology, 109, 186-204.  80  Butler, T., Giordano, S., & Neren, S. (1985). Gender and sex-role attributes as predictors of utilization of natural support systems during personal stress events. Sex Roles, 13, 515523. Blazer, D. (1982). Social support and mortality in an elderly community population. American Journal of Epidemiology, 115, 684-694. Bloom, J.R. (1990). The relationship of social support and health. Social Science and Medicine. 30, 635-637. Butler, T., Giordano, S., & Neren, S. (1985). Gender and sex role attributes as predictors of utilization of natural support systems during personal stress events. Sex Roles. 13, 515524. Case, R., Moss, A., Case, N., McDermott, M., & Eberly, S. (1992). Living alone after myocardial infarction. JAMA. 267. 515-520. CasseL J. (1976). The contribution of the social environment to host resistance. American Journal of Epidemiology. 104. 107-123. Christenfeld, N., Gerin, W., Linden, W., Sanders, M., Mathur, J., Deich, J., & Pickering, T. (1997). Social support effects on cardiovascular reactivity: Is a stranger as effective as a friend? Psychosomatic Medicine. 59, 388-398. Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic Medicine. 38,300-314. Cohen, S. (1988). Psychosocial models of the role of social support in the etiology of physical disease. Health Psychology. 7, 269-297. Cohen, S. & McKay, G. (1984). Social support, stress, and the buffering hypothesis: A theoretical analysis. In A. Baum, S.E. Taylor, & J.E. Singer (Eds.), Handbook of Psychology and Health (pp. 253-267). Hillsdale, N.J: Lawrence Erlbaum Associates, Inc.  81  Cohen, S., Mermelstein, R., Kamarck, T., & Hoberman, H.N. (1985). Measuring the functional components of social support. In I. Sarason & B. Sarason (Eds.), Social Support: Theory. Research, and Applications (pp. 73-94). Dordrecht, Netherlands: Martinus Nijhoff. Cohen, S. & Wills, T A . (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin. 98. 310-357. Conn, M. & Peterson, C. (1989). Social support: seek and ye shall find. Journal of Social and Personal Relationships. 6, 345-358. Davidson, D. & Shumaker, S. (1987). Social support and cardiovascular disease. Arteriosclerosis. 7, 101-104. Dunkel-Schetter, C , Folkman, S., & Lazarus, R. (1987). Correlates of social support receipt. Journal of Personality and Social Psychology. 53. 71-80. Edens, J., Larkin, K , & Abel, J. (1992). The effect of social support and physical touch on cardiovascular reactions to mental stress. Journal of Psychosomatic Research. 36. 371-382. Engebretson, T., Mathews, K., & Scheier, M. (1989). Relations between anger expression and cardiovascular reactivity: Reconciling inconsistent findings through a matching hypothesis. Journal of Personality and Social Psychology. 57, 513-521. Folkman, S., Lazarus, R.S., Gruen, R. & DeLongis, A. (1986). Appraisal, coping, health status, and psychological symptoms. Journal of Personality and Social Psychology. 50, 571-579. Fontana, A.F., Kerns, R.D., Rosenberg, R.L., & Colonese, K.L. (1989). Support, stress, and recovery from coronary heart disease: A longitudinal causal model. Health Psychology. 8 75-193. Ganster, D. & Victor, B. (1988). The impact of social support on mental and physical health. British Journal of Medical Psychology. 61. 17-36.  82  Gerin, W., Milner, D., Chawla, S., & Pickering, T. (1995). Social support as a moderator of cardiovascular reactivity in women: A test of the direct effects and buffering hypotheses. Psychosomatic Medicine. 57, 16-22. Gerin, W., Pieper, C , Levy, R., & Pickering, T. (1992). Social support in social interaction: A moderator of cardiovascular reactivity. Psychosomatic Medicine. 54, 324-336. Glynn, L., Christenfeld, N., Sanders, M., & Gerin, W. (1996). The parameters of social support: effects of contingency and gender on cardiovascular reactivity. Annals of Behavioral Medicine. 18, S095. Gottlieb, B. (1983). Social support as a focus for integrative research in psychology. American Psychologist. 38, 278-287. Hardy, J. & Smith, T. (1988). Cynical hostility and vulnerability to disease: Social support, life stress, and physiological response to conflict. Health Psychology, 7, 447-459. Haynes, S. & Feinleib, M. (1980). Women, work, and coronary heart disease: Prospective findings from the Frarningham Heart Study. American Journal of Public Health. 70, 133-141. House, J. (1982). Work Stress and Social Support. Reading, MA: Addison-Wesley. House, J., Landis, K., & Umberson, D. (1988). Social relationships and health. Science, 112. 540-544. House, J. Robbins, C , & Metzner, H. (1982). The association of social relationships and activities with mortality: Prospective evidence from the Tecumseh Cornmunity Health Study. American Journal of Epidemiology, 116, 123-140. House, J., Umberson, D., & Landis, K. (1988). Structures and processes of social support. Annual Review of Sociology, 14, 293-318. Houston, K. & Vavak, C. (1991). Cynical hostihty: Developmental factors, psychosocial correlates, and health behaviors. Health Psychology, 10, 9-17.  83  Jenkins, C , Miller, T., Hewitt, L., Wallace, J., & Pool, J. (1998). A 3-dimensional analysis of modes of cardiovascular adaptation: concepts, methods and preliminary findings. Journal of Clinical and Investigative Medicine. 21. 228-250. Joseph, J. (1980). Social affiliation, risk factor status, and coronary heart disease: A cross-sectional study of Japanese-American men. Unpublished Ph.D. Thesis, University of California, Berkeley. Kamarck, T., Annunziato, B., & Meriwether, A. (1995). Affiliation moderates the effects of social threat on stress-related cardiovascular responses: Boundary conditions for a laboratory model of social support. Psychosomatic Medicine. 57. 183-194. Kamarck, T., Manuck, S., & Jennings, R. (1990). Social support reduces cardiovascular reactivity to psychological challenges. Psychosomatic Medicine. 52, 42-58. Kaplan, G., Salonen, J., Cohen, R., Brand, K , Syme, L., & Puska, P. (1988). Social connections and mortality from all causes and from cardiovascular disease: Prospective evidence from Eastern Finland. American Journal of Epidemiology. 128. 370-380. King, M., Burrows, G., & Stanley, G. (1983). Measurement of stress and arousal: Validation of the Stress/Arousal Adjective Checklist. British Journal of Psychology. 74, 473479. Kirschbaum, C , Klauer, T., Filipp, S.H., & Helhammer, D.H. (1995). Sex-specific effects of social support on Cortisol and subjective responses to acute psychological stress. Psychosomatic Medicine, 57, 23-31. Kors, D., Linden, W., & Gerin, W. (1997). Evaluation interferes with social support: Effects on cardiovascular stress reactivity in women. Journal of Social and Clinical Psychology. 16, 1-23. Knox, S. (1993). Perception of social support and blood pressure in young men. Perceptual and Motor Skills. 77, 132-134.  84  Krantz, D., Helmers, K , Bairey, N., Nebel, L., Hedges, S., & Rozanski, A. (1991). Cardiovascular reactivity and mental stress-induced myocardial ischemia in patients with coronary artery disease. Psychosomatic Medicine. 53. 1-12. Krantz, D.S. & Manuck, S.B. (1984). Acute psychophysiologic reactivity and risk of cardiovascular disease: A review and methodologic critique. Psychological Bulletin, 96, 435464. Lazarus, R.S. &Folkman, S. (1984). Stress, Appraisal, and Coping. New York: Springer. Lenz, J. (1995). Cardiovascular response to agreement and disagreement: Towards explaining the beneficial effect of social support. Unpublished Ph.D. thesis, University of British Columbia, Vancouver. Lepore, S. (1995). Cynicism, social support, and cardiovascular reactivity. Health Psychology. 14, 210-216. Lepore, S.J., Allen, K M . , & Evans, GAV. (1993). Social support lowers cardiovascular reactivity to an acute stressor. Psychosomatic Medicine, 55, 518-524. Linden, W., Chambers, L., Maurice, J., & Lenz, J. (1993). Sex differences in social support, self-deception, hostility, and ambulatory cardiovascular activity. Health Psychology. 12, 376-380. Manuck, S. (1994). Cardiovascular reactivity and cardiovascular disease: "Once more into the breach". International Journal of Behavioral Medicine. 1, 4-31. Manuck, S., Kamarck, T., Kasprowicz, A., & Waldstein, A. (1993). Stability and patterning of behaviorally evoked cardiovascular reactivity. In: Blascovich, J. & Katkin, E. (Eds.), Cardiovascular Reactivity to Psychological Disease (pp. 111-134). Washington DC: American Psychological Association. Manuck, S., Kasprowicz, A., Monroe, S., Larkin, K , & Kaplan, J. (1989). Psychophysiologic reactivity as a dimension of individual differences. InN. Schneiderman, S. Weiss, & P. Kaufman (Eds.), Handbook of Research Methods in Cardiovascular Behavioral Medicine (pp. 365-382). New York: Plenum Press. 85  Manuck, S., Kasprowicz, A., & Muldoon, M. (1990). Behaviorally-evoked cardiovascular reactivity and hypertension: Conceptual issues and potential associations. Annals of Behavioral Medicine. 12. 17-29. Medalie, J. & Goldbourt, V. (1976). Angina pectoris among 10,000 men: Psychosocial and other risk factors as evidenced by a multivariate analysis of a five year incidence study. American Journal of Medicine. 60, 910-921. Mischel, W. (1973). Toward a cognitive social learning reconceptualization of personality. Psychological Review. 80, 252-283. Mischel, W. (1968). Personality and Assessment. New YorkWiley. Orth-Gomer, K. & Johnson, J. (1987). Social network interaction and mortality: a six year follow-up study of a random sample of the Swedish population. Journal of Chronic Disease. 40, 949-957. Reed, D., McGee, D., Yano, K., & Feinleib, M. (1983). Social networks and coronary heart disease among Japanese men in Hawaii. American Journal of Epidemiology. 117. 384396. Rozanski, A., Bairey, N , Krantz, D., Friedman, J., Resser, K., Morell, M., HiltonChalfen, S., Herstrin, L., Bietendorf, J., & Berman, D. (1988). Mental stress and the induction of silent myocardial ischemia in patients with coronary artery disease. New England Journal of Medicine. 318. 1005-1012. Ruberman, W., Weinblatt, E., Goldberg, J.D., & Chaudhary, B.S. (1984). Psychosocial influences on mortality after myocardial infarction. New England Journal of Medicine. 311. 552-559. Schoenbach, V., Kaplan, B., Friedman, L., & Kleinbaum, D. (1986). Social ties and mortality in Evans County Georgia. American Journal of Epidemiology. 123. 577-591. Seeman, T. & Syme, L. (1987). Social networks and coronary artery disease: A comparison of the structure and function of social relations as predictors of disease. Psychosomatic Medicine. 49, 341-354. 86  Sheffield, D. & Carroll, D. (1996). Task-induced cardiovascular activity and the presence of a supportive or undermining other. Psychology and Health. 11, 583-591. Sheffield, D. & Carroll, D. (1994). Social support and cardiovascular reactions to active laboratory stressors. Psychology and Health, 9. 305-316. Shumaker, S.A. & Brownell, A. (1984). Toward a theory of social support: Closing conceptual gaps. Journal of Social Issues. 40, 11-36. Shumaker, S. A.. & Hill, D.R. (1991). Gender differences in social support and physical health. Health Psychology. 10. 102-111. Snydersrnith, M.A. & Cacioppo, J.T. (1992). Parsing complex social factors to determine component effects: Autonomic activity and reactivity as a function of human association. Journal of Social and Clinical Psychology. 11, 263-278. Spitzer, S.B., Llabre, M.M., Ironson, G.H., Gelhnan, M.D., & Schneiderman, N. (1992). The influence of social situations on ambulatory blood pressure. Psychosomatic Medicine. 54. 79-86. Stevens, J. (1996). Applied Multivariate Statistics for the Social Sciences. Erlbaum Associates: New Jersey.  <  Suls, J. & Wan, C. (1993). The relationship between trait hostility and cardiovascular reactivity: A quantitative review. Psychophysiology, 30, 615-626. Thoits, P. (1986). Social support as coping assistance. Journal of Consulting and Chnical Psychology. 54, 416-423. Umberson, D. (1987). Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior, 28, 306-319. Uchino, B.N., Cacioppo, J.T., & Kiecolt-Glaser, J.K. (1996). The relationship between social support and physiological processes: A review with emphasis on underlying mechanisms and implications for health. Psychological Bulletin. 119, 488-531.  87  Uchino, B.N., Cacioppo, J., Malarkey, W., Glaser, R., & Kiecolt-Glaser, J. (1995). Appraisal support predicts age-related differences in cardiovascular function in women. Health Psychology. 14, 556-562. Unden, A.L., Orth-Gomer, K., & Elofsson, S. (1991). Cardiovascular effects of social support in the work place: Twenty four-hour ECG monitoring of men and women. Psychosomatic Medicine. 53, 50-60. Weiss, J.M. (1968). Effects of coping responses on stress. Journal of Comparative and Physiological Psychology. 65. 251-260. Wills, T. A. (1985). Supportive functions of interpersonal relationships. In S. Cohen & S.L. Syme (Eds.), Social Support and Health (pp. 61-82). Orlando FL: Academic Press.  88  Appendix 1  1. Consent Form - Screening Phase of Study (Note: Original was printed (single-spaced) on Department of Psychology letterhead.) Study: Physiological Responses during Cognitive Challenge and Speaking Questionnaire Phase Principal Investigator: Dr. Wolfgang Linden, Dept. of Psychology, U.B.C, phone #: 822-4156 Co-investigator: Deborah Kors, Dept. of Psychology, U.B.C, phone #: 822-3800 Enclosed is a questionnaire that we would like you to complete. The purpose of the questionnaire is to examine how people cope with a variety of stressful situations, and this questionnaire is the first phase of a study called "Physiological Responses during Cognitive Challenge and Speaking" that is being conducted by Deborah Kors, a Ph.D. student, in conjunction with her research advisor, Dr. Wolfgang Linden, in the department of psychology. Instructions are provided on the questionnaire which tell you how to complete it. After completing the questionnaire, please place it in the sealed envelope, and return it to the experimenters who gave you the questionnaire or you can return the questionnaire directly to the Behavioral Cardiology Laboratory, room 1714, Kenny Building. The questionnaire will take up to 20 minutes to complete, and you will receive course credit for your time involved in completing this questionnaire. Once the completed questionnaire is returned to us, we treat this as your consent to participate in the study. As in all research studies, you are free to refuse participation and may withdraw at any time. If you do so, this will not affect your course standing. There is no need to give your full name or other specific identifying information anywhere, and all materials are treated with confidentiality. Only numerical codes are used, and the results can not be traced back to you. If you are interested in being contacted about the possibility of participating in the second phase of this study (which takes place in our laboratory), for additional credit points, please leave yourfirstname, as well as a phone number where we can reach you. We will contact a certain percentage of mdividuals who havefilledout this questionnaire. The second phase of the study is completely separate from thefirstphase and is entirely voluntary. Whether you choose to participate or not will not affect your class standing and will not affect your receiving course credit for the first phase. Thank you for your time.  89  Appendix 2  2. Consent Form - Laboratory Phase of Study (Alone Condition) (Note: Original was printed (single-spaced) on Department of Psychology letterhead.)  CONSENT F O R M Study: Physiological Responses During Cognitive Challenge and Speaking Principal Investigator: Dr. Wolfgang Linden, Dept. of Psychology, U.B.C. phone #:. 822-4156; 822-3800 Co-investigator: Deborah Kors, Dept. of Psychology, U.B.C. phone #: 822-3800 Purpose: The purpose of this study is to examine blood pressure and heart rate responses during a mental arithmetic task and a speech task. I, , agree to participate in a research project entitled "Physiological Responses during Cognitive Challenge and Speaking" which is to be conducted in the Behavioral Cardiology Laboratory in the Psychology Department at the University of British Columbia, as a Ph.D. dissertation project conducted by Deborah Kors, M.A., with Dr. Wolfgang Linden as the faculty advisor. This study will involve the completion of various questionnaires during a srxty-minute lab session with an additional 15 ruinutes needed for explanations. The study requires me to relax for 20 minutes while my blood pressure and heart rate are monitored. I will then be asked to do a verbal math task. After the math task, I will be asked to rest quietly and to fill out some more questionnaires. Then, I will prepare a speech on an important social topic and deliver it while the speech is recorded on videotape and blood pressure and heart rate are monitored. Finally, I will rest again and fill out afinalset of questionnaires. After that, any questions about the study will be answered. I understand that multiple cardiovascular functions will be monitored throughout the session. All monitors are non-invasive in nature, are harmless, and lead to no foreseeable risk to my health or safety. I understand that my participation in this study is entirely voluntary and that I may refuse to participate or may withdrawfromthe study at any time without influence on my class standing.  90  All information collected is strictly confidential. All information will be coded by number, analyzed in group form and will remain strictly anonymous. There will be no identification of me personally on any permanent records. Additionally, all records will be stored in a locked filing cabinet. Compensation:  For my participation in this study, I will be compensated with 1 and 1/2 course credit points awardedfromthe Department of Psychology. Contact:  If I have a problem with this experiment, I understand that I may contact either Deborah Kors at 822-3800 or Dr. Linden at 822-4156. If I have any concerns about my treatment or rights as a research participant, I may contact the Director Of Research Services at the University of British Columbia, Dr. Richard Spratley at 822-8598. Consent:  Signature below indicates the following: (1) I consent to participate in this study. (2) The procedures to be followed have been explained to me. (3) Questions, if any, have been answered to my satisfaction. (4) I have read and have understood the content of this consent form. Research Participant  Witness  Date:  Laboratory contact number: 822-3800  91  Appendix 3  3. Consent Form - Laboratory Phase of Study (Support Condition) (Note: Original was printed (single-spaced) on Department of Psychology letterhead.)  CONSENT F O R M Study: Physiological Responses During Cognitive Challenge and Speaking Principal Investigator: Dr. Wolfgang Linden, Dept. of Psychology, U.B.C. phone #: 822-4156; 822-3800 Co-investigator: Deborah Kors, Dept. of Psychology, U.B.C. phone #: 822-3800 Purpose: The purpose of this study is to examine blood pressure and heart rate responses during a mental arithmetic task and a speech task, when performed in the presence of another individual.  I, , agree to participate in a research project entitled "Physiological Responses during Cognitive Challenge and Speaking" which is to be conducted in the Behavioral Cardiology Laboratory in the Psychology Department at the University of British Columbia, as a Ph.D. dissertation project conducted by Deborah Kors, M.A., with Dr. Wolfgang Linden as the faculty advisor. This study will involve the completion of various questionnaires during a sixty-minute lab session with an additional 15 minutes needed for explanations. The study requires me to relax for 20 minutes while my blood pressure and heart rate are monitored. I will then have a 1 in 2 chance of performing two tasks while my blood pressure and heart rate are taken. If chosen, I will be asked to do a verbal math task in the presence of the other participant. After the math task, I will be asked to rest quietly and to fill out some more questionnaires. Then, I will prepare a speech on an important social topic and deliver it infrontof the other participant while the speech is recorded on videotape and blood pressure and heart rate are monitored. Finally, I will rest again and fill out afinalset of questionnaires. After that, any questions about the study will be answered. I understand that multiple cardiovascularftmctionswill be monitored throughout the session. All monitors are non-invasive in nature, are harmless, and lead to no foreseeable risk to my health or safety.  92  I understand that my participation in this study is entirely voluntary and that I may refuse to participate or may withdraw from the study at any time without influence on my class standing. All information collected is strictly confidential. All information will be coded by number, analyzed in group form and will remain strictly anonymous. There will be no identification of me personally on any permanent records. Additionally, all records will be stored in a locked filing cabinet. Compensation: For my participation in this study, I will be compensated with 1 and 1/2 course credit points awarded from the Department of Psychology. Contact: If I have a problem with this experiment, I understand that I may contact either Deborah Kors at 822-3800 or Dr. Linden at 822-4156. If I have any concerns about my treatment or rights as a research participant I may contact the Director of Research Services at the University of British Columbia, Dr. Richard Spratley at 822-8598. Consent: Signature below indicates the following: (1) I consent to participate in this study. (2) The procedures to be followed have been explained to me. (3) Questions, if any, have been answered to my satisfaction. (4) I have read and have understood the content of this consent form. Research Participant  Witness  Date:  Laboratory contact number: 822-3800  93  Appendix 4 4. Demographic Questionnaire Demographic Questionnaire  Participant # Gender  Date  F ( ) M ( )Age  What ethnic group do you identify yourself with? H o w many generations has your family been in North America?  1.  D o you have any chronic diseases?  Yes(  ) No ( )  If yes, please list any chronic diseases  2.  D o you have high blood pressure?  Yes(  )  3.  Does anyone in your family have high blood pressure?  Yes(  ) No ( )  4.  D o you smoke?  Yes(  ) No ( )  5. ;  D o you regularly consume tea, coffee, cola drinks, chocolate products?  Yes(  ) No ( )  D o you regularly exercises on a weekly basis?  Yes(  ) No ( )  6.  No(  )  H o w many times per week?  7.  What medications are you currently taking on a regular basis? - include both prescription and over-the-counter (e.g. aspirin)  If you are female: 8.  A r e you currently taking oral contraceptives?  Yes ( ) No ( )  9.  A r e you or could you be pregnant?  Yes ( ) No ( )  10.  H o w many days has it been since your last menstrual period?  94  Appendix 5 Questionnaire No.  Date:  •  C O P I N G S T R A T E G Y INDICATOR Dr. James H. Amirkhan Department of Psychology California State University, Long Beach  SEX: •  Male  •  Female  AGE: HOUSEHOLD INCOME: • Less Than $15,000 • $25,000 to $34,999 • $15,000 to $24,999 • $35,000 to $44,999 NUMBER OF PERSONS IN HOUSEHOLD: HIGHEST COMPLETED EDUCATION: • Grammar School • High School • College •  • •  $45,000 to $60,000 Greater than $60,000  Graduate/Professional  OCCUPATION:  -  •  We are interested in how people cope with the problems and troubles in their lives. Listed below are several possible ways of coping. We would like you to indicate to what extent you, yourself, used each of these coping methods. All of your responses will remain anonymous. Try to think of one problem you have encountered in the last six months or so. This should be a problem that was important to you, and that caused you to worry (anything from the loss of a loved one to a traffic citation, but one that was important to you). Please describe this problem in a few words (remember, your answer will be kept anonymous):  With this problem in mind, indicate how you coped by checking the appropriate box for each coping behavior listed on the following pages. Answer each and every question even though some may sound similar. Did you remember to write down your problem? If not please do so before going on.  95  K e e p i n g t h a t stressful e v e n t in m i n d , i n d i c a t e t o w h a t e x t e n t y o u . .  1. Let your feelings out to a friend?  •  A lot  •  A little  •  Not at all  2. Rearranged things around you so that your problem had the best chance of being resolved?  •  A lot  •  A little  •  Not at all  •  A lot  •  A little  •  Not at ail  4. Tried to distract yourself from the problem?  •  A lot  D A little  •  Not at all  5. Accepted sympathy and understanding from someone? 6. Did all you could to keep others from seeing how bad things really were?  •  A lot  •  A little  •  Not at all  •  A lot  •  A little  •  Not at all  7. Talked to people about the situation because talking about it helped you to feel better?  •  A lot  •  A little  •  Not at all  • •  A lot  A little A little  • •  Not at all  A lot  • •  10. Daydreamed about better times?  •  A lot  •  A little  •  Not at all  11. Tried different ways to solve the problem until you found one that worked?  •  A lot  0 A little  •  Notatall  12. Confided your fears and worries to a friend or relative?  •  A lot  •  A little  •  Notatall  13. Spent more time than usual alone?  •  A lot  D A little  •  Not at all  14. Told people about the situation because just talking about it helped you to come up with solutions?  •  A lot  •  A little  •  Notatall  •  A lot  •  A little  •  Notatall  16. Turned your full attention to solving the problem?  •  A lot  •  A little  •  Not at all  17. Formed a plan of action in your mind?  •  A lot  •  A little  •  Notatall  3. Brainstormed all possible solutions before deciding what to do?  8. Set some goals for yourself to deal with the situation? 9. Weighed your options very carefully?  Not at all  15. Thought about what needed to be done to straighten things out?  96  18. Watched television more than usual?  •  A lot  •  A little  •  Not at al  19. Went to someone (friend or professional) in order to help you feel better?  •  A lot  •  A little  •  Not at al  •  A lot  •  A little  •  Not at al  21. Avoided being with people in general?  •  A lot  •  A little  •  Not at al  22. Buried yourself in a hobby or sports activity to avoid the problem? 23. Went to friend to help you feel better about the problem?  •  A lot  •  A little  •  Not at al  •  A lot  •  A little  •  Notatal  •  A lot  •  A little  •  Notatal  •  A lot  D A little  •  Not at al  •  A lot  •  A little  •  Not at a  •  A lot  •  A little  •  Not at a  28. Identified with characters in novels or movies?  •  A lot  •  A little  •  Not at a  29. Tried to solve the problem?  •  A lot  •  A little  •  Not at a  •  A lot  •  A little  •  Not at a  31. Accepted help from a friend or relative?  D A lot  •  A little  •  Not at a  32. Sought reassurance from those who know you best? 33. Tried to carefully plan a course of action rather than acting on impulse?  •  A lot  •  A little  •  Not at a  •  A lot  •  A little •  Not at a  20. Stood firm and fought for what you wanted in the situation?  24. Went to a friend for advice on how to change the situation? 25. Accepted sympathy and understanding from friends who had the same problem? 26. Slept more than usual? 27. Fantasized about how things could have been different?  30. Wished that people would just leave you alone?  You may STOP here. Thank you for your cooperation!  97  Appendix 6 6. Pilot measure of support seeking Social Support Inventory Please circle the answer that is most descriptive of yourself in each situation. In all situations, support is defined as having someone to listen to your problem in order to help you feel better about it (i.e. emotional support, empathy, or sympathy).  1. You find out that you have just failed an exam. This comes as quite a disappointment, as you have worked quite hard to do well on the exam. How likely is it that you would actively seek out support in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  How likely is it that you would accept support that is provided to you in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  2. You are not getting along with a close friend, and this is upsetting to you as you value the friendship. How likely is it that you would actively seek out support in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  How likely is it that you would accept support that is provided to you in this situation? •1 almost never  2 once in a while  3 sometimes  4 often  98  5 almost always  3. You are having to work on a class project with a class member of the opposite sex. You are extremely nervous about this, because you have a crush on this person (i.e., are interested in the person and find them very attractive) and are worried that you will act stupid and "blow it." How likely is it that you would actively seek out support in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  How likely is it that you would accept support that is provided to you in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  4. You are planning to try out for a sports team and are quite nervous about it, as it is very important for you to get on the team. How likely is it that you would actively seek out support in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  How likely is it that you would accept support that is provided to you in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  5 almost always  5. You have to go back to see your doctor to get some test results (i.e. from blood tests) that you are nervous about receiving. You have the option of bringing a friend with you to the doctor's office to be there with you when you get the results. How likely is it that you would actively seek out support (i.e. ask someone to come with you to the doctor's office) in this situation? 1 almost never  2 once in a while  3 sometimes  4 often  99  5 almost always  How likely is it that you would accept support (i.e. accept an offer from someone to come with you to the doctor's office) that is provided to you in this situation? 1  almost never  2  once in a while  3  sometimes  4  often  5  almost always  6. You are experiencing a daily hassle (or annoyance), such as losing your car keys, getting a parking ticket, arriving late to an appointment, or spilling food on your clothing. How likely is it that you would actively seek out support in this situation? 1  almost never  2  once in a while  3  sometimes  4  often  5  almost always  How likely is it that you would accept support that is provided to you in this situation? 1  almost never  2  once in a while  3  sometimes  4  often  5  almost always  7. You are facing a major life stressor, such as a death of a friend or family member, a relationship break up, a serious illness, etc.  How likely is it that you would actively seek out support in this situation? 1  almost never  2  once in a while  3  sometimes  4  often  5  almost always  How likely is it that you would accept support that is provided to you in this situation? 1  almost never  2  once in a while  3  sometimes  4  often  100  5  almost always  ISEL  Appendix 7  Thh ictlr n marlr wp nf • fat nf tfit»—**** * statement we woddfikeyoa to cketejn^ FALSE (FF) if the statement ii not tree about JOB. wfc  ch m  * «may v  be tree about yon. For each  You mayfindthm nui7 of oV stsiemexto mricUyi»nethapn*ab^ oiiestioos wffl U dtfrxiA to as^ circle only one of ike alternatives for< Please read each stem quickly bat caxefallybefareiespo^ ngbt or wrong i I know someone who would loan me SSOsoIrakigoawiyfor the weekend. FT  FF  Tnere are peopk at school or m town who I reg^ watkwin, FT  PF  I know someone who I see or talk to often with whom I would feel perfectly comfortable talking about problems that I aught have budgeting mytimebetween school and my social Efe. FT  FF  Most people who know me well think highly of me. FT  FF  I doot know anyone who would five me some old hunkure if I moved into my own apartment FT FF 1 know someone who I sec or talk to often wkh whom I wouldfeelperfecdy comfortable talking aboot wmahy transmitted < FT 7.  Idcarlon^grfar^todothi FT  8.  FF  Mxntpeople thmklhaveapxxlsexmsaf hmaor. FT  9.  FF  FF  Idont know anyone at school or m town who would get assignmentsforme from my teachers if I FT  PF  101  10.  I h u g oat in afriend*t room or apartment qua* a lot. FT  1L  I wfl base a better fkfnrt thaa sari otter people w O . PT  12.  Even if I  PF  Most people are more attractive thaa I an. FT  18.  PF  People hang out ha my room or apartment during die day or in the evening. FT  17.  PF  There bat anyone at school or in town with whom I woold fed perfectly comfortable talking aboat problems I might have getting along with my parenti. FT  16.  FF  Lately, when INe been troubled, I keep dungs to myseJL FT  15.  PF  I am not a member of aay social groups (such at drarch group*, dobs, team, etc). FT  14.  PF  needed k my family would (or could) aot five n e money tor tnkioa and book*. PT  13.  FF  PF  I know someone who would give me tome old dishes if I moved into my own apartment. FT  FF  19.  There *aart anyone at school or in town wkh problem* I might have Baking friend*. PT FF  20.  Moat of myfriendsarc more popular than I am. FT  21.  FF  Lately, I often fed lonely, Dx I don't have anyone to reach oat t a FT  22.  whom I woold fed perfectly omlurtabk talking aboot aay  PF  I don't know anyone at school or is town who woold help me study for aa exam by spending several hoars reading me questions. FT  FF  102  Motto/ nryfriendsthink that Tm i PT  PF  I doot ta& to a member of my fasnly at least once a< PT  PF  I know someone who wonH kaa ate $100tohdppay nry tnkioa. FT  FF  There ami anyone at school or mtownvita whom I vwUfedperfecdy ccoibf^^ PT  PF  Moat of myfriendshaw nort control over vast happens to them than L PT  FF  I can fet a date who I enjoy spendmg tame vita whenever I want. FT  FF  I dont know anyone at school or intownwho wooldkaanKdievcwfor accwpfeoflioan. PT PF I know someone who I tee or talk to often with whom I wouldfedperfectly comfortable any tonal problemsJ might 1 FT PF I doot know anyone who would loan me several hundred doQari to pay a doctor biB or dental bnl FT  FF  Most of ary Mend% aic swore swow^ FT  FF  I know someone at school or intownwho wooUbrmgarymeaktonryroom FT  FF  There kVt anyone at school or hi town with whom I wouldfedperfectly comfortable talking abont ffcrTxTifrieswith my social Ha. FT  FF  Morfafnryfxiendskjmaotadje^toa FT  FF  103  36.  I dont usually spend two evenings on the weekend doing something with others. PT  37.  I know someone who I see or talk to often with whom I would fed perfectly comfortable talking about any problems I might have with drags. FT  38.  PF  FF  I know someone who I see or talk to often with whom I would fed perfectly comfortable talking about any problems I might hne adjustmgtocoflege fife. FT  48.  PF  I belong to a group at school or in town that meets regularly or does things together regularly. FT  47.  PF  If I needed k, nry farnih/woukl provide me with an allowance and spending money. PT  46.  PF  If I decided at dinner time to take a study break this evening and go to a movie, I could easily find someone, to go with me. PT  45.  PF  Most of my friends dont do as well as I do in school. PT  44.  tJK*i»ti  If I wanted a date for a party next weekend, I know someone at school or in town who wouldfixme up. PT  43.  PF  I dont fedfriendlywith any *"«^*«™g aratanrt, professors, campus or student PT  42.  PF  I dont know anyone at school or in town who makes my problems dearer and easier to understand. PT  4L  PF  I dont havefriendsat school or in town who would comfort me by showing some physical affection. PT  40.  PF  Most of my friends are more interesting than I am. PT  39.  PF  PF  I know someone who I see or talk to often with whom I would fed perfectly comfortable talking about any problems I might have nKCting people. PT  PF  104  Appendix 8 8. Stress/Arousal Check List  SACL Please circle one of the following (++, +, ?, or -) to indicate how you currently feel. definitely yes Calm Contented Active Vigorous Comfortable Lively Uneasy Tired Sleepy Worried Distressed Uptight Drowsy Tense Relaxed Passive Energetic Alert Bothered Aroused  ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++  slightly yes  + + + + + + + + + + + + + + + + + + + +  not sure or don't understand ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?  definitely not  SCORING INSTRUCTIONS - SACL  If a (++) or (+) has been circled for a positive adjective then score 1, otherwise 0. If a (?) or (-) has been scored for a negative adjective then score 1, otherwise 0. Scores for all adjectives are added to obtain a total score for that factor.  105  Appendix 9 9. Ranking of Possible Speech Topics CONTROVERSIAL ISSUES QUESTIONNAIRE  Below is a list of issues that people have different opinions about. Please select and rank order your top three choices. Number them 1-2-3 with number 1 being your top choice. Base your rankings on how strongly you feel about the topic and your knowledge of the topic. (So, #1 would be a topic that you have the strongest opinions about and also feel that you have the most knowledge of) If something else comes to mind that you feel quite strongly about, feelfreeto write it in the OTHER category. Abortion Euthanasia Capital Punishment Smoking Ban in Vancouver Restaurants Violence on TV. Environmental Concerns (e.g. logging, air pollution, water pollution) Immigration Transportation in Vancouver Animal Rights (e.g. animal experimentation, not using animal parts for people's clothing, etc.) Quebec Separation Other  106  Appendix 10 10. Reactions to task (Alone condition) Note: There is a separate post-task evaluation for the math task and for the speech task.  Post-Task Evaluation For each question below, please rate how you felt during the math task/speech task by marking positions on the lines below. Please put a single line or slash mark on the line to indicate your response.  1. How stressed did you feel during the task? not at all stressed  very stressed  2. How would you rate your performance during the task? poor  excellent  107  Appendix 11 11. Reaction to task and reaction to support (support condition). Note: There is a separate post-task evaluation for the math task and for the speech task.  Post-Task Evaluation For each question below, please rate how you felt during the math task/speech task by marking positions on the lines below. Please put a single line or slash mark on the line to indicate your response. 1. How stressed did you feel during the task? not at all stressed  very stressed  2. How would you rate your performance during the task? poor  excellent  3. How supported by the other student did you feel during the task? not at all supported  very supported  4. How evaluated or judged by the other student did you feel during the task? not at all evaluated  very evaluated  5. How happy were you to have someone support you during the task? not at all happy  very happy  108  Appendix 12 12. Evaluations of the Supporter (support condition)  Evaluation of Supporter Questionnaire:  1. How much did you like the other student who supported you? did not like at all  liked very much  2. How much would you want to become friends with this person (if you were wanting to make more friends)? not at all  very much so  3. How supportive do you feel that this person is of her other friends? not at all supportive  very supportive  4. Do you feel that this student is..  Unfriendly  Friendly  Rejecting  Accepting  Cold  Warm  109  Appendix 13  13. Manipulation Check for support conditions Manipulation check for support statements (filled out by experimenter) Math task statements 1 2 3 4 5  Speech preparation statements 1 2 3 Speech Task statements 1 2 3 .  4 5  110  Appendix 14 14. Helpful Hints for Preparing the Speech SUGGESTIONS FOR SPEECH Remember to begin by stating your opinion clearly and strongly - Tell the important facts. - Make arguments in support of your opinion. - Argue against opposing opinions. Show how inconsistent or inaccurate or uncaring opposing points of views often are. Don't hesitate to be really critical. - If others held your opinion, how would the world (Canada, B.C., Vancouver, or other parts of the world) be a better place? Give some examples. - If your opinion is ignored and the opposing view is held to be true, how would this make the world a worse place? Give examples. - How do others often misunderstand the topic or misinterpret it? - Explore the impact of this topic from different angles Moral Legal Social Political Financial Logical - Personal experience - What have you experienced on this issue? How has this belief changed your life? Do you act differently because of your belief? - Why? - Describe discussions or arguments that you have had on this topic. Experience of others: How has this issue affected other people that you know? Learning about the Issue: What have you read about this issue? What have you seen in the media (i.e. newspapers, or heard on radio or television) about it? How did youfirstlearn about this issue? Tell it like a story. If you get stuck, repeat your main point or go back to earlier ideas that you have mentioned  111  Appendix 15  15. Experimenter rating of participant's degree of comfort during the speech task.  Speech Rating  5 = speaks with tremendous ease/comfort throughout the speech 4 = speaks with ease throughout the speech 3 = speaks with a mixture of ease and some discomfort throughout the speech 2 = speaks with some discomfort throughout the speech 1 = speaks with a tremendous amount of discomfort or appears to freeze for long periods, time throughout the speech  112  Appendix 16 16. Data Collection Sheet  DATA SHEET FOR PARTICIPANT Date of visit Time Participant #. Initials Gender F( ) M( ) Condition: Alone ADAPTATION  Support SBP  DBP  HR  Minute 4 Minute 9 (Minute 10 - in support condition - experimenter brings supporter into room) Minute 14 Minute 19 Minute 20- TURN ON VCR - expter enters room WITH ANSWER SHEET and describes math task (and support instructions). TURN ON TV. - tell participant to put down magazine and that task will begin in 10 seconds SBP  DBP  Minute 22 Minute 24  HR —  .  Minute 26 Minute 28 Minute 30 Minute 30 - (over intercom) - experimenter gives instructions about returning to questionnaires  113  RECOVERY  SBP  DBP  HR  Minute 31 Minute 35 Minute 39 minute 40 - expter enters room -gives instructions for speech preparation and speech task. Gives speech aide and blank paper. (Gives instructions to supporter) SPEECH PREPARATION  SBP  DBP  HR  Minute 42 Minute 44 Tell Participant to stop preparation and in 10 Seconds I Will tell her/him to begin. 10 sees, later - ANNOUNCE T H A T T H E Y CAN NOW BEGIN T H E SPEECH SPEECH TASK  SBP  DBP  HR  Minute 46 Minute 48 Minute 50 Ov er intercom - instruct participant (and supporter) to go back to final set of questionnaires. RECOVERY  SBP  DBP  HR  Minute 51 Minute 55 Minute 59 MINUTE 60 - IF FINISHED WITH QUESTIONNAIRES, UNHOOK PARTICIPANT F R O M BP CUFF AND DEBRIEF.  114  Appendix 17  17. Experimenter script for alone condition:  PROCEDURE: 1. arrival of participant - ask the participant if he/she was able to not smoke, drink/ingest caffeine, exercise, or drink alcohol for the past 2 hours, (check when the last time was that she/he had caffeine today. Tell the participant that you're running about 5 minutes late and ask them if they can wait down the hall in the waiting area - personally escort them and point out the chair to sit on. 2. Five minutes later, bring the participant back to the lab, explain the study (that we are interested in examining cardiovascular responses while one is performing two tasks - a 10 minute math task and a 5 minute speech task...), and have them sign consent forms. Have them fill out the controversial issues questionnaire - indicate that as they will be giving a speech, we like to have an idea about topics that are important to people. Tell them to rank order the issuefromhighest to lowest preference based on the strength of their opinions and their knowledge of the topic. Tell them that you will let them know later what speech topic that they will discuss. 3. bring into lab room and hook up to blood pressure monitor Tell them to fill out questionnaires for the next 20 minutes. If theyfinishearly, they can relax or read magazines. 4. After 20 minutes, enter room and give math task instructions: Tell Participant: "You will be doing a 10 rninute math task which will be presented on TV via VCR. The task has 50 equations, and questions are delivered at a rate of one question every 12 seconds. During the task, you will need to read the question outloud and then write your answer on the answer sheets. You will need to calculate the answer in your head, as no written calculations are allowed. As the questions come up quickly, you will need to work very quickly. If at the end of the 12 second interval, if you are not able to answer the previous question, you will need to move on to the next problem." "Approximately, every 1-2 minutes, there will be a short rest interval where no problem is presented and you will hear a bell (erratic sounding) to cue you and will also see a blank screen. You are not allowed to still work on previous problems during this rest interval."  115  "Your performance will be evaluated by your accuracy and by the number of questions that you are actually able to do. And there will be a prize for the participant who demonstrates the best performance." "At the end of the task, I will tell you (over the intercom) to go back to the questionnaires but please wait for these instructions. Also, try not to move your non-dominant arm as you do the task. Any questions? The task will come on in about 10 seconds." 5. Math task 6. recovery period - (over intercom) tell participant that they can put answer sheet down - and go back to questionnaires. If they finish the questions early, they can relax or read magazines. 7. After 10 minutes, enter room and tell participant about speech task. Bring speech aide and blank note paper for speech notes. Tell the participant: "You will be preparing a speech on the controversial issue which was your first choice, which is . You will have 5 minutes to prepare the speech and will then need to give a 5 minute speech (from sitting in your chair), while your speech is being videotaped and also observed and evaluated by me from behind the one way mirror/intercom. It is important to clearly express your opinions during this speech, telling me exactly how you feel on the issue. So, right at the beginning of your speech, please tell where you stand on the issue. Then, after stating your position, please explain why you feel the way you do about this .issue. Your speech will be evaluated by your ability to present your position as clearly as possible and by the quality of the arguments that you use to defend your position, rather than on the actual stand that you take. So, try to give examples and arguments that support your opinion. It is also important to try to speak as continuously as possible for the fullfiveminute period." "You will have 5 minutes to prepare your speech, and here is some paper to write some ideas down. You can't write out an entire speech in 5 minutes - so it's best to just write out a few key points. Here is a speech aide which provides some suggestions for going about this. But keep in mind that we don't want you to read the speech; rather we want you to look up at the camera while speaking and just use your notes as reference points." Give participant paper and speech aide. "Also, as I will take bp readings during the speech preparation (and speech), if you take notes, again try not to move your non-dominant arm as you do so. I will give you a 10 second warning (by intercom) at the end of the speech preparation and then will (over intercom) tell you exactly when to begin your speech. Do you have any questions?" Appear to turn on video camera as you leave the room.  116  8.  5 minute speech preparation phase - 2 readings  9. 5 minute speech - 3 readings 10. RECOVERY - (over intercom) tell participant that she/he can stop speaking and to go back to final set of questionnaires and then just go to magazines. 11. unhook participant from blood pressure cuff and debrief.  117  Appendix 18 18. Experimenter script for support condition:  PROCEDURE: 1. arrival of participant - ask the participant if he/she was able to not smoke, drink/ingest caffeine, exercise, or drink alcohol for the past 2 hours, (check when the last time he/she had caffeine today.) Tell the participant that you're running about 5 minutes late and ask them if they can wait down the hall in the waiting area - personally escort them and point out the chair to sit on. After the confederate arrives, tell her that you are running just sughtly behind schedule and that the other person is waiting down the hall and could she also wait down there as well (escort them about halfway down) ...During the 5 minute period, the confederate makes conversation with the participant. 2. 5 minutes later - bring participant and confederate back to lab, explain the study, and sign consent forms. Study explanation: "We are interested in exannhing cardiovascular response (i.e. bp and hr) when one person performs two tasks while being supported by another person. The two tasks are a 10 minute mental arithmetic task and a 5- minute speech task. We are also interested in bp and hr. response of the support person. So, shortly you both will choose envelopes to determine who today will be the performer and who will be the supporter. In the meantime, here are the consent forms for you both to look over and please sign if you feel okay with everything. (Please ask me anything that you don't understand)." Have them fill out the controversial issues questionnaires - same instructions as with alone condition. 3. ROLE ASSIGNMENT - "Now we will determine who will be perfonning the task and who will be the supporter. I have two envelopes here, one with the word "performer" and one with the word "supporter" in it. I will shuffle them and then you will choose the envelopes. Confederate - why don't you choose first." (Confederate must deliberate and think for a second before choosing - and will always choose the envelope directly under the expter's thumb.) 4. description of baseline period "We will start out with a 20 minute baseline period before the task. But to get accurate baselines I need to monitor you two separately for the first 10 minutes and then you will be in  118  the room together in the second 10 rninutes. So, Participant, you will be in this room (expter. points to room) and Confederate. I will take you to in this room (expter points to room), but eventually you will join the participant in the other room. Both of you will be filling out some questionnaires during this time and then can relax and read magazines when you're finished with your questionnaires. I just ask that when you are together in the room, that you refrain from talking to each other as this will interfere with getting an accurate baseline." Bring participant into lab room and hook up to blood pressure monitor Tell participant to fill out questionnaires for the next 20 minutes. If they finish early, they can relax or read magazines. 5.10 minutes later, bring confederate into lab room (with her questionnaires) and hook her up to blood pressure monitor. Remind participant and confederate to not talk with one another and that they can each either relax or read magazines if they finish before the end of 20 minutes. 6. After 20 minutes, enter room and give math task instructions: To participant: "You will be doing a 10 minute math task which will be presented on TV via VCR. The task has 50 equations, and questions are delivered at a rate of one question every 12 seconds. During the task, you will need to read the question outloud and then write your answer on the answer sheets. You will need to calculate the answer in your head, as no written calculations are allowed. As the questions come up quickly, you will need to work very quickly. If at the end of the 12 second interval, if you are not able to answer the previous question, you will need to move on to the next problem." "Approximately, every 1-2 minutes, there will be a short rest interval where no problem is presented and you will hear a bell (erratic sounding) to cue you and will also see a blank screen. You are not allowed to still work on previous problems during this rest interval. " "Your performance will be evaluated by your accuracy and number of questions that you are actually able to do. And there will be a prize for the participant who demonstrates the best performance." "At the end of the task, I will tell you (over the intercom) to go back to the questionnaires but please wait for these instructions. Also, try not to move your non-dominant arm as you do the task. Any questions? The task will come on in about 10 seconds." To confederate: "Your role during the task is to support Participant as she/he does this task. You may say or do whatever seems natural to you but it's probably best to save anything you choose to say for the 6 second rest intervals (which are cued by a bell). Okay? Also, as we're interested in assessing your mood during the task, we ask that you fill out these two mood check lists as Participant does the math task. Any questions?"  119  To both: "At the end of the task, I will tell you (over the intercom) to go back to the questionnaires - but please wait for these instructions. The task will come on in about 10 seconds. 7. recovery period - (over intercom) tell participant that they can put answer sheet down - and go back to questionnaires. If they finish the questions early, they can relax or read magazines. To confederate that she also should go back to questionnaires. 8. After 10 minutes, enter room and tell participant about speech task. Bring speech aide and blank note paper for speech notes. Tell the participant: "You will be preparing a speech on the controversial issue which was their first choice, which is . You will have 5 minutes to prepare the speech and will then need to give a 5 minute speech (from sitting in your chair), while your speech is being videotaped and also observed and evaluated by mefrombehind the one way mirror/intercom. It is important to clearly express your opinions during this speech, telling me exactly how you feel on the issue. So,rightat the beginning of your speech, please tell where you stand on the issue. Then, after stating your position, please explain why you feel the way you do about this issue. Your speech will be evaluated by your ability to present your position as clearly as possible and by the quality of the arguments that you use to defend your position, rather than on the actual stand that you take. So, try to give examples and arguments that support your opinion. It is also important to try to speak as continuously as possible for the full five minute period." "You will have 5 minutes to prepare your speech, and here is some paper to write some ideas down. You can't write out an entire speech in 5 minutes, so it's best to just write down a few key points. Here is a speech aide which provides some suggestions for going about this. But keep in mind we don't want you read the speech; rather, we want you to look up at the camera while speaking and just use your notes as reference points." Give participant paper and speech aide. " Also, as I will take bp readings during the speech preparation (and speech), if you take notes, again try not to move your non-dominant arm as you do so. I Will give you a 10 second warning (by intercom) at the end of the speech preparation and then will (over intercom) tell you exactly when to begin your speech. Do you have any questions?" To confederate: "Again, your role is to support Participant as she/he both prepares a speech and gives a speech. You may say or do whatever seems most natural to you. However, we ask that you refrainfromactually helping her/him in planning the speech but rather just act as a support person."  120  "Also, we would you to fill out this questionnaire during her/his speech preparation (also evaluating your present state)- so it will be this one. You will need to complete this within the 5 minute speech preparation period, and you are not to work on anything during her speech just listen and be supportive." Appear to turn on video camera as you leave the room. 9.  5 min. speech prep - 2 readings  10. 5 minute speech - 3 readings 11. RECOVERY - (over intercom) tell participant that she/he can stop speaking and to go back tofinalset of questionnaires and then just go to magazines. Tell confederate to also go back to questionnaires. 12. unhook participant and confederatefromblood pressure cuff. Confederate remains in lab room, while participant is taken into other room for debriefing. Confederate is brought into the other room, after the confederate's role is explained by the experimenter to the participant.  121  Appendix 19 19. List of Support Statements (for support condition)  SUPPORTIVE STATEMENTS MATH TASK: (10 minutes) - There will be 5 six-second rest intervals, during which the supportive statements will be provided. VERBAL STATEMENTS: 1. One of the following is required: - Hang in there. - You can do it - Just keep on trying. 2. One or more of the following is required: - I don't know if I_could think that quickly. -I don't know that I could think off the top of my head like that. - I don't think I could do problems in my head, like that. - I think Pd have a hard time with this. -1 think I'd find this difficult. 3. One is required - It looks like you're really trying hard (or working hard) - It looks like you're getting through them. - It looks like you're moving right along. 4. Optional (only one) (Needs to be asked with genuine concern.) - Is it going okay? - How's it going" 5. Required (but only one) - Wow, I think you must be halfway through (said at 3rd rest interval) - Hang in there, I think you only have about 3 minutes left (at 4th rest interval) - Hang in there -1 think you only have 1 minute left (at 5th rest interval) - Hang in there -1 think you're almost done (at 5th rest interval) (Please begin the statement, with "I think" or something similar, so that you are believable.)  122  6. Optional (if participant appears to be obviously struggling) - Don't give up - Try to hang in there - Just keep on trying - Come on, you can do it. - Don't worry - you're probably doing better than you think. -1 think I'd have a hard time with this.  NON-VERBALS DURING MATH TASK: PLEASE GIVE THESE THROUGHOUT THE TASK, REGARDLESS OF WHETHER YOU THINK THE PARTICIPANT NOTICES OR NOT -open posture (i.e. arms not crossed against chest), with a slightly forward leaning posture (to convey interest) -smiling intermittently -nodding intermittently -thumbs up or a cheering behavior with hands You will be given a questionnaire (i.e. checklist) to fill out during the task - in order to legitimize your role and to prevent the participantfromfeeling awkward in knowing that you are watching her/him - so fill out the questionnaire, but also look at participantfromtime to time and attend to what the participant is doing, without staring at the participant.  SPEECH PREPARATION (5 minutes) As the participant will be working to prepare the speech and, not actually speaking, it should be fairly easy to find appropriate times to say the supportive statements. You will need to provide 3 statements, spaced about 1-2 minutes apart. (So please discreetly look at your watch.) You will be given a questionnaire to fill out during this time. VERBAL STATEMENTS: -Try not to worry too much about this OR Try to relax about this.. -Don't worry about what she (i.e., the experimenter) will think. -Don't worry, it's only for 5 minutes - it will go by quickly OR It will be all over before you know it. -Just do the best you can. -How are you feeling about this? - It looks like you've worked hard to prepare yourself (would say this towards the end of the preparation period) -1 think you're probably more prepared than you think (USE THIS FOR SOMEONE WHO SEEMS VERY NERVOUS.)  123  NOTE: If the participant asks for help, please say - "I'd love to help, but I don't think I'm allowed to help and you can probably do better on your own." NON-VERBALS - Attentive to participant, open posture, occasional smiling (a warm smile), an occasional nod, or whatever else seems appropriate.  SPEECH TASK (5 minutes) You will need to say 5 statements during the speech period, approximately 1 minute apart, so please discreetly look at your watch, so that you are aware of the time. VERBAL STATEMENTS: 1. One of the following is required - Hang in there. -You can do it. 2. One of the following is required: - Good point. - Interesting idea. - That's interesting -1 never thought of it that way. -I like the way you're presenting the idea. 3. One or more of the following are Required - I'd have a hard time with this - I'd have a hard time speaking for 5 minutes -1 don't know that I could think off the top of my head like that (only if doing okay) -1 don't know that I could think on my feet like that, (only if doing okay) 4. Optional (Obviously not for someone who is struggling!) .-Wow, you really have a knack for this. - It looks like you're really able to think on your feet. 5. Only one of these statements should be said (required) - Hang in there -1 think you're halfway through. - Hang in there -1 think you're almost finished. - Hang in there - you only have 1 minute (or 2 minutes) to go.  124  6. Optional (if participant is having a difficult time) -  Don't give up. Try to hang in there. Just keep on trying. Come on, you can do it. Just take your time and try to relax. Try not to worry too much about this.  NON VERBAL BEHAVIORS DURING SPEECH - open body posture, slightly forward leaning - appear interested (no doodling in this task), yet no staring - eyes widen with interest - srniling - nodding (lots of this) - rooting/cheering with hands.  125  Appendix 20 20. Debriefing Form Physiological Responses during Cognitive Challenge and Speaking Study Explanation Research suggests that social support may reduce the risk of cardiovascular disease. Specifically, it is speculated that strong social ties may have a direct influence on physiology, by decreasing cardiovascular response (i.e. blood pressure and heart rate response) to a stressor. However, laboratory studies that have examined an individual's cardiovascular response to a stressor in the presence of a supportive individual have been somewhat inconsistent regarding the physiological benefits derived from the presence of the supporter. We believe that an important determinant of the physiological benefits of support is whether the individual actually wants social support. That is, individuals' tendencies to seek social support in their daily lives may influence their physiological response to support when it is provided in the laboratory. Thus, it is expected that individuals who are high social support seekers will derive greater physiological benefitsfromsocial support provided during a lab stressor than those who are low social support seekers. In contrast, individuals who are low social support seekers may actually show reduced physiological arousal when they encounter a stressful situation alone, as opposed to when they are given social support. Thus, the goals of the present study were: 1) To examine differences in cardiovascular reactivity to a social support condition and an alone condition, as a function of the degree of support seeking. 2) To determine if gender differences exist among these conditions. The present study was designed to answer these questions in the following manner: Following initial screening by a questionnaire, high support seeking and low support seeking participants were randomly assigned to either an alone condition or a support condition. The alone condition involved performing two stressful tasks while in the room alone, and, hence, without any social support. The support condition entailed performing two stressful tasks while in the presence of another participant, whose role was to provide verbal and non-verbal social support. The other "participant" in the study was a U.B.C. student who is employed in this laboratory to play the role of a "supporter." She has been trained to give precise verbal and non-verbal response to speakers while mamtaining specified eye contact, posture, distance, etc. To make the speech task more stressful, all participants have been informed that the speech is being videotaped. However, none of the speeches have been videotaped, as the video camera was never actually turned on. Because other students in your classes may be participating in this study, we ask your cooperation in not discussing what you have learned by participating and in not showing this sheet.  126  Thank you once again for participating in this study. If you would like to know the results of this study, a copy of the report will be available in six to nine months from the following address: Wolfgang Linden, Ph.D. Department of Psychology 2136 West Mall University of British Columbia Vancouver, B.C. V6T 1Z4 Phone: 822-3800 If you are interested in this area of research and would like to read more about it, you could start with: Cohen, S. (1988). Psychosocial models of the role of social support in the etiology of physical disease. Health Psychology. 7, 269-297. Kamarck, T., Manuck, S., & Jennings, J. (1990). Social support reduces cardiovascular reactivity to psychological challenge: A laboratory model. Psychosomatic Medicine. 52. 42-58.  127  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099440/manifest

Comment

Related Items