Subliminal activation of social ties moderates cardiovascular reactivity during acute stress

The quality and quantity of one's personal relationships have been reliably linked to important physical health outcomes, perhaps through the mechanism of physiological stress responses. Most studies of this mechanism have focused on whether more conscious interpersonal transactions influence cardiovascular reactivity. However, whether such relationships can be automatically activated in memory to influence physiological processes has not been determined. In this study, we subliminally activated relationship positivity and negativity and had individuals engage in self-relevant stressors. Results revealed that priming negative relationships was associated with greater threat, lower feelings of control, and higher diastolic blood pressure reactivity during the stressor tasks. Moreover, priming relationships high in positivity and negativity (ambivalent ties) was associated with the highest heart rate reactivity and greatest respiratory sinus arrhythmia withdrawal during the stressor tasks. Results of this study were discussed in light of mechanisms linking relationships with health, and the specific effects of positivity and negativity in relationships.

SUBLIMINAL ACTIVATION OF SOCIAL TIES MODERATES CARDIOVASCULAR REACTIVITY DURING ACUTE STRESS by McKenzie Carlisle A thesis submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Master of Science Department of Psychology The University of Utah December 2010 Copyright © McKenzie Carlisle 2010 All Rights Reserved The University of Utah Graduate School STATEMENT OF THESIS APPRO V AL The thesis of McKenzie Carlisle has been approved by the following supervisory committee members: Bert Uchino , Chair Biz /10 Date Approved Timothy Smith , Member '8/ Z /10 Dale Approved David Sanbonmatsu , Member ~/z 110 Date Approved and by _ ________ ~C::...ycn:.::t=h=ia::...:B=er:....tgL_ ________ ' Chair of the Department of Psychology and by Charles A. Wight, Dean of The Graduate School. ABSTRACT The quality and quantity of one's personal relationships have been reliably linked to important physical health outcomes, perhaps through the mechanism of physiological stress responses. Most studies of this mechanism have focused on whether more conscious interpersonal transactions influence cardiovascular reactivity. However, whether such relationships can be automatically activated in memory to influence physiological processes has not been determined. In this study, we subliminally activated relationship positivity and negativity and had individuals engage in self-relevant stressors. Results revealed that priming negative relationships was associated with greater threat, lower feelings of control, and higher diastolic blood pressure reactivity during the stressor tasks. Moreover, priming relationships high in positivity and negativity (ambivalent ties) was associated with the highest heart rate reactivity and greatest respiratory sinus arrhythmia withdrawal during the stressor tasks. Results of this study were discussed in light of mechanisms linking relationships with health, and the specific effects of positivity and negativity in relationships. TABLE OF CONTENTS ABSTRACT....................................................................................................................... iv ACKNOWLEDGMENTS ................................................................................................. vi INTRODUCTION .............................................................................................................. 1 METHOD ........................................................................................................................... 9 Participants......................................................................................................................... 9 Procedure ............................................................................................................................ 9 Assessments ...................................................................................................................... 13 RESULTS ......................................................................................................................... 16 Manipulation Checks ........................................................................................................ 16 Preliminary Analyses ........................................................................................................ 16 Does Subliminal Priming of Relationships Influence Cardiovascular Reactivity? ......... 18 What Are the Psychological Processes That Co-vary With Subliminal Priming of Relationships? ................................................................................................................... 20 What Are the Potential Pathways Responsible for Subliminal Priming Effect of Relationships? ................................................................................................................... 21 DISCUSSION................................................................................................................... 22 APPENDIX....................................................................................................................... 30 REFERENCES ................................................................................................................. 31 ACKNOWLEDGMENTS The author wishes to thank Bert N. Uchino, David M. Sanbonmatsu, Timothy W. Smith, Matthew Cribbet, Wendy Birmingham, Kathleen C. Light, and Allison Vaughn for their contributions to this project. This research was generously supported by grant number R01 HL085106 from the National Heart, Lung, and Blood Institute. INTRODUCTION Social relationships are an omnipresent part of life. Consistent with the importance of social ties, both the quantity and quality of our relationships have been reliably associated with serious health outcomes, including morbidity and mortality from various diseases (Barth, Schneider, & von Kanel, 2010; Holt-Lunstad, Smith, & Layton, in press; House, Landis, & Umberson, 1988). The House et al. (1988) review of large prospective studies indicates that socially isolated individuals may be at increased risk for earlier mortality, even after controlling for variables such as age and initial health status. Similarly, higher levels of perceived availability of social support are associated with reduced risk (Holt-Lunstad et al., in press; Uchino, 2004). In general, it has been shown that the link between social support versus isolation and health is comparable to standard risk factors, such as physical activity and smoking (Holt-Lunstad et al., in press; House et al., 1988). In addition to these effects of social support, a smaller literature suggests that conflict in close relationships increases risk of serious illness and earlier mortality (De Vogli, Chandola, & Marmot, 2007; Matthews & Gump, 2002). Of key importance is the question of what mechanisms may account for the epidemiological links connecting social relationships and long-term health consequences (Cohen, 1988; Kiecolt-Glaser & Glaser, 1989). One potential physiological mechanism is based on the buffering model, which posits a key role for reactivity to stress. The reactivity hypothesis suggests that psychosocial stress contributes to the development of 2 cardiovascular disorders through excessive physiological response (Krantz & Manuck, 1984). Importantly, a recent meta-analysis provides support for the view that reactivity influences the development of cardiovascular disease (Chida & Steptoe, 2010). Thus, the buffering model predicts that support from our social ties protects, or "buffers" a person from possible pathogenic cardiovascular reactions associated with stressful life events and circumstances (Cohen & Wills, 1985). According to Cohen (1988), stress-buffering mechanisms could act at the level of stress appraisals, or possibly at the onset of a stressful event, by reducing or even eliminating the stress reaction. Through these mechanisms, social support can provide a solution or reduce the perceived significance of the problem, which may decrease reactivity to stress (Cohen & Wills, 1985). Consistent with this possibility, a meta-analysis conducted by Thorsteinsson and James (1999) concluded that social support has an attenuating effect on cardiovascular reactivity produced by stressor, and more recent reviews reached similar conclusions (Uchino, 2006). In contrast, interpersonal conflict and negative social interactions heighten cardiovascular reactivity, and do so similarly for previously unacquainted dyads (e.g., Gallo, Smith & Kircher, 2000) and for partners in close relationships (e.g., Smith et al., 2009). It is important to note that much of the prior work on social support and physiological stress responses has examined more explicit manipulations of support (e.g., Thorsteinsson & James, 1999; Uchino, 2009). Studies have shown that the provision of social support can attenuate cardiovascular responses to stressful tasks (Lepore, 1998; Uchino, Cacioppo, & Kiecolt-Glaser, 1996), although under certain situations (e.g., Lepore, 1995; Taylor et al., 2010), received social support is not associated with reduced 3 CVR. On the other hand, perceptions of support are more consistently related to beneficial health outcomes than is received support (Barrera, 2000; Uchino, 2004). This could be because received support can convey the message that the recipient is incapable of managing difficulties on their own, thereby threatening self-esteem or perceptions of independence (Bolger, Zuckerman, & Kessler, 2000; Martire, Stephens, Druley, & Wonjo, 2002; Nadler & Fisher, 1986). Perceived support, in contrast, provides benefits without these drawbacks (Uchino, 2009). The role of perceptions of support, rather than the actual receipt of support, in stress buffering has been demonstrated in studies that show simply calling to mind feelings of connectedness and support can attenuate CVR to a stressor task (Ratnasingam & Bishop, 2007; Smith, Ruiz, & Uchino, 2004). For instance, Smith et al. (2004) used a supraliminal prime by having participants write about a supportive tie (e.g., what you appreciate about this person) or casual acquaintance and then had them perform psychological stressors. Results revealed that writing about a supportive tie was associated with lower cardiovascular reactivity to a subsequent speech stressor compared to writing about a casual acquaintance (also see Ratnasingam & Bishop, 2007). These findings are consistent with the possibility that conscious activation of mental models associated with specific relationships can moderate cardiovascular response. One important question that arises from these prior studies is how automatic are the cognitive processes linking relationship representations to cardiovascular function? Given that in the two previously mentioned studies (i.e., Ratnasingam & Bishop, 2007; Smith et al., 2004) no actual provision of support occurred, it is possible that subliminal (nonconscious) activation of relationship representations could impact how one feels 4 about the self, with subsequent implications for physiological reactivity. This possibility is consistent with the general view that people construct working models of their relationships with others as an automatic organizational guide in navigating complex and important social worlds: who am I and how do I feel when with a particular other person? These relational schemas include self-and-other similes, as well as scripts for expected patterns of interaction (Baldwin, 1992). Ogilvie and Ashmore (1991) define their "self-with- other" as "a mental representation that includes the set of personal qualities (traits, feelings, and the like) that an individual believes characterizes his or her self when with a particular other person" (p. 290). Consistent with these models, activation of a relational schema could cause spreading-activation throughout linked cognitive structures. During this process, other concepts that share associative relations and/or semantic features would also become activated (McNamara, 1994) and a person could then experience themselves in the way that they normally do when with the relationship partner. This automatic process can also occur in the absence of the actual relational tie (Andersen & Berk, 1998; Baldwin, 1990). For instance, Baldwin (1994) found that when participants were subliminally primed with the name of a positive relationship tie, they reported higher self-esteem during subsequent stressor tasks when compared to priming a more negative relationship. Though not focusing on relationship primes, Levy, Hausdorff, Hencke, and Wei (2000) extended this work to the physiological domain when they subliminally exposed older individuals to either positive or negative aging stereotypes, and then administered psychological stressors. Levy et al. predicted that priming negative aging stereotypes would adversely influence self-relevant processes that are important in coping with stress. 5 Participants who were exposed to the negative aging stereotypes showed heightened cardiovascular response to stress when compared to participants exposed to positive aging stereotypes. A first aim of the current study was thus to examine if subliminal activation of relationships can influence health-relevant physiological processes. We focused on SBP, DBP, and HR given that these measures have been linked to significant health outcomes (Chida & Steptoe, 2010). We also examined respiratory sinus arrhythmia (RSA) as an index of vagal control of the heart, which is receiving greater attention as a health-relevant biological index (see Bernston et al., 1994; Thayer, Hansen, Saus-Rose, & Johnsen, 2009). We tested the relative effects of activating positive and negative relationship representations on cardiovascular reactivity to a subsequent stressor. If subliminal activation of positive relationship representations attenuates these stress responses and activation of negative representations heightens them, such effects would suggest a novel and potentially far-reaching mechanism by which social relationships influence susceptibility to disease. That is, if subliminal activation of working models alters physiological stress responses, then relationship quality could influence health-relevant physiological processes well beyond circumstances involving actual or even conscious recollection of positive or negative relationship experiences. The quality of one's social life could influence pathophysiological processes and resulting health risks much more pervasively, through the myriad subtle reminders of relationships encountered during everyday experience. A second aim of our study was to examine these questions in the context of a more general model on the health effects of relationships (Uchino, Holt-Lunstad, Uno, & 6 Flinders, 2001). Most prior research has examined the stress-buffering influence of supportive relationships (Thorsteinsson & James, 1999). Of course, not all personal relationships are uniformly supportive (Braiker & Kelly, 1979). Relationships can also be a significant source of interpersonal stress, which as noted previously could lead to increases in reactivity (Nealey-Moore, Smith, Uchino, Hawkins, & Olson-Cerny, 2007; Rook & Pietromonaco, 1987) and serious health problems (De Vogli et al., 2007; Matthews & Gump, 2002). In fact, our conceptual framework that considers the separability of positive and negative aspects of relationships (Uchino et al., 2001) suggests that relationships that combine positivity and negativity may be the most deleterious. According to our model, a person falling in the high positivity/low negativity category would be a strong source of social support (e.g., the best friend you can always count on). In contrast, a network member who falls in the category of low positivity/high negativity would represent an aversive network tie (e.g., an unreasonable manager). Network members who exhibit low positivity/low negativity are people whom we come into contact with very low frequency or depth (e.g., the next-door neighbor). Relationship ties that fall into the high positivity/high negativity category are ambivalent ties. These ties are sources of both positivity and negativity during social interactions (e.g., an over-bearing parent). Much of the previous research on social relationships and health has only looked at relationship positivity (support), and a smaller literature has examined the effects of negativity. However, little research has examined the joint or interactive effects of these two common relationship experiences. Our model thus 7 provides a more comprehensive framework for examining social relationships and health (Uchino et al., 2001). The category of relationships we refer to as ambivalent seems to be uniquely related to health. In terms of mental health, ambivalent ties are associated with higher levels of perceived stress, depression, and lower satisfaction with life when compared to other relationship ties (Uchino et al., 2001; Uchino, Holt-Lunstad, Smith, & Bloor, 2004). As for physical health, when directly interacting with ambivalent ties, people show increased cardiovascular reactivity during laboratory stress, heightened cardiovascular reactivity during support-seeking, and higher ambulatory blood pressure during daily life compared to other relationship ties (Holt-Lunstad, Uchino, Smith, Cerny, & Nealey- Moore, 2003; Holt-Lunstad, Uchino, Smith, & Hicks, 2007; Uno, Uchino, & Smith, 2002). Thus, despite the high positivity that exists in ambivalent relationships, it appears that peoples' health may not be benefiting from these social ties. In fact, our prior work suggests that ambivalent ties may be even more detrimental than more purely negative ties (see Holt-Lunstad et al., 2003; Uchino et al., 2004). In a recent study, we tested the effects of supraliminal activation of the relationships in our relationships model on cardiovascular reactivity (Bloor, Uchino, Hicks, & Smith, 2004), by asking participants to speak about different types of social relationships (e.g., supportive, aversive, ambivalent). Speaking about purely negative ties resulted in the highest levels of cardiovascular reactivity. Although these results are inconsistent with our larger program of research suggesting ambivalent ties are associated with the highest reactivity, we speculated and found ancillary support for the possibility that due to the nature of the task (thinking and then speaking about the relationship), 8 individuals may have been focusing less on the negative aspects of ambivalent relationships and more on the positive aspects (Bloor et al., 2004). Thus, it is unclear if these results would be similar if using a subliminal manipulation of relationship representations in which automatic activation of relational schemata may preclude focusing less on its negative aspects. In the present study, we used a subliminal priming technique to activate relationship representations related to supportive, aversive, ambivalent, and indifferent network ties. In contrast to prior work, by utilizing a subliminal priming procedure, we evaluated whether relationships that are automatically activated in memory influence reactivity, rather than due to any conscious appraisal on the part of the individual or through direct behavioral interactions. We predicted that subliminal activation of ambivalent ties would heighten cardiovascular reactivity, as would negative ties, albeit to a lesser extent. We also predicted that activation of supportive ties would reduce reactivity, consistent with the buffering model and prior research on the effects of social support on physiological stress responses. We also predicted that these effects of subliminal activation on cardiovascular reactivity would be accompanied - and perhaps mediated - by parallel effects on appraisals and self-relevant processes such as state self-esteem. METHOD Participants Seventy healthy women and 36 healthy men (N = 106), with a mean age of 29.8 years were recruited using a university participant pool and advertisements in community papers. Following Cacioppo et al. (1995), the following inclusion criteria were utilized to select a relatively healthy sample: (a) no cardiovascular prescription medication use (b) no preexisting hypertension or history of chronic disease with a cardiovascular component (e.g., diabetes, kidney or heart disease), and (c) no recent history of psychological disorder(s), including those for which currently being medicated (e.g., major depressive disorder, bipolar disorder). All participants had normal or corrected-to-normal vision. Procedure This was a two-session study, in which participants completed Session One, and within two days came back for Session Two. Sessions were split because of possible carry-over effects of priming from filling out the relationship questionnaires. Participants were randomly assigned to a mixed design, with relationship category being between participant (supportive, ambivalent, aversive, indifferent) and epoch being within participants (1st primes, 1st stressors, 2nd primes, 2nd stressors). In Session One, participants filled out informed consent and the social relationships index (SRI, described below). For Session Two, disposable spot electrodes 10 were placed according to published guidelines (Sherwood et al., 1990) and an occluding blood pressure cuff fitted to the nondominant arm. Following a 10-minute resting baseline in which measurements of cardiovascular function were assessed, participants completed a practice trial of the priming task, followed by the actual priming task (see Bargh & Pietromonaco, 1982), and then performed two stressor tasks (i.e., a math task and speech task, see below). These tasks were used because they are well-documented as effective evaluative threats (Dickerson & Kemeny, 2004) and are sensitive to self-relevant subliminal manipulation (Levy et al., 2000). Measurements of cardiovascular functioning were taken during each task. Prior to both math and speech tasks, one-item measures of perceived threat, control, and coping appraisals were taken. Following the last task, a state anxiety and state self-esteem measure was also given. These latter two assessments were only given at the end of the block of stressors (i.e., after both the math and speech tasks) because they are longer assessments and we sought to minimize the time between the primes and subsequent stressor tasks. The entire procedure was then repeated (i.e., priming task, math task, speech task) to increase the reliability of our assessments (Kamarck, 1992). Lastly, participants were debriefed and compensated. Priming Procedure For the priming tasks, individuals were seated in front of a computer and told to focus on a "marker" (a plus (+) sign) appearing in the center of the screen. Participants were instructed to indicate as quickly and accurately as possible if a "flash" appeared above or below the marker by pressing the corresponding direction key on a keyboard (see Bargh & Pietromonaco, 1982; Devine, 1989; Levy et al., 2000). The priming stimuli consisted of names of individuals generated from the SRI that most represented their 11 random assignment to relationship conditions (i.e., supportive, ambivalent, aversive, or indifferent). During this task, participants were randomly presented with 75 trials, or "flashes": 25 filler items, and names of up to 5 individuals in a particular relationship category appearing 10 times each for a total of 50 relevant name trials. The intertrial-interval was 1000 ms. The priming stimuli were identical for both priming tasks (i.e., the same relationship names and the same filler words). The filler items were 25 words referring to unrelated objects, places, and events that were used to diminish the likelihood that participants would detect or recognize the presentation of the experimental stimuli or relevant names. The priming stimulus was randomly presented for 43 ms, immediately followed by a backwards mask (Michaels & Turvey, 1979) of random upper- and lower-case letters, appearing both above and below the marker for 100 ms. Masking was employed to diminish the opportunity for the prime words to persist in the participant's visual iconic memory store (Bargh & Chartrand, 2000). The stimulus presentation time was selected on the basis of prior research (e.g., Bargh & Chartrand, 2000) and pretesting in our laboratory. The exposure time was actually briefer than that utilized in previous studies using these procedures (e.g., Levy et al., 2000). To verify that participants were not consciously aware of the relationship ties presented in the task, a pilot study (n = 10) was conducted. In this pilot, participants were primed with relationship ties as described above. However, right after the prime, participants were asked to identify anything that they saw (i.e., identification task). Of the ten pilot participants, none of them correctly reported the relationship tie that was primed using the above procedure. However, to further ensure that the primes were subliminal, a manipulation check was administered in the main study, in which following 12 the second priming task participants were told that the "flashes" they just saw in the prior task had actually been words; participants were then asked to write down any words they may have seen and were told that they were welcome to guess. Psychological Stressors The laboratory stressors consisted of a mental arithmetic challenge and personal speech stressor (see Cacioppo et al., 1995; Levy et al., 2000). The mental arithmetic task utilized serial subtractions from a 3-digit number. Participants were given the number to start with (i.e., 688 in the first math task and 955 in the second math task) and proceeded to subtract 7 from this number continuously for 1 minute. If the participant lost his/her place or got an answer wrong, the correct answer was given and the participant continued in this fashion for the remainder of the minute. In order to choose personally relevant life events for participants' speeches, participants were asked at the start of Session Two to list five recent negative experiences (see Holt-Lunstad et al., 2007) that were personally relevant, and rated each event on a 1 (not at all negative) to 5 (extremely negative) scale in comparison to all other negative events. The two events rated the highest in negativity were used as the topics for the tasks (i.e., one event for each speech task). For the speech task, the experimenter prompted the participant to talk for 1 minute describing the assigned event, 1 minute about the emotional content of the event, and one minute about how they handled the situation or would have done things differently. Prior to both the math and speech tasks, participants completed brief measures of threat and challenge appraisals and perceived control over the task. Following the completion of each stressor block (i.e., math and speech), participants filled out a state anxiety and state self-esteem scale. 13 Assessments Cardiovascular Measures A Dinamap Model 100 was used to measure SBP and DBP. The Dinamap uses the occillometric method to calculate blood pressure. Blood pressure assessments were obtained using a properly sized occluding cuff positioned on the upper left arm of the participant according to manufacturer's specifications. Mean SBP and DBP for each epoch (i.e., baseline, primes, stressors) were averaged across minutes to increase the reliability of these assessments (Kamarck, 1992). A Mindware 2000D Module was used to measure the ECG which was also used to calculate RSA.1 Seven spot-electrodes were placed according to manufacturer and published guidelines (Hoetink et al., 2002; Sherwood et al., 1990). The ECG was digitized at 1000 Hz and each waveform was verified or edited prior to analyses. Respiratory sinus arrhythmia (RSA) provided a noninvasive measure of parasympathetic control of the heart and was calculated based on the digitized interbeat intervals that were checked and edited for artifacts using the detection algorithm of Bernston, Quigley, Jang, and Boysen (1990). Following linear de-trending (Bernston, Cacioppo, & Quigley, 1995; Litvack, Oberlander, Carney, & Saul, 1995), the heart-period time series was band-pass-filtered from 0.12 to 0.40 Hz (Neuvo, Cheng-Yu, & Mitra, 1984). The power spectrum of heart-period time series was calculated using a Fast Fourier Transform and scaled to ms2/Hz. RSA was calculated as the natural log of the area under the heart-period power 1We also collected impedance-derived assessments of cardiac output, total peripheral resistance, and preejection period. However, we encountered high rates of missing data for the dZ/dt signal, which is used to calculate these measures, and thus did not have significant power to look at these assessments. Exploratory analyses of these variables predictably showed no group differences. 14 spectrum within the corner frequencies of the band-pass filter (Litvack et al., 1995). RSA was calculated on a minute-by-minute basis and aggregated across minutes within each epoch to increase measurement reliability. Social Relationship Index (SRI) The SRI was developed as a self-report version of the social support interview (Fiore, Becker, & Coppel, 1983; Kiecolt-Glaser, Dura, Speicher, Trask, & Glaser, 1991; Uchino, Kiecolt-Glaser, & Cacioppo, 1992). This form of the SRI instructs individuals to list significant other, parents, and then 10 "other" family members, friends, co-workers, and social acquaintances. These network members are rated in terms of how helpful and upsetting they are on a 1 (not at all) to 6 (extremely) point scale. The SRI allows an operationalization of different categories of social relationships as primary sources of positivity (i.e., supportive), primary sources of negativity (i.e., aversive), sources of both positivity and negativity (i.e., ambivalent), or low levels of positivity and negativity (i.e., indifferent). Prior work has shown that these network measures were temporally stable with significant 2-week test-retest correlations (see Campo et al., 2009). Threat and Challenge Appraisals Prior to each math and speech task, participants completed a measure of challenge and threat appraisals utilized by Tomaka, Blascovich, Kibler, and Ernst (1997). Participants were asked to rate on a 6-point Likert scale "how threatening do you expect the task to be" and "how able are you to cope with the task". 15 Perceived Control Prior to each math and speech task, perception of perceived control was assessed via items used in prior social psychophysiological studies (Gerin, Litt, Deich, & Pickering, 1995). Participants were asked, "How much control do you feel you have over this task" on a 10-point Likert scale. State Anxiety Scale The short-form of the Spielberger State-Trait Anxiety Scale was administered to participants following baseline assessment and after completion of each set of psychological stressors (ie., math and speech task; Marteau & Bekker, 1992). Participants were asked to rate their current feelings on a 1 (not at all) to 4 (very much) point scale. The internal consistency of the scale in prior work has been high (α > .78). State Self-Esteem Scale (SSES) The SSES (Heatherton & Polivy, 1991) was administered to participants following baseline assessment and after completion of each set of psychological stressors. The SSES consists of 20 items modified from the widely used Janis-Field Feelings of Inadequacy Scale and is sensitive to both naturally-occurring and laboratory-based threats. This scale has good psychometric properties (e.g., α = .92). RESULTS Manipulation Checks To check participants' awareness of the subliminal primes, immediately following the second priming task, participants were given a brief recall task (see procedure) asking them to list any words that they could recall seeing during the "computer" (priming) task. Consistent with our pilot study, across the entire study, less than 1% of primes were correctly recalled by participants. This did not differ significantly as a function of our experimental conditions. Current guidelines for priming research assert that 5% of participants evidencing genuine awareness of a relation between prime and experimental tasks is high (Bargh & Chartrand, 2000). Because awareness levels for the current study are within most standards for subliminal protocols, and thus possibly due to chance (Bargh & Chartrand, 2000), it is unlikely that there was any degree of "true" awareness of the purpose of the current manipulation. Importantly, deleting individuals who reported at least one correct prime did not alter the pattern of findings reported below. Preliminary Analyses To verify our relationship category manipulation, we ran separate tests for average helpful (positivity) and upset (negativity) ratings of the primed ties. A 2 (Positive Primes: low, high) × 2 (Negative Primes: low, high) ANOVA revealed the expected main effects of positive primes on helpfulness ratings, F(1, 99) = 1073.43, p < .001, and negative primes on upset ratings, F(1, 99) = 371.45 p < .001. We also found a 17 positive primes main effect on upset ratings, F(1, 99) = 27.76, p < .01, and negative primes main effect on helpfulness ratings, F(1, 99) = 14.15, p < .01. In these main effects, high positive primes were associated with lower upset ratings and high negative primes with lower helpfulness ratings. Finally, there were significant positive primes × negative primes interactions on both helpfulness, F(1, 99) = 16.10, p < .01, and upset ratings, F(1, 99) = 18.74, p < .01. As shown in Figure 1, individuals who were in the low negative / high positive condition (supportive) had the highest helpful ratings (top panel), whereas individuals in the low positive / high negative condition (aversive) had the highest upset ratings (bottom panel). Figure 1. Relationship positivity and negativity interaction means for helpfulness (top panel) and upset (bottom panel) ratings. 18 Does Subliminal Priming of Relationships Influence Cardiovascular Reactivity? All physiological measurements were first reduced and averaged into 1-minute segments, followed by an average value being obtained for each epoch to increase the reliability of these assessments (Kamarck, 1992). Change scores were then computed as an index of reactivity (average task epoch - average baseline epoch), with baseline values statistically controlled in the analyses (Llabre, Spitzer, Saab, Ironson, & Schneiderman, 1991). We found no positive or negative prime effects that differed by type of task (i.e., math, speech) or epoch (i.e., first stressor block, second stressor block), so we aggregated across those variables to increase measurement reliability (Kamarck, 1992). In addition, preliminary analyses revealed no significant priming effects that differed by gender. We thus ran a 2 (Positive Primes: low, high) × 2 (Negative Primes: low, high) ANOVA with positive primes and negative primes as between participant factors. Our analyses showed a main effect for negative primes on DBP, F(1, 97) = 3.95, η2 = .04, p < .05, such that participants primed with relationships entailing high negativity had greater DBP reactivity during the tasks than participants primed with relationships entailing low negativity (high negativity LSM = 7.74, low negativity LSM = 5.76). A marginal main effect for negative primes on SBP was also found, F(1, 97) = 3.33, η2 = .03, p = .07, where participants primed with relationships entailing high negativity had greater SBP reactivity across tasks than participants primed with relationships entailing low negativity (high negativity LSM = 14.01, low negativity LSM = 11.54). No positivity main effects were found on any of the cardiovascular assessments (p's > .28). 19 Consistent with our predictions, analyses of HR changes revealed an interaction between positive and negative primes, F(1, 92) = 5.54, η2 = .06, p < .05. More specifically, individuals primed with high positive / high negative (i.e., ambivalent) relationship ties had the highest average HR change from baseline to task (see Figure 2, top panel). Follow-up comparisons revealed that reactivity in the group primed with ambivalent ties was significantly greater than that of supportive (p = .01) and even aversive (p = .02) ties. Figure 2. Relationship positivity and negativity interaction means for heart rate (top panel) and RSA (bottom panel) changes during stress. 20 Following up on these analyses, we looked at RSA as an underlying parasympathetic determinant of HR. Importantly, we found an interaction between positive and negative primes on RSA reactivity, F(1, 93) = 3.82, η2 = .04, p = .05. Consistent with our HR changes, individuals primed with high positive / high negative ties showed the greatest decrease in RSA consistent with exaggerated parasympathetic withdrawal to the stressor task (see Figure 2, bottom panel). Follow-up comparisons revealed that the vagal withdrawal in the ambivalent primes condition was significantly greater than that seen during the supportive primes (p = .03) and marginally greater than during the aversive primes (p = .08) condition. What Are the Psychological Processes That Co-vary With Subliminal Priming of Relationships? Important psychological measures for this study were perceived threat, coping, control, state anxiety, and state self-esteem. Again, no effects involving task or time qualified any priming effects so we averaged over these variables to increase measurement reliability. A 2 (Positive Primes: low, high) × 2 (Negative Primes: low, high) ANOVA was conducted, with these variables as between participant factors. Analyses revealed a main effect of negative primes on perceived threat to the stressor tasks, F(1, 97) = 11.66, η2 = .12, p < .01, such that participants primed with high negative ties had greater perceptions of threat towards the stressor tasks (LSM = 2.82) compared to participants primed with low negative ties (LSM = 2.19). A significant main effect of negative primes was also found for perceptions of control over the stressor task, F(1, 97) = 4.21, η2 = .04, p < .05, where individuals primed with high negative ties 21 reported feeling the least control (LSM = 7.17), compared to those primed with low negative ties (LSM = 7.94). No other main effects or interactions were significant. What Are the Potential Pathways Responsible for Subliminal Priming Effect of Relationships? The main effects of negative primes on perceived threat and control suggest that they may be acting as psychological mediators between negative primes and DBP reactivity. Following outlined procedures for assessing mediation (Baron & Kenny, 1986), and having already established the path from the independent variable (negative primes) on the mediator (perceived threat, control) to be significant, we examined if these mediators predicted the outcome measure (DBP). A 2 (Positivity Primes: low, high) × 2 (Negativity Primes: low, high) ANCOVA was conducted, with the positive primes and negative primes as between participant factors, and perceived threat or control as a covariate. Neither perceived threat, F(1, 95) = 1.12, p = .29, nor perceived task control, F(1, 95) = 1.57, p = .21, predicted DBP reactivity. Thus, for the current study, perceived threat and control did not appear to mediate the link between negative primes and DBP. Because of suggested links between RSA and self-regulatory processes (e.g., Thayer et al., 2009) we also examined if the positive primes × negative primes interaction on HR change was mediated by RSA change. The path from the independent variable (positive and negative primes) was established earlier, so we next checked the ability for RSA change to influence HR change. We ran a 2 (Positive Primes: low, high) × 2 (Negative Primes: low, high) ANCOVA, while also co-varying RSA change. Analyses revealed a significant effect of RSA change on HR change, F(1, 90) = 32.92, p < .01. Importantly, consistent with mediation, the previously significant positive prime X 22 negative prime interaction on HR reactivity was rendered non-significant (p = .12) when controlling for RSA reactivity. The sobel test of the indirect effect was significant (t = 1.90, p = .03, one-tailed). DISCUSSION Prior work on social relationships and cardiovascular function has demonstrated the benefits of positive social support and detriments of negative relationships on health (Berkman, 1995; De Vogli et al., 2007; Friedman et al., 1995). However, little is currently known about the relative influences of positive and negative aspects or relationships or the mechanisms that link them to physical health outcomes. The current study used a subliminal priming technique to investigate a nonconscious cognitive link to cardiovascular reactivity. Our primary aim was to examine whether relationships could be automatically activated through subliminal priming (as opposed to any conscious process or explicit behavior), and if this activation could lead to changes in cardiovascular reactivity. This is the first study we know of that shows subliminal activation of relationship representations can alter cardiovascular functioning. We showed that DBP was significantly affected as a function of high negative relationship prime categories. We also found that high positive/high negative primes (ambivalent category primes) elicited the highest HR change and greatest decrease in RSA, presumed to reflect parasympathetic withdrawal leading to greater sympathetic predominance over vagal control of HR. These findings support the notion that relationship schemata can be automatically activated and contribute to changes in cardiovascular function. Although cardiovascular reactivity is linked to future disease risk (Chida & Steptoe, 2010), the direct health relevance of our 24 findings would be strengthened by future studies including other measures of disease risk (e.g., inflammation). In light of ample research that has shown beneficial effects of social support on health outcomes, one important question is why were we unable to find stronger effects for positive or supportive ties? One explanation is that positive primes may have made social evaluation more salient (Taylor et al., 2010), which could offset any stress-buffering influences. Another possibility is that positivity (support) may require more explicit processing (i.e., more conscious or supraliminal processing) to have physiological benefits. All studies that we are aware of showing beneficial main effects of social support on physiological processes used supraliminal, or more explicit manipulations of support (e.g., directly providing emotional support). It has been shown that supraliminal vs. subliminal priming rely on different neurological substrates and hence are not entirely the same. Functional magnetic resonance imaging (fMRI) studies suggest that supraliminal primes activate a much broader range of processors including more anterior cortical sites than subliminal activation (Koulder, Dehaene, Jobert, & Le Bihan, 2007). The fMRI work of Koulder et al. (2007) showed that supraliminal primes increased occipito-temporal activity, leading to activation in a widely distributed set of parieto-frontal areas, thus demonstrating greater deployment of attentional/strategical processes that aid in the ability to compete for global conscious access and induce priming in multiple distant areas. There is also evidence that subliminal processing is related to a more primitive "quick and dirty" process as compared to a slower, more conscious evaluative process as is associated with supraliminal priming (LeDoux, 1997; Liddell, Williams, Rathjen, Shevrin, & Gordon, 2004). Thus, it is possible that our 25 inability to reproduce beneficial effects of support may be due to these different neural substrates involved in conscious vs. nonconscious processing, although future research directly contrasting these processes would be necessary In contrast, we found that high negative primes were associated with greater blood pressure reactivity. These data are consistent with a negativity bias in which such stimuli elicit stronger responses compared to positive stimuli (see Cacioppo & Bernston, 1994; Taylor, 1991). This effect has previously been shown in the relationship literature in which "not being nasty matters more than being nice" in terms of blood pressure reactivity (Ewart, Taylor, Kraemer, & Agras, 1991). Our data further suggest that this bias is prevalent even at a nonconscious level. The uniformity of the links between negative relationships and different levels of processing may account for the deleterious influences of negative social ties on health outcomes (De Vogli et al., 2007). Consistent with our expectations, we also found evidence that priming ambivalent ties was associated with greater HR and RSA changes during stress. One interesting finding was that the link between ambivalent relationship primes and HR reactivity was mediated by RSA changes. Two different interpretations of this finding are possible. First, Thayer et al. (2009) propose that measures of cardiac vagal tone (e.g., RSA) quantify a person's capacity to self-regulate because of its ability to reflect central-peripheral neural feedback mechanisms related to organization of physiological resources and appropriate response in the service of goal-directed behavior. Increases in vagal tone have been associated with increased self-regulation (Butler, Wilhelm, & Gross, 2006; Smith et al., in press), so if RSA is a unitary dimension, then decreases in RSA should be associated with decreased self-regulation. Consistent with this possibility, vagal 26 withdrawal has been linked to less activation of the prefrontal cortex, which may impair self-regulatory processes (Thayer et al., 2009). Applied to the current study, participants primed with ambivalent ties may be showing a self-regulatory deficit during the stressful tasks. Our findings linking social ties to RSA-related self-regulatory processes are also consistent with models emphasizing that activation of relationship representations can influence self-relevant processes (Aron, Aron, Tudor, & Nelson, 1991; Ogilvie & Ashmore, 1991). A second possible interpretation of our findings is based on Porges' (2007) Polyvagal Theory, where a decrease in RSA appears to reflect a switch-over to a fight/flight response. This would represent a change to a more primitive type of system to deal with threats, where the first action would be to let up on the quicker parasympathetic "brake" to confront stressful events. In this sense, our RSA withdrawal associated with ambivalent primes would represent this early stage response of preparing for something threatening or stressful. One potential problem with this explanation is we did not find differences in threat appraisals when participants were primed with ambivalent ties. However, several models of emotion processing indicate that the neural mechanisms for appraising signals of threat may be initiated automatically and without the need for conscious detection of the threatening signals (e.g., Halgren & Marinkovic, 1995). Prior research in our laboratory using the same relationship framework had participants recall and speak about specific relationships (consistent with a supraliminal priming technique; Bloor et al., 2004). Results showed the greatest cardiovascular response when women talked about aversive relationships compared to talking about 27 supportive and indifferent relationships, but did not demonstrate an effect of talking about an ambivalent relationship. In contrast, the present study showed that reactivity was highest when primed with an ambivalent tie. One explanation for these inconsistent findings is that the processes of supraliminal vs. subliminal priming rely on different neurological substrates as noted above (Koulder et al., 2007; LeDoux, 1997; Liddell et al., 2004). We should also note that the paradigm by Bloor et al. (2004) is quite different than more traditional supraliminal primes, which are shorter in duration (e.g., 500 ms). Thus, it is possible that aversive effects in the Bloor et al. study, and null ambivalent effects, are due to six minutes of mulling over their relationship with a highly negative individual. An ambivalent tie, while also entailing negativity, also has positive aspects to draw on. In fact, ancillary analyses revealed evidence for such an interpretation as individuals speaking about ambivalent ties made as many positive comments about them as supportive ties, but less negative comments compared to aversive ties (Bloor et al., 2004). This type of "reflective and selective" focus on relational processes is less likely to occur during automatic or subliminal activation (Koulder et al., 2007). We were also interested in the more precise psychological pathways by which priming relationships influence cardiovascular function. We found that high negative primes were associated with more perceived threat and less perceived control over the tasks compared to low negative primes. We did not, however, find any significant interaction effect of high positive/high negative primes for any psychological assessment, suggesting that with the current paradigm, no effect specific to ambivalent ties was influencing these processes. In addition, the appraisal measures did not statistically mediate the link between negative primes and DBP reactivity. These results are 28 consistent with the uncoupling that can occur between self-report and physiological processes (e.g., Gerin, Pieper, Levy, & Pickering, 1992). Recent research is emphasizing links between physiology and more specific emotional processes (e.g., social emotions, Dickerson & Kemeny, 2004) and may be particularly important for future research in this area. Unexpectedly, we were not able to replicate the overall state self-esteem effects found in Baldwin's (1994) priming study. Furthermore, Baldwin reported that most of their self-esteem effect was attributable to the social evaluation subscale (Heatherton & Polivy, 1991). Inspection of the same subscale for the current study was also non-significant. In Baldwin's study, participants had a 3-minute memory task between the actual priming and then filled out the self-esteem measure, whereas in our study the time frame between the priming and completing the self-esteem measure was considerably longer. Thus, the null findings for some of our psychological measures could be due to the delay of priming task to completion of the self-esteem and anxiety measure. Although stronger priming effects on state self-esteem might be possible if given to participants earlier, our main goal was to demonstrate an influence on cardiovascular reactivity; thus, it was more important to minimize the time between the primes and physiological recordings during the stressful tasks. One other potential limitation of our study is that we did not control respiration rate or depth, which some have argued can alter the validity of RSA (Beda, Jandre, Phillips, Giannella-Neto, & Simpson, 2007). However, RSA is often not highly susceptible to such artifacts within many behavioral contexts (Denver, Reed, & Porges, 2007; Kotani, Tachibana, & Takamasu, 2007). In fact, a comparison of different 29 quantification methods (corrected and uncorrected for respiration) concluded that such assessments ". . . are almost equivalent as indices of cardiac vagal tone and would appear to ease concerns about the inferiority of any of the procedures." (Grossman, Beek, & Wienties, 1990). The links between RSA and heart rate (and its mediational influence) also gives us more confidence in the validity of our assessments (Porges, 2007). The limitations of this study notwithstanding, one additional mechanism linking relationships to health outcomes appears to be related to the activation of internal representations of individuals in our social networks. The current findings are important, in that mental representations of relationships appear to influence individuals in the absence of actual interpersonal interactions. From photographs around the home (Baldwin, 1990), to a familiar name on a billboard (Baldwin, 1994) or meeting someone new (Andersen & Baum, 1994), our everyday lives are filled with potential stimuli for non-conscious activation of mental representations of relationships. Our results open a novel research domain to examine the physical costs and benefits of conscious vs. non-conscious activation of relational schemata, which can inform contemporary theories linking relationships to physical health outcomes. APPENDIX Table 1 Means and Standard Deviations for Major Dependent Variables Relationship type Variable Indifferent Supportive Ambivalent Aversive SBP 11.38 (7.82) 11.19 (5.60) 13.4 (6.95) 15.08 (7.01) DBP 5.75 (5.45) 5.37 (4.85) 7.73 (4.74) 8.11 (5.16) HR 7.61 (6.68) 6.13 (5.63) 10.6 (8.32) 6.02 (5.55) RSA -0.10 (0.82) 0.27 (0.77) -0.30 (1.10) 0.07 (0.81) Threat 2.25 (0.98) 2.13 (0.88) 2.74 (0.98) 2.91 (0.85) Control 8.17 (1.84) 7.72 (1.77) 6.87 (2.14) 7.48 (1.71) Coping 5.00 (1.00) 5.16 (0.78) 4.88 (0.76) 4.83 (0.71) Self-Esteem -0.03 (0.22) 0.03 (0.25) -0.07 (0.19) -0.01 (0.23) Anxiety 0.55 (0.58) 0.56 (0.66) 0.53 (0.42) 0.57 (0.45) REFERENCES Andersen, S. M., & Baum, A. (1994). Transference in interpersonal relations: Inferences and affect based on significant other representations. Journal of Personality, 62, 459-497. Andersen, S. M., & Berk, M. S. (1998). The social-cognitive model of transference: Experiencing past relationships in the present. Current Directions in Psychological Science, 7, 109-115. Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships as including other in the self. Journal of Personality and Social Psychology, 60, 241-253. Baldwin, M. W. (1990). Priming relationship schemas: My advisor and the Pope are watching me from the back of my mind. Journal of Experimental Social Psychology, 26, 435-454. Baldwin, M. W. (1992). Relational schemas and the processing of social information. Psychological Bulletin, 112, 461-484. Baldwin, M. W. (1994). Primed relational schemas as a source of self-evaluative reactions. Journal of Social and Clinical Psychology, 13, 380-403. Barrera, M. (2000). Social support research in community psychology. In J. Rappaport & E. Seidman (Eds.), Handbook of community psychology (pp. 215-245). New York, NY: Kluwer Academic/Plenum. Bargh, J. A., & Chartrand, T. L. (2000). The mind in the middle: A practical guide to priming and automaticity research. In H. T. Reis and C. M. Judd (Eds.), Handbook of research methods in social and personality psychology (pp. 253- 285). New York, NY: Cambridge University Press. Bargh, J. A., & Pietromonaco, P. (1982). Automatic information processing and social perception: The influence of trait information presented outside of conscious awareness on impression formation. Journal of Personality and Social Psychology, 43, 437-449. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. 32 Barth, J., Schneider, S., & von Kanel, R. (2010). Lack of social support in the etiology and prognosis of coronary heart disease: A systematic review and meta-analysis. Psychosomatic Medicine, 72, 229-238. Beda, A., Jandre, F. C., Phillips, D. W., Giannella-Neto, A., & Simpson, D. M. (2007). Heart rate and blood pressure variability during psychophysiological tasks involving speech: Influence of respiration. Psychophysiology, 44, 767-778. Berkman, L. F. (1995). The role of social relations in health promotion. Psychosomatic Medicine, 57, 245-254. Bernston, G. G., Cacioppo, J. T., Binkley, P. F., Uchino, B. N., Quigley, K. S., & Fieldstone, A. (1994). Autonomic cardiac control: III. Psychological stress and cardiac response in autonomic space as revealed by pharmacological blockades. Psychophysiology, 31, 599-608. Bernston, G. G., Cacioppo, J. T., & Quigley, K. S. (1995). The metrics of cardiac chronotropism: Biometric perspectives. Psychophysiology, 32, 162-171. Bernston, G. G., Quigley, K. S., Jang, J., & Boysen, S. (1990). An approach to artifact identification: Application to heart period data. Psychophysiology, 27, 586-598. Blazer, D. (1982). Social support and mortality in an elderly community sample. American Journal of Epidemiology, 115, 684-694. Bloor, L. E., Uchino, B. N., Hicks, A., & Smith, T. W. (2004). Social relationships and physiological function: The effects of recalling social relationships on cardiovascular reactivity. Annals of Behavioral Medicine, 28, 29-38. Bolger, N., Zuckerman, A., & Kessler, R. C. (2000). Invisible support and adjustment to stress. Journal of Personality and Social Psychology, 79, 953-961. Braiker, H. B., & Kelley, H. H. (1979). Conflict in the development of close relationships. In R. L. Burgess and T. L. Huston (Eds.), Social exchange in developing relationships (pp. 135-168). New York, NY: Academic Press. Butler, E. A., Wilhelm, F. H., & Gross, J. J. (2006). Respiratory sinus arrhythmia, emotion, and emotion regulation during social interaction. Psychophysiology, 43, 612-622. Cacioppo, J. T., & Bernston, G. G. (1994). Relationship between attitudes and evaluative space: A critical review, with emphasis on the separability of positive and negative substrates. Psychological Bulletin, 115, 401-423. 33 Cacioppo, J. T., Malarkey, W., Kiecolt-Glaser, J. K., Uchino, B. N., Sgoutas-Emch, S. A., Sheridan, J. F., Bernston, G. G., & Glaser, R. (1995). Heterogeneity in neuroendocrine and immune responses to brief psychological stressors as a function of autonomic cardiac activation. Psychosomatic Medicine, 57, 154-164. Campo, R. A., Uchino, B. N., Holt-Lunstad, J., Vaughn, A. A., Reblin, M., & Smith, T. W. (2009). The assessment of positivity and negativity in social networks: The reliability and validity of the social relationships index. Journal of Community Psychology, 37, 471-486. Chida, Y., & Steptoe, A. (2010). Greater cardiovascular responses to laboratory mental stress are associated with poor subsequent cardiovascular risk status. Hypertension, 55, 1026-1032. Cohen, S. (1988). Psychosocial models of the role of social support in the etiology of physical disease. Health Psychology, 7, 269-297. Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310-357. Denver, J. W., Reed, S. F., & Porges, S. W. (2007). Methodological issues in the quantification of respiratory sinus arrhythmia. Biological Psychology, 74, 286- 294. Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5-18. De Vogli, R., Chandola, T., & Marmot, M. G. (2007). Negative aspects of close relationships and heart disease. Archives of Internal Medicine, 167, 1951-1957. Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressor and cortisol responses: A theoretical integration and synthesis of laboratory research. Psychological Bulletin, 130, 355-391. Ewart, C. K., Taylor, C. B., Kraemer, H. C., & Agras, W. S. (1991). High blood pressure and marital discord: Not being nasty matters more than being nice. Health Psychology, 10, 155-163. Fiore, J., Becker, J., & Coppel, D. B. (1983). Social network interactions: A buffer or a stress? American Journal of Community Psychology, 11, 423-439. Friedman, H. S., Tucker, J. S., Schwartz, J. E., Tomlinson-Keasey, C., Martin, L. R., Wingard, D. L., & Criqui, M. H. (1995). Psychosocial and behavioral predictors of longevity: The aging and death of the "Termites." American Psychologist, 50, 69-78. 34 Gallo, L. C., Smith, T. W., & Kircher, J. C. (2000). Cardiovascular and electrodermal responses to support and provocation: Interpersonal methods in the study of psychophysiologic reactivity. Psychophysiology, 37, 289-301. Gerin, W., Litt, M. D., Deich, J., & Pickering, T. G. (1995). Self-efficacy as a moderator of perceived control effects on cardiovascular reactivity: Is enhanced control always beneficial? Psychosomatic Medicine, 57, 390-397. Gerin, W., Pieper, C., Levy, R., & Pickering, T. G. (1992). Social support in social interaction: A moderator of cardiovascular reactivity. Psychosomatic Medicine, 54, 324-336. Grossman, P., Beek, J., & Wienties, C. (1990). A comparison of three quantification methods for estimation of respiratory sinus arrhythmia. Psychophysiology, 27, 702-714. Halgren, E., & Marinkovic, K. (1995). Neurophysiological networks integrating human emotions. In M. Gazzaniga (Ed.), The cognitive neurosciences (pp. 1137-1151). Cambridge: MIT Press. Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895-910. Hoetink, A. E., Faes, T. J., Schuur, E. H., Gorkink, R., Goovaerts, H. G., Meijer, J. H., and Heethaar, R. M. (2002). Comparing spot electrode arrangements for electric impedance cardiography. Physiological Measurement, 23, 457-467. Holt-Lunstad, J., Smith, T. B., & Layton, B. (in press). Social Relationships and Mortality: A Meta-Analysis. PLoS Medicine. Holt-Lunstad, J., Uchino, B. N., Smith, T. W., Cerny, C. B., & Nealey-Moore, J. B. (2003). Social relationships and ambulatory blood pressure: Structural and qualitative predictors of cardiovascular function during everyday social interactions. Health Psychology, 22, 388-397. Holt-Lunstad, J., Uchino, B. N., Smith, T. W., & Hicks, A. (2007). On the importance of relationship quality: The impact of ambivalence in friendships on cardiovascular functioning. Annals of Behavioral Medicine, 33, 1-12. House, J. S., Landis, K. R., & Umberson, D. (1988). Social relationships and health. Science, 241, 540-545. Kamarck, T. (1992). Recent developments in the study of cardiovascular reactivity: Contributions from psychometric theory and social psychology. Psychophysiology, 29, 491-503. 35 Kiecolt-Glaser, J. K., Dura, J. R., Speicher, C. E., Trask, J. O., & Glaser, R. (1991). Spousal caregivers of dementia victims: Longitudinal changes in immunity and health. Psychosomatic Medicine, 53, 345-362. Kiecolt-Glaser, J. K., & Glaser, R. (1989). Interpersonal relationships and immune function. In L. Carstensen & J. Neale (Eds.), Mechanisms of psychological influence on physical health (pp. 43-59). New York, NY: Plenum. Kotani, K., Tachibana, M., & Takamasu, K. (2007). Investigation of the influence of swallowing, coughing and vocalization on heart rate variability with respiratory-phase domain analysis. Methods of Information in Medicine, 46, 179-185. Koulder, S., Dehaene, S., Jobert, A., & Le Bihan, D. (2007). Cerebral bases of subliminal and supraliminal priming during reading. Cerebral Cortex, 17, 2019-2029. Krantz, D., & Manuck, S. (1984). Acute physiologic reactivity and risk of cardiovascular disease: A review and methodologic critique. Psychological Bulletin, 96, 435- 464. Ledoux, J. E. (1997). Emotion, memory and the brain. Scientific American Special Issue, 7, 68. Lepore, S. J. (1995). Cynicism, social support, and cardiovascular reactivity. Health Psychology, 14, 210-216. Lepore, S. J. (1998). Problems and prospects for the social support-reactivity hypothesis. Annals of Behavioral Medicine, 20, 257-269. Levy, B. R., Hausdorff, J. M., Hencke, R., & Wei, J.Y. (2000). Reducing cardiovascular stress with positive self-stereotypes of aging. Journal of Gerontology: Psychological Sciences, 55B, 205-213. Liddell, B. J., Williams, L. M., Rathjen, J., Shevrin, H., & Gordon, E. (2004). A temporal dissociation of subliminal versus supraliminal fear perception: An event-related potential study. Journal of Cognitive Neuroscience, 16, 479-486. Litvack, D., Oberlander, T., Carney, L., & Saul, J. (1995). Time and frequency domain methods for heart rate variability: A methodological comparison. Psychophysiology, 32, 492-504. Llabre, M. M., Spitzer, S. B., Saab, P. G., Ironson, G. H., & Schneiderman, N. (1991). The reliability and specificity of delta versus residualized change as measures of cardiovascular reactivity to behavioral challenges. Psychophysiology, 28, 701- 711. 36 Marteau, T. M., & Bekker, H. (1992). The development of a six-item short form of the state scale of the Spielberger state-trait anxiety inventory (STAI). British Journal of Clinical Psychology, 31, 301-306. Martire, L. M., Stephens, M. A. P., Druley, J. A., & Wonjo, W. C. (2002). Negative reactions to received spousal care: Predictors and consequences of miscarried support. Health Psychology, 21, 167-176. Matthews, K. A., & Gump, B. B. (2002). Chronic work stress and marital dissolution increase risk of post-trial mortality in men from the Multiple Risk Factor Intervention Trial. Archives of Internal Medicine, 162, 309-315. McNamara, T. P. (1994). Priming and theories of memory: A reply to Ratcliff and McKoon. Psychological Review, 101, 185-187. Michaels, C. F., & Turvey, M. T. (1979). Central sources of visual masking: Indexing structures supporting seeing at a single, brief glance. Psychological Research, 41, 1-61. Nadler, A., & Fisher, J. D. (1986). The role of threat to self-esteem and perceived control in recipient reaction to help: Theory development and empirical validation. In L. Berkowitz (Ed.), Advances in experimental social psychology (pp. 81-122). New York, NY: Academic. Nealey-Moore, J. B., Smith, T. W., Uchino, B. N., Hawkins, M. W., & Olson-Cerny, C. (2007). Cardiovascular reactivity during positive and negative marital interactions. Journal of Behavioral Medicine, 30, 505-519. Neuvo, Y., Cheng-Yu, D., & Mitra, S. (1984). Interpolated finite impulse response filters. IEEE Trans. ASSP, 32, 563-570. Ogilvie, D. M., & Ashmore, R. D. (1991). Self-with-other representation as a unit of analysis in self-concept research. In R. C. Curtis (Ed.), The relational self: Theoretical convergences in psychoanalysis and social psychology (pp. 282-314). New York, NY: Guilford. Porges, S. W. (2007). The polyvagal perspective. Biological Psychology, 74, 116-143. Ratnasingam, P., & Bishop, G. D. (2007). Social support schemas, trait anger, and cardiovascular responses. International Journal of Psychophysiology, 63, 308- 316. Rook, K. S., & Pietromonaco, P. (1987). Close relationships: Ties that heal or ties that bind. Advances in Personal Relationships, 1, 1-35. 37 Sherwood, A., Allen, M. T., Fahrenberg, J., Kelsey, R. M., Lovallo, W. R., & van Doornen, L. J. P. (1990). Methodological guidelines for impedance cardiography. Psychophysiology, 27, 1-23. Smith, T. W., Cribbet, M. R., Nealey-Moore, J. B., Uchino, B. N., Williams, P. A., & MacKenzie, J. (in press). Matters of the variable heart: Respiratory sinus arrhythmia response to marital interaction and associations with marital quality. Journal of Personality and Social Psychology. Smith, T. W., Ruiz, J. M., & Uchino, B. N. (2004). Mental activation of supportive ties, hostility, and cardiovascular reactivity to laboratory stress in young men and women. Health Psychology, 23, 476-485. Smith, T. W., Uchino, B. N., Berg, C. A., Florsheim, P., Pearce, G., Hawkins, M., Henry, N. J. M., Beveridge, R. M., Skinner, M. A., Ko, K., & Olsen-Cerny, C. (2009). Conflict and collaboration in middle-aged and older couples: II: Cardiovascular reactivity during marital interaction. Psychology and Aging, 24, 274-286. Taylor, S. E. (1991). Asymmetrical effects of positive and negative events: The mobilization-minimization hypothesis. Psychological Bulletin, 110, 67-85. Taylor, S. E., Seeman, T. E., Eisenberger, N. I., Kozanian, T. A., Moore, A. N., & Moons, W. G. (2010). Effects of a supportive or an unsupportive audience on biological and psychological responses to stress. Journal of Personality and Social Psychology, 98, 47-56. Thayer, J. F., Hansen, A. L., Saus-Rose, E., & Johnsen, B. H. (2009). Heart rate variability, prefrontal neural function, and cognitive performance: The neurovisceral integration perspective on self-regulation, adaptation, and health. Annals of Behavioral Medicine, 37, 141-153. Thorsteinsson, E. B., & James, J. E. (1999). A meta-analysis of the effects of experimental manipulations of social support during laboratory stress. Psychology and Health, 14, 869-886. Tomaka, J., Blascovich, J., Kibler, J., & Ernst, J. M. (1997). Cognitive and physiological antecedents of threat and challenge appraisal. Journal of Personality and Social Psyhcology, 73, 63-72. Uchino, B. N. (2004). Social support and physical health: Understanding the health consequences of our relationships. New Haven, CT: Yale University Press. Uchino, B. N. (2006). Social support and health: A review of physiological processes potentially underlying links to disease outcomes. Journal of Behavioral Medicine, 29, 377-387. 38 Uchino, B. N. (2009). Understanding the links between social support and physical health: A life-span perspective with emphasis on the separability of perceived and received support. Perspectives on Psychological Science, 4, 236-255. 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. Uchino, B. N., Holt-Lunstad, J., Smith, T. W., & Bloor, L. (2004). Heterogeneity in social networks: A comparison of different models linking relationships to psychological outcomes. Journal of Social and Clinical Psychology, 23, 123-139. Uchino, B. N., Holt-Lunstad, J., Uno, D., & Flinders, J. B. (2001). Heterogeneity in the social networks of young and older adults: Prediction of mental health and cardiovascular reactivity during acute stress. Journal of Behavioral Medicine, 24, 361-382. Uchino, B. N., Kiecolt-Glaser, J. K., & Cacioppo, J. T. (1992). Age-related changes in cardiovascular response as a function of a chronic stressor and social support. Journal of Personality and Social Psychology, 63, 839-846. Uno, D., Uchino, B. N., & Smith, T. W. (2002). Relationship quality moderates the effect of social support given by close friends on cardiovascular reactivity in women. International Journal of Behavioral Medicine, 9, 243-262.