Do Slumped and Upright Postures Affect Stress Responses? A Randomized Trial

Health Psychology Do Slumped and Upright Postures Affect Stress Responses? A Randomized Trial Shwetha Nair, Mark Sagar, John Sollers, III, Nathan Consedine, and Elizabeth Broadbent Online First Publication, September 15, 2014. http://dx.doi.org/10.1037/hea0000146

CITATION Nair, S., Sagar, M., Sollers, J., III, Consedine, N., & Broadbent, E. (2014, September 15). Do Slumped and Upright Postures Affect Stress Responses? A Randomized Trial. Health Psychology. Advance online publication. http://dx.doi.org/10.1037/hea0000146

Health Psychology 2014, Vol. 33, No. 9, 000

© 2014 American Psychological Association 0278-6133/14/$12.00 http://dx.doi.org/10.1037/hea0000146

Do Slumped and Upright Postures Affect Stress Responses? A Randomized Trial Shwetha Nair, Mark Sagar, John Sollers, III, Nathan Consedine, and Elizabeth Broadbent

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

The University of Auckland Objective: The hypothesis that muscular states are related to emotions has been supported predominantly by research on facial expressions. However, body posture also may be important to the initiation and modulation of emotions. This experiment aimed to investigate whether an upright seated posture could influence affective and cardiovascular responses to a psychological stress task, relative to a slumped seated posture. Method: There were 74 participants who were randomly assigned to either a slumped or upright seated posture. Their backs were strapped with physiotherapy tape to hold this posture throughout the study. Participants were told a cover story to reduce expectation effects of posture. Participants completed a reading task, the Trier Social Stress speech task, assessments of mood, self-esteem, and perceived threat. Blood pressure and heart rate were continuously measured. Results: Upright participants reported higher self-esteem, more arousal, better mood, and lower fear, compared to slumped participants. Linguistic analysis showed slumped participants used more negative emotion words, first-person singular pronouns, affective process words, sadness words, and fewer positive emotion words and total words during the speech. Upright participants had higher pulse pressure during and after the stressor. Conclusions: Adopting an upright seated posture in the face of stress can maintain self-esteem, reduce negative mood, and increase positive mood compared to a slumped posture. Furthermore, sitting upright increases rate of speech and reduces self-focus. Sitting upright may be a simple behavioral strategy to help build resilience to stress. The research is consistent with embodied cognition theories that muscular and autonomic states influence emotional responding. Keywords: self-esteem, stress, emotions, posture, embodied cognition

emotion (Niedenthal, 2007). To further enhance our understanding of the effects of muscular movements on emotions, calls have been made for more attention to be paid to general somatic features, such as posture (McIntosh, 1996). Observations of patients with depression support the hypothesis that posture is related to emotions. Slumped posture is a physical feature of patients with major depression (Michalak et al., 2009), and the Diagnostic and Statistical Manual of Mental Disorders (5th ed., American Psychiatric Association, 2013) includes stooped posture when sitting as a sign of psychomotor disturbance in the criteria for a major depressive episode. One of the psychological explanations for how posture may be linked to emotions is through self-observation and attribution (Bem, 1972). Thus, if we are slumped we may infer that we are sad, and if we are sitting upright we may infer that we are confident. Literature on body image has shown that we do have an awareness of our body postures which lends support to the notion that we may infer our own emotions through observations of posture (Schilder, 1999). A handful of studies has suggested that adopting an upright spinal posture may lead to less negative and more positive emotions than a slumped posture, but this research is limited. Few studies have specifically examined stooped and upright sitting posture. In Riskind and Gotay’s (1982) study, sitting erect did not result in any differences in self-reported mood compared to sitting slumped, although it did result in longer persistence at trying to solve an unsolvable task (interpreted as lower learned helplessness). In other research when participants received successful

The question of whether muscular changes can influence emotions has been studied by psychologists for over 100 years. William James (1890) first proposed that muscular and autonomic changes were responsible for the generation of emotions. Most subsequent research on muscular changes has focused on the face. The facial feedback hypothesis proposes that facial proprioception is the experience of emotion (Tourangeau & Ellsworth, 1979). Data consistent with this view is seen in three lines of research, (1) facial configuration can correspond to emotions, (2) facial feedback can modulate emotions, and (3) facial expression can initiate emotions (McIntosh, 1996). Further support for the hypothesis comes from the use of Botox injections to reduce frown lines, which has been shown to reduce depression in people with major depression (Hexsel et al., 2013). The role of sensorimotor abilities has received renewed attention in recent work on embodied cognition theories, which propose that the activation of sensory, motor, and affective states play a role in shaping cognition and

Shwetha Nair, Department of Psychological Medicine, The University of Auckland; Mark Sagar, Bioengineering Institute, The University of Auckland; John Sollers III, Nathan Consedine, and Elizabeth Broadbent. Correspondence concerning this article should be addressed to Elizabeth Broadbent, Department of Psychological Medicine, Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, 85 Park Road, Grafton, Auckland, New Zealand. E-mail: [email protected] 1

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

2

NAIR, SAGAR, SOLLERS, CONSEDINE, AND BROADBENT

results in an upright position compared to a slumped position, there were no differences in any specific positive or negative emotions (cheerful, relaxed, amused, depressed, angry, or aggressive) with the exception of pride (Stepper & Strack, 1993). Likewise, sitting in an erect posture did not affect participants’ mood or valence of thoughts compared to a slumped posture, although it did create more confidence in thoughts (Briñol, Petty, & Wagner, 2009). Related work has found evidence that other types of postures can affect physiological responses. Receiving an insult when lying down compared to standing up, did not result in any differences in self-reported anger but did reduce left lateral frontal activity in EEG, suggesting that supine positions can reduce approachmotivational responses (Harmon-Jones & Peterson, 2009). Adopting an expansive whole body posture produced feelings of power, increased testosterone, lowered cortisol, and increased tolerance of risk compared to a contracted posture (Carney, Cuddy, & Yap, 2010), while generating words related to disappointment led to shortened posture relative to pride words (Oosterwijk, Rotteveel, Fischer, & Hess, 2009). One of the reasons that studies on the effects of posture on emotions have had limited success, may be due to the methods employed and the unstructured way emotions have been assessed. The structure of affect has been described as being composed of two dimensions, pleasure and activation, and affective states fall in a circumflex pattern around the axes (Barrett & Russell, 1999). Using measures based on this model may offer insights into the effects of posture on affect along these dimensions. Linguistic analysis may be an additional method by which to assess changes in emotional or affective states that does not rely on self-report. The use of language has been linked to the activation of autonomic and muscular systems in embodied cognition research (Niedenthal, 2007). Patterns of word use are thought to reflect the thoughts and feelings of an individual (Pennebaker, Chung, Ireland, Gonzales, & Booth, 2007). Greater use of first-person singular pronouns has been linked with negative affective states and depressive symptoms (Chung & Pennebaker, 2007; Rodriguez, Holleran, & Mehl, 2010; Rude, Gortner, & Pennebaker, 2004). In theory, people who are sad or depressed use more first-person singular pronouns because they are more self-focused. According to self-regulatory perseveration theory, the loss of self-worth leads to excessive self-focus regarding the loss, which results in intensified negative affect, and adoption and maintenance of a negative self-image (Pyszczynski & Greenberg, 1987). Together with muscular changes, autonomic changes also are theorized to affect emotion. Theories of how autonomic changes are linked with emotions span psychological, brain-behavioral, and physiological levels (involving cardiovascular, respiratory, and electrodermal variables). A recent review found evidence for both emotion-specific autonomic responses and similarity in responses across different emotions (Kreibig, 2010). Heart rate, for example, was increased across emotions such as anger, crying sadness, and joy, but reduced in emotions that included an element of passivity such as noncrying sadness and contentment. Including autonomic measures is therefore important in studies to fully investigate the effects of posture on emotional processes. This study extends the methods used in previous work by using a differentiated circumplex-based measure of felt affect, assessing cardiovascular activation, and including linguistic analysis. Because a lack of blinding may limit prior studies (e.g., a slumped

“sad” posture did not increase self-reported sadness in the total sample, but only in the subset of participants who interpreted the posture as sad; Duclos et al., 1989), the current study reduced demand characteristics by maintaining participant blindness. Equally, prior studies of posture and emotion have had limited success in that most did not elicit strong emotions. This study addresses this limitation by eliciting a response to a stressful task. The question of whether postural interventions can reduce the stress response is of particular interest because it may have beneficial implications for health. Psychological stress negatively impacts immune function and health outcomes and contributes to depression (Glaser & Kiecolt-Glaser, 2005). Initial evidence supports the hypothesis that muscular feedback can affect stress responses. First, the manipulation of facial expressions has been shown to reduce both affective and cardiovascular stress responses (Kraft & Pressman, 2012). Holding the expression of a smile during stressful tasks resulted in a lower heart rate during the recovery period compared to holding a neutral expression. Holding a smile also resulted in less of a decrease in positive affect during the stressor compared to the neutral expression, in those participants not explicitly aware they were smiling. Second, a supine position led to lower anticipatory anxiety and blood pressure prior to a speech task than standing, although there were no differences in anxiety during the speech performance itself (Lipnicki & Byrne, 2008). Third, holding an expansive posture for 2 min prior to a speech task resulted in higher observer ratings of confidence than a contracted posture (Cuddy, Wilmuth, & Carney, 2012). This experimental study examines the effects of upright versus slumped seated posture on affective states and cardiovascular responses during a controlled stressor. The Trier Social Stress Test (Kirschbaum, Pirke, & Hellhammer, 1993) is an acute laboratorybased stress task. It results in subjective increases in stress, social insecurity, and anxiety, as well as increases in endocrine, cardiovascular, and immunological parameters (Kirschbaum et al., 1993; Hellhammer & Schubert, 2012). Our overall hypothesis was that upright posture would reduce the stress response compared to slumped posture. In particular, it was hypothesized that participants in an upright posture would have higher self-esteem, lower negative affective states, higher positive affective states, and lower fear. This was based on findings that upright and more expansive postures can increase pride and confidence (Briñol et al., 2009; Cuddy et al., 2012; Stepper & Strack, 1993), and that slumped posture is related to depression (Michalak et al., 2009). According to the findings of Chung and Pennebaker (2007), lower negative affect and higher positive affect in the upright posture group would be manifest in greater use of words that reflected positive feelings (such as nice and love) and fewer words that reflected negative feelings (such as sad, and first person singular pronouns). It was further hypothesized that upright participants would have reduced cardiovascular responses (lower heart rate, blood pressure, and pulse pressure) to the stress task compared to participants in a slumped posture. This hypothesis was based on findings showing a smile can reduce heart rate during stressor recovery (Kraft & Pressman, 2012). Following the previous findings that upright posture increased task persistence (Riskind & Gotay, 1982), we further hypothesized that upright posture would increase task persistence as evidenced by greater total word count.

POSTURE AND STRESS

3

Method

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Participants There were 74 healthy participants who were recruited from the University of Auckland and the general public in Auckland, New Zealand. To be included in the study, participants had to be able to read and write English, and be over the age of 18 years. All participants were screened via telephone interview or by e-mail. The exclusion criteria for this study were regularly visiting a physiotherapist for any musculoskeletal related problems or a current diagnosis of psychopathology, cardiac conditions (hypertensive), or dermatological conditions. Those with training or expertise in physiotherapy also were excluded. The sample was mostly female (55%) with a mean age of 26 years (range 18 to 67 years). The study was approved by the University of Auckland Human Participants Ethics Committee, and written informed consent was obtained from all participants. Participants were told a fictitious cover story to reduce expectation effects of posture: People who visit physiotherapists sometimes report effects from tape placed on their backs in their daily physical, mental, and social tasks. This study is to investigate the effects of tape on your upper back on your physiology, mood, and performance of a speech task.

They were not told that they were randomized to one of two conditions to manipulate posture. Participants were debriefed regarding the study purpose at the end of the trial.

Procedure Participants were invited to come to the university (by Shwetha Nair) where they completed a baseline questionnaire to assess demographics, mood, and self-esteem. They were then fitted with a finger cuff placed tightly around their middle finger to measure blood pressure and heart rate continuously throughout the study. Their posture was measured at baseline by asking participants to look at the camera while sitting on a stool for 2 min (purportedly to calibrate the camera). Participants were then randomly assigned (by Shwetha Nair) using a computer-generated randomized list (made by Elizabeth Broadbent). Group allocation was concealed by using consecutively numbered sealed envelopes that were opened only after baseline assessments were made. Participants allocated to the slumped group were asked to sit on the stool in a slumped condition (bowed head, rounded shoulders, and stooped back); while participants allocated to the upright group were asked to sit on the stool in an upright condition (erect back, straight shoulders, and back) using the instructions shown in Figure 1. Physiotherapy tape was laid over the skin while participants sat in their respective postures and LED markers were placed on the shoulders and head for purposes of postural analysis. Videos capturing the postural angles were taken at four time points: baseline, reading task, speech preparation, and speech task. A computer vision based tracking system developed at the Laboratory for Animate Technologies in the Auckland Bioengineering Institute was used to determine the location of the LED markers as a function of time.1 The marker coordinates were used to calculate the angles. Immediately after the postural intervention, participants again completed the mood questionnaire that was followed by the read-

Figure 1. Postural instructions. (A) Images shown during the postural instructions for the slumped condition. (B) Images shown during the postural instructions for the upright condition.

ing task. They then completed a speech task, rested for 5 min, and completed a final questionnaire assessing mood, self-esteem, and perceived threat. The tape was then removed. Participants were in the postures for approximately 30 min in total (5 min for the mood questionnaire, 3 min for the reading task, 5 min for the speech preparation, 5 min for the speech task, 5 min resting, and 7 min while the final questionnaire was completed). Reading task. Participants were asked to read some basic instructions on “how to wash clothes” aloud for 3 min. This was used to measure the individual’s physiology while speaking before the introduction of the stressful speech task. Because psychomotor retardation in depression can manifest as slow talking and a lack of speech and ideas, this also served as a behavioral measure of psychomotor retardation by counting the number of words that the participant read during the time allowed. Stress task. The stress task was based on the standard Trier Social Stress Test (Kirschbaum et al., 1993), a well validated protocol for inducing psychological stress, using a variant of the public speaking task. Participants were told to imagine they were being interviewed for their dream job and had 5 min to tell the interview panel why they were the best candidate. They had 5 min 1 At baseline the LED markers were not on the participants but the software was capable of assessing posture by using virtual markers.

4

NAIR, SAGAR, SOLLERS, CONSEDINE, AND BROADBENT

to prepare but were not allowed to make any written notes. The speech was video-recorded. Participants were instructed that the speech would be evaluated by a panel of judges and the best participant would win a $200 shopping voucher. Speeches were transcribed and analyzed using text analysis software.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Measures Participants’ were asked about their age, gender, ethnicity, marital status, and general health. Affect. State emotion was assessed with the Affect Valuation Index (AVI; Tsai, Knutson, & Fung, 2006). Twenty-five items were rated on a 1 (very slightly or not at all) to 5 (extremely or all of the time) scale, mapping onto valence-arousal dimensions within the affective circumplex (Barrett, 1996). Following the expectation that postural manipulations should impact arousal and thus be differentially manifest in subclasses of affect (rather than only in positive affect/negative affect), we scored the AVI into the established eight categories: high arousal positive (enthusiastic, excited, strong), positive (happy, satisfied, content), low arousal positive (calm, rested, relaxed, peaceful), low arousal (quiet, still, passive), low arousal negative (dull, sleepy, sluggish), negative (sad, lonely, unhappy), high arousal negative (fearful, hostile, nervous), and high arousal (aroused, surprised, astonished). The measure has moderate to high test–retest reliability, high internal consistency, and adequate discriminant validity (Sims, Tsai, Koopmann-Holm, Thomas, & Goldstein, 2014; Tsai et al., 2006; Tsai, Miao, Seppala, Fung, & Yeung, 2007). The scale was reliable in this sample; Cronbach’s alphas for all the subscales were above .70, except for the high arousal negative subscale (␣ ⫽ .63). Affect also was assessed using Linguistic Inquiry and Word Count (LIWC) (Pennebaker et al., 2007). In the speech task, participants must justify why they are a good candidate for their ideal job. The speech content therefore contains many personal thoughts and feelings, which can be analyzed for affect. The text analysis application, LIWC, was used to explore the emotional components in the content of the participants’ speeches. LIWC analyses text on a word-by-word basis and categorizes content into four high-level categories: linguistic dimensions (e.g., pronoun use, unique words), psychological processes (e.g., cognitive, social and emotional), relativity (e.g., time, space), and personal concerns (e.g., occupation, leisure). The LIWC dictionary contains more than 2,100 words and word stems and measures up to 74 linguistic dimensions. The LIWC is highly reliable over different types of writing and over time (Pennebaker & King, 1999). The categories of first person singular pronouns (I, me, mine), and affective processes (broken into positive emotions, negative emotions, anxiety, anger, and sadness) were most relevant to the hypothesis that posture would affect emotional responses to the stressor and were analyzed for differences between groups. Cronbach’s alphas for these categories are reported as: first person singular pronouns, ␣ ⫽ .62; affective processes, ␣ ⫽ .97; positive emotion, ␣ ⫽ .97; negative emotion, ␣ ⫽ .97; anxiety, ␣ ⫽ .89; anger, ␣ ⫽ .92; and sadness, ␣ ⫽ .91 (Pennebaker et al., 2007). Perceived threat. This questionnaire was used to assess whether the postural manipulation could alter how people perceived threatening situations. It contains a set of hypothetical situations that include five physical danger situations (e.g., “climbing a mountain in the Himalayas”) and four social danger situa-

tions (e.g., “being interviewed on a live TV show”) (Rapee, 1997). Participants were asked about the probability, consequences, amount of control, and fear they would feel for each situation. Rapee (1997) showed that perceptions of probability, consequences, and control were all related to fear. The subscales were reliable in this study; Cronbach’s alphas were all above .75 except for fear in physical situations (␣ ⫽ .63) and control in physical situations (␣ ⫽ .60). Self-esteem. The 20-item State Self-Esteem Scale (SSES; Heatherton & Polivy, 1991) was chosen because it is sensitive to laboratory manipulations of self-esteem. Example items include, “I feel confident about my abilities” and “I feel good about myself,” were rated on a scale ranging from 1 (not at all) to 5 (extremely). Research suggests that the SSES is psychometrically sound and valid, is responsive to temporary changes in self-evaluation, and has good internal consistency (Cronbach’s ␣ ⫽ .92). In this sample, ␣ ⫽ .69. It was administered at baseline and at the end of the experiment. Task persistence. Total word count during the reading task and the stress task, was used as a measure of task persistence and psychomotor retardation. Cardiovascular response. The Nexfin (BMEYE B.V., Amsterdam, The Netherlands) monitor was used to noninvasively measure continuous blood pressure and heart rate. This has been shown to be accurate (Martina et al., 2012). The device uses an internal calibration method and does not require calibration with an external device (Wesseling, de Wit, van der Hoeven, van Goudoever, & Settels, 1995). Differences between groups over time in heart rate, stroke volume, systolic and diastolic blood pressure, and pulse pressure, were analyzed. Pulse pressure is the difference between systolic and diastolic blood pressure and is a measure of the force the heart generates when it contracts.

Statistical Analysis IBM SPSS statistics 19.0 was utilized for analysis. A significance level of p ⬍ .05 was maintained. Mixed analyses of variance were used to test for main effects and interactions for each of the independent variables (group and task) on dependent variables of postural angles, mood states, levels of self-esteem, LIWC, and physiological variables. Independent samples T tests were used to analyses differences between groups in perceived threat situations. All analyses included all participants; except pairwise deletion was used for those variables with missing data (physiology data and reading task—see Results for details). The effect sizes from previous research on the manipulation of posture on cortisol and feelings of power were large (d ⫽ .95 and .91, respectively; Cuddy et al., 2012). This suggests the effects of posture on physiological and psychological outcomes are substantial. Choosing a moderate to large effect size, a power analysis indicated that 68 participants would be required to detect an effect size d ⫽ .70 of posture on affect and cardiovascular outcomes, with alpha of .05, and power of .80, using a two-tailed test.

Results Thirty-nine participants were randomized to the upright group and 35 were randomized to the slumped group. All participants received the postural intervention as intended. Due to problems

POSTURE AND STRESS

with equipment, five participants’ physiological data were not recorded (four in the slumped condition and one in the upright condition), and one reading task was not recorded (slumped condition). All data was collected between July and November 2012. There were no significant differences between groups on age, gender, or ethnicity.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Postural Manipulation The difference in head, neck, back, and shoulder angles at baseline and during the stress task were compared between groups. There were no significant differences in posture between groups at baseline (p ⬎ .05). There were significant main effects of group for every angle, (p ⬍ .001), and significant main effects of time for every angle except shoulder angle (p ⬍ .001). There were significant Group ⫻ Time interaction effects for every angle (see Table 1). During the stress task, the slumped group faced down more, had a more bent neck and back, and more rounded shoulders than the upright group.

Affect There were significant main effects of time for all of the mood states except high arousal negative, and significant main effects of group for high arousal positive, low arousal, low arousal negative, negative, and high arousal negative (see Table 2, Figure 2). There were significant Group ⫻ Time interactions for high arousal positive, low arousal, low arousal negative, high arousal negative, and a marginally significant interaction for positive state. Contrasts revealed that high arousal positive state dropped for the slumped group from baseline to immediately after the postural intervention, F(1, 70) ⫽ 8.10, p ⫽ .006, r ⫽ .32, and from baseline to after the stress task, F(1, 70) ⫽ 6.79, p ⫽ .011, r ⫽ .30. There was a similar drop in positive state for the slumped group between baseline and immediately after the postural intervention, F(1, 70) ⫽ 5.05, p ⫽ .028, r ⫽ .26; from baseline to after the stress task was marginally significant F(1, 70) ⫽ 3.83, p ⫽ .054, r ⫽ .23. Low arousal decreased in the upright group from baseline to immediately after

5

the postural intervention, F(1, 70) ⫽ 13.31, p ⬍ .001, r ⫽ .40. Likewise low arousal negative dropped from baseline to immediately after the postural intervention in the upright group, F(1, 70) ⫽ 11.29, p ⬍ .001, r ⫽ .37, and to after the stress test, F(1, 70) ⫽ 7.47, p ⫽ .008, r ⫽ .31. High arousal negative increased in the slumped group compared to the upright group from baseline to immediately after the postural intervention, F(1, 70) ⫽ 10.20, p ⫽ .002, r ⫽ .36, and after the stress test, F(1, 70) ⫽ 6.64, p ⫽ .012, r ⫽ .29. In the speech task, the slumped group used more first-person singular pronouns, more affective words, more negative emotion and sadness words, and fewer positive emotion words (see Table 3).

Perceived Threat There were no significant differences between groups on their perceptions of the probability of the threat occurring, or their perceptions of consequences or control in either the physical threat or the social threat situations. There was a significant difference between groups in their ratings of fear in the social threat situations only; the slumped condition rated feeling more fear in social threat situations (M ⫽ 14.03, SD ⫽ 4.46, 95% CI [12.50, 15.56]) compared to the upright condition (M ⫽ 10.56, SD ⫽ 3.57, 95% CI [9.41, 11.71]), t(72) ⫽ ⫺3.71, p ⬍ .001, (Cohen’s d ⫽ 0.86).

Self-Esteem Results indicated a significant Time ⫻ Group interaction in self-esteem, F(1, 70) ⫽ 11.28, p ⫽ .001, r ⫽ .37, (see Figure 3). Self-esteem was similar between the slumped and upright condition at baseline, but the slumped condition had reduced ratings of self-esteem compared to the upright condition after the stress task.

Task Persistence In the speech task, the slumped group spoke significantly fewer words in total (see Table 3). The slumped group also read fewer

Table 1 Manipulation Check That the Postural Intervention Changed Postural Angles Postural angle Head

Group

Time point

M

[95% CI]

F

df

p

r

Upright

Baseline Stress task Baseline Stress task Baseline Stress task Baseline Stress task Baseline Stress task Baseline Stress task Baseline Stress task Baseline Stress task

22.399 21.475 27.062 47.588 46.761 43.007 49.145 8.249 82.056 85.905 81.065 70.893 62.972 72.374 61.851 53.631

[16.85, 27.95] [18.09, 24.86] [21.10, 33.02] [43.96, 51.22] [42.87, 50.64] [39.63, 46.38] [44.97, 53.32] [4.63, 11.87] [80.22, 83.89] [84.19, 87.62] [79.09, 83.04] [69.05, 72.73] [60.63, 65.32] [69.84, 74.91] [59.33, 64.37] [50.91, 56.35]

18.71

69

⬍.001

.46

94.88

69

⬍.001

.76

69.08

69

⬍.001

.71

65.15

69

⬍.001

.41

Slumped Neck

Upright Slumped

Back

Upright Slumped

Shoulder

Upright Slumped

Note. Postural angles are shown before the intervention (baseline) and after the intervention (during the stress task). Results of Group ⫻ Time interaction tests are shown.

NAIR, SAGAR, SOLLERS, CONSEDINE, AND BROADBENT

6

Table 2 The Effects of Time, the Postural Intervention, and Time ⫻ Posture Interactions on Mood

Time

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Posture

Time ⫻ Posture interaction

Octant

Items

F

df

p

r

HAP POS LAP LA LAN NEG HAN HA HAP POS LAP LA LAN NEG HAN HA HAP POS LAP LA LAN NEG HAN HA

Enthusiastic, excited, strong Happy, satisfied, content Calm, rested, relaxed, peaceful Quiet, still, passive Dull, sleepy, sluggish Sad, lonely, unhappy Fearful, hostile, nervous Aroused, surprised, astonished Enthusiastic, excited, strong Happy, satisfied, content Calm, rested, relaxed, peaceful Quiet, still, passive Dull, sleepy, sluggish Sad, lonely, unhappy Fearful, hostile, nervous Aroused, surprised, astonished Enthusiastic, excited, strong Happy, satisfied, content Calm, rested, relaxed, peaceful Quiet, still, passive Dull, sleepy, sluggish Sad, lonely, unhappy Fearful, hostile, nervous Aroused, surprised, astonished

5.03 9.57 34.57 10.17 4.01 5.97 0.91 4.76 4.73 3.43 0.27 8.90 11.60 6.14 10.20 0.31 6.20 3.02 0.26 6.88 8.28 2.57 6.09 6.23

1.65 2.00 2.00 2.00 1.46 2.00 2.00 1.32 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.65 2.00 2.00 2.00 1.46 2.00 2.00 1.32

.01 ⬍.001 ⬍.001 ⬍.001 .03 .003 .40 .02 .03 .07 .61 .004 .001 .02 .002 .58 .005 .05 .77 .001 .002 .08 .003 .58

.87 .91 .97 .91 .85 .87 .56 .88 .91 .88 .46 .95 .96 .93 .95 .49 .89 .78 .34 .88 .92 .75 .87 .91

Note. Time is from baseline to immediately after the postural intervention, to after the stress task. HAP ⫽ high arousal positive; POS ⫽ positive; LAP ⫽ low arousal positive; LA ⫽ low arousal; LAN ⫽ low arousal negative; NEG ⫽ negative; HAN ⫽ high arousal negative; HA ⫽ high arousal.

words in the reading task (M ⫽ 388.44, SD ⫽ 84.74, 95% CI [359.96, 413.92]) than the upright group (M ⫽ 503.14, SD ⫽ 81.20, 95% CI [477.66, 528.62]), t(69) ⫽ 5.82, p ⬍ .001 (Cohen’s d ⫽ 1.38).

Cardiovascular Response There were statistically significant main effects of time for all cardiovascular outcomes (p ⬍ .001), however there were no statistically significant main effects of group. There were no significant Group ⫻ Time interaction effects for stroke volume, diastolic or systolic pressure, or heart rate. There was a significant Group ⫻ Time interaction for pulse pressure, F(1, 64) ⫽ 3.39, p ⫽ .005, r ⫽ .15. Contrasts conducted during the speech task between the slumped (M ⫽ 46.90, SD ⫽ 8.45) and upright group (M ⫽ 49.08, SD ⫽ 8.75) showed that pulse pressure was higher in the upright condition, F(1, 64) ⫽ 1.50, p ⫽ .02. Likewise, at the recovery point, contrasts revealed that pulse pressure was higher in the upright condition (M ⫽ 47.11, SD ⫽ 8.24), than the slumped condition (M ⫽ 42.23, SD ⫽ 10.56), F(1, 64) ⫽ 7.48, p ⫽ .008.

Discussion This is the first study to show that holding an upright seated posture during a psychological stressor can have protective effects on mood compared to a slumped posture. The upright participants reported feeling more enthusiastic, excited, and strong, while the slumped participants reported feeling more fearful, hostile, nervous, quiet, still, passive, dull, sleepy, and sluggish. The upright participants also reported higher self-esteem and reduced fear

compared to slumped participants. They spoke more words in total, used fewer first person personal pronouns, fewer negative emotion, and more positive emotion words. Furthermore, a bigger pulse pressure response was seen in the upright participants during the stressor and during recovery, than in the slumped participants. To our knowledge this is the first study to analyze word use during a stressful speech task. The effects of posture on the use of personal pronouns and emotion words provides additional, convergent evidence to self-report measures that an upright posture can improve mood. Furthermore, the effects of posture on the total number of words spoken may reflect increased alertness and arousal in the upright posture compared to the slumped posture, which aligns with self-reports of feeling less sleepy and sluggish. Previous research has shown that upright posture causes increased temperature (Nicholson & Barbara, 1987), decreased venous pressure, and increased heart rate, as well as elevated epinephrine and norepinephrine (Caldwell, Prazinko, & Caldwell, 2003). The observed effects of upright posture on positive affect and increased word use also align with theories that positive affect can boost cognitive functioning (Fredrickson, 2001). There was a bigger increase in pulse pressure for participants in the upright condition during the stress task and this was maintained in the recovery period, which was contrary to the hypothesis. In general blood pressure increases in response to a stressor have been related to increases in cardiac output and decreases in systemic vascular resistance (Goldstein & Shapiro, 1988). These results seem to reflect increased physiological arousal in the upright group compared to the slumped group. This increased arousal may be driving an active coping response in the upright group,

7

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

POSTURE AND STRESS

Figure 2. The effects of posture on mood states. Before the postural intervention (baseline), immediately after the postural intervention (after taping), and after the stress task. (A) High arousal positive state; (B) positive state; (C) low arousal positive state; (D) low arousal state; (E) low arousal negative state; (F) negative state; (G) high arousal negative state; (H) high arousal state. Means and standard errors (error bars) reported.

compared to the slumped group who may be exhibiting more of a helplessness type response consistent with previous work (Riskind & Gotay, 1982). These results add to the existing literature on embodied cognition by finding that slumped and upright posture can influence

affective states (results from after the postural intervention but before the stressor) and modulate affective responding (results from after the stressor). The results align with recent work showing that the facial expression of a smile can influence both cardiovascular and affective responses to stress (Kraft & Pressman, 2012).

NAIR, SAGAR, SOLLERS, CONSEDINE, AND BROADBENT

8

Table 3 The Effects of the Postural Intervention on Language Use During the Speech Task LIWC dimension Word count Pronouns, total First-person singular pronouns Affective processes

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Positive emotions Negative emotions Anxiety Anger Sadness Note.

Posture

M (SD)

[95% CI]

t

df

Sig (2-tailed)

Cohen’s d

Upright Slumped Upright Slumped Upright Slumped Upright Slumped Upright Slumped Upright Slumped Upright Slumped Upright Slumped Upright Slumped

561.97 (203.60) 392.74 (166.73) 26.67 (18.50) 23.60 (16.33) 8.48 (2.90) 13.17 (4.92) 4.72 (2.57) 8.81 (6.03) 6.67 (3.61) 4.31 (1.71) 1.84 (2.79) 5.67 (5.65) 0.19 (0.26) 1.37 (3.89) 0.10 (0.19) 0.16 (0.40) 0.06 (0.22) 1.57 (1.96)

[498.07, 625.87] [336.70, 448.78] [20.86, 32.48] [18.11, 29.09] [7.57, 9.39] [11.52, 14.82] [3.91, 5.53] [6.78, 10.84] [5.54, 7.80] [3.74, 4.88] [0.96, 2.72] [3.77, 7.57] [0.11, 0.27] [0.06, 2.68] [0.04, 0.16] [0.02, 0.30] [–0.01, 0.13] [0.91, 2.23]

3.83

70.0

⬍.001

0.91

0.74

70.0

.461

0.18

0.63

70.0

⬍.001

1.16

–3.67

43.53

.001

0.88

3.59

53.98

.001

0.84

–3.57

46.99

.001

0.86

–1.77

33.26

.08

0.43

–0.88

70.0

.38

0.19

–4.47

33.75

⬍.001

1.08

LIWC ⫽ Linguistic Inquiry and Word Count.

The mechanisms by which posture affects emotions are likely to be through the activation of the skeletomuscular, neuroendocrine, and autonomic nervous systems, and also through self-perception. More research that encompasses these processes is needed to better elucidate specific mechanisms. There is a relative lack of research on slumped versus upright postures on emotions. Other than a handful of earlier investigations carried out with upright versus slumped postures (Ahn, Teeters, Wang, Breazeal, & Picard, 2007; Briñol et al., 2009; Duclos et al., 1989; Riskind, 1984; Riskind & Gotay, 1982), postural research has primarily focused on comparing sitting, supine, and standing positions, or particular poses of anger, persistence, sadness, fear, and power. The use of a cover story about the effects of physiotherapy tape is a strength compared to previous studies because it reduced the likelihood that the observed effects were simply due to

Figure 3. The effects of the postural intervention on self-esteem. Before the intervention (baseline) and after the stress task. Means and standard errors (error bars) reported.

expectations or the placebo effect. The use of objective computer software to analyze postural angles is a further strength. The study has a number of limitations. Similar to most previous studies, the use of two opposite postures precludes the determination of whether slumped posture is having negative effects or upright posture is having positive effects on mood. In this study, all angles (head, neck, back, and shoulder) significantly changed in the slumped group relative to baseline, whereas in the upright group, there were significant changes in the back and shoulder angles only relative to baseline. This suggests that the slumped group had worse posture compared to normal and the upright group had better posture compared to normal at least in the back and shoulders. Including a usual posture control group would allow us to study any natural postural changes that may occur over time as a result of stress. This also would allow the determination of whether slumped posture is having negative effects or upright posture having positive effects, relative to a control group. Another limitation is the lack of a measure of body image, which would have helped show whether the intervention affected body perceptions. We did not assess depressive symptoms at baseline and therefore cannot rule out possible baseline group differences in subclinical depression that may have impacted the outcomes. However, the use of randomization to groups and the finding that there were no group differences in any of the affect valuation subscales at baseline gives us confidence that this is very unlikely. Future research in this area should assess depressive symptoms at baseline. Finally, most of the measures had good internal consistency (Cronbach’s ␣ ⬎ .70) but four had alphas between .60 and .70, which is considered acceptable. The use of physiotherapy tape was used both as a cover story and as a way to maintain posture in both groups. The tape may have drawn the participants’ attention to their backs but this would have occurred in both groups, so it is not likely to be responsible for the differences between groups. Future studies could use an alternative method to maintain posture.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

POSTURE AND STRESS

The effects of postural manipulations on stress responses are particularly relevant to psychology because stress has been implicated in the etiology of depression. Although major depressive episodes are often preceded by stressful events, not all stressful events lead to depression (Hammen, 2005). Certain psychological factors may make a person more vulnerable to the development of depression. Low self-esteem can increase the risk that a severe stressor will trigger a depressive episode (Bifulco, Brown, Moran, Ball, & Campbell, 1998). On the other hand, self-enhancing cognitions and psychological resources (including self-esteem, optimism, mastery, happiness, and extraversion) can buffer stress responses (Taylor, Lerner, Sherman, Sage, & McDowell, 2003). The affirmation of personal values prior to a stressful task has been shown to reduce stress reactions (Creswell et al., 2005). The results of this study show that holding an upright posture can maintain self-esteem and buffer stress responses. Although speculative at this stage, future clinical implications may include using postural interventions to supplement current treatments for affective issues in depressed populations. The next stage in this line of research could be to test the effects of manipulating posture on mood as a proof of concept in people who score in the range of mild to moderate depression, before moving to trial the intervention with those who meet major depression criteria. In conclusion, this study shows that postural interventions causally affect mood, behavior, and physiology in response to a shortterm stressor. These findings are consistent with theories of embodied cognition and emotion. Future studies should investigate the psychological effects of postural interventions using a longer time period and in populations with depression.

References Ahn, H. I., Teeters, A., Wang, A., Breazeal, C., & Picard, R. (2007). Stoop to conquer: Posture and affect interact to influence computer users’ persistence. In A. Paiva, R. Prada, & R. W. Picard (Eds.), Affective computing and intelligent interaction (pp. 582–593). Berlin, Germany: Springer Berlin Heidelberg. doi:10.1007/978-3-540-74889-2_51 American Psychiatric Association. (2013). Diagnostic and statistical manual of mental health disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. Barrett, L. F. (1996). Hedonic tone, perceived arousal and item desirability: Three components of self reported mood. Cognition and Emotion, 10, 47– 68. doi:10.1080/026999396380385 Barrett, L. F., & Russell, J. A. (1999). Structure of current affect. Current Directions in Psychological Science, 8, 10 –14. doi:10.1111/1467-8721 .00003 Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 6, pp. 2– 62). New York, NY: Academic Press. Bifulco, A., Brown, G. W., Moran, P., Ball, C., & Campbell, C. (1998). Predicting depression in women: The role of past and present vulnerability. Psychological Medicine, 28, 39 –50. doi:10.1017/S0033291797005953 Briñol, P., Petty, R. E., & Wagner, B. (2009). Body posture effects on self-evaluation: A self-validation approach. European Journal of Social Psychology, 39, 1053–1064. doi:10.1002/ejsp.607 Caldwell, J. A., Prazinko, B., & Caldwell, L. (2003). Body posture affects electroencephalographic activity and psychomotor vigilance task performance in sleep-deprived subjects. Clinical Neurophysiology, 114, 23– 31. doi:10.1016/S1388-2457(02)00283-3 Carney, D. R., Cuddy, A. J., & Yap, A. J. (2010). Power posing: Brief nonverbal displays affect neuroendocrine levels of risk tolerance. Psychological Science, 21, 1363–1368. doi:10.1177/0956797610383437

9

Chung, C. K., & Pennebaker, J. W. (2007). The psychological functions of function words. In K. Feilder (Ed.), Social communication (pp. 343– 359). New York, NY: Psychology Press. Creswell, J. D., Welch, W. T., Taylor, S. E., Sherman, D. K., Gruenewald, T. L., & Mann, T. (2005). Affirmation of personal values buffers neuroendocrine and psychological stress responses. Psychological Science, 16, 846 – 851. doi:10.1111/j.1467-9280.2005.01624.x Cuddy, A. J., Wilmuth, C. A., & Carney, D. R. (2012). The benefit of power posing before a high-stakes social evaluation (Harvard Business School Working Paper, No. 13– 027). Retrieved from http://nrs.harvard .edu/urn-3:HUL.InstRepos:9547823 Duclos, S. E., Laird, J. D., Schneider, E., Sexter, M., Stern, L., & Van Lighten, O. (1989). Emotion-specific effects of facial expressions and postures on emotional experience. Journal of Personality and Social Psychology, 57, 100 –108. doi:10.1037/0022-3514.57.1.100 Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56, 218 –226. doi:10.1037/0003-066X.56.3.218 Glaser, R., & Kiecolt-Glaser, J. K. (2005). Stress-induced immune dysfunction: Implications for health. Nature Reviews Immunology, 5, 243– 251. doi:10.1038/nri1571 Goldstein, I. B., & Shapiro, D. (1988). Cardiovascular responses to mental arithmetic and handgrip during different conditions of postural change. Psychophysiology, 25, 127–136. doi:10.1111/j.1469-8986.1988 .tb00974.x Hammen, C. (2005). Stress and depression. Annual Review of Clinical Psychology, 1, 293–319. doi:10.1146/annurev.clinpsy.1.102803.143938 Harmon-Jones, E., & Peterson, C. K. (2009). Supine body position reduces neural response to anger evocation. Psychological Science, 20, 1209 – 1210. doi:10.1111/j.1467-9280.2009.02416.x Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale of measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895–910. doi:10.1037/0022-3514.60.6.895 Hellhammer, J., & Schubert, M. (2012). The physiological response to Trier Social Stress Test relates to subjective measures of stress during but not before or after the test. Psychoneuroendocrinology, 37, 119 –124. doi:10.1016/j.psyneuen.2011.05.012 Hexsel, D., Brum, C., Siega, C., Schilling-Souza, J., Forno, T. D., Heckmann, M., & Rodrigues, T. C. (2013). Evaluation of self-esteem and depression symptoms in depressed and nondepressed subjects treated with onabotulinumtoxinA for glabellar lines. Dermatologic Surgery, 39, 1088 –1096. doi:10.1111/dsu.12175 James, W. (1890). The principles of psychology. New York, NY: Holt. doi:10.1037/11059-000 Kirschbaum, C., Pirke, K. M., & Hellhammer, D. H. (1993). The “Trier Social Stress Test”—A tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28, 76 – 81. doi: 10.1159/000119004 Kraft, T. L., & Pressman, S. D. (2012). Grin and bear it the influence of manipulated facial expression on the stress response. Psychological Science, 23, 1372–1378. doi:10.1177/0956797612445312 Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A review. Biological Psychology, 84, 394 – 421. doi:10.1016/j.biopsycho .2010.03.010 Lipnicki, D. M., & Byrne, D. G. (2008). An effect of posture on anticipatory anxiety. International Journal of Neuroscience, 118, 227–237. doi:10.1080/00207450701750463 Martina, J. R., Westerhof, B. E., van Goudoever, J., de Beaumont, E. M., Truijen, J., Kim, Y. S., . . . van Lieshout, J. J. (2012). Noninvasive continuous arterial blood pressure monitoring with Nexfin. Anesthesiology, 116, 1092–1103. doi:10.1097/ALN.0b013e31824f94ed McIntosh, D. N. (1996). Facial feedback hypotheses: Evidence, implications, and directions. Motivation and Emotion, 20, 121–147. doi:10.1007/ BF02253868

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

10

NAIR, SAGAR, SOLLERS, CONSEDINE, AND BROADBENT

Michalak, J., Troje, N. F., Fischer, J., Vollmar, P., Heidenreich, T., & Schulte, D. (2009). Embodiment of sadness and depression—Gait patterns associated with dysphoric mood. Psychosomatic Medicine, 71, 580 –587. doi:10.1097/PSY.0b013e3181a2515c Nicholson, A. N., & Barbara, M. S. (1987). Influence of back angle on the quality of sleep in seats. Ergonomics, 30, 1033–1041. doi:10.1080/ 00140138708965993 Niedenthal, P. M. (2007). Embodying emotion. Science, 316, 1002–1005. doi:10.1126/science.1136930 Oosterwijk, S., Rotteveel, M., Fischer, A. H., & Hess, U. (2009). Embodied emotion concepts: How generating words about pride and disappointment influences posture. European Journal of Social Psychology, 39, 457– 466. doi:10.1002/ejsp.584 Pennebaker, J. W., Chung, C. K., Ireland, M., Gonzales, A., & Booth, R. J. (2007). The development and psychometric properties of LIWC2007. Austin, TX: University of Texas and University of Auckland. Available at http://homepage.psy.utexas.edu/HomePage/Students/Chung/Publications_ files/PennebakerChungIrelandGonzales&Booth2007_LIWC.pdf Pennebaker, J. W., & King, L. A. (1999). Linguistic styles: Language use as an individual difference. Journal of Personality and Social Psychology, 77, 1296 –1312. doi:10.1037/0022-3514.77.6.1296 Pyszczynski, T., & Greenberg, J. (1987). Toward an integration of cognitive and motivational perspectives on social inference: A biased hypothesis-testing model. Advances in Experimental Social Psychology, 20, 297–340. doi:10.1016/S0065-2601(08)60417-7 Rapee, R. M. (1997). Perceived threat and perceived control as predictors of the degree of fear in physical and social situations. Journal of Anxiety Disorders, 11, 455– 461. doi:10.1016/S0887-6185(97)00022-4 Riskind, J. (1984). They stoop to conquer: Guiding and self regulatory functions of physical posture after success and failure. Journal of Personality and Social Psychology, 47, 479 – 493. doi:10.1037/0022-3514 .47.3.479 Riskind, J., & Gotay, C. J. (1982). Physical posture: Could it have regulatory or feedback effects on motivation and emotion? Motivation and Emotion, 6, 273–298. doi:10.1007/BF00992249 Rodriguez, A. J., Holleran, S. E., & Mehl, M. R. (2010). Reading between the lines: The lay assessment of subclinical depression from written

self-descriptions. Journal of Personality, 78, 575–598. doi:10.1111/j .1467-6494.2010.00627.x Rude, S. S., Gortner, E. M., & Pennebaker, J. W. (2004). Language use of depressed and depression-vulnerable college students. Cognition and Emotion, 18, 1121–1133. doi:10.1080/02699930441000030 Schilder, P. (1999). The image and appearance of the human body. Oxon, England: Routledge. Sims, T., Tsai, J. L., Koopmann-Holm, B., Thomas, E. A., & Goldstein, M. K. (2014). Choosing a physician depends on how you want to feel: The role of ideal affect in health-related decision making. Emotion, 14, 187–192. doi:10.1037/a0034372 Stepper, S., & Strack, E. (1993). Proprioceptive determinants of emotional and non-emotional feelings. Journal of Personality and Social Psychology, 64, 211–220. doi:10.1037/0022-3514.64.2.211 Taylor, S. E., Lerner, J. S., Sherman, D. K., Sage, R. M., & McDowell, N. K. (2003). Are self-enhancing cognitions associated with healthy or unhealthy biological profiles? Journal of Personality and Social Psychology, 85, 605– 615. doi:10.1037/0022-3514.85.4.605 Tourangeau, R., & Ellsworth, P. C. (1979). The role of facial response in the experience of emotion. Journal of Personality and Social Psychology, 37, 1519–1531. doi:10.1037/0022-3514.37.9.1519 Tsai, J. L., Knutson, B., & Fung, H. H. (2006). Cultural variation in affect valuation. Journal of Personality and Social Psychology, 90, 288 –307. doi:10.1037/0022-3514.90.2.288 Tsai, J. L., Miao, F., Seppala, E., Fung, H., & Yeung, D. (2007). Influence and adjustment goals: Sources of cultural differences in ideal affect. Journal of Personality and Social Psychology, 92, 1102–1117. doi:10.1037/0022-3514 .92.6.1102 Wesseling, K. H., de Wit, B., van der Hoeven, G. M. A., van Goudoever, J., & Settels, J. J. (1995). Physiocal, calibrating finger vascular physiology for Finapres. Homeostasis, 36, 76 – 82.

Received February 5, 2014 Revision received July 4, 2014 Accepted July 7, 2014 䡲