Consistent with the growing popularity of both body and movement therapies and the literature supporting their effectiveness, physical movements and positions appear to impact individuals' emotional experiences (Rossberg-Gempton & Poole, 1993). Open and closed postures, for instance, have been found to impact the emotional states of the people assuming them (Pool, 1997). Crossed arms and legs, and leaning back are generally considered closed postures, while open postures are the opposite. In one study, three different open postures significantly decreased positive feelings (happiness, agreeableness, interest, and surprise), while three different closed postures significantly increased negative feelings (fear, anger, sadness, and disgust), with the latter finding having over twice the effect size as the former (Rossberg-Gempton & Poole, 1993). The open and closed postures were defined by the positioning of subjects' arms, legs, and heads, as well as the angle of the leans of their torsos, but not by spinal alignment. The authors reported that a likely reason no increase in positive feelings or decrease in negative feelings was found was because the beginning emotional states of most participants was positive, leaving less room for growth in positive emotions or opportunity for decreasing negative emotions.
Another similar study showed that open postures were associated with increased positive emotions, whereas closed postures elicited negative emotions (Rossberg-Gempton et al., 1992). Because of the effect posture may have on emotion, and because studies show that high distress levels in therapists may prevent the growth of clients, or even lead to negative changes in them, a therapist's posture may indirectly affect therapeutic outcome (Begin and Garfield, 1994). According to Begin and Garfield (1994), the well-being and adjustment of the therapist plays a role in optimal therapeutic results. These effects of posture relate to “postures as felt,” a term which refers to the experience of a posture (James, 1890). “Postures as seen,” in contrast, relates to the observation of postures (James, 1890). These terms are useful because postures appear to affect both the one assuming the posture, and those observing him or her.
Nonverbal communication is an example of the utilization of “Postures as seen.” Much is already known about how people express themselves without words and embody their emotions. From facial expressions to subtle postural shifts, people use physical cues to guide social interaction, both consciously and unconsciously. Research shows, for instance, that postural congruence, or the amount of similarity between the postures assumed by two individuals, appears to affect how much the individuals like each other (Maxwell & Cook, 1997). More similar postures between the individuals leads to or reflects increased liking between them (Maxwell & Cook, 1997). Research on “Postures as seen,” also shows that particular body language is more socially acceptable to people in certain contexts than others (Harrigan et al., 1991). People judged individuals seen rubbing their hands together as more socially acceptable when told that the people were patients or job applicants, roles in which anxiety might be expected. In contrast, nose rubbing was considered more socially acceptable with friends than strangers. The controls in this study, who did not move, were viewed as more “calm and dominant, but less expressive, warm, and interested” (Harrigan et al., 1991, p. 585). Harrigan et al. (1991) also found that people seen touching their arms were considered “anxious, submissive, cold, and bored, but also as expressive” (p. 604). These findings indicate that people respond to nonverbal expression, and that contexts may play a role in their responses.
Something as subtle as integrated body movements, judged by Posture-Gesture Mergers (PGMs), corresponds with more truthful and relaxed verbal communication (Winter, 1989). PGMs are defined as when a person's postures (involving the whole body) lead into gestures (involving only part of the body) or the reverse, showing a consistency between what the two kinds of body languages are saying. An example of a PGM could be a situation where both a person's overall posture and his/her gestures suggest the person is frustrated. Four experiments have demonstrated the effects of PGMs (Winter, 1989). In one of the four studies, trained observers found more PGMs when they analyzed videos of subjects in a situation when they were told to tell the truth than one in which they were instructed to lie (Winter, 1989). Further data on the effects of “posture as seen,” is evident in another study in which particular body movements, thought to correspond with particular emotions, were videotaped. These videos were used to determine whether participants judged the people in the videos as having the emotions their postures were predicted to express (Meijer, 1989). The study found that participants did attribute the anticipated emotions to individuals in corresponding postures, and that combinations of movements caused specific attributions to be made about the movers. The movement of the torso was found to be particularly potent in expressing emotions. The author concluded that torso movement “may reflect the subject's overall readiness to interact and to deal with his or her surroundings,” considerations relevant to a psychotherapy session (Meijer, 1989, p. 265).
Some studies on the effects of posture have already been done in a therapeutic context. The observed postures of therapists and couples in couple's therapy were shown to reflect the degree to which the group or each individual was ready to interact and committed to involvement in the therapy session (Roten et al., 1999). Particular therapist postures have been shown to lead to better therapeutic outcomes than other postures (Carhuff, 1983). In particular, leaning forward and maintaining an open posture, here defined as sitting with one's feet on the floor, four to seven inches apart, with one's arms resting on the arms of the chair, is more effective than leaning back or maintaining a closed posture, defined as crossing one's arms and legs. One study showed that although therapists' open postures did not increase client evaluation of therapists' attractiveness or trustworthiness, it did significantly increase their evaluations of therapists' expertness, communicating a sense of confidence and involvement with clients (Ridley & Asbury, 1988). Trustworthiness, attractiveness, and expertness were measured by the Counselor Rating Form (CRF) (Barak & LaCrosse, 1975). Therapists also gestured more when assuming open postures than closed. The authors state that this finding may suggest an increase in spontaneity and freedom associated with open postures.
If part of the reason “open” postures appear to increase the effectiveness of therapy is because they are related to clients' perceptions of therapists as more open (and possibly, then, more available or receptive), then other postures considered open may have the same effect. Though in microcounseling open and closed postures refer to the difference between crossed and uncrossed arms or legs, (Carhuff, 1983; Ridley & Asbury, 1988) other professions appear to define them in other ways. In dance and yoga, for instance, an open posture is one that reflects the proper alignment of one's spine so that the tips of its three curves would all just touch an imaginary vertical line drawn down the center of the body (Olsen, 1998, p. 51). It would also include the most efficient placement of the cranium on its axis, so it is balanced in relation to gravity (Olsen, 1998). Because particular open postures for therapists have already been found more effective in treating clients, postures considered “open” in other fields may yield the same effect and thereby increase the specificity of advice yielded to therapists about effective postures for them. No research, however, is currently available to determine whether a therapist's postural alignment affects therapeutic process and outcome.
Despite the lack of research on spinal alignment in relation to therapists' effectiveness, consideration of the research on spinal alignment in general reflects the rationale for the present study. Showing that postures do not just reflect internal states, but can produce them, Riskind & Gotay (1982) found that tasks designed to cause learned helplessness were more effective when performed by people in slouched spinal positions than erect spinal positions. Those sitting up straight continued working on tasks without giving up for longer than those sitting in slouched positions (Riskind & Gotay, 1982). In another study, clinical case studies were used by Lowell Ward (1981) to create postural-personality profiles reflecting the correlation he found between particular spinal alignments and particular personality types or profiles, including sexual abuse, over-defensiveness, parental conflict, and level of ego strength (Ward, 1981: as cited in Koren & Rosenwinkel, 1992). Another study furthered this research by correlating spinal alignment with personality types as defined by the Minnesota Multiphasic Personality Inventory (MMPI) (Koren & Rosenwinkel, 1992). This study found that the angle of the atlas (where the head connects to the spine) predicted subjects' levels of hypochondriasis, paranoia, and hysteria. Spinal length was correlated with the masculinity/femininity scale (p < .001). These findings were true in the analysis of both subjects' sitting and standing postures. It seems from these several studies that spinal alignments correspond to personal traits and characteristics, possibly communicating these to others.
The muscle relaxation allowed by spinal alignment, explained below in the discussion of how spinal alignment functions physically, may also play a role in therapy. Spinal alignment is used in practices such as meditation and in the treatment of lower back pain caused by improper postures (Olsen, 1998). Poor postural alignment can be generally defined as “when the head is held forward in relation to the trunk or when the shoulders appear slouched forward;” (Raine & Twomey, 1997, p. 1215). Posture in which one's head juts forward is “linked to musculoskeleton dysfunction and pain including craniofacial pain, headache, and shoulder pain” (Raine & Twomey, 1997, p. 1215). Alignment of one's spine allows the force of gravity to travel down the body without placing stress on particular areas of the back or neck. The bones may rest on each other, requiring minimal muscular work. Thus, the nervous and endocrine systems may interpret, respond, and direct unhindered, while the organs can function with minimum restriction and compression (Olsen, 1998). Efficient postures, then, allow release, as opposed to the holding of postural tension. Muscle relaxation, as allowed by efficient body postures, has been found to relate to increased mental performance: attention, memory, and thinking (Legostaev, 1996). These aspects of mental functioning appear important for a therapist to utilize during a treatment session with a client. In addition, posture and muscular relaxation training have been shown to significantly increase visual acuity, visual field, and hearing acuity (Konno, 1997). These heightened sensory levels may also play a role in a therapist's effectiveness. In addition, muscle relaxation has been shown to decrease or alleviate anxiety and guilt feelings (Laird, 1984; Rasid & Parish, 1998). Decreasing these negative emotions may prove beneficial to therapy, as suggested by the already-mentioned research showing that therapists' emotional states impact therapeutic outcome.
Consideration of postural alignment has already been taken into account in one therapeutic technique. Systems-Centered Psychotherapy (SCT), a technique growing in popularity among therapists in the last ten years, requires therapists and clients both to sit “centered,” which is defined as sitting up straight, over one's sitz bones (the part of the pelvis one can feel on the chair or floor when sitting up straight) with one's feet planted on the ground (Agazarian, 1997). SCT theory holds that in addition to decreasing anxiety and tension, an erect, but relaxed, posture provides clients more energy to put toward therapeutic exploration (Agazarian, 1997). In support of this theory, one medical study found that participants going from erect to slouched postures uniformly decreased in their principal resonance frequency (Kitazaki, 1998). Results suggest that spinal alignment affects the vibrational level of the body. If principal resonance translates into working energy, which the author does not specify, this finding may indicate alignment could provide the therapist more energy available for attending to clients. In addition, body alignment is considered helpful in effective group therapy for alleviating postural pains and preventing immediate transferences that people invite through their usual postures (Agazarian, 1997). For example, someone maintaining an especially authoritative posture may elicit either obedience or rebellion from others. Supportive of this stance, Berne claimed he could tell when a client assumed a role by watching his/her body: “there is a characteristic posture, gesture, mannerism, tic, or symptom which signifies [a client] is living in his script or has gone into his script” (Berne, 1971, p. 315).
By focusing on how a therapist's spinal alignment might elicit different ratings from observers, the study utilized “postures as seen.” Also related to “postures as seen,” observers were asked to rate therapist energy level in addition to the traditionally measured constructs of trustworthiness, attractiveness, and expertness. The differences in ratings of perceived therapist energy level when in and out of postural alignment will provide a first step in determining whether a therapist's energy level may be higher in spinal alignment than in other postures. The hypothesis of this study was that people will rate a therapist in a simulated videotape as more trustworthy, attractive, expert and energetic when the therapist is in postural alignment than when not in postural alignment.
Though the sample does not appear diverse in racial/ethnic identification, it appears diverse in participant interests. Thirty different majors of study were reported by participants. The largest groups were education majors, with 22% reporting a major in Elementary Education and 17% reporting Education. Nine percent of the sample reported a major in Psychology, 5% reported having an undecided major, and the remaining 47% of the sample reported majors in the following subjects: Chemistry, Spanish, Art, Exercise Science, Communications Studies, Business, Speech Pathology, Accounting, Cinema, Speech and Hearing, Nursing, Pre-medicine, Pharmacology, Engineering, Biology, Journalism, Pre-law, Political Science, Marketing and Management, Electrical Engineering, Finance, Biomedical Engineering, Computer Science, Religion, Biochemistry, Athletic Training, and History. Thirty-three percent of participants answered “yes” to the question: “Have you been in therapy before or sought advice about a personal issue from a mentor or spiritual advisor?,” while 65% said “no,” and 2% did not answer. It seems previous therapy might inform participants in their ratings, but little is known about how this variable might relate to ratings. Age and year in school of participants are presented in Tables 1 and 2, respectively.
The predictive validity of the CRF was supported by a study showing significant correlations (.37 to .56) between initial client ratings of therapists' trustworthiness, attractiveness, and expertness according to the CRF and the client's therapeutic outcome (goal attainment) (LaCrosse, 1980). These client's ratings after therapy (at the same time the therapeutic outcomes were measured) correlated more highly with therapeutic outcome than the initial ratings. These correlations of .47 to .62 provide concurrent validity evidence (LaCrosse, 1980). Further validity evidence is apparent in the finding of positive correlations (.23 to .67 with a median of .47) between clients' ratings of therapists on the CRF and clients' desire to self-refer to those therapists (Barak & Dell, 1977).
Factor analysis supports the construct validity of the CRF and its ability to distinguish between trustworthiness, attractiveness, and expertness (LaCrosse, 1977). Some argue, however, that because of the constructs' high correlations with each other, one over-arching “general satisfaction” factor (a unitary dimension, “charisma”") would be more appropriate (LaCrosse, 1977). Some studies find a one-factor CRF equally able to discriminate client's perceptions as the three-factor (Heppner & Claiborn, 1989). Mixed data and opinion exist in regard to this question, and it is suggested that the factors be considered both independently and together in future studies, advice which was followed in the present study, where all three scales and over-all scores were utilized (Heppner & Claiborn, 1989). The measure given also included a final, one-item, rating scale (analogous to the ones that make up the CRF) between energetic and not energetic, along which the viewed therapist was rated.
None of the five t-tests revealed significant differences between the scores corresponding to V1 and V2. Because adjusting for inflated type one error from multiple t-tests would only make rejecting the null hypotheses more difficult, thereby leading to the same conclusions as were already found, adjustments were unnecessary. Tests for equal variances failed to reject the null that variances are equal, suggesting that assumption of the tests was met. Scores (generally falling in the center of the scale) were not sufficiently high to create a ceiling affect. The CRF data showed a Cronbach Alpha Coefficient of .90, in the central part of the range of CRF internal consistency alphas reported by other studies (Epperson & Pecnik, 1985).
The average rating of therapist energy level differed between videos by .178 with a 95% confidence interval ranging from -.806 to .4506 (p = .5756). Those watching the hunched therapist gave a slightly higher energy level rating on average, the opposite direction relationship than predicted, but not to a level that this difference would not be likely found by chance. For therapist expertise, average scores between videos differed by 1.883, with a 95% confidence interval ranging from -3.525 to 7.2911 (p = .4912). Expertise is the only construct on which the predicted direction of relationship between ratings and posture was shown (aligned posture being rated higher than not aligned), but, like the other results, not significantly so. Therapist trustworthiness ratings between videos varied by 1.872 with a 95% confidence interval ranging from -6.85 to 3.106 (p = .4573), while attractiveness ratings between videos differed by 4.091 with a 95% confidence interval ranging from -8.824 to .6409 (p = .0894). Both of these results were in the opposite direction as predicted, but again not at significant levels. Therapist attractiveness ratings were the closest to being significantly impacted by therapist alignment, and would have been found just significant at the .05 level if the t-test had been unidirectional (in the direction opposite that predicted). Overall CRF ratings of the counselor were again in the opposite direction than predicted, but not significantly so. The average of the total CRF scores from V1 to V2 differed by .178, with a 95% confidence interval ranging from -15.69 to 3.405 (p = .2048).
Clients might still, for instance, evaluate a therapist in spinal alignment higher on other constructs not tested by the present study, such as felt presence with the client or particular personality characteristics (discussed previously to be correlated with spinal posture), than a therapist who is slightly hunched over. These different perceptions could affect treatment outcome. Other ways spinal alignment as a “posture as seen” might affect the therapeutic process, as discussed in the introduction, are in decreasing immediate transference or in changing the types of emotions clients attribute to therapists. Or, spinal alignment in the “posture as felt” domain could affect treatment outcome by its effects on the therapist (increasing energy, visual acuity, attention, memory, thinking, hearing, and emotional state), as explained in the introduction but not tested by the current study. Despite its not disproving the hypothesis that spinal alignment plays a role in therapeutic outcome, the present study fails to provide support for the theory that it does so through “posture as seen,” by not providing evidence that therapist spinal alignment affects client ratings of the therapist on CRF constructs and energy level. It remains the assertion of this researcher, however, that, in light of the way past research in this area fits together as described in the introduction, the effects of therapist spinal alignment on therapeutic outcome merits continued research.
Another limitation of the present study is its failure to consider how a therapist's natural posture plays a role in how his/her spinal alignment affects therapy. The therapist in the videos used to collect data is accustomed to sitting in a slightly hunched posture. Though not affecting the validity of the videos in representing the postures desired according to definition, the fact that the aligned posture was not natural to the therapist could have increased the awkwardness of maintaining the aligned posture over the other posture. It may be true that a therapist who naturally sits in aligned posture may receive a boost in treatment efficacy because of the way that posture affects therapy, while a therapist with another natural spinal posture who attempts to sit in spinal alignment would not reap the same benefits because it is overridden by the opposite direction effect of the awkwardness of a posture to which one is not accustomed.
Future work in this area might first want to address some of the design weaknesses discussed in the Limitations n. From there, it may also be useful to study how a therapist's “usual” spinal posture plays into how different spinal postures affect their clients. Studies may also want to target one of the other possible mechanisms discussed, but not measured in this study, by which a therapist's spinal alignment might affect therapeutic outcome. To explore how increases in energy, attention, memory, and perception associated with spinal alignment might cause alignment to affect therapeutic outcome might necessitate the conductance of field studies in which therapeutic outcome is measured across many therapists whose natural spinal postures assumed when doing therapy are recorded. With its possibly differential effect on those of different ages and cultures, and the many mechanisms by which it could have its effect, the relationship, if one exists, between therapist spinal alignment and therapeutic outcome appears complex.
Agazarian, Y. (1997). Systems-Centered Therapy for Groups. NY & London: Guilford Press.
Barak, A., & Dell, D. (1977). Differential Perceptions of Counselor Behavior: Replication and Extention. Journal of Counseling Psychology, 24, (4), 288-292.
Barak, A., & LaCrosse, M. (1975). Multidimentional Perception of Counselor Behavior. Journal of Couseling Psychology, 22, (6), 471-476.
Barak, A., & LaCrosse, M. (1975). The Counselor Rating Form, Revised Form.
Begin, A. E., & Garfield, S. L. (Eds.). (1994). Handbook of psychotherapy and behavior change (4th ed.). New York, NY: John Wiley & Sons, 238.
Berne, E. (1971). What do you say after you say hello? New York: Grove Press.
Carkhuff, R. R. (1969). Helping and Human Relations. (Vol 1.). New York: Holt, Rinehart and Winston, Inc.
Egan, G. (1982). The Skilled Helper (2nd ed.). Monterey, CA: Brooks Cole Publishing Co.
Epperson, D., & Pecnik, J. (1985). Counselor Rating FormShort Version: Further Validation and Comparison to the Long Form. Journal of Counseling Psychology, 32, (1), 143-146.
Frank, J. D. (1973). Persuasion and Healing (2nd ed.). Balitimore: Johns Hopkins Press, 1973.
Harrigan, J., Lucic, K., Kay, D., McLaney, A., & Rosenthal, R. (1991). Effect of Expresser Role and Type of Self-Touching on Observers' Perceptions. Journal of Applied Social Psychology, 584-609.
Heppner, P., & Claiborn, C. (1989). Social Influence Research in Counseling: A Review and Critique. Journal of Counseling Psychology, 36, (3), 365-387.
James, W. (1932). A study of the expression of bodily posture. Journal of General Psychology, 7, 405-406.
Kitazaki, S., & Griffin, M. (1998). Resonance behaviour of the seated human body and effects of posture. Journal of Biomechanics, 31, 143-149.
Konno, Y. (1997, May). The effects of relaxation and postural training on external perception: Improvement of visual acuity, visual field, and hearing acuity. Japanese Psychological Research, 39, (2), 119-123.
Koren, T., & Rosenwinkel, E. (1992). Spinal Patterns as Predictors of Personality Profiles: A Pilot Study. International Journal of Psychosomatics, 39, 10-17.
LaCrosse, M. B. (1977). Comparative Perceptions of Counselor Behavior: A Replication and Extension. Journal of Counseling Psychology, 24, (6), 464-471.
LaCrosse, M. B. (1980). Perceived counselor social influence and counseling outcomes: Validity of the Counselor Rating Form. Journal of Counseling Psychology, 27, 320-327.
Laird, J. (1974). Self-attribution of emotion: The effects of expressive behavior on the quality of emotional experience. Journal of Personality and Social Psychology, 29, 405-486.
Legostaev, G. N. (1996, Sep-Oct). Changes in mental performance after voluntary relaxation. Human Physiology, 22, (5), 637-638.
Maxwell, G., & Cook, M. (1985, June). Postural congruence and judgments of liking and perceived similarity. New Zealand Journal of Psychology, 14, (1), 20-26.
Meijer, M. (1989, Winter). The Contributions of General Features of Body Movement to the Attribution of Emotions. Journal of Nonverbal Behavior, 247-267.
Mohacsy, I. (1995). Nonverbal communication and its place in the therapy session. The Arts in Psychotherapy, 22, (1), 31-38.
Olsen, A. (1998). Bodystories. Barrytown, NY: Station Hill Opening.
Ponterotto, J., & Furlong, M. (1985). Evaluating Counselor Effectiveness: A Critical Review of Rating Scale Instruments. Journal of Counseling Psychology, 32, (4), 597-616.
Raine, S., & Twomey, T. (1997, November). Head and Shoulder Posture Variations in 160 Asymptomatic Women and Men. Archives of Physical Medicine and Rehabilitation, 78, 1215-1223.
Rasid, Z., & Parish, T. (1998, Spring). The effects of two types of relaxation training on students' levels of anxiety. Adolescence, 33, (129), 99-101.
Ridley, N., & Asbury, F. (1988, March). Does counselor body posture make a difference? The School Counselor, 35, (4), 253-258.
Riskind, J., & Gotay, C. (1982). Physical posture: Could it have regulatory or feedback effects on motivation and emotion? Motivation and Emotion, 6, (3), 273-298.
Rossberg-Gempton, I., & Poole, G. (1993). The Effect of Open and Closed Postures on Pleasant and Unpleasant Emotions. The Arts in Psychology, 20, 75-82.
Rossberg-Gempton, I. E., Dickinson, J., Kristiansen, L. G., & Allin, S. (1992). Body postures and affect. Submitted for publication.
Roten, Y., Darwish, J., Stern, D.J., Fivaz-Depeursinge, E., & Corboz-Warnery, A. (1999) Nonverbal Communication and Alliance in Therapy: The Body Formation Coding System. Journal of Clinical Psychology, 55, (4), 425-438.
Winter, D., Widell, C., Truitt, G., & George-Falvey, J. (1989). Empirical Studies of Posture-Gesture Mergers. Journal of Nonverbal Behavior, 207-222.
Young, M. (1998). Learning the Art of Helping. Upper Saddle Rv, NJ: Prentice Hall Inc.
|Received: December 14, 2002
Accepted: February 2, 2003
Revised: February 3, 2003
Copyright © Reysen Group 2002