The (Un)Emotional Male: Physiological, Verbal, and Written Correlates of Expressiveness.

ROBERTSON, JOHN M. ; WOODFORD, JOYCE ; LIN, CHI-WEI 等

In North American culture, it frequently is asserted that men find it difficult to talk about their emotions. For more than 25 years, this topic has been the subject of both theoretical exploration (David & Brannon, 1976; Friedman & Lerner, 1986; Goldberg, 1976; Pleck, 1981) and social science research. Most scholarly observers in this area have described men as having considerable difficulty in verbally expressing their emotions to others (e.g., Brooks & Gilbert, 1995; Sher, 1993; Wilcox & Forrest, 1992). Based on this theoretical work, clinicians have been encouraged to deemphasize the focus on talking about feelings in therapy and to offer men more opportunities to be active. Examples of such activities include workshops, consultations, self-help reading, classes, or other behavioral approaches (Bernstein, 1995; Hurst, 1997; Kelly & Hall, 1992; Miller, 1995; Robertson, 1989; Shay, 1996; Wanigaratne & Barker, 1995).

These studies seem to suggest that most men are hypoemotional. Repeatedly, it is asserted that most men have great difficulty in identifying their own emotions, in sending emotional messages, or in accurately interpreting the emotions of others. The term alexithymia has been used to describe this inability to adequately express affect (Levant, 1992). Studies that emphasize the hypoemotionality of men typically compare men with women and then report significant gender differences. To illustrate, men have been reported to have lower levels of happiness, life satisfaction, depression, fear, and sadness than women (Nolen-Hoeksema, 1987; Scherer, Wallbott, & Summerfield, 1986; Wood, Rhodes, & Whelan, 1989).

However, more recent work has questioned some of the empirical basis for this perspective. For example, Heesacker and his colleagues have suggested that it may be mostly a stereotype to suggest that men are hypoemotional (Heesacker, Wester, Vogel, Wentzel, Mejia-Millan, & Goodholm, 1999). Although the view that men are hypoemotional (and that women are hyperemotional) has a long history (cf., Parsons & Bales, 1955), Heesacker cited several studies that show very little affective difference between men and women. Specifically, men and women have shown no significant differences in their abilities to send or receive emotional messages from others (Conner & Heesacker, 1999), in their expressions of anger (Burrowes & Halberstadt, 1987), in their private experiences of emotion (LaFrance & Banaji, 1992), or in the facial expressions and visual behavior that they use to respond to emotional questions (Cherulink, 1979).

Other studies have found similar results. Eagly and Steffen (1986) found that although typical women are judged to be more emotionally expressive than typical men, when subjects were told that men and women were in identical roles, subjects reported no differences in their expectations of men and women. Similarly, in a study that required men to report on their own emotions (fear, joy, sadness, and love), results indicated that when men report less intense emotional experiences, they are more likely to hold stereotypical views about emotional differences between men and women (Grossman & Wood, 1993). The Heesacker review concluded that "research on emotion fails to support the perception of vast affective differences between men and women" (Heesacker et al., 1999, p. 483,484). This conclusion echoes the findings of others who have reviewed the literature (Canary & Emmers-Sommers, 1997; LaFrance & Banaji, 1992).

Although these two strands of literature may seem somewhat incompatible, it also can be argued that they are complementary. Granted, many North American men may have difficulty expressing their emotions verbally; this has been noted for several decades. However, it may also be true that men do not necessarily have fewer emotions or have any less capability than women of expressing them. It may be that men are less verbally expressive of their emotions because they have been socialized to say less, or because they have a limited capacity to express themselves, not because they have little emotional arousal in their lives.

To explore this possibility, it would be useful to look for within-group differences among men, rather than to compare, men with women. Two types of questions can be asked. First, do all men experience significant emotional arousal to the events in their lives? Are there significant differences in levels of arousal among men? If so, can this be shown empirically? Second, are some men more verbally expressive of their emotional arousal than other men? If so, what makes the difference? Do the themes of traditional masculine socialization help us understand any within-group differences on the emotional expressiveness of men?

In considering these questions, two clusters of studies seem helpful to review. The first group suggests that high levels of emotional expressiveness in men are unlikely because of the internal strain it creates. Pleck and his associates (1993) developed a set of statements illustrating the "masculine ideology" of North American culture. Included were items involving respect from others, self-confidence, physical toughness, reluctance to talk about problems, avoidance of feminine-like behavior, low interest in housework, and readiness for sexual activity. Other attempts to describe central masculine socialization themes have included such constructs as achievement/success/status; toughness/aggressiveness; dominance/control; analysis/rationality; and avoidance of emotionality/femininity/homosexuality (David & Brannon, 1976; O'Neil, 1981). Researchers have investigated the conflicts created for men by these expectations, using such models as "gender-role conflict" (O'Neil, Helms, Gable, David, & Wrightsman, 1986), "sex-role strain" (Pleck, 1981), and "gender-role stress" (Eisler & Skidmore, 1987). A basic idea in this body of work is that men have been socialized in ways that make the verbal expression of emotions unlikely because of the "conflict," "strain," or "stress" that such expressions might evoke.

The second group of relevant studies suggests that men are less emotionally expressive as adults because they have fewer opportunities to develop those skills. Cross-cultural studies have shown that the child-care role (requiring high levels of emotional investment and expressiveness) are much less frequently held by men than by women; Crano and Aronoff (1978) looked at 186 societies and found that young boys through the age of five received much less emotional support from fathers than from mothers. Similarly, personal and emotional care of the frail elderly is less frequently a responsibility taken by men (Dwyer & Seecomb, 1991); men are expected to provide less emotionally involving services, such as household maintenance and repairs or transportation to appointments. Given the differentiation of these roles by sex, men simply have fewer opportunities to develop and use skills of emotional expressiveness.

Based on these two ideas (that traditional masculine ideology makes the verbal expression of emotions stressful, and that men simply have few opportunities to develop the skills of emotional expression), it is plausible to suggest that the problem may not be the inherent hypoemotionality of men; it may rather be that men have been socialized not to verbalize their feelings.

To empirically test this theory, two suppositions must be examined: that virtually all men are significantly physiologically aroused by emotion-inducing events; and that men vary in how they wish to express this emotional arousal, based on (perhaps) masculine idealogy. To date, these links have not appeared in the empirical literature. However, recent studies have examined physiological data in ways that begin to explore this possibility. In brief, this body of research hints that many men actually may be experiencing higher levels of emotional arousal than they are reporting verbally. For example, Grossman and Wood (1993) found that when men and women were not given normative instructions about responding to emotion-inducing slides, men verbally reported less emotional intensity than women (a finding consistent with socialized expectations), but they also found that when the instructions normalized emotional responsiveness, no sex differences were found in physiological responses to the slides. Further, there is evidence that men are highly reactive physiologically to stressful events. Gottman and Levenson (1988) reviewed a series of studies showing that men do indeed experience physiological arousal when presented with emotion-inducing events (i.e., changes in blood pressure, basal skin conductance, heart rate, release of hormones, and corticosteroid excretion).

The purpose of this study was to explore any associations between two variables of the male experience: what men actually experience physiologically when emotionally aroused, and how they report those experiences. To accomplish this, we presented men with three emotion-inducing events and then compared what they reported about the experience with measures of their actual physiological responses. We also were curious about the role of masculine gender-role issues in this equation. Based on the literature (Bernstein, 1995; Hurst, 1997; Kelly & Hall, 1992; Miller, 1995; Robertson, 1989; Shay, 1996; Wanigaratne & Barker, 1995; Wilcox & Forrest, 1992), we hypothesized that men with more traditional gender-role expectations would not have lower levels of emotional arousal, but instead would be more reluctant to express verbally their internal emotional arousal. At the same time we expected that a more structured exercise would elicit more expressiveness from these men than would invitations to simply speak about their experiences.

Because we were interested primarily in within-group differences, no comparison groups of women were used. That is, we were interested in examining the stereotype that most men are hypoemotional--that they do not experience significant levels of emotional responsiveness or that they do not express their emotions effectively. Further, we were interested in whether or not masculine socialization might be a variable that would help explain any within-group differences among men with regard to emotional expressiveness.

The importance of examining this question is illustrated by the variety of intrapsychic, interpersonal, and physical health problems men experience (Cleary, 1987; Courtenay, 2000; Meinecke, 1981). In particular, emotional inexpressiveness has been associated with marital problems, alcohol abuse, workplace stress, reluctance to seek counseling, poor health habits, the use of violence, Type A behavior, loneliness, low use of social support networks, and sexual difficulties for men (Eisler & Blalock, 1991; Gordon & Meth, 1990; Pasick, Gordon, & Meth, 1990). If we can begin to show that the problem is not a lack of emotional arousal or responsiveness among men, but rather a lack of opportunity or a lack of social support for expressing themselves, then these latter factors might be directly addressed.

One final comment concerns the problem of defining basic emotions. Although a review of this definitional problem is beyond the scope of this study (e.g., Ekman & Davidson, 1994), it is appropriate to describe briefly the components of an emotion as conceptualized for the present work: (a) an emotion requires an antecedent event (intrapsychic, interpersonal, or environmental) that stimulates physiological arousal; (b) the arousal functions to focus attention on life tasks that need to be addressed and provides motivation to make appropriate adaptive responses (Davidson, 1994; Frijda, 1994; Gray, 1994; Levenson, 1994; Robertson & Freeman, 1995). Put in real life terms, when a bear suddenly appears on a mountain pathway in front of a man, physiological reactions occur (changes in heart rate, skin temperature, etc.). The purpose of this arousal is to focus attention on the problem and energize appropriate adaptive responses. The emotion includes the arousal, the appraisal, and the motivation to make an adaptive response. For this particular illustration, the emotion might be named fear or panic. Using this definition, most emotions can be named with a single descriptive word, and these emotion words can be rated with regard to intensity (e.g., Averill, 1975).

METHOD

RESEARCH PARTICIPANTS

Participants were recruited from several male-dominated occupations and were all employed near a major university in the Midwest. Two themes consistent with traditional masculine socialization formed the selection criteria: the men were directly responsible for the physical or emotional well-being of a fairly large number of people; and they were required to respond immediately and act as problem-solvers to persons in acute distress. Men theorized to have met these criteria included police officers, sheriff deputies, jailers, fire fighters, members of the clergy, and men with administrative responsibilities that included significant safety concerns. Generally, they were recruited through the organizations in which they were employed (e.g., announcements in police departments, fire stations, and so forth). For participating in a study of "Stress Management Factors," the men were offered free admission to a four-hour workshop on the management of stressful emotions at work.

The 69 participants had a mean age of 35.71 (SD =11.81). The response rate (the proportion of actual participants to those invited to participate) was relatively high (just over 80%); superior officers and supervisors strongly encouraged the men to participate. The participants had completed an average of 4.25 years of education beyond high school. They identified themselves ethnically by reporting that their ancestors came mostly from Europe (85%), Africa (4%), or Hispanic countries (4%). The rest (6%) reported mixed ancestry or left the question blank. They reported their current relationship status as married (49%), single with no steady relationship (32%), divorced (10%), and steady but unmarried relationship (9%).

Because traditional counseling invites men to be emotionally expressive, the men were asked about their preferences or experiences with counseling. Most had never talked with a professional helper for personal concerns (56%), and only a fifth (21%) of the entire sample indicated that they would be more likely in the future to seek help from a professional than from family members, relatives, or friends. Yet, 43% had sought help from a professional counselor or a member of the clergy, and had attended an average of 5.93 sessions. This latter percentage is higher than we expected, given the usual counseling utilization rates of men. Perhaps the higher levels of risk and demand in these occupations made them more likely to seek help.

INSTRUMENTS

Instruments were selected to measure the following variables: (a) physiological arousal during an emotion-inducing event; (b) verbal and structured emotional expressiveness immediately following the emotion-inducing event; and (c) the presence of gender-role stress.

To measure psychophysiological responses to emotion-inducing events, participants were attached to the Focused Technology F1000 Instrumentation System (BioMedical Instruments, 1994). A battery of responses was collected, including skin temperature and electrodermal response (EDR) data at 7.2 samples per second, electromyographic (EMG) data at 10 samples per second, and heart rate with electrodes on the upper body. Responses were measured during three emotion-inducing events, involving visual/auditory experiencing (a nine-minute videotape), cognitive functioning (a one-minute mental arithmetic task), and sensory functioning (a one-minute cold pressor test).

Videotape Segment. Numerous studies have successfully induced emotional responses with the use of film/videotape stories. Reviews of these studies (Gerrards-Hesse, Spies, & Hesse, 1994; Martin, 1990) have noted that viewing videotaped scenes does indeed induce physiological arousal. For example, segments of the motion picture "The Champ" have been used to elicit sadness (Marston, Hart, Hileman, & Faunce, 1984), and the movie "Run" has been used to induce anxiety (Isen & Gorgolione, 1983).

For the present study, segments of the movie "My Life" (Netter & Rubin, 1993) were selected because the story deals with themes many theorists have indicated are difficult for traditionally socialized men to express (father-son relationships, sadness, and death). The father-adult son relationship in the story is a subplot of the larger story about the terminally ill adult son who wanted to preserve his life story on videotape for his own newly born son. The father-adult son interactions in the movie were edited together to form a video that lasted nine minutes and 40 seconds.

Immediately following presentation of the videotape segment, participants were asked "What words would you use to describe your reaction to the video?" Answers were recorded on audiotape and then transcribed. Graduate students were employed to count the number of emotion words used by participants and to indicate the emotionality rating for each word using the model developed by Averill (1975). Previous studies have used the words of research participants to evaluate the presence of emotions (e.g., Greenwald, Cook, & Lang, 1989; Lang, 1984, 1985), drawn conclusions based on the counting of English words (Breland, 1996; Thorndike, 1921), and used response time as a measure of emotionality (Fitzgibbons & Simons, 1992; Hess, Kappas, McHugo, Lanzetta, & Kleck, 1992; Lang, Greenwald, Bradley, & Hamm, 1993).

Mental Arithmetic Task. Earlier studies have shown that mental arithmetic procedures can increase activity in muscle tension, skin temperature, heart rate, electrodermal response, and systolic blood pressure (e.g., Anderson, 1989; Arena, 1984). After reviewing the protocols of several investigators (Carlson, Collins, Stewart, Porzelius, Nitz, & Lind, 1989; Donat & McCullough, 1983; Subotnik & Shapiro, 1984), a one-minute arithmetic task was devised for the present study: participants were given a three digit number on a 4x6 card and asked to count aloud backwards by increments of 13. They were instructed to work as rapidly as possible while the investigator wrote down their responses.

Cold Pressor Test. Psychophysiological sensitivity has been found with the use of the cold pressor procedure (e.g., Reeves & Shapiro, 1983; Subotnik & Shapiro, 1984). Participants in the present study were asked to immerse one hand in a three-gallon tank filled with crushed ice and water for as long as they could withstand the pain, up to a maximum of 60 seconds. Water was kept constant at a temperature of less than .5 degrees Centigrade. Participants were informed that they could remove their hand at any time the sensations became too painful.

Emotional Assessment Scale (EAS). After each administration of the three emotion-inducing events, participants were given a structured pencil and paper assessment that was generally completed in less than one minute. The EAS (Carlson et al., 1989) consists of 24 words that describe eight emotional states (anger, anxiety, disgust, fear, guilt, happiness, sadness, and surprise). The eight selected emotions were based on the work of Izard and others (Ekman, 1972; Izard, 1977, 1991). Three descriptors for each emotion were chosen using the work of Nunnally (1981), who developed a factor-analytic list of words most often used for each emotion category.

To complete the exercise, participants used a visual analogue scale. Participants were asked to measure the intensity of each emotional descriptor by drawing a slash on a 100-millimeter line. The line ranged from points labeled "Least Possible" (zero) to "Most Possible" (100 millimeters). The instrument has been used to measure momentary and immediate changes in emotional states. The EAS was scored by measuring the number of millimeters from the left endpoint up to the slash mark. Scores for each emotion, therefore, ranged from 0-100.

Carlson et al. (1989) reported that reliability was measured using both inter-item (.70 to .91) and split-half (.94) procedures. Because the EAS is designed to measure an emotional state at a particular point in time, computing reliability over several administrations was not appropriate. Validity was examined by comparing several of the subscales with existing measures of emotions, such as the Profile of Mood States (McNair, Lorr, & Doppleman, 1981), the Beck Depression Inventory (Beck, 1972), and the State form of the State-Trait Anxiety Inventory (Speilberger, Gorsuch, & Luschene, 1970). Results were supportive, with positive correlations found in the expected directions for such subscales as sadness (.74), anxiety (.78), anger (.69), and happiness (-.36).

Semantic Atlas of Emotional Concepts. Intended to be a complete list of emotional terms in the English language, this instrument was used to measure both the frequency and the intensity of the emotion words used by men in our study. Averill and his colleagues (1975) first reviewed 18,000 psychological terms compiled by Allport and Odbert (1936), as well as three other emotion word lists. If two of five judges agreed that a term had emotional connotations, Averill kept the term on the list. A total of 717 terms were thus retained. Further work using university undergraduates as subjects eliminated non-emotional terms and reduced the list to 558 words.

For each word, an emotionality rating was developed. Several factors were combined (activation/energy, evaluation/pleasantness, importance/depth of experience, and degree of control). Estimates of reliability for these ratings were obtained by randomly splitting the subject pool into halves. Correlations between the split-sample scores were ranged from .87 to .98 for the four factors. An overall emotional rating was calculated, based on a scale of -3 (less emotionality) to +3 (more emotionality). To illustrate, the term "furious" had an emotionality rating of +2.25, and the term "listless" had a rating of -1.65.

For the present study, a word used by a research participant was called an emotion if it appeared in Averill's atlas, and the word's intensity was recorded using Averill's emotionality rating.

Masculine Gender-Role Stress (MGRS). Developed as a measure of male gender-role stress (Eisler & Skidmore, 1987), the MGRS consists of 40 items factor analyzed into five subscales: Physical Inadequacy (not being competitive in sports or sexual rivalries); Emotional Inexpressiveness (difficulties in expressing feelings or responding to emotions in others); Subordination to Women (in athletic or sports settings); Intellectual Inferiority (indecisiveness, inability to handle situations); and Performance Behavior (work and sexual adequacy). The underlying theory is that men will experience stress if they believe they are not behaving as men should behave or if a situation forces them to act in ways others define as feminine.

Higher scores reflect comparatively higher levels of gender-role stress and have been correlated positively with stress in expressing emotions, higher levels of anxiety and anger, poorer health habits, and higher systolic blood pressure changes in a cold pressor test (Eisler & Blalock, 1991; Eisler, Skidmore & Ward, 1988). The scale has shown high internal consistency (alpha coefficients in .90s), and test-retest reliability over two weeks was .93. Construct validity has been supported by comparing the MGRS with other traits theoretically linked with masculine stress, such as inexpressiveness, anger, state anxiety, stress reactions, and adverse health habits (Eisler et al., 1988; Saurer & Eisler, 1990).

PROTOCOL

Participants were scheduled for individual 90-minute sessions, with the entire data set being collected during a four-week period to minimize the potential confound of contemporary events. They came to a university health center where the project was explained. The consent form was signed, and the men were instructed to thoroughly wash their arms up to the elbow. Sensors were attached to the forefinger of the nondominant hand, the palm of the hand, the forearms, and clavicles. Participants were then seated in a recliner chair, with a visual screen shielding the equipment, a video monitor, and an audiotape recorder. They were connected to the F1000 system and directed through the following sequence of events: a baseline rest period (10 minutes); viewing the videotape stimulus (9 minutes); rest and recovery (five minutes); performing the arithmetic task (counting backwards by 13s for one minute); rest and recovery (five minutes); completing the cold pressor task (placing hand in ice water for up to one minute); final rest and recovery (sitting quietly and relaxing for five minutes). They were then ushered into a separate room and given the MGRS. Finally, they were debriefed by a licensed psychologist who inquired about any discomfort the procedure might have induced.

RESULTS

PHYSIOLOGICAL AROUSAL

Table 1 shows the means and standard deviations of the physiological measures for each event in the protocol. It is clear that each of the three stressors (videotape, math test, and cold pressor) induced measurable physiological responses. Skin temperature was reduced, and significant changes were measured for EDR, EMG, and heart rate. The widest range of responses occurred on the EMG, with more moderate changes occurring with skin temperature, EDR, and heart rate.

Table 1
Physiological Measures of Responsiveness to Emotion-Inducing Stimuli

 Temperature EDR (sweat)

Event M SD M SD

Baseline Rest 92.8 3.9 4.6 4.8
Video Stimulus 91.1 4.4(**) 6.2 5.2(**)
Verbal Response 90.1 4.7(**) 8.8 7.1(**)
EAS--Video 89.9 4.7(**) 8.3 6.2(**)
Rest Period 90.8 4.8(**) 5.8 5.2(**)
Math Stimulus 89.7 4.6(**) 9.1 6.9(**)
EAS--Math 89.0 5.6(**) 8.5 6.1(**)
Rest Period 90.4 5.1(**) 5.9 5.1(**)
Cold Pressor 89.3 5.0(**) 6.9 5.7(**)
EAS--Cold Pressor 88.4 5.0(**) 6.3 5.3(**)
Rest Period 89.2 5.8(**) 5.5 4.9(**)

 EMG (muscle) Heart Rate

Event M SD M SD

Baseline Rest 8.0 4.1 69.2 9.9
Video Stimulus 16.8 8.3(**) 68.3 9.2
Verbal Response 20.2 10.3(**) 78.0 12.0(**)
EAS--Video 30.2 8.3(**) 73.2 10.3(**)
Rest Period 9.8 6.2(*) 67.4 12.1
Math Stimulus 15.4 6.3(**) 78.5 12.8(**)
EAS--Math 30.1 9.3(**) 73.5 11.0(**)
Rest Period 8.2 3.9 68.5 10.2
Cold Pressor 30.4 11.2(**) 74.0 9.6(**)
EAS--Cold Pressor 29.8 9.3(**) 70.1 10.1
Rest Period 8.7 5.9 66.0 10.4(**)

(**) p < .01

(*) p < .05

Note: Temperature is reported in degrees Fahrenheit. EDR
(electrodermal response) is reported in micro ohms. EDR numbers
express units of change in conductance, relative to an EDR
reference value; higher numbers reflect greater arousal. EMG
(electromyographic) units are micro volts per event; higher
numbers generally reflect greater arousal. Heart rate is
reported in beats per minute. EAS refers to the Emotional
Assessment Scale (Carlson, et al., 1989). Each data point
is compared to the baseline rest event.

It should also be noted from Table 1 that the simple request to report on their responses to the stimuli also elicited measurable arousal. This held true for both forms of expression (i.e., for both the open-ended verbal responses and the structured exercise [EAS]). Every time the men were asked to describe their reactions to one of the stimuli, their bodies responded with measurable arousal.

These results were consistent throughout the sample. Men who scored high on the MGRS (indicating higher levels of gender-role stress), for example, displayed physiological arousal patterns virtually indistinguishable from men who scored lower on the MGRS. A median split of MGRS scores divided the men into two groups (high gender-role stress and low gender-role stress). There was no significant difference between these two groups on any of the physiological measures for any event in the entire protocol, with p-values ranging from .371 to .830.

A primary finding, then, was that all men responded physiologically to the emotion-inducing stimuli, regardless of their scores on the gender-role stress measure. Further, it should be noted that there were no significant differences within the sample by occupation.

VERBAL AND STRUCTURED EXPRESSIVENESS

Tables 2 and 3 report results regarding expressiveness. To measure verbal responsiveness, participants were asked to talk about the stimulus they had just experienced (e.g., "What words would you use to describe your reactions to the video?"). Audiotaped responses were analyzed with regard to response time, total number of words used, number of emotion words used, and an emotionality rating of the emotion words. In addition, after the men had completed their responses to all three stressors, the participants were asked how they generally handled stress in their lives. Again, audiotaped responses were analyzed, calculating both the response time and the number of words used in responding to the question. These six elements are reported in Table 2.

Table 2
Verbal Measures of Responsiveness to Emotion-Inducing Stimuli

Measure M SD

Response time in reacting to video stimulus(a) 6.0 4.5
Number of words used in reacting to video 51.2 51.7
Number of emotion words used in responding to video(b) 3.1 2.7
Emotionality rating of emotion words used for video(c) 2.3 2.5
Response time in reporting stressful reactions(a) 4.3 2.5
Number of words used in describing stressful reactions 109.8 86.1

Note: (a) Response time was the number of seconds between the
interviewer's question and the participant's answer; faster
responses times were interpreted as more expressive. (b) Words
were accepted as emotion words if they appeared in the atlas of
English emotion words compiled by Averill (1975) (c) The Averill
(1975) atlas developed an emotionality rating based on a
seven-point scale from -3 to +3, with higher numbers indicating
stronger emotionality.
Table 3
Written Measures of Responsiveness to Emotion-Inducing Stimuli

 Videotape Math Task Cold Pressor

Emotion M SD M SD M SD

Anger 25.3 25.8 19.6 24.7 12.5 17.9
Anxiety 26.0 24.5 39.5 29.3 22.9 21.2
Disgust 17.8 20.6 15.6 21.6 8.4 13.1
Fear 19.0 20.7 20.0 22.0 10.4 14.9
Guilt 22.7 23.6 24.3 24.9 9.8 12.9
Happiness 20.6 23.9 11.2 13.8 8.4 11.4
Sadness 33.1 19.0 17.0 21.6 6.9 9.3
Surprise 17.6 19.9 28.3 27.3 29.4 25.4

Stimulus Totals 22.8 22.3 21.9 23.2 13.6 15.8

 Average Score
 by Emotion

Emotion M SD

Anger 19.1 22.8
Anxiety 29.5 25.0
Disgust 13.9 18.4
Fear 16.5 19.2
Guilt 18.9 20.5
Happiness 13.4 16.4
Sadness 19.0 16.6
Surprise 25.1 24.2

Stimulus Totals 19.4 20.4

Note: Participants reported the intensity of each emotion by drawing
a slash on a 100-millimeter line (range = 1 to 100). Higher numbers
indicate greater self-reported intensity for each emotion (Carlson et
al., 1989).

The men tended to use rather few emotion words when reporting their reactions to the video. However, the emotion words that they did use were quite intense (M = +2.28, on a seven-point scale from -3 to +3, with higher numbers representing more intense emotionality). Variability in verbosity was wide, as the average number of total words used (emotion words plus all other words) was 51.16, with a standard deviation of 51.65. Individual differences in overall wordiness were quite prominent. The variability on the use of emotion words was much smaller.

Results on the structured measure of expressiveness (EAS) indicated that men were quite willing to report emotional responses with the use of pencil and paper. Table 3 indicates that all eight of the EAS emotions were reported as present at some point during the protocol. Within a range of 1 to 100, the videotape (M = 22.8, SD = 22.3) and the math test (M = 21.9, SD = 23.2) evoked more intense overall ratings than the cold pressor (M = 13.6, SD = 15.7). With regard to individual emotions, the highest rankings across the three stressors were given to anxiety (M = 29.5, SD = 25.0) and surprise (M = 25.1, SD = 24.2), and the lowest ratings were given to disgust (M = 13.9, SD = 18.4) and happiness (M = 13.4, SD = 16.4). These findings are not surprising, given that none of the emotion-inducing events was designed to induced happiness or disgust. The high standard deviations for all these ratings again suggested wide individual variation.

For the videotape stimulus, the most intense ratings were for sadness (M = 33.1, SD = 19.0), anxiety (M = 26.0, SD = -24.5), and anger (M = 25.3, SD = 25.8). The math test elicited reports of anxiety (M = 39.5, SD = 29.3) and surprise (M = 28.3, SD = 27.3). The cold pressor generated the most intense responses on the same two dimensions as the math test, though in the reverse order: surprise (M = 29.4, SD = 25.4) and anxiety (M = 22.9, SD = 21.2).

It is noteworthy that the emotion with the highest rating in the EAS exercise was anxiety, given the theoretical construct that traditional masculinity emphasizes success, independence, and a reluctance to acknowledge fears. It may suggest that when men are given a structured way of responding to emotional issues, they are willing to respond, even if it means acknowledging emotional states generally minimized by traditional masculine socialization. Supportive of this idea was the finding that men scoring highest on the gender-role stress measure (upper half of the MGRS) actually reported greater emotional intensity on the structured exercise (EAS) than did men in the lower half of the MGRS, t(67) = -2.27, p [is less than].02. In this sample, then, men with higher levels of gender-role stress were quite willing to identify their emotional states on paper in a structured exercise--even more willing to do so than men reporting less gender-role stress.

It is apparent that our measures of verbal and structured expressiveness were tapping different dimensions. The correlation between verbal expressiveness and scores on the EAS was rather low (r = .06). The implication is that other variables may explain the difference. In the present study, masculine gender-role stress served as the independent variable, revealing the preferred modes of emotional expressiveness (structured or verbal) in men.

GENDER=ROLE STRESS

The MGRS score for this sample of men (M = 80.16, SD = 20.88) was similar to another sample of adult men reported in the literature: M = 83.2, SD = 21.0 (Watkins, Eisler, Carpenter, Schechtman, & Fisher, 1991). The mean age of both groups was about 35. No significant differences were found by comparing MGRS scores for the various occupational groups within the present sample.

The low overall correlation between verbal and structured expressiveness in the entire sample leaves open the possibility that masculinity themes may be related to a preference for either structured or verbal forms of expressiveness. To test this question, a t-test for paired samples was performed, using gender-role stress as an independent variable for verbal and structured expressiveness.

Verbal expressiveness scores were computed by transforming into z-scores the different units of measurement (see Table 2) generated by the audiotaped responses to the videotape. We combined the intensity of emotions rating with the response time and the proportion of overall words that related to emotion; the result was a single z-score measure of verbal expressiveness. Similarly, the structured exercise scores (EAS) were transformed to z-scores. The t-test for paired samples compared the value of verbal expressiveness of each participant with his overall EAS score, and yielded a paired mean difference for each man. Two groups were formed by a median split of scores on the MGRS, indicating high and low gender-role stress.

Results of the two-tailed t-test comparisons were consistent with theoretical expectations (Table 4). In the high gender-role stress group, the mean z-score for structured expressiveness (M = .3223) was higher than the mean score for verbal expressiveness (M = -.1259). The paired difference (M = .4482) was significant, t(34) = 2.17, p [is less than] .05. The low gender-role stress group had higher verbal expressiveness scores (M = .1335) than structured expressiveness scores (M = -.3525). The paired difference (M = .4860) was significant, t(32) = -2.00, p [is less than].05. This pattern indicates that high verbal expressiveness was associated with low gender-role stress, whereas high structured expressiveness was associated with high gender-role stress. It seems appropriate to suggest that gender-role stress may predict the preferred mode of emotional expressiveness. Men with more gender-role stress related to views of traditional masculinity are more likely to be comfortable expressing themselves in structured ways, and men with less masculine gender-role stress are likely to be more comfortable expressing themselves verbally.

Table 4
Masculine Gender Role Stress as a Predictor of Verbal versus Written
Expressiveness

Masculine Gender-Role Stress (MGRS) Verbal Written t p

High Masculine Stress -.1259 .3223 2.17 .018
Low Masculine Stress .1335 -.3525 -2.00 .027

Note: Values are expressed as combined z-scores, transformed from
measures of verbal expressiveness (response time, number of emotion
words, and emotionality of words) and from structured expressiveness
(Emotional Assessment Scale, from Carlson et al., 1989). High and low
masculine groups were formed by a median split of the MGRS (Eisler &
Skidmore, 1987).

DISCUSSION

What accounts for the belief that men do not easily express themselves emotionally? Is it that many men experience very low levels of physiological arousal to emotional stimuli? Is it primarily because men are socialized to say little about their emotions? Or is it some combination of both?

Two findings in the present study seem relevant to this question. The first is that all men in our sample were physiologically aroused by emotional stimuli; arousal was not absent among men in traditionally masculine occupations or for men who hold views of masculinity that lead to stress. The second finding is that regardless of gender-role stress issues, men are aware of their emotional arousal and are willing to report it; what varies are their preferences about the most comfortable format for expressing those emotions. Men with higher levels of stress related to traditional gender roles appear to prefer structured approaches, and men with less gender-role stress appear to prefer verbal expressions.

The first finding that all the men in our sample were physiologically aroused may not be especially surprising, as it is consistent with the summary findings of Gottman and Levenson (1988), who pointed to a series of studies showing high levels of physiological arousal in men. This finding speaks directly to the question posed at the outset of the present study. The explanation for the emotional inexpressiveness of some men is not that they experience low physiological arousal and therefore have little to report. Men do respond physiologically to emotional stimuli, and the arousal does not appear to be related to any predictive variable in our sample (gender-role stress, occupation, age, and so forth).

The second finding is consistent with the literature indicating that many men prefer exercises and structured tasks more than open-ended talk about their inner feelings (Hurst, 1997; Robertson, 1989). It is also consistent with findings from studies that show men still are far more likely in family settings to offer services such as household maintenance, repairs, or transportation to appointments, and less likely to be involved in family roles that require high levels of emotional investment (Dwyer & Seecomb, 1991).

Perhaps the best answer to the question that prompted this study is that both "nature" and "nurture" contribute to an understanding of men and emotions. With regard to nature, we found that emotional inexpressiveness in men was not due to an absence of emotional arousal. In fact, as Gottman and Levenson have argued (1988), the fact that men have very high levels of physical arousal to stress may actually contribute to less verbal expressiveness. They noted that while experiencing conflict with their partners, many men report feeling flooded or overwhelmed and therefore less able to be verbal. Physiological arousal may indeed be a factor contributing to low levels of emotional expressiveness in men--not because the arousal levels are low, but because they are comparatively high (Gottman & Levenson, 1988).

Regarding the nurture contribution, we found wide variability in the format preferred by men to express the emotions that they do have. Some men preferred to express themselves in an activity, and others preferred words. These differences seem more likely to be influenced by socialization. Given the differences between gender-role expectations in this culture, many men simply have fewer opportunities to develop their skills of emotional expressiveness. It is plausible to argue that many of these differences among men are learned (Eagly & Steffen, 1986; Grossman & Wood, 1993).

It is tempting to speculate about implications for counseling that might stem from these results. For example, when working with male clients struggling with affect, therapists may be able to increase their effectiveness by giving attention to the client's gender role perspectives, and exploring any stress reactions related to those expectations. Many men may freeze at versions of the question, "How do you feel?" But those same men may cooperate willingly and effectively if given a structured way of reporting their feelings. In the context of the present study, the task of identifying the intensity of their emotions on 100-millimeter lines was effective; other creative ways may be equally as effective. This approach is certainly consistent with theoretical expectations, considering the traditional male emphasis on problem-solving, task-completion, and structured activities. Given that more traditionally defined men find talk therapy difficult, if not ineffective, it follows that the use of written assessments, structured expressive modalities, and task direction may provide more satisfying results to the client.

At the same time, it must be noted that men with lower levels of gender-role stress actually prefer to express their emotions through words rather than a structured exercise. Again, there are implications for therapists. Knowing something about the gender-role views of a particular male client may be useful in treatment.

Another implication of these findings is related to the concept of stress for men. The present study was described to the research participants as a study about responses to stress; about 80% of the men who were invited actually participated. Participants often volunteered comments to the experimenters about how pleased they were to be able to participate in a study about how men respond to stress. Seeking to reduce stress is one of the more common presenting complaints of men coming to therapy. "Feeling stressed" covers a wide array of separate physical and emotional problems. When men do not know how to identify or recognize underlying issues, most symptoms can be described as stress. Stress generally arises for men when they do not believe problem-solving options exist or when they are uncertain about how to maximize healthy and productive responses that might be available. One approach to men, then, may be to focus on physical and emotional self-awareness and how these are indications of stress. When placed in this more socially acceptable category of stress, men may be more willing to consider behavior and attitude choices designed to reduce their stress.

This study underlines the need to explore the importance of gender-role issues in thinking about the emotional components of specific presenting issues brought by men to counseling--men with relationship concerns, men as parents, men with sexual concerns, male friendship development, male interactions at work, and so forth. Given that men who experience more stress related to traditional masculinity may respond more readily to treatment that includes exercises or tasks, therapists can approach these issues with a wider array of treatment options. For these men, psychotherapy that emphasizes behavioral changes, activities, and exercises may be more appealing.

Two potential limitations in this study must be noted. First, we selected working men from a narrow set of male-dominated occupations (e.g., fire fighters, police officers, and safety officials). To what degree these findings generalize to men with other occupational interests or personality styles is unknown. Further, because we were primarily interested in within-group differences among men, no women were included in the sample. Comparative observations are therefore restricted. A second limitation is that the techniques used to induce stress (video, math test, and cold pressor) are not in themselves tasks that men perform daily. Although they were effective in meeting internal validity demands, they may not generalize to other forms of stress which men encounter.

This study also raises questions for further reflection and study. Can the use of biofeedback techniques facilitate successful self-recognition of unexpressed emotionality in men? What motivates men to desire more effective verbal skills? The use of physiological measures certainly can assist in uncovering more information about the emotional content of various presenting issues. It may be that physiological arousal patterns differ with regard to presenting issues. One way to test this would be to invite male research participants to report their emotional responses (both verbally and in written form) regarding a variety of presenting issues while measuring their physiological arousal.

These and other questions remain. Nonetheless, for therapists working with the emotional content of men in therapy, this study does suggest that virtually all men experience physiological arousal as a response to emotional stimuli (there are no unemotional men), and that given the right format, men are willing to share those feelings.

The authors wish to acknowledge the contributions of Arthur J. Rathbun, Cheryl Harper, Ann Johnson, and Linda L. Gaffney for their technical and editorial assistance; and of Wendy Grove, Zaneta Barte, and Cheryl Harper in collecting the data.

REFERENCES

Allport, G. E., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study. Psychological Monographs, 47 (1, Whole No. 211).

Anderson, C. D. (1989). Expression of affect and physiological response in psychosomatic patients. Journal of Psychosomatic Research, 25(3), 143-149.

Arena, J. G. (1984). Inter- and intrasession reliability of psychophysiological post-stress adaptation periods. Journal of Behavioral Assessment, 6(3), 247-260.

Averill, J. R. (1975). A semantic atlas of emotional concepts. Abstracted in Catalog of Selected Documents in Psychology, 5, 330. American Psychological Association and Johnson Associates: Greenwich, CT.

Beck, A. A. (1972). Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press.

Bernstein, J. S. (1995). "Real men" don't talk: Attempting psychodynamic group therapy with inner-city boys. Journal of Child and Adolescent Group Therapy, 5(2), 83-106.

Bio-medical Instruments, Inc. (1994). Focused Technology F1000 Instrumentation System. Warren, MI: Focused Technology.

Breland, H. M. (1996). Word frequency and word difficulty: A comparison of counts on four corpora. Psychological Science, 7(2), 96-99.

Brooks, G. R., & Gilbert, L. A. (1995). Men in families: Old constraints, new possibilities. In R. F. Levant & W. S. Pollack (Eds.), A new psychology of men (pp. 252-279). New York: Basic Books.

Burrowes, B. D., & Halberstadt, A. G. (1987). Self- and family-expressiveness styles in the experience and expression of anger. Journal of Nonverbal Behavior, 11, 254-268.

Canary, D. J., & Emmers-Sommers, T. M. (1997). Sex and gender differences in personal relationships. New York: Guilford Press.

Carlson, C. R., Collins, F. L., Stewart, J. F., Porzelius, J., Nitz, J. A., & Lind, C. O. (1989). The assessment of emotional reactivity: A scale development and validation study. Journal of Psychopathology and Behavioral Assessment, 11(4), 313-324.

Cherulnik, P. D. (1979). Sex differences in the expression of emotion in a structured social encounter. Sex Roles, 5, 413-424.

Conner, K., & Heesacker, M. (1999). Sex, gender, and emotions: Accuracy of the dyadic communication of emotions. Manuscript submitted for publication.

Crano, W. D., & Aronoff, J. (1978). A cross-cultural study of expressive and instrumental role complementarity in the family. American Sociological Review, 43, 463-471.

Cleary, P. D. (1987). Gender differences in stress related disorders. In R. C. Barnett, L. Biener, & G. K. Baruch (Eds.), Gender and stress (pp. 39-72). New York: Free Press.

Courtenay, W. H. (2000). Engendering health: A social constructionist examination of men's health beliefs and behaviors. Psychology of Men & Masculinity, 1(1), 4-15.

David, D. S., & Brannon, R. (1976). The forty-nine percent majority: The male sex role. New York: Random House.

Davidson, R. J. (1994). Complexities in the search for emotion-specific physiology. In P. Eckman & R. J. Davidson (Eds.), The nature of emotion (pp. 237-242). New York: Oxford University Press.

Donat, D. C., & McCullough, J. P. (1983). Psychophysiological discriminants of depression at rest and in response to stress. Journal of Clinical Psychology, 39(3), 315-320.

Dwyer, J. W., & Seecomb, K. (1991). Elder care as family labor. Journal of Family Issues, 12, 229-247.

Eagly, A. H., & Steffen, V. J. (1986). Gender stereotypes, occupational roles, and beliefs about part-time employees. Psychology of Women Quarterly, 10(3), 252-262,

Eckman, P. (1972). Facial expressions of emotion. In J. K. Cole (Ed.), Nebraska Symposium on Motivation. Lincoln: University of Nebraska Press.

Eckman, P., & Davidson, R. J. (1994). The nature of emotion. New York: Oxford University Press.

Eisler, R. M., & Blalock, J. A. (1991). Masculine gender role stress: Implications for the assessment of men. Clinical Psychology Review, 11, 45-60.

Eisler, R. M., & Skidmore, J. R. (1987). MGRS: Scale development and component factors in the appraisal of stressful situations. Behavior Modification, 11, 123136.

Eisler, R. M., Skidmore, J. R., & Ward, C. H. (1988). Masculine gender-role stress: Predictor of anger, anxiety, and health-risk behaviors. Journal of Personality Assessment, 52(1), 133-141.

Fitzgibbons, L., & Simons, R. F. (1992), Affective response to color-slide stimuli in subjects with physical anhedonia: A three-systems analysis. Psychophysiology, 29(6), 613-620.

Friedman, R. M., & Lerner, L. (1986). Toward a new psychology of men: Psychoanalytic and social perspectives. New York: Guilford.

Frijda, N. H. (1994). Emotions are functional, most of the time. In P. Eckman & R. J. Davidson (Eds.), The nature of emotion (pp. 112-122). New York: Oxford University Press.

Gerrards-Hesse, A., Spies, K., and Hesse, F. W. (1994). Experimental inductions of emotional states and their effectiveness: A review. British Journal of Psychology, 85, 55-78.

Goldberg, H. (1976). The hazards of being male: Surviving the myth of masculine privilege. New York: Signet.

Gordon, G., & Meth, R. L. (1990). Men as husbands. In R. L. Meth and R. S. Pasick (Eds.), Men in therapy (pp. 54-77). New York: Guilford.

Gottman, J. M., & Levenson, R. W. (1988). The social psychophysiology of marriage. In P. Noller and M. A. Fitzpatrick (Eds), Perspectives in marital interaction (pp. 182-199). Philadelphia: Multilingual Matters Ltd.

Gray, J. A. (1994). Three fundamental emotion systems. In P. Eckman & R. J. Davidson (Eds.), The nature of emotion (pp. 243-246). New York: Oxford University Press.

Greenwald, M., Cook, E., & Lang P. J. (1989). Affective judgment and psychophysiological response: Dimensional covariation on the evaluation of pictorial stimuli. Journal of Psychophysiology, 3, 51-64.

Grossman, M., & Wood, W. (1993). Sex differences in intensity of emotional experience: A social role interpretation. Journal of Personality and Social Psychology, 65(5), 1010-1022.

Heesacker, M., Wester, S. R., Vogel, D. L., Wentzel, J. F., Mejia-Millan, C. M., & Goodholm, Jr., C. R. (1999). Gender-based emotional stereotyping. Journal of Counseling Psychology, 46(4), 483-495.

Hess, U., Kappas, A., McHugo, G. J., Lanzetta, J. T., & Kleck, R. E. (1992). The facilitative effect of facial expression on the self-generation of emotion. International Journal of Psychophysiology, 12, 251-265.

Hurst, M. A. (1997). The best fit in counseling men: Are there solutions to treating men as the problem? Unpublished doctoral dissertation, Ball State University, Muncie, Indiana.

Isen, A. M., & Gorgolione, J. M. (1983). Some specific effects of four affect-induction procedures. Personality and Social Psychology, 45, 20-31.

Izard, C. E. (1977). Human emotions. New York: Plenum.

Izard, C. E. (1991). The psychology of emotions. New York: Plenum.

Kelly, K. & Hall A. (1992). Mental health counseling for men [Special issue]. Journal of Mental Health Counseling, 14(3).

LaFrance, M., & Benaji, M. (1992). Toward a reconsideration of the gender-emotion relationship. In M. S. Clark (Ed.), Review of personality and social psychology (Vol. 14, pp. 178-201). Newbury Park, CA: Sage.

Lang, P. J. (1984). Cognition in emotion: Concept and action. In C. Izard, J. Kagan, & R. Zajonc (Eds.), Emotion, cognition and behavior (pp. 193-226). New York: Cambridge University Press.

Lang, P. G. (1985). The cognitive psychophysiology of emotion: Fear and anxiety. In A. H. Tuma & J. D. Maser (Eds 3, Anxiety and the anxiety disorders (pp. 131-170). Hillsdale, NJ: Erlbaum.

Lang, P. J., Greenwald, M. K., Bradley, M. M., & Hamm, A. O. (1993). Looking at pictures: Affective, facial, visceral, and behavioral reactions. Psychophysiology, 30, 261-273.

Levant, R. F. (1992). A gender-aware divorce prevention program. Men's Studies Review, 9(2), 15-20.

Levenson, R. W. (1994). Human emotions: A functional view. In P. Eckman & R. J. Davidson (Eds.), The nature of emotion (pp. 123-126). New York: Oxford University Press.

Marston, A., Hart, J., Hileman, C., & Faunce, W. (1984). Toward the laboratory study of sadness and crying. American Journal of Psychology, 97, 127-131.

Martin, M. (1990). On the induction of mood. Clinical Psychology Review, 10, 669697.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1981). The profile of mood states. San Diego: Educational and Industrial Testing Service.

Meinecke, C. E. (1981). Socialized to die younger? Hypermasculinity and men's health. The Personnel and Guidance Journal, December, 241-245.

Miller, L. (1995). Tough guys: Psychotherapeutic strategies with law enforcement and emergency services personnel. Psychotherapy, 32(4), 592-600.

Netter, G. (Producer), & Rubin, B. J. (Director). (1993). My life. [Film]. (Available from Columbia Tristar, 3400 Riverside Drive, Burbank, California, 91505-4627.

Nolen-Hoeksema, S. (1987). Sex differences in unipolar depression: Evidence and theory. Psychological Bulletin, 101, 259-282.

Nunnally, J. (1981). The affect inventory. Unpublished manuscript. Nashville, TN: Vanderbilt University.

O'Neil, J. M. (1981). Patterns of gender role conflict and strain: Sexism and fear of femininity in men's lives. The Personnel and Guidance Journal, 60, 203-210.

O'Neil, J. M., Helms, B. J., Gable, R. K., David, L., & Wrightsman, L. S. (1986). Gender-role conflict scale: College men's fear of femininity. Sex Roles, 14, 335-350.

Parsons, T., & Bales, R. F. (1955). Family, socialization and interaction process. New York: Free Press.

Pasick, R. S., Gordon, S., & Meth, R. L. (1990). Helping men understand themselves. In R. L. Meth and R. S. Pasick, Men in therapy (pp. 152-180). New York: Guilford.

Pleck, J. H. (1981). The myth of masculinity. Cambridge, MA: MIT Press.

Pleck, J. H., Sonenstein, F. L., & Ku, L. (2. (1993). Masculinity ideology and its correlates. In S. Oskamp and M. Costanzo (Eds.), Gender and social psychology (pp. 85-110). Newbury Park, CA: Sage.

Reeves, J. L., & Shapiro, D. (1983). Heart-rate reactivity to cold pressor stress following biofeedback training. Biofeedback and Self-regulation, 8(1), 87-99.

Robertson, J. M. (1989). Men who avoid counseling: Identifying personality correlates and preferences for assistance alternatives. Unpublished doctoral dissertation, University of California, Santa Barbara.

Robertson, J. M., & Freeman, R. (1995). Men and emotions: Developing masculine-congruent views of affective expressiveness. Journal of College Student Development, 36(6), 606-607.

Saurer, M. K., & Eisler, R. M. (1990). The role of masculine gender role stress in expressivity and social support network factors. Sex Roles, 23(5-6), 261-171.

Scher, M. (1993). Men and boys, myths and mythic proportions: A rejoinder to Kelly and Hall. Journal of Mental Health Counseling, 15, 290-297.

Scherer, K. R., Wallbott, H. G., & Summerfield, A. B. (1986). Experiencing emotion: A cross-cultural study. Cambridge, England: Cambridge University Press.

Shay, J. J. (1996). "Okay, I'm here, but I'm not talking:" Psychotherapy with the reluctant male. Psychotherapy 33(3), 503-513.

Spielberger, C. D., Gorsuch, R. L., & Lushene, R. (1970). The State-Trait Anxiety Inventory Manual. Palo Alto: Consulting Psychologists Press.

Subotnik K. L., Shapiro, D. (1984). Heart rate biofeedback and cold pressor pain: Effects of expectancy. Biofeedback and Self-regulation, 9(1), 55-75.

Thorndike, E. L. (1921). The teacher's word book. New York: Teachers College, Columbia University.

Wanigaratne, S., & Barker, C. (1995). Clients' preferences for styles of therapy. British Journal of Clinical Psychology, 34(2), 215-222.

Watkins, P. L., Eisler, R. M., Carpenter, L., Schechtman, K. B., & Fisher, E. B. (1991). Psychosocial and physiological correlates of male gender role stress among employed adults. Behavioral Medicine, 17(2), 86-90.

Wilcox, D. W., & Forrest, L. (1992). The problems of men and counseling: Gender bias or gender truth? Journal of Mental Health Counseling, 14(3), 291-304.

Wood, W., Rhodes, N., & Whelan, M. (1989). Sex differences in positive well-being: A consideration of emotional style and marital status. Psychological Bulletin, 106, 249-264.

Correspondence concerning this article should be sent to John M. Robertson, University Counseling Services, Lafene 238, Kansas State University, Manhattan, Kansas 66502 or johnrobe@ksu.edu.

John M. Robertson, Ph.D., is a psychologist at University Counseling Services and a member of the graduate faculty for the Department of Counseling and Educational Psychology at Kansas State University. He also provides clinical services for Southwind Counseling Services in Manhattan, Kansas. His professional interests focus on issues related to the psychology of men, including relationships, emotionality, abuse, gender-role transitions, and help-seeking behavior. (johnrobe@ksu.edu)

Joyce Woodford, M.S., is a senior clinician at University Counseling Services, Kansas State University. Her therapy and research interests are help-seeking, gender differences, and issues of peace. Her work has included foci on women's issues, childhood abuse survival, and relationships. She also has an interest in the contemporary subject of social interest through Adlerian Individual Psychology. She is completing a doctorate in clinical psychology at the Fielding Institute. (joycew@ksu.edu)

Kimberly Kay Danos received her Bachelor of Arts and Science with majors in biology and Asian studies from Regis University in 1993. She continued her education at Kansas State University completing her Master of Science in counseling and personnel services in 1997. Danos enhanced her graduate education with special emphasis on counseling, biofeedback, and stress management. She was an active member of the Kansas State University counseling services staff and research team from 1995 to 1999. In 1999, Danos accepted a position with the Human Services Planning Alliance in Des Moines, Iowa. She now focuses her research and work on early childhood issues. Her research continues to encompass gender studies, applied methods of biofeedback, effects of stress on brain development, the psychology of poverty, and violent deviant behavior. (kkdanos@aoi.com)

Mark A. Hurst is a licensed psychologist conducting private practice and consultation in Olympia, Washington. Raised in a family of boatbuilders, tugboaters, and school teachers in the rural Pacific Northwest, he earned a master's degree in clinical psychology from Eastern Washington University and was awarded the doctorate by Ball State University in Muncie, Indiana. Dr. Hurst serves as adjunct faculty at the Evergreen State College, where he sponsors student-directed internships and creates social psychology and men's studies courses. His other projects address human development and transition, media literacy, and film production about men's inner lives. (mkhurst@home.com)

Chi-Wei Lin, Ph.D. Kansas State University 1999, is director of the student counseling center and assistant professor in the College of Humanities and Education at Chung Yuan Christian University. Lin's clinical and research interests include personality disorders, anxiety disorders, at-risk youth, and voluntary services. (chiwei@cycu.edu.tw)