INTRODUCTION Communication especially the ability to communicate emotions

INTRODUCTION Communication, especially the ability to communicate emotions has been associated with individuals’ ability to cope with stress. Talking with others is a normal and healthy way to cope with upheaval and there is evidence that withholding desires to talk about emotions may have a detrimental effect on health (e. g. , Pennebaker, 1989). There is also evidence that people with good communication skills develop stronger and more effective social support networks (Sarason et al. , 1985). Self-disclosure is an important factor in enduring friendships, and reciprocal self-disclosure extends and deepens the relationship. However, not everyone establishes mutually rewarding personal relationships, and to a degree, this reflects poor communication (Hargie & Tourish, 1997). Extensive research has been carried out to assess what variables may moderate cardiovascular reactivity (CVR) to psychological stress. The majority of this research has been conducted in laboratories, which allow for strict control of the situation and standardization of conditions across participants and occasions. Most stressors, including mental arithmetic tasks, RT tasks, video games and speech tasks have been standardized in order to remove individual differences to increase the validity of the research. While CVR research has investigated under which circumstances speech tasks elicit stress, this area has not been standardized in order to minimize individual differences in verbal ability (Hughes, 2001). Hughes (2001) conducted a study to investigate the possible stress buffering effect of communication ability (CA) on CVR under two stress conditions, arithmetic and a short term memory task. Participants then filled out a standardized questionnaire that measured CA. The results showed that those high in effective CA exhibited reduced levels of pulse reactivity to the more stressful maths task. This suggests that CA may play some role in the relationship between the experience of a stressor and the CV response to stress, even when the stressor is a non-speech based one. PURPOSE The objective of this study was to investigate whether the stress-buffering effect of high CA found by Hughes (2001) for an arithmetic task would translate to a speech-based stressor. Specifically, the study was conducted to assess whether levels of CA would affect cardiovascular Communication ability as a moderator of cardiovascular reactivity to a speech task Susie Kola & Jane C. Walsh Department of Psychology, National University of Ireland, Galway responses to a task directly related to the act of communicating, in other words, a speech task. METHOD Design The study employed a 2 (high and low CA) x 3 (baseline, task, and recovery) mixed design. The unrelated measures IV was communication ability. Two measures of CA were considered; effective and dominant CA. The DVs measured were heart rate, systolic and diastolic blood pressure. Participants Fifty-six undergraduate female psychology students, with a mean age of 19. 45 years (SD=4. 97) participated. Exclusion criteria included the use of oral contraceptives, medication affecting cardiovascular functioning, and history of hypertension. Materials Mc. Manus, Kidd and Aldous’s (1997) re-standardized version of the Norton Communicator Style Measure (Norton, 1978) was used to assess high and low CA. It contains 18 items and measures CA on three scales: effective communication, dominant communication and non-verbal communication. Items are assessed on a four-point scale, from ‘Describes me very badly’ to ‘Describes me very well’. Mc. Manus et al. report that each measure is highly reliable (Effective α =. 79, Dominant α =. 76, Non-verbal α =. 74). The speech task employed was based on the evaluative speaking task (Saab et al. , 1989). Participants were asked to prepare and deliver a speech about a hypothetical situation. The participants were told that her speech would be tape recorded and later evaluated for style, content and articulation. The tape recorder was present to increase the demands and the perception of stress, by highlighting performance aspects of the task. Procedure Pre-screening took place among 300 undergraduate psychology students by administration of Mc. Manus et al. ’s (1997) version of the Norton Communicator Style Measure. A random selection of those that scored in the 33 rd and 66 th percentiles were contacted and invited to participate, thus creating two categories: high and low effective and dominant CA. For the main study, participants were tested individually in a cubicle in a psychometrics laboratory. Participants were asked to avoid smoking, drinking caffeinated beverages, or taking exercise an hour before the study. Each participant relaxed in a chair for an initial tenminute resting period. They were instructed to relax in silence without much movement, and were provided with magazines to read. Measures of heart rate (HR) and blood pressure (BP) were taken at end of minutes 3, 6, and 9. This was followed by a 5 -minute pre-task period with CV readings taken at the end of minutes 1, 2. 5, and 4. The participants were then given a 3 minute preparation period to arrange their responses for the speech task, and were provided with pen and paper. During the 5 -minute presentation period, CV measures were recorded at the end of minutes 1, 2. 5, and 4. Following presentation, the participant again relaxed during a 5 -minute recovery period, where CV measures were obtained at the same intervals. Based on the three sets of readings for each period, mean HR, mean SBP and mean DBP were calculated for pre-task, experimental task and recovery periods. RESULTS A series of 2 x 3 mixed ANOVAs were carried out to assess whethere would be any mean differences in pulse or BP reactivity for high and low effective and dominant communication ability. The results for Effective CA (ECA) and HR revealed a significant main effect for time F(1. 40, 71. 51) = 25. 08, p =. 000. There was no significant interaction between time and ECA, F(1. 40, 71. 51) =. 24, p =. 708. No significant differences were found between high and low ECA, F(1, 51) =. 19, p =. 170. For ECA and SBP, a significant main effect for time was found, F(1. 45, 73. 70) = 100. 08, p =. 000. No significant interaction between time x ECA was found, F(1. 45, 73. 70) =. 06, p =. 892. Similarly, there were no significant differences between high and low ECA for SBP, F(1, 51) =. 004, p =. 951. For additional information contact Susie Kola: S. Kola 1@nuigalway. ie Results cont’d For ECA and DBP, a significant main effect for time was found, F(1. 36, 69. 17) = 126. 34, p =. 000. No significant interaction was found between time x ECA, F(1. 36, 69. 17) =. 17, p =. 759. There were no significant differences found between high and low ECA for DBP, F(1, 51) =. 99, p =. 323. All analyses for Dominant CA (DCA) revealed a significant main effect for time for all the analyses (p =. 000 for each for HR, SBP and DBP). There were no interaction effects found for DCA for any of the dependent variables, nor were there any significant differences found between high and low DCA for any of the dependent variables. DISCUSSION The study was conducted to assess whether CA would have a stress-buffering effect on CVR to a speech task. The stressor was successful in eliciting a stress response, as evidenced by the significant main effect for time in all analyses. No significant differences were found between levels of CA for CVR or cardiovascular recovery. CA did not act as a buffer for CVR in the present study, which may indicate that the expected stress buffering of CA was moderated by some other factor or factors. The degree of reactivity during speaking is determined by a wide range of factors, e. g. , the context, the presence of other people, the speaker’s emotional state, personality characteristics, and presence of social support (Lyons et al. , 2000). In the current sample, variables such as extraversion/introversion, trait anxiety, communication apprehension or evaluation apprehension may have influenced the degree of reactivity. It is also possible that in the present sample, high communicators displayed exaggerated CV responses due to a greater task engagement, thus minimizing differences between high and low communicators. If good communicators participate in more conversations, and have a higher degree of comfort when communicating, they may engage more in the task than individuals who are not comfortable in this type of situation. Implications The present study has shown that the relationship between CA and CVR is more complex than a straightforward linear relationship. Future research is needed before the role of CA on CVR has been clearly defined. Finding out under which circumstances CA moderates stress responses would be useful in the employment of speech-based stressors in CV research, to reduce the likelihood of false inferences being drawn.