حساسیت در روابط متقابل و واکنش پذیری به درگیری همسر در زنان سالمند سالم

Abstract This study investigated whether interpersonal sensitivity was associated with reactivity to spousal conflict beyond neuroticism. Interpersonal sensitivity is defined as an excessive sensitivity to the behavior and feelings of others. Participants were 49 healthy older women who were interviewed weekly for 12–20 consecutive weeks. Interpersonal sensitivity and neuroticism were measured in mailed questionnaires. Spousal conflict, positive affect, and negative affect were assessed weekly in phone interviews. Hierarchical linear regression analyses were conducted predicting positive and negative affect from interpersonal sensitivity, spousal conflict, and the sensitivity×spousal conflict interaction. Neuroticism and age were used as control variables. The results indicated that women higher in interpersonal sensitivity had more negative affect and less positive affect in times of spousal conflict than women lower in interpersonal sensitivity. The findings suggest that interpersonal sensitivity may play a unique role in reactivity to interpersonal conflict.

Introduction Personality may play an important role in reactivity to stressful events (Bolger & Zuckerman, 1995). Researchers have examined a variety of personality characteristics that may moderate the relationship between stress and illness (Antonovsky, 1987, Kobasa, 1979 and Smith and Anderson, 1986). However, neuroticism is the characteristic that has most consistently been related to reactivity to stress (Bolger and Schilling, 1991, Larsen and Ketelaar, 1991 and Marco and Suls, 1993). Indeed, neuroticism has threatened to dominate the study of personality and health and obscure the importance of more subtle individual differences (Kasl & Rapp, 1991). The purpose of this study was to examine interpersonal sensitivity as a personality characteristic that is related to neuroticism but may add something unique. The definition of interpersonal sensitivity adopted for this article is that given by Boyce and Parker (1989). They developed a measure of interpersonal sensitivity defined as “an undue and excessive awareness of, and sensitivity to, the behaviour and feelings of others” (p. 342). In their view, interpersonal sensitivity characterizes individuals who are preoccupied with interpersonal relationships and are particularly sensitive to interactions involving criticism or rejection. While this definition of interpersonal sensitivity is similar to others that have arisen in social science and medical research, there are some important differences. For example, the interpersonal sensitivity subscale of the SCL-90 is defined in terms of low self-esteem rather than sensitivity to criticism or rejection (Derogatis, Lipman & Covi, 1973). Also, social psychologists have studied interpersonal sensitivity as a characteristic involving the accuracy of interpersonal perception (Snodgras, Hecht & Ploutz-Snyder, 1998). Boyce and Parker (1989) have found that interpersonal sensitivity is strongly related to neuroticism, but believe that it represents a refined risk factor for health problems. There is reason to suspect that interpersonal sensitivity may influence health by increasing reactivity to stressful events, especially those that are interpersonal in nature. First, the definition of interpersonal sensitivity emphasizes sensitivity to criticism and rejection. Heightened sensitivity to negative social interactions involving criticism or rejection could result in increased negative affect and decreased positive affect. Sustained changes in affect due to ongoing conflict could have a significant impact on both psychological and physical health. Second, Boyce, Parker, Barnett, Cooney and Smith (1991) found that interpersonal sensitivity predicted an increase in depression following a stressful event with a significant interpersonal dimension. They assessed interpersonal sensitivity in 140 mothers prior to giving birth and assessed depression at 3 and 6 months following birth. They found that women high in interpersonal sensitivity were much more likely to become depressed than those low in sensitivity. While this study provides some evidence that interpersonal sensitivity may play a role in reactivity to stress, there are several studies suggesting that neuroticism may increase reactivity. Larsen and Ketelaar (1991) found that 359 undergraduates high in neuroticism were more emotionally reactive to a negative mood induction than those low in neuroticism. Marco and Suls (1993) examined the relationship between daily stressors and mood in 40 male community residents. They found that individuals high in trait negative affectivity were more reactive to stressors than those low in trait negative affectivity. Finally, Bolger and his colleagues found that neuroticism was related to increased reactivity to stressful events in both a community sample and in a sample of undergraduates (Bolger and Schilling, 1991 and Bolger and Zuckerman, 1995). The central question of this study was whether interpersonal sensitivity would play a role in reactivity to interpersonal stressors beyond neuroticism. Interpersonal stressors are defined as negative social interactions including conflict, criticism, arguing, and the withdrawal of affection. Because neuroticism reflects a general tendency to experience negative affective states (Watson & Clark, 1984), we reasoned that the role it plays in reactivity might be similar across all domains of stressful events. Because interpersonal sensitivity is specifically relevant to the domain of personal and social relationships (Boyce & Parker, 1989), we reasoned its role in reactivity might be stronger than neuroticism with regard to interpersonal stress. Therefore, we predicted that interpersonal sensitivity would be related to reactivity to interpersonal stress beyond neuroticism. Specifically, we predicted that individuals high in interpersonal sensitivity would demonstrate increased reactivity to interpersonal stress after controlling for neuroticism. This hypothesis was tested by determining whether interpersonal sensitivity moderated the relationship between spousal conflict and affect in 49 healthy older women. These women were interviewed over a period of several weeks in order to examine week-to-week fluctuations in spousal conflict and affect. We chose these short-term fluctuations as our focus because we were interested in understanding the process by which individuals move from health to illness. Our specific hypotheses were that: (1) women high in interpersonal sensitivity would have more negative affect in times of spousal conflict than women low in interpersonal sensitivity and; (2) women high in interpersonal sensitivity would have less positive affect in times of spousal conflict than women low in interpersonal sensitivity.

Results Table 1 contains the descriptive statistics and correlations among the study variables. The mean levels of spousal conflict, negative affect, and positive affect were used in the correlation analyses. Interpersonal sensitivity was positively related to neuroticism and negative affect and negatively related to positive affect. Hierarchical linear modeling (HLM) was used to test the hypotheses because the data has a hierarchical structure with 12–20 weekly observations nested within each of 49 participants. While the hierarchical linear equations are presented below, a full explanation of the hierarchical models used in this study can be found in Bryk and Raudenbush (1992) and Singer (1998). Table 1. Descriptive statistics and correlations among study variables Variable 1 2 3 4 5 Interpersonal sensitivity – Neuroticism 0.570** – Spousal conflict 0.085 0.253 – Negative affect 0.367** 0.444** 0.291* – Positive affect −0.335* −0.666** −0.093 −0.196 – Mean 3.27 1.83 1.39 3.39 1.46 Standard deviation 0.96 2.61 0.35 0.67 0.37 * P<0.05. * *P<0.01. Table options HLM equations were developed in which interpersonal sensitivity, spousal conflict, and the sensitivity×conflict interaction were the predictor variables and the affect measures were the criterion variables. The control variables were neuroticism, the neuroticism×spousal conflict interaction, and age. Age was included because it was the only demographic characteristic related to positive or negative affect. An autoregressive covariance matrix was specified in order to examine week-to-week fluctuations in affect. This is similar to controlling for the dependent variable during the previous week but has the advantage of taking into account variance on all previous scores. The result is that the dependent variable represents change relative to all previous scores. The HLM equations that were specified can be divided into two levels. On the first level, we estimated an equation for each participant of the relationship between spousal conflict and affect. This level 1 equation can be written as follows: Full-size image (<1 K) where the level of affect is predicted by an intercept (β0), spousal conflict (β1), and random error (r). On the second level, we estimated equations to predict differences between persons in intercepts. This level 2 equation can be written as follows: Full-size image (<1 K) where each person's level one intercept (β0p) is predicted by an intercept (γ00), interpersonal sensitivity (γ01), neuroticism (γ02), age (γ03), and random error (u0). Similarly, we estimated equations to predict differences between persons in level one intercepts. This level 2 equation can be written as follows: Full-size image (<1 K) where each person's level one slope (β1) is predicted by an intercept (γ10), interpersonal sensitivity (γ11), neuroticism (γ12), and random error (u1). Separate analyses were conducted with negative affect and positive affect as the dependent variable. In the first analysis, change in negative affect was predicted from spousal conflict, interpersonal sensitivity, neuroticism, age, the sensitivity×conflict interaction, and the neuroticism×conflict interaction. The results are displayed in Table 2. The interpersonal sensitivity×conflict interaction was a significant predictor of increases in negative affect supporting the first hypothesis. The interaction was positive indicating that women high in interpersonal sensitivity had more negative affect in times of spousal conflict than women low in interpersonal sensitivity. Interpersonal sensitivity was not directly related to change in negative affect. Table 2. Hierarchical linear regression predicting negative affect from interpersonal sensitivity, neuroticism, spousal conflict, sensitivity×conflict, neuroticism×conflict, and age Predictor variables Estimatea S.E. Estb d.f. t Spousal conflict 0.038 0.006 614 6.34** Interpersonal sensitivity 0.036 0.051 45 0.71 Neuroticism 0.346 0.131 45 2.64* Age −0.018 0.006 45 −2.83** Sensitivity×conflict 0.015 0.007 614 2.12* Neuroticism×conflict 0.004 0.013 614 0.30 a Estimate, units of change in negative affect for every one unit of change in each predictor variable. b S.E. Est, standard error of the estimate. * P<0.05. * *P<0.01. Table options In the second analysis, change in positive affect was predicted from spousal conflict, interpersonal sensitivity, neuroticism, age, the sensitivity×conflict interaction, and the neuroticism×conflict interaction. The results are displayed in Table 3. The sensitivity×conflict interaction was a significant predictor of decreases in positive affect supporting the second hypothesis. The interaction was negative indicating that women high in interpersonal sensitivity had less positive affect in times of spousal conflict than women low in interpersonal sensitivity. Interpersonal sensitivity was not directly related to changes in positive affect. Table 3. Hierarchical linear regression predicting positive affect from interpersonal sensitivity, neuroticism, spousal conflict, sensitivity×conflict, neuroticism×conflict, and age Predictor variables Estimatea S.E. Estb d.f. t Spousal conflict −0.006 0.008 614 −0.81 Interpersonal sensitivity 0.031 0.093 45 0.33 Neuroticism −1.240 0.236 45 −5.24 and Age −0.008 0.012 45 −0.72 Sensitivity×conflict −0.023 0.009 614 −2.53** Neuroticism×conflict 0.008 0.012 614 0.42 a Estimate, units of change in negative affect for every one unit of change in each predictor variable. b SE Est, standard error of the estimate. * P<0.05. * *P<0.01.