امکان همزیستی مشترک خوش بینی و بدبینی بالا وجود دارد؟ یافته های حاصل از بیماران مبتلا به آرتریت مقابله با درد

Abstract Data from 120 elderly osteoarthritis patients showed that dispositional optimism and pessimism interacted in their associations with the frequency of use of pain-coping strategies: participants higher on both dispositions reported more intense coping. Controlling for age, pain level, history of prior serious disease, or negative and positive affectivity did not affect the results. Possible explanations for the co-existence of high optimism and pessimism under certain conditions are proposed.

. Introduction The construct of dispositional optimism, or generalized outcome expectancies, was defined by Scheier and Carver (1985) on the basis of their broader theory of self-regulation (Carver & Scheier, 1981). Extensive research has provided evidence of the benefits of being optimistic (Andersson, 1996), yet there are two open debates that arose from inconsistencies in the findings: the dimensionality of optimism and pessimism and the independence of optimism from affect. The findings to date raise the possibility that there is no single resolution to these debates, but rather that both issues depend on the context and population in which optimism is studied. Regarding the dimensionality of optimism, the relatively low correlation between optimism and pessimism in older samples compared to younger ones suggests that the distinction between them may be more pronounced as one ages. Regarding its independence from affect, most of the studies that provided evidence for it were based on adults coping with health stressors. In contrast, many of the studies in which negative affectivity accounted for the effects of optimism were based on student or community samples. These findings suggest a unique role for optimism when active coping is required. The main objective of the current study is to test the proposition that optimism and pessimism are distinct and can even co-occur among older people. A secondary objective is to replicate the finding that among people coping with health problems, the effects of optimism cannot be explained by negative (or positive) affect. Scheier and Carver (1985) developed the Life Orientation Test (LOT) and later a revised version (LOT-R, Scheier, Carver, & Bridges, 1994). Both measures include positively- and negatively-phrased items aimed at assessing a single bipolar dimension of optimism. Exploratory and confirmatory factor analyses (CFA) showed that two-dimensional models fit the LOT items (e.g., Chang, D’Zurilla, & Maydeu-Olivares, 1994). However, when CFA was used allowing for correlated errors among similarly valenced items, these tests yielded good fit for both one-factor and two-factor solutions (e.g., Scheier et al., 1994). Additional studies supported the two-factor model with evidence of differences in the associations of the LOT optimism and pessimism subscales with personality traits, coping strategies, mental and physical health (e.g., Marshall, Wortman, Kusulas, Hervig, & Vickers, 1992; see also book edited by Chang, 2001). Further support for a two-dimensional model involves the correlation between optimism and pessimism (Dember, Martin, Hummer, Howe, & Melton, 1989): If they were indeed two ends of one dimension, their correlation should have been high, in the order of the internal reliabilities reported for the full LOT, i.e., in the 70s (Scheier & Carver, 1985; Scheier et al., 1994). While many studies reported correlations in the 50s (e.g., Chang & Sanna, 2001), studies of older adults typically reported lower correlations, under 30 (e.g., Mahler & Kulik, 2000; Mroczek, Spiro, Aldwin, Ozer, & Bosse, 1993; Plomin et al., 1992). A low correlation could be an artifact, resulting from people who are in the middle range of both optimism and pessimism and therefore lower the correlation between these subscales, or from a response set such as an acquiescence bias. However, another explanation could be that at old age, some people are indeed relatively high on both optimism and pessimism. As people age, it seems reasonable that some would gradually accept the possibility of adverse events occurring, regardless of their basic level of optimism. When one considers the content of the LOT-R optimism and pessimism items, which do not fully mirror one another, this interpretation seems possible. Several researchers have proposed approaches that could be related to the possible co-occurrence of high optimism and pessimism. Norem and Cantor (1986) proposed that some people cope with anxiety related to future goals by adopting the strategy of defensive pessimism: they report low expectations for an upcoming performance even though they have had earlier successes. Defensive pessimism is defined and measured as domain-specific and quite malleable and is therefore different from Scheier and Carver’s (1985) concept of generalized expectancies. However, it is possible that some people use defensive pessimism across domains and time and respond to the LOT with both high (defensive) pessimism and high optimism (stemming from prior successes). This possibility was tested in a sample of students and was not supported by the data ( Hummer, Dember, Melton, & Schefft, 1992). The researchers concluded that people who are high on both traits “may have assimilated to a certain degree their pessimistic assertions into their internal experience, with the result being that they differ from both optimists and genuine pessimists” (p. 48). Second, Wallston (1994) proposed that there are cautious optimists, who show optimistic beliefs and corresponding behaviors, yet acknowledge the realistic possibility that some bad things could happen, or that some good things will not turn out as well as expected. Indeed, studies that reported the distribution of the LOT scores, showed that very few people are “real” pessimists, i.e., agree that things are not going to work out for them. Most of the people who reported low optimism or high pessimism were close to the middle of the response scale (e.g., Mroczek et al., 1993) and therefore can more accurately be labeled “realists”—people who accept that bad things would happen. The main objective of the current study was to explore the possibility that there is a unique subset of people who are high on both optimism and pessimism, among a sample of older adults coping with osteoarthritis (OA), a painful but not life-threatening joint disease. If dispositional optimism is bipolar and relatively high scores on both of its subsets represent an artifact, then in tests of their associations with OA pain-coping strategies, there should be additive effects of optimism and/or pessimism. If relatively high scores on both subsets characterize a unique group of people, then there may even be significant interaction effects of optimism and pessimism on coping. The second objective was to determine whether the effects of optimism, pessimism, or their interaction on pain-coping strategies are eliminated when negative and positive trait affects are controlled for. The rationale for this test is the debate about the independence of dispositional optimism from negative affectivity, first raised by Smith, Pope, Rhodewalt, and Poulton (1989). Scheier et al. (1994) have replied to this criticism with data showing independent effects of optimism in its relations with some of the coping strategies tested, but not with physical symptoms. Other studies that examined the relationship of optimism and affectivity to coping provided mixed evidence: some supported the independence of optimism from affectivity (Aspinwall & Taylor, 1992), while others found that when negative affectivity was controlled, the effect of optimism was eliminated (Boland & Cappeliez, 1997)

. Results The correlation between the LOT optimism and pessimism subscales was r = −0.49 (p < 0.001), which, similar to previous reports, is much lower than their internal reliability coefficients. Mean scores on the 1–5 scale were 3.64 (±0.74) for the optimism subscale and 2.21 (±0.64) for the pessimism subscale. Only 14% of the sample had a mean optimism score lower than 3 (the mid-point of the 1–5 response scale) and only 10% had a mean pessimism score higher than 3. For descriptive purposes, the sample was first divided into two groups on each disposition, using median splits, and these divisions were combined to create four groups (2 × 2): 28% were undifferentiated, reporting low optimism and low pessimism (LL), 28% were “true” optimists (TO), reporting high optimism and low pessimism, 29% were “true” pessimists (TP), reporting high pessimism and low optimism, and 17% reported high levels of both optimism and pessimism (HH). The four groups were compared on age, gender, self-rated health and arthritis pain level. Mean age was 76 in the LL group, 78 in both the TO and the TP groups, and 81 in the HH group (non-significant differences). There were no differences in the gender composition of the four groups. The more optimistic half of the sample reported significantly better self-rated health and lower pain levels. The groups were also compared on their optimism and pessimism scores (see Fig. 1). Using ANOVA, we found significant main effects for optimism and pessimism, i.e., the HH reported lower levels of optimism than the TO and lower levels of pessimism than the TP. However, these differences were very small compared with the differences in optimism and pessimism between the high and low halves of the sample on each subscale (i.e., the differences in optimism between the HH and the TO were much smaller than the differences in optimism between the HH and TO on the one hand and the LL and the TP on the other hand; similarly, the differences in pessimism between the HH and the TP were small compared to the differences in pessimism between these two groups and the LL and TO groups). Mean levels of optimism (1a) and pessimism (1b) reported by the four groups ... Fig. 1. Mean levels of optimism (1a) and pessimism (1b) reported by the four groups created by median splits on optimism and on pessimism. In both figures the main effects are significant and the interaction is not (analysis of variance results: (1a) optimism F(1, 116) = 161.01, p < 0.0005, pessimism F(1, 116) = 8.52, p = 0.004; (1b) optimism F(1, 116) = 12.57, p = 0.001, pessimism F(1, 116) = 202.55, p < 0.0005). Figure options 3.1. Testing for the effects of optimism and pessimism on pain-coping strategies Linear regression models using optimism, pessimism, and the interaction between them to predict the use of each of the coping strategies were conducted, with an additional step in which trait anxiety and happiness were controlled (see Table 1; the first step is not shown because the results did not differ substantially from the final one). These models were also repeated with controls for age, arthritis pain level, and history of cancer (not shown), with no change in the results. A significant optimism by pessimism interaction was found for three of the coping strategies (and a borderline effect for a fourth strategy). In order to interpret these interactions, mean rates of use of the coping strategies in the four groups are presented in Fig. 2. The figure shows greater use of coping among participants who were high on both optimism and pessimism, especially in regard to diverting attention, reinterpreting pain sensations, and coping self-statements. Table 1. Summary of regression analyses for optimism, pessimism, and their interaction as predictors of the use of pain-coping strategies Coping strategy Diverting attention Reinterpreting pain sensations Coping self-statements Ignoring pain sensations Praying or hoping Catastrophizing Increasing activity Optimism 0.16 0.42** 0.28* 0.44** 0.21† 0.02 0.06 Pessimism 0.15 0.11 −0.09 0.09 0.04 0.03 −0.02 Opt X Pess 0.33** 0.23* 0.21* 0.04 0.15 0.12 0.18† Trait Anxiety 0.17 0.09 0.02 −0.07 0.35** 42*** 0.13 Trait Happiness −0.01 −0.29* −0.22† −0.14 −0.03 −0.06 0.02 R2 0.11* 0.14** 0.11* 0.15** 0.09* 0.19*** 0.04 †p < .10, *p < .05, **p < .01, ***p < .001. Table options Mean levels of use of pain coping strategies in the four groups created by ... Fig. 2. Mean levels of use of pain coping strategies in the four groups created by median splits on optimism and pessimism.