تفاوت های فردی در عملکرد، حجم کار و استرس در توجه پایدار: خوشبینی و بدبینی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|38258||2009||8 صفحه PDF||سفارش دهید||4942 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Personality and Individual Differences, Volume 47, Issue 5, October 2009, Pages 444–451
Abstract The relationship between optimism, pessimism and vigilance was investigated as a function of the degree to which different display formats facilitated performance across types of perceptual discrimination. Pessimism was associated with display and task dependent differences in workload, stress, and coping strategy. Optimism by pessimism interaction was observed for stress (Tense Arousal). Neither trait was associated with performance differences. Pessimism, but not optimism, was related to coping strategy independent of experimental condition. The results of this study were more consistent with a coping/cognitive resources perspective on optimism and pessimism than with an explanation based on learned helplessness theory. Further, the data supported the contention that optimism and pessimism are correlated but distinct constructs. The results also underscore the importance of considering both task parameters and person characteristics when evaluating the performance, workload, and stress of sustained attention.
1. Introduction Vigilance, or sustained attention, refers to the ability to monitor displays over time. Vigilance performance declines with time on watch, in part because cognitive resources available for task performance are depleted at a rate faster than they can be replenished (Parasuraman, Warm, & Dember, 1987). The resource theory explanation is supported by the finding that perceived workload and stress increase as a function of increased task demands (Warm, Matthews, & Finomore, 2008). Several studies of the workload of sustained attention have employed the NASA-Task Load Index (TLX; Hart & Staveland, 1988), a well-regarded measure that provides a global index and the relative contributions of six sources of workload (Mental Demand, Physical Demand, Temporal Demand, Performance, Effort, and Frustration). Research has shown that task characteristics that impair performance also induce high workload, and that the Mental Demand and Frustration subscales are the largest contributors to these effects ( Warm et al., 2008). Stress has been measured using the Dundee Stress State Questionnaire (DSSQ; Matthews et al., 2002), which consists of eleven factor-analytically determined scales grouped into three secondary factors of cognitive state: Task Engagement, reflected by scales of Energetic Arousal, Concentration, and two Motivation scales (Intrinsic and Success); Distress, consisting of primary factors of Tense Arousal, Hedonic Tone, Self-Esteem, and Control and Confidence; and Worry, reflected by scales of Self-Focused Attention and two forms of Cognitive Interference (Task-Related and Task-Irrelevant). Several studies have shown that vigilance is associated with declines in Task Engagement and increased Distress, and that task factors that impair performance also increase the stress of vigilance ( Warm et al., 2008). Further, the limited control observers typically have over the task environment may also be a significant source of stress in vigilance ( Hancock, 1998). Although the effects of vigilance on performance, workload, and stress are robust, large within-group variability is typically observed. Research on the individual differences variables driving this variability have produced mixed results (Berch & Kanter, 1984), and the interactive effects of person and task characteristics have yet to be clearly identified (Szalma, 2008). One skill that may differentiate good performers from poor ones is the capacity to cope with high workload and stress. Traits that influence vigilance may therefore include those related to stress and coping, such as optimism and pessimism. Optimism and pessimism have been defined in terms of differences in expectancies regarding the future, with the former associated with more favorable expectancies than the latter (Scheier, Carver, & Bridges, 1994). Further, optimism and pessimism have been found to be associated with differences in performance and stress response. For instance, using a double median-split approach to categorize individuals as ‘optimists’ or ‘pessimists’, Helton, Dember, Warm, and Matthews (1999) reported that although there were no significant differences between trait groups in overall performance, pessimists achieve a steeper vigilance decrement and higher levels of post-task stress relative to optimists. Efforts to replicate the performance results have been mixed (e.g., Helton et al., 2005 and Szalma et al., 2006), but subsequent experiments confirmed that pessimism is associated with higher levels of stress in vigilance (e.g., Szalma et al., 2006). One of the major theoretical approaches to explain differences in performance and stress response as a function of optimism and pessimism has been learned helplessness theory (Abramson, Seligman, & Teasdale, 1978), which argues that differences occur because pessimistic individuals have learned to habitually disengage or ‘give up’ in difficult or demanding situations or when failure occurs. Further, these outcomes are related to two expectations: Outcome (hopeless expectancies) and control (helpless expectancies). These expectancies operate by a diathesis–stress mechanism: Individuals who are pessimistic are more vulnerable to helpless and hopeless responses in stressful situations (Gillham, Shatte, Reivich, & Seligman, 2002). On the basis of learned helplessness theory, Gillham et al. (2002) argued that the positive expectations of optimistic people should facilitate motivation to maintain performance in the face of difficult situations, but that pessimistic expectancies should reduce effort and impair performance. However, it is possible that individual differences in performance and stress may be due to the different styles of coping (Scheier et al., 1994) and differences in cognitive resources available for task performance (Szalma, 2008). Optimism has been associated with lower stress levels, and greater active or problem/task-focused coping and less avoidant coping, while pessimism has been associated with higher levels of stress, and more emotion-focused and avoidant coping (Scheier et al., 1994). As a result of more active coping, individuals high in optimism may devote more of their resources to task performance, while individuals high in pessimism may have fewer resources to allocate to the task because they are diverting some of their resources to either emotion-focused or avoidant coping efforts to deal with the stress posed by the task demands. Alternatively, it is also possible that more pessimistic individuals have learned ways of engaging in compensatory effort in order to maintain performance. If this were the case, one would expect attenuated performance differences but higher perceived workload and stress as a function of increased pessimism. The resource theory perspective leads to the prediction that task difficulty should moderate the relation between optimism, pessimism, and performance, workload, and stress, such that individuals higher in pessimism and lower in optimism should exhibit greater performance decrements and increased workload and stress as task difficulty is increased. However, task characteristics that facilitate performance (e.g., render the perceptual discrimination easier) should have a larger positive effect on individuals higher in pessimism and lower in optimism, because such individuals presumably have fewer cognitive resources to devote to the task and will therefore benefit more from a manipulation that reduces the resources required for performance. By contrast, the helplessness theory leads to the prediction that imposition of a difficult vigilance task, in which observers have little or no control over task parameters (Hancock, 1998), should elicit helplessness appraisals across task conditions, so that individuals higher in pessimism will show similar patterns of performance, workload, and stress response regardless of task/display characteristics. In vigilance research one of the most potent determinants of task difficulty is signal salience. High salience has been found to improve performance and relieve the workload and stress of sustained attention (Warm et al., 2008). One way in which signals can be made more salient is via the use of configural displays, which utilize easily perceived features that improve performance for tasks requiring integration of information (Bennett & Flach, 1992). Such displays work in part because the elements form an easily perceivable, integrated feature that ‘pops out’ and is much more salient than displays with separated elements without such feature integration. A previous study found that use of a configural display was associated with an attenuated vigilance decrement, possibly due to enhanced signal salience (Szalma et al., 2006). Hence, use of these display formats for tasks requiring integration of information may improve performance and reduce workload and stress. By contrast, cases in which display features do not support the discrimination requirements of the task should have substantially lower signal salience, and individuals higher in pessimism and lower in optimism may show greater vulnerability to performance impairment and increased workload and stress. The current study evaluated this possibility by manipulating display format and the degree to which it facilitated the perceptual demands of the task (i.e., the difficulty of the discrimination). Based on resource theory, it was expected that in the more demanding task conditions (in which the display format is not well suited for the perceptual discrimination required) pessimism should predict more emotion-focused coping and avoidant-coping, higher stress levels, and greater perceived workload. Higher levels of optimism should predict greater task-focused and less emotion-focused and avoidant coping, and lower levels of workload and stress. In the easier conditions the benefits of a display format that facilitates performance should be greater for those higher in pessimism and lower in optimism. Based on previous research (Helton et al., 1999), if there are performance differences as a function of traits it will likely manifest in changes over time, such that increased pessimism should be associated with a steeper decrement and optimism should be related to an attenuated decrement in the more demanding conditions
نتیجه گیری انگلیسی
. Results The descriptive statistics for Optimism (M = 53.83; SD = 5.14; coefficient α = .71) and Pessimism (M = 37.97; SD = 6.66; coefficient α = .80), and the correlation between the scales (r = −.52, p < .001), were similar to those obtained from prior samples using this instrument ( Dember, 2002). Task and trait effects were evaluated using hierarchical regression ( Pedhazur, 1997). The variables entered at each step are shown in Table 1. Note that step 1 is equivalent to an ANOVA of independent variable effects, which are reported elsewhere (Szalma, 2002). Significant product vectors were tested using the Johnson-Neyman procedure for simultaneous regions of significance using a criterion of α = .15 ( Pedhazur, 1997). To control for Type I error, evaluation of the product vector regression coefficients was done using the modified Bonferroni correction described by Jaccard and Turrisi (2003). Statistics for the steps with statistically significant ΔR2 values and product vectors are summarized in Table 2. Table 1. Summary of hierarchical regression variables. Step in regression Variables added 1 Task, display, task × display 2 Pessimism, optimism 3 Task × pessimism, task × optimism, display × pessimism, Display × optimism, pessimism × optimism 4 Task × display × pessimism, task × display × optimism 5 Task × display × pessimism × optimism Table options Table 2. Summary of optimism and pessimism regressions (N = 96). Criterion Step Variable R2 ΔR2 p β p Global workload 3 T × P .29 .14 .042 −2.06 .005 Mental demand 3 PG × P .32 .16 .013 −2.00 .018 Effort 3 T × P .29 .16 .020 −2.38 .020 Pre–post stress Distress Tense Arousal 3 Opt × P .25 .20 .007 −1.98 .006 Worry Cognitive interference Task-Related 3 T × P .21 .18 .015 −1.74 .022 Note: T = Task type; P = Pessimism; Opt = Optimism; PG = Polygon graph display. Table options 3.1. Performance Regressions of performance (sensitivity and response bias) were performed in two ways: (1) an overall score based on the entire vigil; and (2) a difference score between the first and last periods on watch. The regressions of overall A′ and change in sensitivity did not result in significant ΔR2 values. Similar regressions for View the MathML sourceβD″ indicated no significant effects for product vectors involving either trait. 3.2. Global workload A significant ΔR2 was observed at step 3, with a significant task by pessimism regression coefficient ( Fig. 2). The midpoint identification task induced less workload than the dot-distance task for individuals with pessimism scores in the middle-to-high range. Global workload as a function of pessimism and task type. Note. The dotted ... Fig. 2. Global workload as a function of pessimism and task type. Note. The dotted vertical line and the arrow indicate the significant region of group differences in global workload. Figure options 3.3. Workload scales Significant regressions were observed for weighted mental demand and weighted effort. The other scales showed no significant effects related to pessimism. 3.4. Mental demand A significant ΔR2 was obtained at step 3, with a significant PG display by pessimism regression coefficient ( Fig. 3). The PG display was less mentally demanding than the BGDB display for individuals in the middle to upper range of pessimism. Weighted mental demand as a function of pessimism and display type. Note. The ... Fig. 3. Weighted mental demand as a function of pessimism and display type. Note. The vertical dotted line and the arrow indicate the region of significant group differences in weighted mental demand. Figure options 3.5. Effort A significant ΔR2 was obtained at step 3, with a significant task by pessimism regression coefficient ( Fig. 4). The midpoint identification task induced less workload than the dot-distance task for individuals with pessimism scores in the middle-to-high range. Weighted Effort as a function of pessimism and task type. Note. The vertical ... Fig. 4. Weighted Effort as a function of pessimism and task type. Note. The vertical dotted line and the arrow indicate the region of significant group differences in weighted effort. Figure options 3.6. Pre–post task stress state A significant ΔR2 was observed for pre–post change in Tense Arousal at step 3, with a significant optimism by pessimism regression coefficient. Hence, the relationship between Tense Arousal and each trait varied as a function of the other trait. Following procedures described in Jaccard and Turrisi (2003), optimism and pessimism data were mean-centered and ‘high’ and ‘low’ levels of optimism defined in terms of scores one standard deviation above and below the optimism mean, respectively. Separate regression functions were computed for pessimism at each level of optimism ( Fig. 5). Across all experimental conditions, pessimism predicted increased pre–post task Tense Arousal only for individuals relatively low in optimism. For individuals relatively high in optimism the reverse trend was observed: higher pessimism scores were associated with a pre–post decline in Tense Arousal. Pre-post change in tense arousal as a function of pessimism at three levels of ... Fig. 5. Pre-post change in tense arousal as a function of pessimism at three levels of optimism. Note. Scores are mean-centered. Each line represents the regression of tense arousal on pessimism when the level of optimism is set at its mean or +/− 1 SD above or below the mean. Each regression line was generated by entering three values for pessimism: −1 (one SD below the mean for pessimism), 0 (pessimism mean), or +1 (1 SD above the mean for pessimism). Figure options For Task-Related Cognitive Interference, a significant ΔR2 was obtained at step 3, with a significant task by pessimism regression coefficient ( Fig. 6). The Johnson-Neyman procedure failed to yield a solution, which can occur when the within-groups error variance is sufficiently large to procedure a negative value for the square root operation of the function ( Pedhazur, 1997). Separate regressions were computed for the two tasks, but neither regression was statistically significant. Pre-post change in Cognitive Interference-task related as a function of ... Fig. 6. Pre-post change in Cognitive Interference-task related as a function of pessimism and task type. Note. No region of significant group differences is indicated because the Johnson-Neyman procedure failed to yield a solution. The separate regression equations for each task type are shown in the Figure. CITR = Cognitiver Interference-Task Related. Figure options 3.7. Coping measures There were no significant product vectors involving the experimental conditions for any of the coping measures. A regression for emotion-focused coping yielded a significant regression coefficient for pessimism, β = .26, p = .02, R2 = .10, but not for optimism or the product vector. Increased pessimism was associated with more emotion-focused coping ( Fig. 7). There were no statistically significant regressions for Task-focused or Avoidant coping. Emotion-focused Coping as a function of Pessimism. Fig. 7. Emotion-focused Coping as a function of Pessimism.