Repressers and low-anxious
It is assumed within most theories of trait anxiety (including those of H.J. Eysenck and Gray) that low scorers on trait anxiety form a homogeneous group. However, this assumption is not incorporated into the four-factor theory. According to that theory, individuals scoring low on trait anxiety should be divided into two groups on the basis of their level of defensiveness. Individuals who are low in trait anxiety but high in defensiveness are categorized as repressers or as having the repressive coping style, whereas individuals who are low in trait anxiety and low in defensiveness are categorized as low-anxious. These categories were first popularized by Weinberger, Schwartz and Davidson (1979), who used the Marlowe-Crowne Social Desirability Scale as a measure of defensiveness. They found substantial differences between repressers and low-anxious individuals when placed in a moderately stressful situation. For example, they found that repressers’ physiological and behavioural responses indicated much higher levels of anxiety than did those of low-anxious individuals.. all three types of measure in a moderately stressful situation in which the participants were videotaped. All of the data were then converted to standard scores for purposes of comparison. As had been found in previous research, repressers had relatively high physiological anxiety but low self-reported state anxiety. In contrast, the low-anxious participants had relatively low levels of physiological anxiety and/or self-reported state anxiety. The pattern was similar when behavioural anxiety (based on ratings of the videotape evidence by independent judges) was considered.
Repressers had relatively high levels of behavioural anxiety, whereas the low-anxious had a low level of behavioural anxiety. Most theoretical interest from the evidence discussed in this section so far is the fact that the repressers showed large discrepancies between their self-reported anxiety on the one hand and their physiological and behavioural anxiety on the other hand. One possible explanation of these discrepancies is simply that repressers deliberately distort their self-reports to claim low levels of experienced anxiety even though they actually experience high levels. If that were the case, then the discrepancies would be illusory rather than genuine. This issue has been addressed in several studies. Levels of trait anxiety on the Spielberger State- Trait Anxiety Inventory on two occasions separated by approximately two months: (a)standard conditions; (b) bogus pipeline conditions. In the second condition, the participants were wired up to an impressive-looking piece of equipment (known as the ‘bogus pipeline’) which they were led to believe could detect lying. The bogus pipeline has proved effective in numerous studies in social psychology in increasing honesty. It has been found that participants admit to much more prejudice under bogus pipeline than under standard conditions. The trait-anxiety scores of repressers were slightly higher under bogus pipeline than under standard conditions, but the difference was not statistically significant. This finding suggests that repressers do not deliberately distort their scores when completing measures of trait anxiety under standard conditions. It is extremely difficult to decide whether repressers’ self-reported levels of state and trait anxiety are genuine or subject to distortion. Accordingly, they devised an indirect approach to the issue, using a paradigm which did not involve obtaining self-report data. In their study, the participants had to perform a verbal reasoning task which was combined with a high or low memory load. It is known that the adverse effects of the high memory load on performance of the verbal reasoning task are much greater for individuals high in trait anxiety than for those low in trait anxiety. It is generally accepted that the main reason for this pattern of findings is that individuals high in trait anxiety engage in more task-irrelevant negative thoughts about themselves and about their inadequate level of performance. Such task-irrelevant thoughts interfere with task performance, and such interference effects are especially strong when there is a concurrent memory load. The key finding obtained by Derakshan and Eysenck (1998) was that the negative effects of high memory load on verbal reasoning performance in repressers were relatively modest, resembling the effects on low-anxious individuals much more than those on high-anxious individuals. These findings support the view that repressers genuinely experience low levels of anxiety when confronting complex tasks.
According to Eysenck’s four-factor theory, the two groups differ in terms of important aspects of cognitive functioning. It is assumed that low-anxious individuals do not have cognitive biases relating to the processing of external and internal stimuli. The absence of cognitive biases helps to explain why low-anxious individuals show minimal discrepancies among measures of self-reported, behavioural and physiological anxiety. In contrast, it is assumed that repressers have opposite cognitive biases with respect to the processing of both external and internal stimuli. More specifically, repressers have opposite attentional and interpretive biases. This means that they tend systematically to avoid attending to threat-related stimuli, and they also tend to interpret ambiguous stimuli and situations in a nonthreatening way. As a result, repressers tend to minimize the threateningness of their own physiological state and behaviour, and this helps to produce discrepancies between self-report, physiological and behavioural measures of anxiety (Derakshan and Eysenck, 2001b). More specifically, repressers’ opposite attentional and interpretive biases lead them to ignore and/or misinterpret the evidence of their own physiological state, future cognitions and behaviour. The trait-anxiety scores of repressers were slightly higher under bogus pipeline than under standard conditions, but the difference was not statistically significant. This finding suggests that repressers do not deliberately distort their scores when completing measures of trait anxiety under standard Fox (1993) used the dot-probe paradigm with participants who were repressers, low-anxious, and high-anxious in order to assess attentional biases for social and physical threat words. For present purposes, the key finding was that the repressers showed an opposite attentional bias for the social threat stimuli, whereas the low-anxious individuals did not exhibit either an attentional bias or an opposite attentional bias. There is relatively little direct evidence that repressers possess an opposite interpretive bias (see Eysenck, 2000). However, potentially important findings were reported by Calvo and Eysenck (2000), who assessed interpretive biases by measuring the speed of naming target words which confirmed or disconfirmed the consequence implied by previous ambiguous sentences. Various time intervals between sentence and target were used in order to plot the time course of activation of threat-related and neutral target words. The low-anxious participants showed no attentional or opposite attentional bias at any time interval. In contrast, the repressers had an interpretive bias when there was a short interval between sentence and target word, but this bias totally disappeared at a longer interval. It could be argued that this reversal (which was not found in the data of the low-anxious participants) reflects the operation of an opposite interpretive bias.
It is assumed within most theories of trait anxiety (including those of H.J. Eysenck and Gray) that low scorers on trait anxiety form a homogeneous group. However, this assumption is not incorporated into the four-factor theory. According to that theory, individuals scoring low on trait anxiety should be divided into two groups on the basis of their level of defensiveness. Individuals who are low in trait anxiety but high in defensiveness are categorized as repressers or as having the repressive coping style, whereas individuals who are low in trait anxiety and low in defensiveness are categorized as low-anxious. These categories were first popularized by Weinberger, Schwartz and Davidson (1979), who used the Marlowe-Crowne Social Desirability Scale as a measure of defensiveness. They found substantial differences between repressers and low-anxious individuals when placed in a moderately stressful situation. For example, they found that repressers’ physiological and behavioural responses indicated much higher levels of anxiety than did those of low-anxious individuals.. all three types of measure in a moderately stressful situation in which the participants were videotaped. All of the data were then converted to standard scores for purposes of comparison. As had been found in previous research, repressers had relatively high physiological anxiety but low self-reported state anxiety. In contrast, the low-anxious participants had relatively low levels of physiological anxiety and/or self-reported state anxiety. The pattern was similar when behavioural anxiety (based on ratings of the videotape evidence by independent judges) was considered.
Repressers had relatively high levels of behavioural anxiety, whereas the low-anxious had a low level of behavioural anxiety. Most theoretical interest from the evidence discussed in this section so far is the fact that the repressers showed large discrepancies between their self-reported anxiety on the one hand and their physiological and behavioural anxiety on the other hand. One possible explanation of these discrepancies is simply that repressers deliberately distort their self-reports to claim low levels of experienced anxiety even though they actually experience high levels. If that were the case, then the discrepancies would be illusory rather than genuine. This issue has been addressed in several studies. Levels of trait anxiety on the Spielberger State- Trait Anxiety Inventory on two occasions separated by approximately two months: (a)standard conditions; (b) bogus pipeline conditions. In the second condition, the participants were wired up to an impressive-looking piece of equipment (known as the ‘bogus pipeline’) which they were led to believe could detect lying. The bogus pipeline has proved effective in numerous studies in social psychology in increasing honesty. It has been found that participants admit to much more prejudice under bogus pipeline than under standard conditions. The trait-anxiety scores of repressers were slightly higher under bogus pipeline than under standard conditions, but the difference was not statistically significant. This finding suggests that repressers do not deliberately distort their scores when completing measures of trait anxiety under standard conditions. It is extremely difficult to decide whether repressers’ self-reported levels of state and trait anxiety are genuine or subject to distortion. Accordingly, they devised an indirect approach to the issue, using a paradigm which did not involve obtaining self-report data. In their study, the participants had to perform a verbal reasoning task which was combined with a high or low memory load. It is known that the adverse effects of the high memory load on performance of the verbal reasoning task are much greater for individuals high in trait anxiety than for those low in trait anxiety. It is generally accepted that the main reason for this pattern of findings is that individuals high in trait anxiety engage in more task-irrelevant negative thoughts about themselves and about their inadequate level of performance. Such task-irrelevant thoughts interfere with task performance, and such interference effects are especially strong when there is a concurrent memory load. The key finding obtained by Derakshan and Eysenck (1998) was that the negative effects of high memory load on verbal reasoning performance in repressers were relatively modest, resembling the effects on low-anxious individuals much more than those on high-anxious individuals. These findings support the view that repressers genuinely experience low levels of anxiety when confronting complex tasks.
According to Eysenck’s four-factor theory, the two groups differ in terms of important aspects of cognitive functioning. It is assumed that low-anxious individuals do not have cognitive biases relating to the processing of external and internal stimuli. The absence of cognitive biases helps to explain why low-anxious individuals show minimal discrepancies among measures of self-reported, behavioural and physiological anxiety. In contrast, it is assumed that repressers have opposite cognitive biases with respect to the processing of both external and internal stimuli. More specifically, repressers have opposite attentional and interpretive biases. This means that they tend systematically to avoid attending to threat-related stimuli, and they also tend to interpret ambiguous stimuli and situations in a nonthreatening way. As a result, repressers tend to minimize the threateningness of their own physiological state and behaviour, and this helps to produce discrepancies between self-report, physiological and behavioural measures of anxiety (Derakshan and Eysenck, 2001b). More specifically, repressers’ opposite attentional and interpretive biases lead them to ignore and/or misinterpret the evidence of their own physiological state, future cognitions and behaviour. The trait-anxiety scores of repressers were slightly higher under bogus pipeline than under standard conditions, but the difference was not statistically significant. This finding suggests that repressers do not deliberately distort their scores when completing measures of trait anxiety under standard Fox (1993) used the dot-probe paradigm with participants who were repressers, low-anxious, and high-anxious in order to assess attentional biases for social and physical threat words. For present purposes, the key finding was that the repressers showed an opposite attentional bias for the social threat stimuli, whereas the low-anxious individuals did not exhibit either an attentional bias or an opposite attentional bias. There is relatively little direct evidence that repressers possess an opposite interpretive bias (see Eysenck, 2000). However, potentially important findings were reported by Calvo and Eysenck (2000), who assessed interpretive biases by measuring the speed of naming target words which confirmed or disconfirmed the consequence implied by previous ambiguous sentences. Various time intervals between sentence and target were used in order to plot the time course of activation of threat-related and neutral target words. The low-anxious participants showed no attentional or opposite attentional bias at any time interval. In contrast, the repressers had an interpretive bias when there was a short interval between sentence and target word, but this bias totally disappeared at a longer interval. It could be argued that this reversal (which was not found in the data of the low-anxious participants) reflects the operation of an opposite interpretive bias.
No comments:
Post a Comment