توجه داخلی در مقابل توجه خارجی در اضطراب اجتماعی : یک تحقیق با استفاده از پارادایم رمان

Several cognitive models propose that social anxiety is associated with increased self-focused attention. Indirect evidence for this hypothesis has been provided by questionnaire studies, and by cognitive psychology paradigms that have demonstrated reduced processing of external information during feared social-evaluative situations. However, no studies have simultaneously measured on-line attention to internal and external events. A probe detection task that aimed to measure the balance of attention between internal and external stimuli was developed. High and low socially anxious individuals were instructed to detect two probes. The external probe was superimposed on pictures of faces (happy, neutral, angry) or household objects that were presented on a VDU. The ‘internal’ probe was a pulse to the finger which participants were led to believe represented significant changes in their physiology. Compared to low speech anxious individuals, high speech anxious individuals showed an internal attentional bias, that was specific to conditions of social-evaluative threat.

Self-focused attention has been implicated as an important cognitive process in both clinical (Ingram, 1990 and Wells and Matthews, 1994) and non-clinical populations (Carver and Scheier, 1981, Duval and Wicklund, 1972 and Fenigstein, Scheier and Buss, 1975). Carver and Scheier, 1981 and Carver and Scheier, 1998 suggest that flexible self-focused attention may be part of a functional process that highlights the discrepancy between one’s present perceived state and desired standards of performance, thereby motivating a change in behaviour to reduce the discrepancy. In the literature on clinical populations, it has been suggested that individuals with psychological disorders demonstrate excessive self-focused attention, which leads to several dysfunctional effects, such as intensifying emotional states, reducing effortful coping and impairing task performance (see Ingram, 1990 and Wells and Matthews, 1994, for reviews). Excessive self-focused attention plays an important role in several cognitive models of social anxiety (e.g. Clark and Wells, 1995, Hartman, 1983, Hope, Gansler and Heimberg, 1989 and Rapee and Heimberg, 1997). For example, Clark and Wells (1995) propose that when patients with social phobia become concerned that they may fail to make their desired impression on other people, their attention shifts from observation of others to detailed monitoring and observation of themselves. They then use the misleading interoceptive information produced by self-monitoring as evidence they are making a negative impression on others. It is suggested that several types of internal information are utilized, including images, thoughts and physiological reactions. Such a processing bias would maintain social anxiety by generating inappropriately misleading information, preventing access to external information that could disconfirm patients’ negative beliefs, and instigating negative behaviors that are noticeable to others. A range of indirect sources of evidence suggests that self-focused attention is associated with social anxiety. Self-report questionnaires such as the public and private self-consciousness scales (Fenigstein et al., 1975) can be used to assess self-focused attention. Several studies have found significant correlations between social anxiety scales and both measures (Hope and Heimberg, 1988, Monfries and Kafer, 1994 and Saboonchi and Lundh, 1997), although two studies have found a correlation with public but not private self-consciousness (Bögels, Alberts and de Jong, 1996 and Smari, Clausen, Hardarson and Arnarson, 1995). In addition, Jostes, Pook, and Florin (1999) found levels of both public and private self-consciousness in individuals with social phobia to be significantly higher than that of patients with bulimia and non-patient controls. High socially anxious individuals also report more self-focused attention during social-evaluative situations. Hope, Heimberg, and Klein (1990) found that high socially anxious individuals score higher on modified state versions of the public and private self-consciousness scales than low socially anxious individuals during a conversation with a stranger. Similarly, shy individuals have been found to spend a higher percentage of time than non-shy individuals focusing on the impression they are making and thinking about what to say next (Melchior & Cheek, 1990). A shortcoming of questionnaire studies is that they rely on subjective ratings, whereas experimental paradigms have the potential to provide more objective measures. For instance, one might infer that individuals are engaging in self-focused attention if they demonstrate reduced processing of external social cues. In the study by Hope et al. (1990), high and low socially anxious females had a conversation with a male confederate on a number of fixed topics. The high socially anxious participants recalled less information and made more errors in recall than the low socially anxious participants. Similarly, a study of public speaking (Daly, Vangelisti, & Lawrence, 1989) demonstrated that high public-speaking anxious individuals recalled fewer elements of the environment present during their speech. Encouragingly, Mellings and Alden (2000) demonstrated a convergence of self-report and objective measures; they found that recall of external social information (details of conversation partner) was poorest in individuals who reported that they were more self-focused during the interaction. In a study of attentional bias, Mansell, Clark, Ehlers, and Chen (1999) found that high socially anxious individuals who were expecting to be socially evaluated selectively avoided emotional facial expressions relative to low socially anxious individuals. Similarly, Chen, Ehlers, Clark, and Mansell (in press) found that patients with generalized social phobia avoided looking at faces vs objects compared to controls. Self-focused attention might be inferred from individuals showing increased accuracy at judging changes in internal cues. In this context, Johansson and Ost (1982) found that heart rate correlated significantly with subjective measures of heart rate perception in social phobics during a feared social situation. Taken together, the above results provide reasonable indirect evidence for increased attention to internal cues in social anxiety. However, these studies have neither directly measured the extent of internally focused attention in social anxiety, nor have they simultaneously investigated the balance between internal and external attention. The current study was designed to incorporate both elements, using a novel probe-detection paradigm that measured internal vs external attention. On the basis of cognitive models of social phobia (Clark and Wells, 1995 and Rapee and Heimberg, 1997), we expected that, under conditions of social-evaluative threat, individuals high in social anxiety would show internal attentional bias, as indicated by faster reaction to a probe occurring on their body than to an external stimulus, whereas individuals with low social anxiety were expected to either show no bias or a bias towards processing external information. No group difference was expected in the absence of social-evaluative threat (as in Mansell et al., 1999). The study also assessed participants’ ratings of their performance during the speech and compared them to observer ratings to explore whether high socially anxious individuals underestimated their performance relative to low socially anxious individuals (Mellings and Alden, 2000, Rapee and Hayman, 1996, Rapee and Lim, 1992 and Stopa and Clark, 1993). A significant correlation between internal attentional bias and underestimation would be consistent with the view that socially anxious individuals use information from their internal cues to deduce how they appear to others (Clark & Wells, 1995). Mansell and Clark (1999) provided indirect support for this hypothesis when they observed a significant correlation between perceived bodily sensations during a speech and subsequent overestimates of anxious appearance and negative global behaviours in high socially anxious participants. The present study investigated the relationship between underestimation of performance and attentional bias directly.