شناخت در اضطراب دوران کودکی :مسائل مفهومی، روش شناسی و توسعه ای

Anxiety disorders are one of the most common psychiatric disorders in the general child population and can have significant impact on immediate and long-term functioning. Despite the common use of cognitive-behavioral interventions that target specific, negative thoughts in anxious children, it is unclear that the extant literature clearly documents cognitive aberrations among these children. In this review, conceptual and methodological issues related to the assessment of cognition in anxious children are highlighted and empirical data addressing these areas are evaluated. Furthermore, data addressing cognitive change as a function of treatment outcome is examined, and the impact of cognitive development as a moderating variable is discussed. Finally, areas for future research are presented.

For the past 30 years, the predominant conceptual model of anxiety disorders has been considered to consist of three components: somatic or physical reactivity, subjective distress, and behavioral avoidance (e.g., Lang, 1968). Over the ensuing years, the term subjective distress has become equated with cognition, cognitive schemata, and the presence of anxious thoughts. Increasingly, models of etiology, maintenance, and treatment of anxiety disorders have suggested a need to attend to the cognitive dimension and, in some cases, the “cognitive model” has been proposed as a primary model for anxiety disorders (e.g., Clark & Beck, 1988). As usually occurs, models of disorders, as they exist in adults, are adopted as the initial basis for understanding these same disorders in children and, thus, there has been a marked increase in cognitive-behavioral research aimed at anxiety disorders in youth. Growing amounts of empirical data address the “cognitive factors” associated with childhood anxiety (e.g., Bogels & Zigterman, 2000, Houston et al., 1984, Kendall & Chansky, 1991, King et al., 1995, Messer & Beidel, 1994, Prins, 1986, Prins & Hanewald, 1999, Zatz & Chassin, 1983 and Zatz & Chassin, 1985) and, although a substantial literature on cognition and childhood anxiety disorders exists, much of it is conflictual in nature. Thus, the purpose of this review is to present and evaluate this literature and propose areas for future investigation. Understanding cognition in childhood anxiety disorders requires attention to conceptual, methodological, and developmental considerations. As such, this review presents some issues that have received inadequate attention thus far. First, with respect to conceptual issues, there are competing definitions of the term “cognition.” For example, within the childhood anxiety literature, the term cognition broadly refers to both actual thoughts reported (e.g., cognitive product; Kendall & Ingram, 1987) and to a basic underlying schema organized around a theme of threat (e.g., cognitive structure; Kendall & Ronan, 1990). For purposes of clarity, we shall refer to the former simply as cognitive content and the latter as cognitive process. In the first part of this review, the literature on both cognitive content and process in anxious children will be examined. To date, most research has focused solely on cognitive content including assessing negative vs. positive cognitions, determining cognitive errors (inaccurate beliefs), evaluating cognitive distortions, and determining state-of-mind ratios (SOM; ratio of positive to negative thoughts). Additionally, we examine whether there is cognitive content specific to childhood anxiety as opposed to a more general cognitive style associated with various affective states. Following the review of cognitive content, research on cognitive process is presented. Specifically, we review the results of investigations that have examined anxious children's interpretations, attributions, and expectancies in relation to specific events, including research examining the role of familial factors (e.g., transmission of information) in maintaining a specific cognitive style among anxious children. Finally, we consider the role of emotional understanding in children's report of cognition. Specifically, research suggesting that young children may not understand the links between cognition and emotion is reviewed, and implications for assessment are considered. This review also addresses available research examining cognitive change in the context of treatment outcome. Based on a marked increase in cognitive-behavioral treatments aimed at childhood anxiety disorders within recent years, there remains a growing need for a better understanding of the changes (if any) associated with specific treatment components. For example, despite the lack of clear evidence that children with anxiety disorders have “distorted” cognitions, many researchers and clinicians have included cognitive interventions as part of their treatment packages. Thus, we have seen a proliferation of treatments designed to alter cognitions in the absence of understanding cognitive content and process, and in the absence of cohesive explanations that meet the usual criteria of theory. In this section, data addressing cognitive change as a result of behavioral and cognitive-behavioral interventions are evaluated. Finally, important methodological considerations such as the suitability of various assessment methods for use with children are addressed. The vagaries of the various assessment methods are relevant for understanding the current, and often conflicting, state of the assessment literature. Conflicting results from various studies of negative cognition in anxiety disorders may be the result of different assessment methods, the specific time at which the assessment occurred, or from the wholesale adoption of adult assessment models for use with children. Without further attention to these issues, the confused and conflicting nature of the literature is likely to continue. Moreover, the role of cognitive development and its impact upon specific methodologies is discussed. Although basic developmental abilities have clear and direct implications for examining cognition in anxious children, few investigations have adequately considered the impact of differences in cognitive maturation when adopting methods of child assessment. Thus, recommendations for future research in these areas also are proposed.

Overall, research on the cognitive aspects of childhood anxiety has produced divergent and sometimes confusing findings. Although some data indicate significant differences in terms of the types of cognitions reported by clinically anxious children, many of the findings have been difficult to interpret due to methodological differences and issues related to statistical vs. clinical significance. Nonetheless, these results have implications for future research in this area. Specifically, the divergent findings highlight the need to create some agreed upon definitions for cognition (e.g., related to content vs. process) and for research that specifically compares the same and different methods of assessment related to both cognitive content and processes. However, while such research will better inform assessment efforts, it should also be noted that examination of these factors in isolation from one another may produce an incomplete understanding of the phenomena of interest. For example, research focusing solely on cognitive content is still limited by an inadequate understanding of the processes that maintain the specific content being examined. Along these lines, models of information processing may play an important role in the regulation of anxiety and may be directly responsible for cognitive content (Daleiden & Vasey, 1997). However, assessment of information processing is generally inaccessible through self-report inventories and, thus, the development of other forms of assessment, designed specifically to address cognitive processing, are necessary. Further, because information-processing capacity is thought to be linked with other metacognitive skills, such as emotional understanding (e.g., Caroll & Steward, 1984 and Harter, 1986), these may be important concepts to consider when examining cognition. The development of emotional and cognitive understanding as related to broader cognitive processes represents an important, yet generally ignored consideration in the area of cognitive assessment. Methodological considerations also merit further attention. Many of the current findings depend heavily on the assessment strategies used, including timing of assessment (i.e., before, during, or after a specific task). However, although an array of assessment methods continues to be used, it is primarily the theoretical conceptualizations that have obfuscated the meaning of empirical findings (i.e., content vs. process). Cognitions are commonly theorized to serve as a causal mechanism underlying these disorders, fostering the belief that these cognitions (or cognitive processes) need to be changed in order to treat anxious symptomatology. However, as reviewed here, although negative cognitions are sometimes reported, investigation into their specific role as a contributory factor in the etiology and maintenance of childhood anxiety disorders has not occurred. Inasmuch as there is evidence that dysfunctional cognition exists as a concomitant of anxiety, the possibility remains that these symptoms may occur as a result of such disorders. For example, some models of childhood anxiety have assumed that chronic overactivity of cognitive schemas organized around threat-relevant stimuli is responsible for the development of anxiety in children (e.g., Kendall, 1985 and Kendall & Ronan, 1990). Other theorists have pointed toward cognitions as moderators in the development of anxiety (e.g., Chorpita & Barlow, 1998). With respect to the latter, Chorpita and Barlow (1998) proposed that early experiences with a lack of control foster a cognitive style characterized by a potential to increase thoughts or feelings of helplessness in anxious children. In other words, although a specific cognitive vulnerability may contribute to childhood anxiety disorders, cognitive vulnerability is conceptualized as a moderator variable, in that certain environmental influences can be modified (e.g., amplified) through a specific cognitive or attributional style (Barlow, Chorpita, & Turovsky, 1996). Thus, even if specific cognitive processes contribute to the maintenance of these disorders by biasing perception of future events, it is not clear that they represent etiological factors in the development of anxiety. Another important issue when evaluating the validity of cognitive data in anxious children is the necessity to consider basic cognitive maturation and different developmental stages (Kendall & Ronan, 1990). For example, just as young children's fears tend to center on physical, external events (Graziano, DeGiovanni, & Garcia, 1979), cognition and worry among young children also involves these stable constructs. With increasing age however, fears and worry become more internal and abstract, mirroring the development of basic cognitive abilities. Although developmental maturation has clear implications for studies of cognition, to date, this basic developmental literature has had limited impact upon cognitive assessment procedures among anxious children. For example, Flavell, Green, Flavell, and Grossman (1997) noted that children become more mentally reflective as they mature, and much of this increased reflection is verbal in nature. According to Flavell et al., knowledge of inner speech is usually acquired during the early school years, perhaps as a result of repetitive cognitive activities such as silent reading or counting. Nonetheless, recent research supports the suggestion that young children are often unaware of their own ongoing “stream of consciousness” Flavell et al., 2001 and Flavell et al., 2000. Thus, an initial question of future investigations should be whether children, particularly young children, have the basic cognitive skills necessary to complete assessment procedures. For example, do younger children have and/or are they able to articulate their cognitions when asked to do so (e.g., during a think-aloud task)? While Flavell et al. reported that (compared to 4 and 5 year-olds) 6 and 7 year-old children were able to understand the concept of “talking to oneself,” clearly not every child acquires this specific cognitive ability at exactly the same age. Thus, some 6 or 7 year-old children may be deficient in detecting—and, as a result, reporting—their inner thoughts. Therefore, specific assessment of this basic ability is probably necessary for most first and second grade children and should be administered prior to assessment. Otherwise, reports of “no thoughts” during a task could result in the interpretation that negative cognitions are not part of the clinical presentation of anxiety in children, rather than that the task was inappropriate and invalid. Of course, making this determination is different from assessing a child's reading level. It requires assessment of a child's ability to reflect upon cognitive processes, clearly an ability different from being able to read. Furthermore, the issue of distinguishing between emotional arousal and specific cognitive content among anxious children merits further consideration. Our own preliminary research indicates that young children (as well as some adolescents) may have difficulty in differentiating emotional arousal from cognitive processes during assessment procedures. Although some preliminary evidence exists that clinically anxious children may possess specific deficits in understanding emotion (Southam-Gerow & Kendall, 2000), it is unclear whether this problem may be specific to anxious children and further, how such a difficulty might relate to their report of cognition. Developmental considerations may also be germane to examinations of cognition among specific childhood anxiety disorders. Although several investigations have examined cognitive-content specificity among anxiety vs. depressive disorders, to date, few investigations have directly examined cognitive content among the specific anxiety disorders. Nonetheless, important distinctions may exist within these subgroups with regard to cognitive development. For example, Weems et al. (2001) reported that certain types of dysfunctional cognitions (i.e., catastrophizing and personalizing) were more strongly associated with anxiety among anxious adolescents than among young children, perhaps as a result of basic cognitive development and young children's inability to conceptualize and understand future events. Such developmental differences may have implications for cognitive content within specific diagnostic categories of anxiety. One example comes from basic developmental research on how children form self-descriptions and on children's conceptualization of the self. The Self-descriptions of children between the ages of 4 and 6 are based on physical appearance, behavior and activities. In contrast, for children between the ages of 7 and 11, self-descriptions begin to incorporate psychological characteristics and social comparisons (Stone & Lemanek, 1990). Likewise, Vasey (1993) noted that prior to age 8, children evaluate themselves primarily with regard to absolute standards (e.g., physical characteristics), compared to children 8 years and older, who appear capable of comparing themselves to others (i.e., social comparison). Meanwhile, yet another milestone in conceptualization of the self appears to occur during adolescence when children become more aware of others' evaluations and are able to think in terms of internal and stable traits. These differences in cognitive development are relevant to childhood social anxiety disorder. Specifically, this would suggest that social-evaluative cognitions are more likely to be present during adolescence and are not likely to be present in young children who might be lacking crucial social comparison skills (Vasey, 1993). In support of this notion, Seidner, Stipek, and Feshbach (1986) found that among children in kindergarten and first grade, embarrassment was a relatively unfamiliar concept. Yet, because children younger than age 8 can be diagnosed with social phobia (based on physical and behavioral symptoms), the specific role of cognition in the expression of the disorder in young children is unclear. Overall, conceptualizing specific cognitions as characteristic of anxiety disorders (or specific anxiety disorders) may be ill advised in the absence of empirical data. Additionally, when considering a specific cognitive style among children with anxiety disorders, it also may be important to include the role of familial factors. Based on consistent findings from recent research, parental behavior appears to play a role in maintaining certain aspects of cognitive style (e.g., perceptions of threat and avoidance behavior). Specifically, Barlow's model (1988;Chorpita & Barlow, 1998) highlights the potential transmission of information through modeling and/or an internalization of parental fearful/controlling behavior. Indeed, excessive parental control has been strongly associated with childhood anxiety Chorpita et al., 1998 and Rapee, 1997. Therefore, interventions that do not consider, and thus, do not attempt to alter fearful or controlling parental behaviors (when necessary) may result in an attenuated treatment outcome. Finally, as a result of the common use of nonspecific cognitive assessment and treatment outcome measures, researchers and clinicians are frequently challenged by a limited understanding of the actual cognitive change (content, process, or both) produced by most CBT strategies. Although cognitive intervention strategies are common elements within most treatment protocols, little is known about whether these strategies provide children with a different mechanism for change (i.e., do they alter cognitive process) than behavioral treatment components, which also appear to change cognitive symptoms. Research is needed to more closely examine the specific use of behavioral vs. cognitive treatment components aimed at altering underlying information-processing systems toward cognitive change (Daleiden & Vasey, 1997). In conclusion, it is important that we do not put “the cart before the horse.” As illustrated by this review, many basic issues regarding the role of cognition in childhood anxiety disorders remain unaddressed and until clarified, assuming the validity of a cognitive model appears unjustified. Furthermore, until such a model is validated, assuming that specific cognitive intervention is a necessary element for the treatment of childhood anxiety disorders also remains tenuous. Much work remains to be done.