Introduction
Epidemiological research has shown that a substantial proportion of children and adolescents suffer from psychiatric disorders [
1‐
3]. The most frequently diagnosed disorders can be roughly divided in two categories: internalizing problems (e.g., anxiety disorders, depression) and externalizing problems (e.g., disruptive behavior disorders) [
4]. Only a small percentage of the children with problems actually receive services, and so many of these untreated children go on to become adults with psychiatric disorders [
5]. During the last decades, clinical psychology researchers have increasingly accepted the notion that many psychiatric problems in adults have their origins during childhood, resulting in an intensification of studies on the etiology of childhood psychiatric disorders [
6]. This research has made clear that childhood disorders do not have their origins in a single factor, but originate from the dynamic interplay of multiple factors. Most studies have focused on vulnerability and risk factors such as reactive temperament, negative learning experiences, stressful life events, and adverse family factors, which are all thought to increase children’s proneness to develop psychiatric problems and disorders [
7‐
9]. Fortunately, there are also protective influences, which may serve to shield children and adolescents against the development of psychopathological symptoms.
One protective variable that is currently receiving an increasing amount of research attention is effortful control. Briefly, effortful control can best be defined as the ability to inhibit a dominant response to perform a subdominant response, and refers to self-regulative processes that enable a person to control or regulate behavior under certain circumstances [
10,
11]. Effortful control is generally thought to consist of two main components: inhibitory control, which pertains to the ability to inhibit one’s behavior if necessary, and attention control, which can be defined as the ability to focus and shift attention when needed. It is generally assumed that the capacity for effortful control processes is innate [
12]. Relatively few data exist on the temporal stability of this regulative temperament factor, but available evidence has revealed reasonable stability from infancy through preschool and into the early school years [
13,
14], and so it can be concluded that effortful control has trait-like qualities [
10]. Meanwhile, it is also clear that this regulative factor further develops as a result of brain maturation, especially in the frontal lobes, in interaction with the environment [
15,
16]. The gradual improvement of effortful control increasingly enables children to regulate emotions and to control their behavior, which may have positive effects on their social interactions with other children [
17,
18]. In children who have little effortful control by nature or who fail to adequately develop this regulative trait, such normal processes can be disturbed and this may increase their vulnerability to develop psychopathological symptoms.
Various studies have demonstrated that low levels of effortful control are associated with higher levels of internalizing and externalizing symptoms. For example, in an investigation by Eisenberg and colleagues [
19], parents and teachers completed rating scales for measuring effortful control in a sample of 4- to 8-year-old children. Some of these children displayed internalizing or externalizing problems, whereas other children did not manifest any psychological problems. Results showed that children with internalizing problems and children with externalizing problems scored relatively low on effortful control as compared to children without psychological problems. Other studies have yielded comparable results: that is, low levels of effortful control are associated with higher levels of psychopathological symptoms [
20‐
22], although the link with externalizing symptoms generally appeared stronger than that with internalizing symptoms [
23‐
25]. However, the latter finding may be primarily due to the fact that the instruments that have been used in the above-mentioned studies were somewhat biased to the assessment of inhibitory control. Research employing measures that tap aspects of attention control has demonstrated that internalizing problems such as anxiety and depression are also convincingly associated with low levels of effortful control [
26‐
31]. In their recently formulated theory on the role of temperament in the etiology of childhood psychopathology, Muris and Ollendick [
32] assume that different aspects of effortful control are allied to specific types of symptoms. More precisely, lack of attention control seems more strongly linked to internalizing symptoms, whereas a deficiency of inhibitory control is more clearly related to externalizing symptoms. Note that these differential relations are in keeping with the clinical observation that internalizing disorders are typically characterized by uncontrollable negative thought, while externalizing disorders are frequently marked by impulsive and disinhibited behavior [
33].
Research on the psychopathological correlates of effortful control has predominantly relied on self-report scales for measuring this regulative trait. Most studies have employed subscales of temperament questionnaires such as the Child Behavior Questionnaire [
34] and the Early Adolescent Temperament Questionnaire [
35], or scales that have been derived from these instruments such as the Attention Control Scale (ACS) [
36] and the Effortful Control Scale (ECS) [
29]. Only a number of investigations have relied on behavioral batteries [
13,
37], in spite of the fact that effortful control shows strong resemblance to what neuropsychologists refer to as ‘executive functions’, which makes that this regulative factor can be readily assessed by means of cognitive performance tests that tap children’s capacity of governing their attention and controlling their behavior. This research has generally confirmed the notion that effortful control is especially associated with externalizing problems, although it should again be noted that most performance tests mainly focus on the control of behavior (inhibitory control) and largely neglect the aspect of attention control. Interestingly, Manly and colleagues [
38,
39] have developed the Test of Everyday Attention for Children (TEA-Ch), which primarily focuses on attention processes and includes various tasks that call on focusing, sustaining, and switching attention.
The current study further examines the link between the regulative trait of effortful control, and in particular attention control, and psychopathological symptoms in children. For this purpose, a large sample of non-clinical children aged 8 to 12 years were tested with the TEA-Ch and completed self-report scales for measuring attention control (ACS) and effortful control (ECS) as well as symptoms of anxiety, depression, and aggression. In this way, we investigated (1) the relationships between TEA-Ch scores and self-report indices of attention control and effortful control, and (2) the links between attention control and effortful control as measured by the TEA-Ch and self-report questionnaires, on the one hand, and psychopathological symptoms in children, on the other hand. It was anticipated that behavioral (TEA-Ch) and self-report (ACS and ECS) indices of attention and effortful control would be positively related. Further, we expected to find negative associations between indices of attention and effortful control (TEA-Ch, ACS, and ECS) and measures of anxiety, depression, and aggression, which of course would provide further support for the notion that these types of psychopathological symptoms are accompanied by lower levels of this regulative trait.
Discussion
The present study examined the relationships between self-report and performance-based measures of attention and effortful control and psychopathological symptoms in a sample of 8- to 12-year-old non-clinical children. The main results can be catalogued as follows. First of all, correlations between self-reports of attention and effortful control and the TEA-Ch were not convincing. In fact only, the ACS was significantly related to the TEA-Ch total score and various of its subtests. Further, self-reported attention and effortful control (ACS and ECS) were clearly negatively related to psychopathological symptoms. However, the correlations between the TEA-Ch and symptom scores were very small and even non-significant after controlling for demographic variables (i.e., gender and age).
The data indicate that there is little overlap between self-report and performance-based indexes of attention and effortful control. Only the ACS displayed some modest but significant associations with various aspects of attention performance as indexed by the TEA-Ch. The finding that the ACS was a better indicator of TEA-Ch test performance than the ECS was not that surprising given the fact that the ACS and the TEA-Ch are both primarily measures of attention control [
36,
38], whereas the ECS attempts to measure attention as well as inhibitory control [
29].
A number of explanations for the observation that self-report and performance-based indexes had so little in common can be put forward. First of all, it may well be the case that the psychometric properties of the ACS and ECS are insufficient and hence the scales might be poor indicators of the regulative traits of attention and effortful control. This idea is partly supported by the rather modest reliability coefficients that were found for both scales. It has been noted that this is probably due to a number of reversed items that are included in these scales [
48,
49]. While it is clear that such items are incorporated to cancel out response tendencies, several authors have indicated that they might undermine the psychometric properties of a scale and this might be especially true in child populations [
50]. Second, it is also possible that the ACS and the ECS do not really assess children’s actual levels of attention and effortful control, but rather reflect children’s personal beliefs about how well they can control cognitive processes. If true, this would seriously question the validity of these scales. However, there is some evidence showing that self-reported attention control are actually predictive of children’s performance in situations that call upon this regulative trait. That is, Muris [
40] found that child- and parent-ratings of attention control (as indexed by the ACS) were predictive of teacher-reported school performance. A third possibility is that TEA-Ch scores do not provide a valid indication of children’s level of attention and effortful control. Although previous research has provided evidence for the validity of this test, as evidenced by significant correlations with measures of intelligence, other measures of attention, and school performance [
38,
39], it should be mentioned that the present study only employed five subtests of the TEA-Ch. Perhaps the administration of the full battery is necessary to really put load on children’s capacity for controlling attention processes. A final, related point pertains to Epstein’s [
51,
52] observation that performance-based assessments are only predictive of traits, if the pertinent behavior is measured in multiple tasks for which scores are combined. The present study used five subtests of the TEA-Ch, but it may well be that this number was too small to get a valid impression of children’s level of attention control.
The links between self-reported attention/effortful control and psychopathological symptoms in children were as anticipated. That is, significant negative correlations were found between scores on the ACS and ECS and levels of anxiety, depression, and aggression [
26‐
31]. Further, the pattern of these correlations was also comparable to that observed in previous studies and indicates that attention control is more convincingly linked to internalizing symptoms (i.e., anxiety and depression) than to externalizing symptoms (i.e., aggression) [
49]. Unfortunately, the present study did not include a pure scale for measuring inhibitory control, which seems to be the regulative trait most clearly related to externalizing symptoms [
32,
53]. The correlations between the TEA-Ch and psychopathological symptoms were small, and even non-significant after controlling for demographic variables (i.e., gender and age). On the one hand, this finding provides further support for the aforementioned notion that self-reports and performance-based measures of attention/effortful control do not assess fully similar constructs. Perhaps scales such as the ACS and ECS not only measure attention/effortful control capacity, but also tap some type of meta-cognition, namely the idea of children that they have less control over attention and other cognitive processes. Interestingly, several authors have noted that children with high levels of psychopathological symptoms report to experience less control over their emotions [
54,
55], and it may well be that this is the result of children’s lowered perceptions of control over various cognitive processes. On the other hand, it should be borne in mind that the TEA-Ch was administered under low-stress conditions. It may well be that children’s problems with regulating negative emotions only become really manifest when their cognitive system is seriously challenged (e.g., by exposing them to a stressful situation). This implies that TEA-Ch scores would only be related to psychopathological symptoms under stressful or taxing circumstances, which is a possibility that can be tested in further research. Note that this comes close to a diathesis-stress account of psychopathology, which implies that emotional and behavioral problems only develop in vulnerable children (i.e., children with high levels of negative affectivity and/or poor regulative skills) when they are confronted with stressful or challenging events [
56].
There are a number of additional findings that require some brief comment. To begin with, there were significant gender differences for symptoms of anxiety and aggression. More precisely, girls reported higher levels of anxiety symptoms, whereas boys reported higher levels of aggression. This finding is in keeping with previous research indicating that girls tend to display more internalizing problems, while boys are inclined to exhibit more externalizing symptoms [
57]. Further, consistent age effects were found on indexes of attention control. In all cases, the correlations were positive: this indicates that attention control improved as children were older, which is of course in keeping with the notion that this regulative trait further develops during middle childhood [
16].
Admittedly, the current study suffered from various limitations. First, the study only obtained child-reported ACS and ECS data. Although substantial parent-child correlations have been obtained for these measures [
31], the inclusion of parent data would certainly have provided additional information and further strengthened our study. Second, fairly little is known on the psychometric properties of the instruments that are currently used for measuring attention and effortful control. More research in this area may yield more reliable and valid scales and tests to assess these regulative temperament factors [
48]. Third, the study relied on non-clinical children who were all reasonably well-functioning, and hence displayed fairly good performance on the TEA-Ch. It may well be that more variation in TEA-Ch scores can be found in a sample of clinically referred youths [
47], which of course is important when one attempts to investigate relationships with psychopathology symptoms. In spite of these shortcomings, the current data provide interesting information on the assessment of attention/effortful control, and on the relation between these regulative traits and psychopathological symptoms in youths. As noted by Hughes and Graham [
58], these issues remain important points on the research agenda of clinical and developmental psychologists in the near future.