Anxiety is characterised by apprehension or worrisome thinking when anticipating a future threat or negative outcomes. It is suggested to be contributed by three interconnecting components: cognition (e.g., worry), affect (e.g., nervousness), and behaviour (e.g., avoidance; Handelzalts & Keinan, 2010). Anxiety may be considered an inevitable part of life and may actually help the individual to prepare for potential future dangers. However, anxiety might become dysfunctional when it is prolonged or becomes more intense than what would be normally expected.
Anxiety is one of the most prevalent mental health concerns in children. According to a recent survey, approximately 20.7% of primary students and 30% of secondary students in Hong Kong experienced some kind of anxiety symptoms (HKSAR Social Welfare Department, 2015). The most common themes of worry as reported by children were found to be school, health and personal harm (Silverman, Greca, & Wasserstein, 1995). In particular, they were often worried about relationships with the family and peers. It has been suggested that the presentation of anxiety may differ between children and adults, due to differences in daily environments and developmental needs (March, Parker, Sullivan, Stallings, & Conners, 1997). For instance, school phobia and separation anxiety (anxiety experienced upon separation from emotionally attached figures) are more common in children than adults (Last, Strauss & Francis, 1987).
Childhood anxiety can have significant academic and psychosocial implications, including school withdrawal, perceived self-competence, and peer relations (Van Ameringen, Mancini, & Farvolden, 2003; Messer & Beidel, 1994). It may also be a predictor of adult depression, substance abuse and other emotional behavioural problems (Bittner et al., 2007). Therefore, it would be important to examine the vulnerabilities that contribute to anxiety symptoms so as to identify possible sources of intervention. As Bronfenbrenner (1979) noted, human development is influenced by the interactions between the individual and different environmental systems. As a result, anxiety may be contributed by the interaction between the child’s personal characteristics and his/her surrounding systems. Two of the major factors are cognitive control/executive functioning and parenting.
The level of executive functioning (EF) has been posited to be related to anxiety. EF is often considered as an umbrella term for higher-order cognitive processes that help coordinate goal-directed behaviour (Banich, 2009). Miyake et al., (2000) proposed the “unity/diversity executive functioning” model as they found that three of the commonly postulated EF abilities were interrelated yet clearly separable. The three EF abilities were mental set-shifting (flexibly switching attention or between goals), updating working memory (monitoring and revising the content of working memory), and inhibition of prepotent responses. EF is believed to be important for self-regulation of behaviour as they help individuals adaptively modify their behaviour in response to changes in environmental demands. Therefore, deficits in EF may increase the exposure to stress and hence increased vulnerability to anxiety.
Individuals with anxiety symptoms have often been found to experience EF difficulties. Studies have found that the inhibition and cognitive shifting abilities among individuals with a high level of anxiety were weaker than those with a low level of anxiety, as indicated by various experimental paradigms, such as the classic Stroop task, Wisconsin Card Sorting Test and Cognitive Assessment System (Toren et al., 2000; Visu-Petra, Miclea, ; Visu-Petra, 2013). Furthermore, among children and adolescents, anxiety has been found to be related to dysregulated brain activity in the prefrontal cortex which is commonly shown to be associated with EF control (Fitzgerald et al., 2013; Sylvester et al., 2013).
Recently there has been a growing number of studies suggesting that EF might actually play an important role in the development and maintenance of anxiety. As anxiety is characterised by worrisome thinking which is considered as repetitive negative thoughts or images that are usually intrusive and difficult to control (Silverman et al., 1995), reduced inhibitory control may provide a gateway for unwanted negative thoughts, according to the gateway mechanism proposed by De Raedt and Koster (2010). Deficits in shifting and cognitive flexibility may result in the experience of “getting stuck” on the thoughts, thus leading to the maintenance of negative repetitive thinking (Sharp, Miller, ; Heller, 2015). Low attentional control might also undermine emotion regulation processes and perceived controllability, and this has been shown to increase anxiety (Opitz, Gross, ; Urry, 2012; Mineka ; Kelly, 1989). Birk, Opitz and Urry (2017) found that subjective anxiety and autonomic arousal were strongly associated with the participants’ attentional control capacities. Hendrawan, Yamakawa, Kimura, Murakami and Ohira (2012) even found that the participants’ performance on EF tests predicted their physiological stress reactivity as measured by cortisol level, heart rate and skin conductance during an acute psychological stress exposure. Apart from cross-sectional studies, evidence in support of the relationship between EF and anxiety is provided by longitudinal research. For example, Kertz, Belden, Tillman and Luby (2016) found that deficits in inhibition and shifting among five-year-olds were associated with increased anxiety level after 3.5 to 7.5 years, with the influence of other factors being controlled for, such as IQ, gender and the anxiety level prior to the study. However, these prospective studies among the school population have been mostly done in western cultures and it remains unknown whether the relationship between EF and childhood anxiety also holds in other cultures such as in East Asians.
Many tasks have been identified to assess EF, such as the Wisconsin Card Sorting Test (WCST; Heaton, Chelune, Talley, Kay, ; Curtiss, 1993), Tower of London (ToL; Culbertson ; Zillmer, 2005), the Letter Memory Task (Morris ; Jones, 1990), and the Stop-signal Task (Logan, 1994). Among the EF measures, the WCST and ToL have been found to be the most frequently used in research and clinical practice in the field of neuropsychology (Etnier ; Chang, 2009).
The WCST is a classical neuropsychological task that was originally developed to assess cognitive strategies in the normal population. It has then gained popularity in clinical settings to study cognitive dysfunction in patients (Rabin, Barr, ; Burton, 2005). In the WCST, participants are required to sort cards based on one of the three characteristics of the card, in which the sorting rule changes regularly and has to be inferred from the examiner’s feedback. It has been primarily conceptualised as a set-shifting task because participants are required to flexibly switch between sorting strategies (Berg, 1948; Puente, 1985). Although some researchers have hypothesised that it also assesses other EF skills such as inhibition and updating (Ozonoff & Strayer, 1997), path analyses showed that set-shifting was the single most significant ability tested by the WCST (Miyake et al., 2000). In particular, the number of perseverative errors made on the WCST has been suggested to directly indicate an individual’s shifting ability or cognitive flexibility (Miyake et al., 2000; Robinson, Heaton, Lehman, ; Stilson, 1980), and therefore, it has been purported to be the best metric of EF from the WCST (Rhodes, 2004).
The ToL is another classical EF task that has been used extensively in both normal and clinical populations (Kim, woo Yi, Jung, ; Nam, 2017; Chang et al., 2011). It requires the participants to move the beads one-by-one from one peg to another so as to achieve their target positions on the “tower” as quickly as possible using the least number of moves. The ToL has often been described to measure executive planning and problem-solving abilities (Chang et al., 2011) because the participants have to anticipate the consequences of action and the transformation required from the initial to goal state beforehand. However, it has been suggested that the participants may tend to rely on the perceptual strategy spontaneously when making the next move, which involves considering whether the current state is perceptually closer to the target, instead of the more demanding strategy with careful planning and extensive goal management (Goel ; Grafman, 1995; Miyake et al., 2000; Murji ; DeLuca, 1998). Therefore, the ToL may place greater emphasis on inhibition than planning, as one may need to inhibit his tendency of making salient and perceptually congruent yet incorrect moves.