The Neuropsychology of Anxiety Disorders:
Affect, Cognition, and Neural Circuitry
Jack B. Nitschke and Wendy Heller
In attempting to construct a neuropsychology of anxiety, findings can be drawn from several related, yet often perceived as separate, domains of research, including cognitive science and neuroscience. The relatively new fields of cognitive neuroscience and affective neuroscience are concerned with very similar questions regarding brain-behaviour relationships as were fundamental to the older field of neuropsychology, and the neuroimaging tools central to those disciplines are no less pertinent to neuropsychology than are traditional neuropsychological test batteries or cognitive/behavioural paradigms. Thus, this review of the neuropsychological findings in anxiety disorders covers a wide array of methods that together inform knowledge of the brain mechanisms involved in the circuitry governing pathological forms of anxiety.
Although often overlooked by neuroscientists studying brain function in anxiety, cognitive research over the past two decades has contributed substantially to knowledge about brain function in anxiety. A large body of work demonstrates that anxiety disorders are characterized by cognitive biases, indicating a heightened response to the possibility of threat (for review, see McNally, 1998). Attentional biases have been elicited very reliably across a variety of paradigms in which potentially threatening information is associated with greater attentional capture in individuals with anxiety disorders than in controls. The interference of this attentional capture with other cognitive processing serves as the operationalization of this bias in research studies. Furthermore, attentional biases have been found to disappear upon remission (for review, see McNally, 1998), suggesting that such biases are state-dependent. Cognitive biases have also been observed in the form of interpretation and memory biases. Across a number of different paradigms involving ambiguous stimuli that can be interpreted as threatening or neutral, anxious people choose the threatening meaning. Accruing evidence suggests that anxiety disorders are also accompanied by enhanced memory for negative or threatening information under certain conditions. These cognitive data suggest dysfunctional activation of a right hemisphere system involved in threat perception (for review, see Nitschke, Heller and Miller, 2000; see also Compton et al., 2000, 2002).
In addition to these cognitive biases, cognitive deficits have been documented in anxiety disorders. One is a tendency to do poorly on tasks that require selective attention and concentration. This deficit has been suggested to reflect a general problem of preoccupation and distraction due to worry or rumination that interferes with other mental processes (for review, see Nitschke, Heller and Miller, 2000). Compromised visual-spatial functioning has also been reported. In addition, individuals with posttraumatic stress disorder often exhibit deficits in explicit memory. Taken together, these cognitive deficits suggest aberrant frontal, anterior cingulate, right parietal, and hippocampal functioning. Building on this cognitive research as well as on behavioural and electroencephalographic (EEG) findings (for review, see Nitschke, Heller and Miller, 2000) and an extensive literature in non-human animals examining fear and anxiety (for reviews, see LeDoux, 1996; Davis and Lee, 1998), haemodynamic neuroimaging research has implicated a number of the suggested regions.
Although emotional, cognitive, and neural commonalities are apparent, the diversity of findings also warrants the importance of respecting unique patterns and heterogeneity both among and within the various anxiety disorders. An observation that has become increasingly salient in the burgeoning neuropsychological literature on anxiety and its disorders is the lack of clarity and specificity about what anxiety is. Views of anxiety range from its usage in contemporary clinical research as a rubric term that encompasses fear, panic, worry, and all the anxiety disorders listed in the DSM-IV to its very specific operationalization referring to context conditioning and long-term sensitization (e.g., Davis and Lee, 1998) to a more generic personality dimension closely linked to neuroticism (e.g., Gray, 1982). Further, the heterogeneity within each of the different anxiety disorders has become increasingly apparent and represents a major problem for investigators attempting to uncover the neurobiological correlates of individual anxiety disorders. Inconsistencies across studies may be explained by the fact that anxiety is not a unitary phenomenon and that different types and symptoms of anxiety are associated with particular cognitive patterns (Heller and Nitschke, 1998; Nitschke, Heller and Miller, 2000). An important mission of neuroscience research in this area is to help unravel the inchoate notions of anxiety that currently exist. Thus, although it is important to look for generalizations regarding the neural mechanisms of anxiety, it is also necessary to consider the possibility of heterogeneity by being as specific as possible regarding the disorder or type of anxiety under investigation.
The aim of this chapter is to assess what is known about the neuropsychology and neural circuitry of anxiety disorders by examining the relevant cognitive research. Structural and functional neuroimaging data will also be reviewed, including morphometric magnetic resonance imaging (MRI), functional MRI (fMRI), positron emission tomography (PET) using various radiotracers such as [18F]fluorodeoxyglucose (FDG) for glucose metabolism and 15O-labelled water for blood flow, single-photon emission computed tomography (SPECT) with l33Xenon or 99mTc-HMPAO, and scalp-recorded EEG. This review of the cognitive and neuroscience literatures reveals that the anxiety disorders engage brain regions involved in threat perception (e.g., right hemisphere