86 COGNITIVE THEORY AND RESEARCH ON ANXIETY
seConDary elaborative reappraisal:
the state of anxiousnessHypothesis 8. Facilitated Threat Elaboration
A selective threat bias will be evident in explicit and elaborated cognitive processes such that
in anxiety memory retrieval, outcome expectancies, and inferences to ambiguous stimuli will
show a preponderance of threat- related themes relative to nonanxious individuals.
As discussed in Chapter 2, the cognitive model of anxiety postulates that a second-
ary, compensatory stage of information processing occurs in response to threat mode
activation (i.e., the immediate fear response). Whereas the earliest moments of anxiety
are dominated by automatic processes that characterize primal threat mode activation,
the later secondary phase primarily involves deliberate and effortful processing that
reflects a conscious strategic approach to anxiety reduction.
The secondary elaborative phase plays a primary role in the persistence of anxiety.
In fact, most cognitive- behavioral interventions of anxiety focus on change at this elabo-
rative phase. The modification of effortful cognitive processing can lead to a significant
reduction even in the more automatic aspects of fear activation. In his review Mansell
(2000) presented clinical and experimental evidence that conscious interpretations can
have a significant positive or negative impact on the automatic processes involved in
anxiety. Psychological intervention that effectively reduces anxious symptoms has been
shown to also lessen automatic attentional bias for threat (see MacLeod, Campbell,
Rutherford, & Wilson, 2004). Nevertheless, we consider conscious effortful informa-
tion processing that involves making judgments, generating expectancies, evaluating or
appraising information, reasoning and decision making, and explicit memory retrieval
important aspects of the threat- biased cognitive architecture of anxiety. As evident from
the review below, there has been much debate in the research literature on the role of
elaborative, strategic processing in anxiety.
Threat- Biased Interpretations
A variety of experimental tasks have been employed to determine if anxious individuals
exhibit a greater tendency to make biased threat- related judgments than nonanxious
individuals. In some studies threat and nonthreat words were presented but evidence for
a clear preference for threat was mixed (e.g., Gotlib et al., 2004; Greenberg & Alloy,
1989). More consistent findings emerged from emotional priming experiments in which
participants are shown positive and negative trait adjectives preceded by a positive or
negative sentence prime. In these studies GAD and panic patients exhibited a preferen-
tial response to primed threat stimuli (e.g., D. M. Clark et al., 1988; Dalgleish, Cam-
eron, Power, & Bond, 1995).
Biased judgment is more accurately investigated with experimental paradigms that
present threatening and nonthreatening ambiguous stimuli, with the prediction that
anxious individuals will endorse the more threatening interpretation. Ambiguous tasks
are more sensitive to evaluation biases because they allow for the possibility of gener-
ating alternative interpretations that vary in their aversiveness (MacLeod, 1999). One
experimental paradigm used to investigate interpretation bias involves auditory presen-
tation of homophones, which are words with identical pronunciation but distinct spell-