Empirical Status of the Cognitive Model 97
to engage in deliberate mental control of unwanted thoughts. Positive correlations have
been reported between the WBSI and various self- report measures of anxiety as well
as measures of obsessionality (e.g., Rassin & Diepstraten, 2003; Wegner & Zanakos,
1994). Moreover, scores on the WBSI are significantly elevated in all the anxiety dis-
orders but then decline in response to effective treatment (Rassin, Diepstraten, Merck-
elbach, & Muris, 2001). A factor analytic study of the WBSI, however, found that an
unwanted intrusive thoughts rather than thought suppression factor correlated with
anxiety and OCD symptoms (Höping & de Jong-Meyer, 2003). Nevertheless, other
clinical studies have indicated that thought suppression is evident in the anxiety disor-
ders. Harvey and Bryant (1998a) found that survivors of motor vehicle accidents with
acute stress disorder (ASD) had higher ratings of natural thought suppression than sur-
vivors without ASD. A study of women who experienced a pregnancy loss revealed that
a tendency to engage in thought suppression predicted PTSD symptoms at 1 month and
4 months postloss (Engelhard, van den Hout, Kindt, Arntz, & Schouten, 2003). Overall
these findings indicate that thought suppression is a coping strategy that is very often
employed by those who are suffering from anxiety.
NEgATIvE EFFECTS oF THoUgHT SUPPRESSIoN
It appears that individuals with an anxiety disorder are as effective as nonclinical or
low anxious individuals in suppressing anxious target thoughts, at least in the short
term (Harvey & Bryant, 1999; Purdon, Rowa, & Antony, 2005; Shipherd & Beck,
1999), although there are other studies that indicate less efficient suppression by diag-
nostically anxious individuals (Harvey & Bryant, 1998a; Janeck & Calamari, 1999;
Tolin, Abramowitz, Przeworski, & Foa, 2002a). Moreover, the experimental evidence
is inconsistent in whether suppression of anxious thoughts such as worries, obsessional
intrusive thoughts, or trauma- related intrusions is more likely to result in postsuppres-
sion rebound. Some studies have reported rebound effects with anxious and obsessional
target thoughts (Davies & Clark, 1998a; Harvey & Bryant, 1998a, 1999; Koster, Ras-
sin, Crombez, & Näring, 2003; Shipherd & Beck, 1999), whereas others have generally
failed to find any rebound suppression effects (Belloch, Morillo, & Giménez, 2004a;
Gaskell, Wells, & Calam, 2001; Hardy & Brewin, 2005; Janeck & Calamari, 1999;
Kelly & Kahn, 1994; Muris, Merckelbach, van den Hout, & de Jong, 1992; Purdon,
2001; Purdon & Clark, 2001; Purdon et al., 2005; Roemer & Borkovec, 1994; Rutledge,
Hollenberg, & Hancock, 1993, Experiment 1). Generally, it appears that postsuppres-
sion rebound of anxious thoughts is no more or less likely in clinically anxious samples
than in nonclinical individuals (see Shipherd & Beck, 1999, for contrary findings).
Even though an immediate postsuppression resurgence of unwanted thought intru-
sions has not been consistently supported, there is evidence that suppression of anxious
thoughts may have other negative effects that are important to the persistence of anxi-
ety. First, it appears that over a longer time period, such as a 4- or 7-day interval, previ-
ous suppression of anxious targets will result in a significant resurgence of unwanted
thoughts (Geraerts, Merckelbach, Jelicic, & Smeets, 2006; Trinder & Salkovskis, 1994).
Abramowitz et al. (2001) suggested that individuals can successfully suppress unwanted
thoughts over short time periods, but as time progresses and individuals relax their
control efforts, a resurgence of target thought frequency is more likely. Second, suppres-
sion does appear to have a direct negative effect on mood, causing anxious and depres-