Cognitive approaches to psychopathology assert that an individual's fe
elings and behaviours are determined by the way they perceive and stru
cture their experiences. Beck's (1976) cognitive model of anxiety prop
oses that pathological anxiety results from unrealistic perceptions of
the danger inherent in a given situation. This elevated sense of dang
er is presumed to occur through one or more of the following cognitive
errors: a) overestimating the probability (risk) of a dangerous event
; b) overestimating the severity (threat) of a feared event; c) undere
stimating coping resources; or d) underestimating rescue factors, Stud
ies of individuals with a range of anxiety disorders confirm the prese
nce of threat and danger related cognitions. However, there is little
empirical research to show that the cognitive errors described above a
re associated with anxiety. The results from two studies examining the
role of risk and threat perception in anxiety are presented. Both stu
dies used the Risk Perception Questionnaire (RPQ) to assess risk and t
hreat perception of positive and negative life events and daily uplift
s and stressors. In the first study university students and community
residents completed the RPQ and the State Trait Anxiety Inventory (STA
I). Correlations between the two measures showed increased anxiety was
associated with decreased expectations of positive life events and da
ily uplifts and increased risk perception for negative life events and
daily stressors. Increased threat scores (risk X cost) for negative l
ife events and daily stressors were associated with increased anxiety
in the community subjects. Mean scores on the four risk perception sub
scales and the two negative threat scales differed significantly betwe
en subjects with high and low levels of trait anxiety. In the second s
tudy, risk and threat perception scores for subjects with DSM-III-R an
xiety disorders were compared with those of a non-anxious clinical sam
ple and community subjects. Increased risk perception of negative life
events differentiated the anxious from the community subjects (non-an
xious clinical subjects showed a nonsignificant elevation on this scal
e). Threat ratings of negative life events discriminated the two clini
c groups from the community subjects. Increased risk perception for da
ily stressors was unique to the anxiety disordered group. Results from
the second study suggest increased risk perception of daily stressors
may be unique to individuals suffering from anxiety disorders, while
increased risk and threat perception of negative life events may be as
sociated with psychopathology generally. The implications of these fin
dings for Beck's model are discussed.