Thinking high but feeling low: An exploratory cluster analysis investigating how implicit and explicit spider fear co-vary

ABSTRACT

Research has demonstrated large differences in the degree to which direct and indirect measures predict each other and variables including behavioural approach and attentional bias. We investigated whether individual differences in the co-variance of “implicit” and “explicit” spider fear exist, and whether this covariation exerts an effect on spider fear-related outcomes. One hundred and thirty-two undergraduate students completed direct and indirect measures of spider fear/avoidance, self-report questionnaires of psychopathology, an attentional bias task, and a proxy Behavioural Approach Task. TwoStep cluster analysis using implicit and explicit spider fear as criterion variables resulted in three clusters: (1) low explicit/low implicit; (2) average explicit/high implicit; and (3) high explicit/low implicit. Clusters with higher explicit fear demonstrated greater disgust propensity and sensitivity and less willingness to approach a spider. No differences between clusters emerged on anticipatory approach anxiety or attentional bias. We discuss results in terms of dual-systems and cognitive–behavioural models of fear.

KEYWORDS:

• Anxiety
• spider fear
• attention bias
• dual-systems model
• implicit associations

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCiD

Adam S. Radomsky http://orcid.org/0000-0003-2807-184X

Notes

1. We use the terms associative/rule-based, consistent with Ouimet et al. (2009) and Sloman (1996). However, different models use different terms, such as impulsive/reflective, associative/propositional, and reflexive/reflective.

2. We did observe an overall main effect of stimulus presentation duration, such that overall, participants were faster to respond following a 600-ms duration than following a 400- ms duration. However, there was no interaction between duration and bias type, suggesting that this pattern reflects faster responding because of more time to process the stimulus, rather than differences in attention bias, per se. Moreover, when we examined the effects of cluster membership on attention bias at different presentation durations, there was no main effect or interaction. As such, we presented the collapsed data for brevity.

Additional information

Funding

This study was supported by the Natural Sciences and Engineering Research Council of Canada (Discovery Grant # RGPIN 249833 2007).