The combined cognitive bias hypothesis in anxiety: A systematic review and meta-analysis

Chantel J. Leung, Jenny Yiend*, Antonella Trotta, Tatia M.C. Lee

*Corresponding author for this work

Research output: Contribution to journalReview articlepeer-review

6 Citations (Scopus)


Cognitive theories have postulated the relational nature of different cognitive biases in the development and maintenance of anxiety disorders. To test this combined cognitive bias hypothesis, this review addressed the following questions: (i) whether different cognitive biases are associated with each other and (ii) whether one bias influences another bias. We identified 36 articles that studied the relationship between cognitive biases (attention, interpretation and memory bias). Of these, 31 studies were entered into two meta-analyses. Sixteen studies were included in the first meta-analysis of the correlation between cognitive bias indices. A further 15 studies were included in another meta-analysis to examine the transfer effects of cognitive bias modification (CBM) to another bias. Both meta-analyses yielded small but significant overall pooled effect sizes after the removal of outliers (r = 0.11 and g = 0.19 respectively). Moderator analyses revealed that the relationship between interpretation and memory bias was significantly stronger than other types of cognitive bias correlations and CBM is more potent in modifying biases when it was delivered in the laboratory compared with online. Our review quantifies the strength of the relationships between biases and transfer effects following CBM, which serves as a basis to further understand the mechanisms underlying biased information processing.

Original languageEnglish
Article number102575
JournalJournal of Anxiety Disorders
Publication statusPublished - Jun 2022


  • Anxiety
  • Attention bias
  • Combined cognitive bias hypothesis
  • Interpretation bias
  • Memory bias


Dive into the research topics of 'The combined cognitive bias hypothesis in anxiety: A systematic review and meta-analysis'. Together they form a unique fingerprint.

Cite this