Test-retest reliability of affective bias tasks

• DOI: 10.31234/osf.io/n2fkh
• Publication Date: 2022-04-07T12:26:00Z
• Authors: Alexandra Claire ...; Katrina Tan; Hoda Jaber Ansari; Michelle Wing; Oliver Joe Robinson
• Keywords: Attentional Bias; Response bias; Cognitive bias modification

BackgroundAffective biases are commonly seen in disorders such as depression and anxiety, where individuals may show attention towards more rapid processing of negative or threatening stimuli. Affective have been shown to change with effective intervention: randomized controlled trials into these the mechanisms that underpin them allow greater understanding how interventions can be improved their success maximized. For informative, we must reliable ways measuring affective bias over time, so detect changing biases. In particular, test-retest reliability our measures puts an upper bound on ability effects: thus, this study, examine two behavioural tasks bias. MethodsWe recruited 58 online study who completed twice, at least 14 days between sessions. We analysed both summary statistics parameters from computational models using Pearson’s correlations intra-class correlations. ResultsStandard statistic had ranging 0.18 (poor) 0.49 (moderate). Parameters modelling were many cases less than statistics. Embedding covariance sessions within generative framework resulted higher stability estimates. ConclusionsIn sum, moderately reliable, but further work improve would still draw inferences trials.