Test-Retest Reliability of Common Behavioral Decision Making Tasks.

Abstract

To examine test-retest reliability of common behavioral decision making tasks. A total of 98 undergraduate students completed two administrations of the Iowa Gambling Task (IGT), Balloon Analogue Risk Task (BART), Columbia Card Task (CCT), and Game of Dice Task (GDT), three weeks apart. The BART, CCT, and GDT showed moderately strong correlations across time. On the IGT, no correlations were seen between Time 1 Trials 1-40 and Time 2 performance; however, weak correlations were observed between Time 1 Trials 41-100 and Time 2 performance. Paired-samples t-tests indicated participants were riskier at Time 1 than Time 2 on the IGT and GDT, but riskier at Time 2 on the BART. The BART, CCT, and GDT showed moderate test-retest reliability, with the IGT showing weak reliability during the decision making under risk trials only. Implications for repeated test administration in clinical and non-clinical settings are discussed.