Reversing the alternate uses task

Divergent thinking tests have been widely used to identify creative potential and have demonstrated meaningful predictive power. However, these measures are often criticized for employing unrealistic tasks that might not elicit creative thinking processes similar to real-world creative problems. The goal of the present study was to construct and validate a new measure of divergent thinking, the Reverse Alternate Uses Test (rAUT), by reversing the prompts of the Alternate Uses Test (AUT) and asking participants to generate objects for mundane uses rather than surprising uses for objects in order to make the items more goal-driven. The sample consisted of 240 undergraduate students who responded to the rAUT. We aimed to examine whether the originality and elaboration components of the new instrument can be reliably scored by an automatic artificial intelligence scoring system (Ocsai). Nine items of rAUT have shown good internal consistency reliability for originality, fluency, and elaboration scores (α = 0.88, α = 0.89, α = 0.88, respectively). Automatic originality scoring system demonstrated high levels of inter-rater reliability with three human raters (α =0.947). To examine the internal validity and latent structure of the rAUT vis-à-vis the AUT, we applied a six-factor confirmatory factor model. The model indicated a good fit, with CFI = 0.907 and RMSEA = 0.05. In line with our predictions, rAUT elicited statistically more original and less elaborated responses than AUT. In line with our predictions, rAUT elicited statistically more original and less elaborated responses than AUT. Noteworthy, time-on-task also showed a statistically higher correlation with originality scores on rAUT, than AUT.