Functional consistency in the face of topographical change in articulated thoughts

Abstract

Studies of verbal reports have always assumed that thoughts verbalized on a given occasion are not the only ones that could have been expressed. This poses a potential problem for a recently developed thinkaloud procedure called Articulated Thoughts in Simulated Situations (ATSS). To address this issue, comparisons were made between the articulated thoughts of subjects over two presentations of a simulated stressful event in the ATSS procedure. Articulated thoughts were analyzed for similarity across presentations both topographically for surface content and inferentially for the amount of irrationality. While the topographic analysis generally suggested a lack of stability across presentations, the inferential analysis revealed little change. The suggestion is made that incomplete verbal protocols are not problematic for ATSS if the coding strategy taps into the underlying structure presupposed by one's theoretical interest.