Explanations for differences in preference reversals between separate and joint evaluations of options are generally based on the assumption that individuals rely on task-given attribute information in their value construction. However, decision-makers also draw on personally-salient, reflexive experiences not explicitly mentioned in the experimental task as inputs, which heretofore have not been investigated systematically. Further, interpretive approaches that might reveal such experiences generally have not been applied in the analysis of value judgments. Therefore, we study value construction processes in three types of preference reversal problems using written protocols. Our results demonstrate that task-based explanations for preference reversals are inversely weakened relative to the problems' complexity, while reflexive, experiential thoughts are increased. As such, the current approach provides new insight into the preference construction process and indicates some alternative reflexive factors and reference points that influence choice. The results also highlight that for any particular choice problem there exist distinct segments of decision-makers who differ in their value construction processes.