ABSTRACT
What might account for the persistent tendency for performance on reasoning tasks to covary after algorithmic limitations have been accounted for? Some clues as to the cause of these relationships are provided by an examination of the types of thinking dispositions that predicted normatives responses after computational limitations have been controlled in the Stanovich and West (1997, 1998b; Sa et al., in press) studies. It is argued in this chapter that many of these reflect tendencies toward the types of decontextualizing operations discussed in chapter 6. Consider some examples of the dimensions tapped (and example items from each) in the questionnaires used in those investigations l :
• Willingness to postpone closure (Stanovich & West, 1997): example item = "There is nothing wrong with being undecided about many issues."
