Process models require process measures

From Dual-Process Theories of the Social Mind. Edited by Jeffrey W. Sherman, Bertram Gawronski, and Yaacov Trope. Copyright 2014 by The Guilford Press. All rights reserved. Chapter 9 Process Models Require Process Measures Jeffrey W. Sherman, Regina Krieglmeyer, and Jimmy Calanchini The goal of dual-­process models (DPMs) is to describe the contributions of two distinct classes of cognitive processes to judgments and behavior. One of the major challenges to achieving this goal is the need for a compre- hensive measurement strategy that provides clear indicators of the dual processes, the conditions under which they operate, and the manner in which they interact to trans- form input into behavior. In this chapter, we review the most common measurement strategies applied to DPMs and some of the difficulties associated with those strate- gies. Many problems for DPMs result from the confounding of the qualitative nature of the dual processes in the models with the distinction between automaticity and control. Other problems arise from the use of behavioral measure outcomes as prox- ies for unmeasured cognitive processes and their operating conditions. We detail the advantages of a formal modeling approach and highlight some of the uses to which this approach has been applied in our own research with the Quadruple Process (Quad) model (Sherman et al., 2008). Finally, we describe theoretical insights related to DPMs that have been gleaned from this research. Measurement Challenges in DPMs Operating Principles versus Operating Conditions Common measurement strategies developed for DPMs and some of their attendant prob- lems can be traced to the historical origins of DPMs. Specifically, many DPMs arose in the wake of the long-­ standing dispute between two competing views of human information processing. The naive scientist view proposes that people try to understand their world much as scientists try to under- stand their topic of study—via careful, ratio- nal attempts to discover the true state of the world (e.g., Heider, 1958; Kelley, 1967). In contrast, the cognitive miser view holds that people have limited processing capacity and, as a result, rely on cognitively efficient mental shortcuts and heuristics that provide sufficiently accurate information for little effort (e.g., March & Simon, 1958; Tversky & Kahneman, 1974). Throughout the 1970s and 1980s, it became apparent that the dis- tinction between effortful, optimizing pro- cesses and effortless, satisficing processes was not an either–­ or proposition. Rather,