Dimensions of Affect Relationships: Models and Their Integrative Implications John W. Reich, Alex J. Zautra, and Mary Davis Arizona State University This article presents data from a number of areas of psychology that have dealt with the issue of whether positive and negative affects are independent—the bivariate view— or whether they operate inversely from each other—the unidimensional, bipolar view. Both models have extensive empirical support. A more integrative view, the Dynamic Model of Affect (DMA), specifies conditions under which both bivariate and bipolar models are valid. It is tailored to analyzing both affect systems functioning concur- rently. The DMA is reviewed and then extended to show how 3 major areas of research can begin to incorporate the more integrative framework of analyzing co-occurring types of affect. Psychological science and practice are under- going a major shift in focus in response to developments in “positive psychology” (Fred- erickson, 2001; Myers, 2000; Peterson, 2000; Ryff & Singer, 1998; Seligman, 1999; Taylor, Kemeny, Reed, Bower, & Grunewald, 2000). This newly emerging emphasis attempts to re- focus psychological science and practice away from one-sided concentration on maladaptive, negative thoughts and behaviors to an openness to positive, affirming, resilience-enhancing as- pects of human adaptation and well-being (Die- ner, 2000; Salovey, Rothman, Detweiler, & Steward, 2000; Seligman & Csikszentmihalyi, 2000). With the newer developments, psychological science and practice now face the task of mov- ing to a new level of analysis, the outlines of which are only now coming into focus. Rather than concentrating on only negative emotions, or only on positive emotions, psychologists are now beginning to investigate how both emotion systems operate conjointly. The co-occurrence of two foci, two emotion systems, presents a challenge to the science to bring to bear con- cepts and techniques to achieve a more integra- tive view of the human organism and perhaps other organisms as well. At this juncture, it is important to define the question further. We should know the nature of the relationship of positive and negative feel- ings if we are going to develop a better under- standing of the processes of human learning and memory, build more powerful models of psy- chotherapy, and articulate more creative means of reducing vulnerability to stress and enhanc- ing resiliency. Better approaches to human de- velopment and cognition, more precise and comprehensive understandings of mental health, and corresponding improvements in treatment and interventions will be more achievable if this fundamental issue is moved to the forefront of our science and practice. Antecedent Research Traditions Two productive research traditions concern- ing the relationships between positive and neg- ative emotions have become well established, each with its own body of supportive research. However, the traditions yield opposite models, or nearly so, of how the affects relate to each other. One tradition posits that they are inde- pendent. This is the bivariate or uncorrelated bidimensional affect approach. The other tradi- tion presumes that they are related, but in- versely so. This is the bipolar or unidimensional affect approach. We first review research sup- porting each approach. Then we review briefly studies that support an integrative model, the Dynamic Model of Affect (DMA), that specifies the conditions under which both perspectives are valid. Linked conceptually with other ap- John W. Reich, Alex J. Zautra, and Mary Davis, Depart- ment of Psychology, Arizona State University. Correspondence concerning this article should be ad- dressed to John W. Reich, Department of Psychology, Ar- izona State University, Tempe, Arizona 85287-1104. E- mail: john.reich@asu.edu Review of General Psychology Copyright 2003 by the Educational Publishing Foundation 2003, Vol. 7, No. 1, 66–83 1089-2680/03/$12.00 DOI: 10.1037/1089-2680.7.1.66 66