The two general activation systems of affect: Structural findings, evolutionary considerations, and psychobiological evidence.

TLDR

Watson and Tellegen (1985) proposed a consensual structure of affect based on Russell’s (1980) circumplex model. The authors examine two embedded dimensions of this structure, labeling them Negative Activation (NA) and Positive Activation (PA). The review shows that a two‑factor model of affect captures robust self‑rated mood structure yet requires refinement, as its dimensions are interdependent and not fully bipolar, positioning it between simple structure and a true circumplex, and the authors link PA and NA to broader approach and withdrawal systems, thereby clarifying affective phenomena such as circadian rhythms, sleep, and mood disorders.

Abstract

D. Watson and A. Tellegen (1985) proposed a consensual structure of affect based on J. A. Russell's (1980) circumplex. The authors' review of the literature indicates that this 2-factor model captures robust structural properties of self-rated mood. Nevertheless, the evidence also indicates that the circumplex does not fit the data closely and needs to be refined. Most notably, the model's dimensions are not entirely independent; moreover, with the exception of Pleasantness-Unpleasantness, they are not completely bipolar. More generally, the data suggest a model that falls somewhere between classic simple structure and a true circumplex. The authors then examine two of the dimensions imbedded in this structure, which they label Negative Activation (NA) and Positive Activation (PA). The authors argue that PA and NA represent the subjective components of broader biobehavioral systems of approach and withdrawal, respectively. The authors conclude by demonstrating how this framework helps to clarify various affect-related phenomena, including circadian rhythms, sleep, and the mood disorders.