INTRODUCTION Emotion has historically played a central role in psychophysiological research, with rich traditions focusing on both central and peripheral nervous system measures. A central tenet of most functionalist/evolutionary theories is that emotions prepare the organism for dealing effectively and efficiently with threats, challenges, and opportunities (Levenson, 1994). Thus, research has arisen around the actions that occur (the “motion” part of emotion), which are largely subserved by the somatic nervous system, and the metabolic support for these actions, which is largely subserved by the autonomic nervous system (ANS). In this chapter, we focus primarily on the role the ANS plays in emotion and the ways it can best be studied. THEORETICAL UNDERPINNINGS Emotion research has been strongly influenced by a number of opposing theoretical positions, some of which have been debated for over a century: (a) discrete versus dimensional; (b) hard-wired versus socially constructed; and (c) universal versus culture-specific. Although psychophysiological research is often viewed as being primarily data-, phenomenon-, and methodology-driven, these theoretical debates have had important influences on the ways this research is conducted. Moreover, psychophysiological studies have provided valuable data that are often used (and misused) in theoretical debates about the nature of emotion. Arguably most important for psychophysiological research is the discrete versus dimensional debate, which centers on the differences among emotion. In discrete emotions theories, a limited number of distinct emotions can be distinguished from each other in terms of structural features (e.g., facial expression, ANS activity) and functions (e.g., preparation for fight, preparation for flight). Discrete emotions are often seen as “natural kinds” (Matsumoto & Willingham, 2006; Panksepp, 2000), with their sources found in the structure of the natural world. In contrast, dimensional theories do not envision distinct emotions, but rather allow for a large (perhaps unlimited) number of emotional states that are located within a dimensional space defined by one or more descriptors. Dimensional approaches that are prominent in the psychophysiological literature include: (a) a two-dimensional model that includes valence (negative-positive) and arousal (low-high), and (b) a single-dimensional model comprised by a motivational or action tendency dimension (approach-avoidance). Although discrete and dimensional models can be combined (e.g., locating a discrete emotion such as fear in negative valence/high arousal dimensional space), these theoretical models have often led to very different research paradigms…