This chapter discusses network theories in psychology (Lang, 1994; Leventhal, 1984; Lewis, 2005) and kindred theories in philosophy (Colombetti, 2009). While evolutionary theories focus on innate mechanisms, network theories shift focus to learning. Network theories propose the following causal-mechanistic explanation of emotion. An emotion is caused by the activation of an emotion network in memory, made up of representations of stimuli, emotional components (appraisals, somatic and motor responses, and feelings), and related information. The network can be activated via the stimulus side or the response side. Network theories come in a biological version, in which networks are built around affect programs, and a non-biological version, in which networks result from mutual interactions between emotional components. The biological version postulates that emotion networks have multiple layers, of which only some (those containing perceptual and/or response nodes) can directly cause emotions. The biological version accounts for the properties of ontogenetic and phylogenetic continuity, Intentionality, phenomenality, bodily aspect, heat, control precedence, and irrationality, even if the (world-directed) Intentionality is rather optional. The non-biological version is less suited to account for continuity. The biological version delivers discrete emotions whereas the non-biological version does not necessarily do so. Empirical research that tests network theories is discussed.