Researchers have wondered how the brain creates emotions since the early days of psychological science. With a surge of studies in affective neuroscience in recent decades, scientists are poised to answer this question. In this target article, we present a meta-analytic summary of the neuroimaging literature on human emotion. We compare the locationist approach (i.e., the hypothesis that discrete emotion categories consistently and specifically correspond to distinct brain regions) with the psychological constructionist approach (i.e., the hypothesis that discrete emotion categories are constructed of more general brain networks not specific to those categories) to better understand the brain basis of emotion. We review both locationist and psychological constructionist hypotheses of brain–emotion correspondence and report meta-analytic findings bearing on these hypotheses. Overall, we found little evidence that discrete emotion categories can be consistently and specifically localized to distinct brain regions. Instead, we found evidence that is consistent with a psychological constructionist approach to the mind: A set of interacting brain regions commonly involved in basic psychological operations of both an emotional and non-emotional nature are active during emotion experience and perception across a range of discrete emotion categories.

Primary affects exist at an ecological-communicative level of analysis, and therefore are not identifiable with specific brain regions. The constructionist view favored in the target article, that emotions emerge from “more basic psychological processes,” does not specify the nature of these processes. These more basic processes may actually involve specific neurochemical systems, that is, primary motivational-emotional systems (primes), associated with specific feelings and desires that combine to form the “cocktail” of experienced emotion.

We agree that conceptualisation is key in understanding the brain basis of emotion. We argue that by conflating facial emotion recognition with subjective emotion experience, Lindquist et al. understate the importance of biological predisposition in emotion. We use examples from the anxiety disorders to illustrate the distinction between these two phenomena, emphasising the importance of both emotional hardware and contextual learning.

We challenge the classic experience/expression dichotomous account of emotions, according to which experiencing and expressing an emotion are two independent processes. By endorsing Dewey's and Mead's accounts of emotions, and capitalizing upon recent empirical findings, we propose that expression is part of the emotional experience. This proposal partly challenges the purely constructivist approach endorsed by the authors of the target article.

Lindquist et al. present a strong case for a constructionist account of emotion. First, we elaborate on the ramifications that a constructionist account of emotions might have for psychiatric disorders with emotional disturbances as core elements. Second, we reflect on similarities between Lindquist et al.'s model and recent attempts at formulating psychiatric disorders as networks of causally related symptoms.

In matters of the mind, the opposition between what is mind-made or inside and natural or outside the mind is bound to misfire. Lindquist et al. build their analysis on a strong contrast between naturalism, which they reject, and psychologism, which they endorse. We challenge this opposition and indicate how adopting psychologism to combat a naturalistic view of emotional mind/brain areas is self-defeating. We briefly develop the alternative view of emotions as mental organs.

Although the target article provides strong evidence against the locationist view, evidence for the constructionist view is inconclusive, because co-activation of brain regions does not necessarily imply connectivity between them. We propose a rigorous approach wherein connectivity between co-activated regions is first modeled using exploratory Granger causality, and then confirmed using dynamic causal modeling or Bayesian modeling.

Existence of similarities of overall brain activation, specifically during emotional and other common psychological operations (discussed by Lindquist et al.), supports a proposal that emotion participates continuously in dynamic adjustment of behavior. The proposed participation can clarify the relationship of emotion to musical experience. Music, in turn, can help explore such participation.

In Vytal and Hamann (2010) we reported a neuroimaging meta-analysis that found that basic emotions can be distinguished by their brain activation correlates, in marked contrast to Lindquist et al.'s conclusions in the target article. Here, I discuss implications of these findings for understanding emotion, outline limitations of using meta-analyses and neuroimaging as the sole basis for deciding between emotion views, and suggest that these views are essentially compatible and could be adapted and combined into an integrated emotion framework.

Lindquist et al. argue that emotional categories do not map onto distinct regions within the brain, but rather, arise from basic psychological processes, including conceptualization, executive attention, and core affect. Here, we use examples from cultural neuroscience to argue that psychological constructionism, not locationism, captures the essential role of emotion in the social and cultural brain.

In this commentary, we do two things. First, we sketch two further routes to psychological constructionism. They are complementary to Lindquist et al.'s meta-analyses and have potential to add new evidence. Second, we look at a challenging kind of case for constructionism, namely, emotional anomalies where there are correlated, and probably relevant, brain anomalies. Psychopaths are our example.

We suggest that, in animals, the core-affect system is linked to partially assimilated behavioral dispositions that act as developmental scaffolds for the ontogenetic construction of emotions. We also propose that in humans the evolution of language altered the control of emotions, leading to categories that can be adequately captured only by emotion-words.

In addition to active wake, emotions are generated and experienced in a variety of functionally different states such as those of sleep, during which external stimulation and cognitive control are lacking. The neural basis of emotions can be specified by regarding the multitude of emotion-related brain states, as well as the distinct neuro- and psychodynamic stages (generation and regulation) of emotional experience.

Lindquist et al. assess the neural evidence for locationist versus psychological construction accounts of human emotion. A wealth of experimental and clinical investigations show that individual differences in emotion and personality influence emotion processing. These factors may also influence the brain's response to emotional stimuli. A synthesis of the relevant neuroimaging data must therefore take these factors into consideration.

The target article by Lindquist et al. considers discrete emotions. This commentary argues that these are but a minor part of human emotional abilities, unifying us with animals. Uniquely human emotions are aesthetic emotions related to the need for the knowledge of “high” cognition, including emotions of the beautiful, cognitive dissonances, and musical emotions. This commentary touches on their cognitive functions and origins.

Lindquist et al. provide a convincing case against what they call the locationist account of emotion. Their quantitative approach elegantly illustrates the shortcomings of this still-entrenched viewpoint. Here, I discuss how a network perspective will advance our understanding of structure-function mappings in general, and the relationship between emotion and cognition in the brain.

Although we agree that a constructivist approach to emotional experience makes sense, we propose that implicit (visceromotor and somatomotor) emotional processes are dissociable from explicit (attention and reflection) emotional processes, and that the conscious experience of emotion requires an integration of the two. Assessments of implicit emotion and emotional awareness can be helpful in the neuroscientific investigation of emotion.

The closeness of somatosensory phenomena and emotional states can be critically extended into a clinical perspective by referring to Tourette's Syndrome (TS). Two examples are discussed in this commentary: (1) the neurodevelopmental approach to the pre- and post-tic sensorimotor urges, and (2) the TS treatment with deep brain stimulation. It is shown that in TS, both views (locationist and constructionist) need to be combined along the lifespan in order to get a more realistic picture of the brain basis of emotion.

Lindquist et al. convincingly argue that the brain implements psychological operations that are constitutive of emotion rather than modules subserving discrete emotions. However, the nature of such psychological operations is open to debate. I argue that considering appraisal theories may provide alternative interpretations of the neuroimaging data with respect to the psychological operations involved.

Lindquist et al. have assumed that functional specialization requires a one-to-one mapping between brain regions and discrete emotions. This assumption is in tension with the fact that regions can have multiple functions in the context of different, possibly distributed, networks. Once we open the door to other forms of functional specialization, neuroimaging data no longer favor constructionist models over natural kind models.