Comment: The Interaction Between Metaphor and Emotion Processing in the Brain

George Lakoff, in his article, “Language and Emotion,” (2016) makes a strong case for the important link between language and emotion processing. Is there any neuroscientific support for this? To date, most neuroscience studies on embodied semantics have focused on whether or not motor regions are active during the process of language related to actions. This is largely due to the fact that we have a good deal of scientific knowledge about motor brain regions and thus we can better predict expected neural activation patterns. The neuroscience of emotion processing is instead more complicated. Emotions such as joy or anger can often activate the same set of brain regions, causing many to speculate that these regions may be involved in more general processes (e.g., salience; Lindquist, Wager, Kober, Bliss-Moreau, & Barrett, 2012). These factors make studying the interaction between language and emotion processing in the brain difficult.

Nevertheless, a few studies have explored the engagement of emotion-related brain regions during language processing. One region that is better understood is the anterior insula. Several studies suggest that the anterior insula may be especially important for processing the emotion of physical disgust (Calder et al., 2007; Calder, Keane, Manes, Antoun, & Young, 2000). Accordingly, several studies have found that disgust language also activates the anterior insula (Citron, 2012; Ponz et al., 2014). However, the real question that Lakoff (2016) poses is not if processing emotion words activate emotion-related brain regions, but that this effect can be seen in metaphorical processing in particular.

Disgust is a particularly interesting emotion to study for figurative language because it has been linked by many studies to moral judgment (i.e., “Lying is putrid”; Schnall, Haidt, Clore, & Jordan, 2008). Lakoff has previously noted the connection between the use of disgust-related metaphors in political discourse, especially for conservatives, and their consequences on reasoning and judgments in the moral domain (Lakoff, 1996). He theorizes that disgust-related metaphors are powerful rhetorical tools impacting decision-making because the abstract domain of morality draws upon our experiences with physical impurity/disgust. Recently, we explored this theory by looking at the neural correlates of how people process disgust language in either a literal (“The slabs of deer meat are rotten”) or metaphoric context (“Congressman Smith’s statements about marriage are rotten”). We hypothesized that such language should be processed by neural systems responsible for processing physical disgust, like the anterior insula and pallidum, which is what we found (Gamez-Djokic et al., 2013). In addition, we found that disgust-related metaphors showed similar activation patterns as processing physical disgust (looking at disgusting pictures). We also found that individuals who are more susceptible to feeling disgust show stronger activation patterns for disgust-related metaphors. This finding may have implications for rhetoric choices by individuals in different political parties, as conservatives have been shown to have a stronger susceptibility to disgust than liberals. A current study is testing this latter hypothesis.

Critically, a recent study showed that taste metaphors (“The break up was bitter for him”) activate emotion-related brain regions more than their literal counterparts (“The break up was bad for him”). This suggests that metaphors may be more emotionally evocative than their literal counterparts possibly in part through their grounding in modality specific representations (Citron & Goldberg, 2014). In fact, it has been proposed that abstract concepts, as compared to concrete concepts, might gain their grounding predominantly from affective experiences (Vigliocco et al., 2014). This is in line with Lakoff’s (2016) predictions about the prevalence of metaphor in talking about our affective lives.

It is also possible for language to dampen activity in emotion-related brain regions. For example, it has been shown that in individuals experiencing emotional reactivity, simply naming the emotion they are experiencing reduces activity in emotion-related brain regions (Lieberman et al., 2007). In that study, individuals with arachnophobia exhibited more dampened activity in the amygdala when they labeled their emotions (i.e., “afraid”) than when they didn’t. This result reveals the powerful interaction between language and emotion in the brain, in this case to dampen emotional reactivity. In fact the practice of speaking about emotions in order to dampen negative emotional responses is the basis of psychotherapy. To test Lakoff’s (2016) theory that metaphor especially impacts emotion processing, one could implement Lieberman’s (2007) paradigm using literal phrases such as “I am extremely afraid” as compared to metaphors such as “I am paralyzed with fear.” Here we would predict that metaphors would have stronger effects than literal phrases.

There is still a great deal of work to be done to test specific linguistic models and to understand the details of the neural basis of metaphor processing and its relation to emotional processing. Collaboration between the linguists and neuroscientists is essential for this endeavor.