Tell it to a child! A brain stimulation study of the role of left inferior frontal gyrus in emotion regulation during storytelling

Abstract

In everyday life we need to continuously regulate our emotional responses according to their social context. Strategies of emotion regulation allow individuals to control time, intensity, nature and expression of emotional responses to environmental stimuli. The left inferior frontal gyrus (LIFG) is involved in the cognitive control of the selection of semantic content. We hypothesized that it might also be involved in the regulation of emotional feelings and expressions. We applied continuous theta burst stimulation (cTBS) over LIFG or a control site before a newly-developed ecological regulation task that required participants to produce storytelling of pictures with negative or neutral valence to either a peer (unregulated condition) or a child (regulated condition). Linguistic, expressive, and physiological responses were analyzed in order to assess the effects of LIFG-cTBS on emotion regulation. Results showed that the emotion regulation context modulated the emotional content of narrative productions, but not the physiologic orienting response or the early expressive behavior to negative stimuli. Furthermore, LIFG-cTBS disrupted the text-level structuring of negative picture storytelling and the early cardiac and muscular response to negative pictures; however, it did not affect the contextual emotional regulation of storytelling. These results may suggest that LIFG is involved in the initial detection of the affective arousal of emotional stimuli.

Introduction

The ability to evaluate and regulate the emotions triggered by environmental stimuli affects the way people feel and express their emotions (Gross and Thompson, 2007). As such, it is critical to behave appropriately in daily-life contexts. Different strategies of emotion regulation have been described (Gross and Thompson, 2007, Gross, 1999, Gross, 1998, Ochsner and Gross, 2005, Ochsner et al., 2012), including forms of: i) attentional control, which determines which stimuli are to be attended; ii) cognitive change, which relies on the ability to attribute a new meaning to the emotion-inducing stimulus (e.g., by reappraising the mental model triggered by the stimulus); and iii) behavioral regulation, which may attenuate or emphasize the expressive behavior related to the emotional response. While behavioral regulation does not necessarily dampen the individual's emotional experience, cognitive emotion regulation (including both attentional control and cognitive change) modulates feelings, particularly by down-regulating them when the stimuli have negative valence. Growing evidence suggests that cognitive regulation, in particular reappraisal, of the emotional reactivity to negative stimuli reduces the subjective experience of negative emotions (Gross, 1998, Hajcak and Nieuwenhuis, 2006, McRae et al., 2010, Ochsner et al., 2004, Ochsner et al., 2002, Phan et al., 2005). Furthermore, people feel less intense negative emotions when they look at pictures that have been previously reappraised (Schulze et al., 2011, Vanderhasselt et al., 2012). Interestingly, Jackson et al. (2000) showed that reappraisal was the strategy spontaneously adopted by individuals who had been asked to decrease the negative experience of experimentally triggered emotions. Moreover, it was associated to decreased activity of the corrugator supercilii, a muscle usually activated while processing stimuli with negative valence (Kim and Hamann, 2012). Thus, cognitive regulation to decrease negative emotions seems to moderate both subjective and physiological responses to negative stimuli.

Cognitive emotion regulation abilities are implemented in an articulated network of cortical and subcortical brain regions, which includes the amygdala (Ochsner et al., 2002, Phan et al., 2005), dorsal anterior cingulate cortex (McRae et al., 2010, Ochsner et al., 2004, Phan et al., 2005), medial (McRae et al., 2010, Ochsner et al., 2004, Ochsner et al., 2002, Phan et al., 2005) and lateral prefrontal cortices (Ochsner et al., 2004, Ochsner et al., 2002, Phan et al., 2005, Vanderhasselt et al., 2012; see also Kalisch, 2009 for a meta-analysis), left insula (Ochsner et al., 2002), inferior parietal cortices (Ochsner et al., 2002, Vanderhasselt et al., 2012), middle temporal and medial occipital gyri (Ochsner et al., 2002). Furthermore, different regulation goals (e.g., to decrease or to increase the intensity of a negative emotion) recruit with different intensities these areas in the left and right hemispheres (e.g., (Korb et al., 2015, Ochsner et al., 2012). Interestingly, many of these areas overlap with the neural networks associated with attentional control (Ochsner and Gross, 2005, Ochsner et al., 2009, Wessing et al., 2013), goal identification (temporo-parietal network) (Van Overwalle, 2009), and semantic control (fronto-temporo-parietal network) (Badre and Wagner, 2007, Jefferies, 2012, Wagner et al., 2001). This might reflect the different cognitive processes involved in cognitive emotion regulation. Causative evidence on the role played by these areas in cognitive emotion regulation is still lacking. In our study, we focused on an epicenter of this network, namely the left inferior frontal gyrus (LIFG), which has been previously implicated in context-driven semantic control (Jefferies, 2012). We tested the hypothesis that LIFG is involved in cognitive regulation to decrease negative emotions by generating a new conceptual model of an emotional situation. Former brain stimulation studies have shown that interference with LIFG activity affects the selection of contextually appropriate words during storytelling (Marini and Urgesi, 2012), but also in other context-relevant conditions such as during synonym judgment (Gough et al., 2005) or semantic control tasks (Badre and Wagner, 2007, Jefferies, 2012, Whitney et al., 2011). In the current study, we explored the hypothesis that interference with LIFG activity, as a region subserving context-driven semantic control, affects cognitive regulation to decrease the emotional responses to negative stimuli. While previous neuroimaging studies (e.g., Kohn et al., 2014) of cognitive emotion regulation have pointed to a mainly bilateral network of left and right hemisphere areas, we focused on LIFG in our storytelling task for its documented role, as compared to the homolog area in the right hemisphere, in semantic control (Jefferies, 2012) and discourse processing (Spalletta et al., 2010, Marini and Urgesi, 2012).

In down-regulation experiments, participants are usually asked to decrease the intensity of their emotional reaction. In some studies, the authors provided general instructions to guide regulation strategies, for example “imagine that what you are about to see is staged/performed by actors rather than real” (Jackson et al., 2000). In other investigations, the authors provided their participants with a set of predefined descriptions of the stimulus, for example “this is a scene from a movie” (Macnamara et al., 2009). In these experimental settings, participants explicitly received relevant information about the nature of the incoming stimuli or instructions on the regulation strategy that they were required to use. This, however, does not reflect typical everyday life situations, where emotion regulation is elicited spontaneously in response to environmental stimuli and is often embedded in situational and relational contexts. Indeed, emotion regulation abilities are not only used to modulate the individual's reactions to negative stimuli, but are also deemed as particularly important in adjusting the appraisal of negative stimuli by other individuals, in particular when interacting with children.

Growing evidence suggests that adults' ability to influence others' experience and expression of emotion in caregiving contexts facilitates children's positive and sensitive responding to emotional stimuli and impacts their psychological and physical development (Crandall et al., 2015, Rutherford et al., 2015). Production of narratives about autobiographical and public events is widely used to convey cultural information to younger individuals, which is considered a main social goal for adults (Kellas, 2013, Mergler et al., 1984). The style of these narrations is modulated not only by internal motivational and subjective factors related to the producer, but also by external factors related to context and listeners (Adams et al., 2002, Pasupathi et al., 2002, Pasupathi et al., 1998). Indeed, parents tend to use either an elaborated and detailed style or a repetitive and topic-switching style when talking about past events to their daughter or son, respectively (Reese and Fivush, 1993). On the same line, both younger and older adults produce more elaborated narratives, make more repetitions and simplify the complexity of the story when talking to a young child as compared to an adult (Adams et al., 2002). Crucially, both the propositional content and the emotional aspect of narratives can be modulated by social context. When older adults talk to young children, they tend to up-regulate positive emotions and down-regulate negative emotions (Pasupathi et al., 2002). Thus, narrating to children triggers specific emotion regulation strategies aimed at influencing children's appraisal of emotional situations. Interestingly, modulation of narratives according to the listener may also affect what the producer remembers about the story (Pasupathi et al., 1998), pointing to the possibility that modulated narrations change the producer's processing of emotional stimuli.

To investigate the role of LIFG in spontaneous use of emotional regulation in ecological contexts, we developed an emotion regulation task that required participants to spontaneously adopt regulation strategies during storytelling. Participants were asked to describe what was going on in a series of pictures with either negative or neutral emotional content: in the unregulated condition, they were asked to describe the story as if they were talking to an adult; in the regulated condition, they were asked to describe it as if they were talking to a child. Therefore, in the critical condition (i.e., narration to a child) the participants were required (but not explicitly asked) to appraise the emotional valence of the stimulus, to down-regulate their consequent emotional reaction according to the interlocutor, and to generate a down-regulated description of the portrayed events. The picture description task employed in this study has been shown to have high ecological validity in the assessment of narrative skills in both healthy individuals (Marini et al., 2005) and brain damaged patients (Galetto et al., 2013). However, since no previous study has used this task to elicit emotional regulation, we preliminarily tested if the emotional content of the narratives could be modulated by the social context of storytelling.

The causative involvement of LIFG in down-regulation of emotional reactivity to negative stimuli was investigated by using continuous theta burst stimulation (cTBS), which is expected to inhibit the area underlying the coil for up to 30 min after the end of the train (Huang et al., 2005). In keeping with previous studies (Adams et al., 2002, Pasupathi et al., 2002), we expected that manipulation of the emotional regulation context of the narration would modulate the content of the storytelling, as well as the associated physiological and expressive behaviors, and that this would occur especially during narration of stories with negative valence. Crucially, we hypothesized that interference with neural activity in LIFG would hamper the process of cognitive emotion regulation and the modulation of linguistic, emotional and physiological responses according to the emotional regulation requirements of the task.