Handbook of emotion regulation ( PDFDrive )

Valuation and Emotion Regulation 31 of the emotion generation sequence they should be involved in the selection of escape target. In the context of the present frame- behaviors—e­specially in the kinds of emo- work, this idea is expressed by suggesting tionally arousing situations humans face in that emotion regulatory processes differ everyday life. Although this hypothesis has in the stage of the PVA sequence at which not been tested, the involvement of prefron- they have their primary impact (see Figure tal regions is suggested by behavioral stud- 2.3). Some strategies influence the situation-­ ies in humans showing that emotion can be dependent perceptual inputs (situation selec- regulated by deliberately changing situation-­ tion, situation modification, and attention dependent stimulus inputs in the service of deployment). Others influence the valua- explicit regulatory goals. For example, either tion step itself (cognitive change). Still oth- physically or mentally, using visual imagery, ers influence the response output associated one can move closer to or further away from with activated action sequences (response an emotion-e­liciting stimulus (e.g., making modulation). By impacting different states oneself feel more positive by approaching a of the PVA cycle, different strategies impact pleasant stimulus or less negative by with- emotional responding in different ways, as drawing from an unpleasant one; Davis, detailed below. Gross, & Ochsner, 2011; Muhlberger, Neu- mann, Wieser, & Pauli, 2008; Williams & Situation selection refers to altering the Bargh, 2008). inputs to the PVA sequence through deci- sions about whether to expose oneself to a Attentional deployment refers to altering given situation/stimulus based on its pro- the inputs to the PVA sequence by increasing jected affective impact. For example, call- or decreasing attention to them (e.g., look- ing to mind the image of the subway might ing away from the teen and at the man or lead to a negative evaluation, and a feeling woman). While this can gate specific stim- of fear. This feeling might motivate a higher-­ uli wholly into or out of the PVA stream, level PVA that would trigger a decision to thereby promoting or preventing responses take an alternative means of transportation to them, we propose that more graded in order to decrease the probability of the changes in attention to stimuli may result negative experiences that one associates with in correspondingly graded levels of acti- taking the subway. More generally, situation vation of their associated PVA sequences. selection can take many forms, for example, This strategy involves interactions between when a socially anxious individual avoids a cognitive control systems and valuation sys- social event. To date, the neural bases of sit- tems, with particular involvement of dorsal uation selection have been studied only with PFC and inferior parietal regions associated avoidance conditioning tasks in which an with selective attention (Pessoa et al., 2003), animal learns to select an action (e.g., run- and in some cases rmPFC regions implicated ning in a wheel when a light is illuminated in attending to and explicitly judging the predicts an upcoming shock) that enables value of stimuli (Bishop, 2007; Egner, Etkin, it to avoid experiencing a noxious stimulus Gale, & Hirsch, 2008; Etkin, Egner, Peraza, (that prevents shock administration). Rodent Kandel, & Hirsch, 2006; Lane et al., 1998; studies have shown this involves modulation Ochsner, Hughes, Robertson, Cooper, & of two core valuation systems, the striatum Gabrieli, 2009). A growing but somewhat and amygdala (Everitt et al., 1999; LeDoux inconsistent imaging literature shows that, & Gorman, 2001), and one human imaging by and large, when a task manipulation study (Delgado et al., 2009) indicates that diminishes attention to an affectively arous- it also engages vlPFC and dorsolateral pre- ing stimulus, activation increases in PFC frontal cortex (dlPFC) regions involved in regions implicated in cognitive control (sug- cognitive control that presumably modulate gesting that cognitive control systems are the core valuation systems. involved in manipulating attention), and activity decreases in regions implicated in Situation modification refers to alter- core (e.g., amygdala, PAG), contextual (e.g., ing the situation one is in, thereby modi- insula) or conceptual (e.g., mPFC) valuation fying inputs to the PVA sequence, and (e.g., Ochsner & Gross, 2005; Pessoa, 2009). changing the emotion (e.g., sitting further While it is clear that the specific valuation away from the teen or exiting the subway). systems modulated by attentional deploy- We would expect that prefrontal systems

Emotion Regulation Strategies Related Phenomena Region Neural Systems Processes Attentional Cognitive Response Affective Affective Expectancies Type of selective Deployment Change Modulation Learning Decisions (e.g., Placebo System attention/ (Reappraisal) (e.g., Fear working (e.g., Extinction) (e.g., Effects) memory Expressive Intertemporal Suppression) Choice) dorsolateral   PFC dorsal  posterior mPFC inferior parietal Control   dorsal ACC performance     monitoring ventrolateral selection/     PFC inhibition       dorsal mPFC Conceptual conceptual/ ? ?  Valuation categorical     valuations    rostral mPFC attention to   valuation    ventromedial Contextual value of PFC/OFC Valuation stimulus in ventral current context striatum amygdala reward/ reinforcement insula value arousal (and Core Valuation threat) value of stimulus Valuation representation and awareness of body states for all types of valuation W1 P1 V1 A1 W1 P1 V1 A1 W1 P1 V1 A1 W1 P1 V1 A1 Placebo P1 V1 A1 W2 P2 V2 A2 W2 P2 V2 A2 W1 P2 V2 A2 PVA Cycle Illustrations W1 P1 Vsafe A1 Option1 P1 V1 A1 Painful Vthreat Stimulus P2 V2 A2 Option2 P2 V2 A2 FIGURE 2.3.╇ Neural systems for valuation and control postulated by the valuation framework pre- sented in the chapter (left-hand columns), as well as the roles these neural systems play in three kinds of emotion regulation strategies (center columns, see text) and three kinds of related phenomena (right- hand columns, see text). Up arrows indicate increased activation, down arrows indicate decreased activation, and “?” indicates involvement in some (but not the majority) of the studies. The final row diagrams, in PVA terms, how each emotion regulation strategy or related phenomenon might oper- ate (see text for details). The three center columns show, for each emotion regulation strategy, how control actions impact either attention paid to particular stimuli at the perception stage (attentional deployment), how one values those stimuli (reappraisal), or what actions one takes as a consequence of this valuation (response modulation). The three right-hand columns show for related phenomena how initial valuations (e.g., threat) may be overridden if one learns new valuations (e.g., safe) for a stimulus (extinction) one may select among choice options as a function of their relative valuations, with control actions coming into play when the choice options are similarly valued and/or in conflict (intertemporal choice), or a placebo may influence the valuation of a painful stimulus via the action of control pro- cesses (placebo effects). 32

Valuation and Emotion Regulation 33 ment depend on the sensory qualities of the systems (e.g., Kober et al., 2010; Ochsner et stimulus, distraction from pain modulates al., 2004; Urry et al., 2006; Wager, David- nociceptive regions of insula and cingulate son, Hughes, Lindquist, & Ochsner, 2008; cortex (Frankenstein, Richter, McIntyre, & reviewed in Kalisch, 2009; Ochsner & Remy, 2001; Tracey et al., 2002), whereas Gross, 2005, 2008) in accordance with one’s distraction from an aversive image modu- regulatory goals. lates the amygdala (McRae et al., 2010; Pes- soa, 2009); for example, cross-study vari- Finally, response modulation refers to tar- ability in attentional deployment strategies geting behavioral manifestations of emotion and the lack of a common metric for deter- (e.g., playing it cool by not showing fear of mining how much any given strategy dimin- the knife-­wielding teen). Human research ishes attention to a stimulus in one study primarily has focused on one exemplar of compared to others (see Ochsner & Gross, this strategy, expressive suppression, which 2005, for a detailed review) have limited the involves hiding behavioral manifestations of conclusions that can be drawn about when emotion (Gross, 1998). Behaviorally, expres- and how specific cognitive control systems sive suppression effectively reduces facial are involved. expressions of emotion, but the effort and attention required to do so trigger autonomic Cognitive change refers to altering the responses, impair memory for visual cues, subjective meaning and/or perceived self-­ and can negatively impact social interactions relevance of the present situation (e.g., (Butler et al., 2003; Gross, 1998; Richards thinking of the knife as a stage prop or that & Gross, 2000). In keeping with these find- knife-­tossing is an innocent way of pass- ings, an initial imaging study showed that ing time). The framework suggests that this suppressing the expression of disgust acti- strategy should involve interactions between vated two key PFC regions associated with cognitive control systems that can be used cognitive control (dlPFC associated with deliberately to change one’s interpretation of maintaining goals, and vlPFC associated a stimulus and valuation systems that trigger with response selection and inhibition more an affective response. Of note here is the fact generally; Aron et al., 2004; Badre & Wag- that the framework predicts that conceptual ner, 2007; Thompson-­Schill et al., 2005), valuation systems can play a role on either and increased activation in core (amgydala) side of this regulatory equation: on the one and contextual (insula) valuation regions hand, being the target of cognitive control associated with detection of threats and systems that seek to change one’s high-level awareness of body states (Goldin, McRae, conceptual valuation of a stimulus, and on Ramel, & Gross, 2008). This supports the the other, assisting those cognitive control idea that expressive suppression, like other systems in reformulating the attributions one strategies, depends on interactions between makes about the nature of one’s own beliefs, cognitive control and valuation regions, and desires, and feelings (e.g., “I’m feeling less may have neural bases similar to those sup- afraid now”)—or those expressed by others porting response inhibition more generally (e.g., “The subway passengers are anxious (Aron et al., 2004). about the crowding, not the teen”)—as one deliberately changes his or her interpretation Applications of of an emotion-­eliciting stimulus. Research the Valuation Perspective on cognitive change—r­eferred to in the lit- The neural systems implicated in emotion erature as “reappraisal”—consistently sup- and emotion regulation play key roles in ports the predictions of the framework: other phenomena that involve valuation and When engaging in a cognitive change strat- cognitive control (Hartley & Phelps, 2010; egy, activation is observed in lPFC and cin- Murray et al., 2007; Pessoa, 2008; Phillips gulate PFC regions associated with cognitive et al., 2008; Rangel et al., 2008). We believe control, as well as mPFC regions associated it is important that any account of emotion with conceptual valuation (albeit primarily and emotion regulation use terminology and when up-r­egulating emotional responses) concepts that are broadly applicable to allied and at the same time increasing or decreas- phenomena as well. ing activity in core (e.g., amygdala, striatum) and/or contextual (e.g., insula) valuation

34 BIOLOGICAL BASES With this in mind, we illustrate in this examples of value updating, namely, extinc- tion of a fear response. We discussed extinc- section the broad applicability of our valu- tion earlier as an example of contextual ation perspective on emotion and emotion valuation in which an organism learns that regulation by showing how it can provide a a previously feared stimulus need no longer framework for describing the mechanisms be feared in the current temporal context. underlying three types of related phenomena In that section, however, we did not explain that traditionally are considered in relatively how the organism learns this contextual separate literatures. This has the dual bene- association. Here we propose that value fits of broadening the framework to account updating is the learning mechanism. for aspects of related phenomena it was not initially formulated to address, and in so The subway example can help make this doing, making the framework more robust concrete. Recall that the knife-­wielding teen and generally applicable. initially elicits a threat valuation and fear response involving amygdala-m­ ediated core Affective or Emotional Learning level PVAs. If the expected outcome does As noted in the earlier section on the PVA not transpire (i.e., the teen takes no harm- processing dynamics, our valuation perspec- ful actions), however, then over time a new tive allows learning to occur by updating contextual-­level PVA is acquired by ventro- the valuations placed on stimuli with each medial/orbitomedial PFC systems indicating iteration of the PVA sequence. To account the teen is not a threat. The longer the teen more broadly for various forms of affective takes no harmful action, the stronger this or emotional learning, we can elaborate the PVA becomes. Ultimately, even though the way in which this updating process occurs. knife still connotes threat at the core level, the contextual PVA wins out for expression When encountering a stimulus, one’s cur- in behavior (see the section on PVA process- rent valuation of it sets an expectation for the ing dynamics). Because the core-level PVA outcome states of the world that should fol- itself remains unchanged, a fear response low from execution of the associated action can be quickly reinstated in the future impulse(s). These outcomes become inputs to should the teen become truly threatening the next PVA, which evaluates discrepancies (Bouton, 2004; LeDoux, 1993). The frame- between the expected and actual outcomes. work can be similarly applied to other cases If this valuation is negative (i.e., when the in which one learns, or already has learned, discrepancy is large and/or important in that a given emotional impulse is inappro- light of currently active goals), this valuation priate or unnecessary in the current context, triggers learning and updating processes as in reversal of learned appetitive or aver- that change links between a stimulus and its sive associations (Bouton, 2004; Corcoran valuation (P-V) or between a valuation and & Quirk, 2007; Schoenbaum et al., 2007). an action (V-A)—or between separate PVA sequences—­so that future valuations are As this example makes clear, affective more accurate (Delgado, Olsson, & Phelps, learning and the types of regulatory strate- 2006; Rangel et al., 2008; Schultz, Dayan, & gies reviewed earlier are not mutually exclu- Montague, 1997). Each change is small, so sive and may in many contexts work together. that one’s value expectations for a stimulus For example, the act of reappraising can be at a given moment in time are a function of seen as a way of cognitively creating a dis- one’s prior experiences with it, biased more crepancy between an expected internal out- heavily toward recent experiences. While come (e.g., a fear response) associated with this value updating process typically is stud- a given percept (e.g., the knife-­wielding teen) ied in the context of conditioning, reward, and the response that actually occurs (e.g., and affective learning, it fits neatly within calmness). This discrepancy could activate our valuation framework as the way that learning processes that weaken core-level changes in the contingency between actions PVA representations of the teen as threat- and outcomes can adaptively alter the valua- ening, build new contextual-­level represen- tions that drive the actions. tations of the teen as nonthreatening, and strengthen conceptual-l­evel PVAs of the teen To illustrate this, we can use our subway as an actor. In this way, reappraisal—­and example to consider one of the most studied by extension other regulatory strategies—­

Valuation and Emotion Regulation 35 can be seen as providing top-down “teach- reward in working memory and inhibit acti- ing” inputs to outcome-­driven regulatory vation of PVAs for the immediately avail- processes that typically are triggered by able choice option (Figner et al., 2011). In external cues (cf. Delgado, Gillis, & Phelps, keeping with this account, human imaging 2008; Delgado, Nearing, LeDoux, & Phelps, studies have shown greater ventral striatal 2008). (VS) and/or vmPFC versus greater dlPFC activity when participants select immediate Affective Decision Making versus delayed rewards (McClure, Ericson, Our valuation perspective also may be Laibson, Loewenstein, & Cohen, 2007; applied to affect-­laden decision making. McClure, Laibson, Loewenstein, & Cohen, Affective decisions require a choice between 2004), and a recent transcranial magnetic options that are associated with different stimulation study showed that disruption of expected rewards or punishments. In our left—but not right—dlPFC led participants framework, these expectations are reflected to “impulsively” select immediate rewards in the values computed for choice options when they had shown a prior preference at various levels of the valuation hierarchy. for the delayed reward (Figner et al., 2011). To the extent that the play of activation Strikingly, this result dovetails with the find- and inhibition across PVA sequences associ- ing that a pathway from left dlPFC to the VS ated with choice options results in a core-, supports the use of reappraisal to diminish contextual-, or conceptual-l­evel valuation craving for desired substances (e.g., sugary/ determining the behavioral output, then the fattening foods) when participants think choice option associated with that valuation about the negative long term (e.g., diabetes) will be selected. However, in some cases, this as opposed to the immediate (e.g., delicious play of activation fails to determine clearly a taste) consequences of consuming them most highly valued selection, and cognitive (Kober et al., 2010). control processes may be engaged in order to construct, hold in mind, and implement top- As these findings make clear, affective down processes that influence PVAs associ- decision making and the regulatory strate- ated with choice options. This commonly gies reviewed earlier may depend upon very happens when choice options are similarly similar neural systems and, as such, the line valued and/or conflict with one another, between them is not always clear. Indeed, but it also may happen when the valuation intertemporal choices—a­nd other choices process itself becomes a target of valuation that require selecting between options con- (e.g., when there is a negative valuation of an sistent with long- versus short-term goals— attractive response option). can be viewed as self-­control tasks (Figner et al., 2011; Hare, Camerer, & Rangel, 2009; To illustrate (see Figure 2.3), consider Wunderlich, Rangel, & O’Doherty, 2009) in how the framework accounts for a com- which the decision to select an option consis- monly studied choice dilemma in behavioral tent with a long-term goal is influenced by economics and neuroeconomics known attention deployment and cognitive change as “intertemporal choice” (or as delay of strategies (Mischel et al., 1989). Our valua- gratification in the developmental litera- tion perspective can also be applied to other ture; Mischel, Shoda, & Rodriguez, 1989). types of choice dilemmas in which control This dilemma involves choosing between and valuation processes interact to deter- a smaller reward available now or a larger mine choice, including risky decision making reward available at some point in the future. (Gianotti et al., 2009), in which the choice is In our framework, selection of the immedi- to be fair toward or to punish others (Knoch ate reward would be promoted by core-level et al., 2008; Knoch, Pascual-­Leone, Meyer, (striatal) or contextual-l­evel (medial/orbital Treyer, & Fehr, 2006), and when the act of frontal) valuation systems that represent choice itself changes our valuations of stim- the reward value of the currently available uli via the value-­updating process (Sharot, stimulus. By contrast, picking the delayed De Martino, & Dolan, 2009; Sharot, Shiner, reward would require the use of lateral pre- & Dolan, 2010) as in cognitive dissonance frontal cognitive control systems in order reduction (Lieberman, Ochsner, Gilbert, to maintain a representation of the delayed & Schacter, 2001; Sharot et al., 2009; van Veen, Krug, Schooler, & Carter, 2009).

36 BIOLOGICAL BASES Expectancies, Beliefs, PAG) (Ploghaus, Becerra, Borras, & Bor- and Placebo Effects sook, 2003; Wager, 2005). Our valuation framework also helps make sense of the growing imaging literatures on Thus, from the viewpoint of the frame- the ways in which expectancies and beliefs of work, expectancies and beliefs operate much various sorts—­including placebo effects—­ like two of the emotion regulation strategies influence responses to various kinds of affec- described earlier—­attention deployment tive stimuli (Wager, 2005). In these tasks, and cognitive change—­in that they alter participants are given one of two kinds of lower-level inputs to PVA systems and/or the explicit expectations. In studies of expectan- valuation process. cies or anticipation, participants are told that an upcoming stimulus—­whether a painful Summary sensation, an image, or something else—will One of the fundamental challenges faced by be of a particular intensity or kind. In pla- any animal is computing and expressing the cebo experiments, participants are told that value of stimuli in an accurate and timely a drug (e.g., a cream or a pill) will increase manner. This is difficult, because the ani- or decrease their subsequent responses to a mal’s internal state and external environment stimulus. In either case, these expectations change over time, and its information acqui- lead participants to experience the stimulus, sition, processing, and response resources when presented, as subjectively more simi- and capabilities are limited. To address these lar to what they expected than would have challenges, humans (and other animals) have been the case had they held no expectations developed a complex set of interacting valua- or beliefs about its nature or the protective tion systems, each of which can be described properties of a drug. in terms of a simplified P-V-A sequence, in which a particular perceptual input is valued From the perspective of our framework, (negatively or positively to a given degree), these phenomena all involve the top-down leading to an impulse to alter ongoing behav- influence of cognitive control systems on ioral or cognitive responses. These P-V-A valuation systems or the influence of higher sequences run in parallel at various levels in level valuation systems on lower level valu- the brain and compete for expression. ation systems. Our interpretation of these effects is consistent with results of imaging This process-o­riented valuation frame- studies of expectancies and placebo effects work suggests a number of directions for on pain responses. Such studies indicate future research. One direction concerns that expectancies and placebo beliefs about the valuation systems. While Figures 2.2A pain are maintained in a combination of and 2.3 feature three kinds of valuation lateral prefrontal/parietal working memory systems (core, contextual, and conceptual), systems and/or medial prefrontal systems future work should clarify the number and (Atlas, Bolger, Lindquist, & Wager, 2010; kind of valuation systems, as well as the Lieberman, Jarcho, Berman, et al., 2004; rules that govern their engagement in par- Wager, 2005; Wager, Atlas, Leotti, & Rill- ticular contexts. A second direction has to ing, 2011) that in the framework could be do with how the often-­competing action described as representing either conceptual-­ impulses associated with different P-V-A level beliefs (e.g., “The cream on my forearm sequences are coordinated. We have empha- should lessen the pain”) or contextual-­level sized the role of competitive activation and expectations about the stimulus or placebo. inhibition, but how this and other processes According to our framework, these systems lead to coordinated and sustained adaptive influence attention to and appraisal of the behavior rather than erratic and conflicting value of stimuli in contextual-l­evel and/ behavior is not yet clear. A third direction or core-level valuation systems (see Figure concerns the inputs and outputs of valuation 2.3), modifying their levels of activation to systems. We have suggested that the class be consistent with top-down beliefs (e.g., of P-V-A sequences whose inputs are other lessening activation of pain-s­ensitive valu- P-V-A sequences, and outputs that include ations systems, including contextual-l­evel the engagement of cognitive control pro- regions (e.g., cingulate and insular cortex) cesses are fundamental to emotion regula- and core-level regions (e.g., amygdala and

Valuation and Emotion Regulation 37 tion and self-c­ontrol more generally. That References said, the range of relevant inputs and out- puts, and the malleability of input–o­utput Anderson, J. R. (1983). A spreading activation relations requires further study. A fourth theory of memory. Journal of Verbal Learning direction concerns the efficacy of various and Verbal Behavior, 22(3), 261–295. forms of value regulation and how they are intermixed in everyday life. Which “pure” Aron, A. R., Robbins, T. W., & Poldrack, R. A. or “hybrid” forms of value regulation are (2004). Inhibition and the right inferior fron- most effective? A fifth direction concerns tal cortex. Trends in Cognitive Sciences, 8(4), translation of what we learn to illuminate 170 –177. individual differences. In our framework, a given emotional response and regulation Atlas, L. Y., Bolger, N., Lindquist, M. A., & profile could involve individual differences Wager, T. D. (2010). Brain mediators of pre- in (1) the initial valuations placed on spe- dictive cue effects on perceived pain. Journal cific classes of stimuli by systems at the core, of Neuroscience, 30(39), 12964–12977. contextual, and/or conceptual levels; (2) the speed with which these valuations are made; Badre, D., & Wagner, A. D. (2007). Left ven- (3) how quickly and easily one resolves con- trolateral prefrontal cortex and the cognitive flicts between them to express emotional control of memory. Neuropsychologia, 45(13), responses; (4) how quickly and effectively 2883–2901. learning processes update these valuations given that some emotional responses may be Barrett, L. F. (2006). Solving the emotion para- more difficult to change than others; (5) the dox: Categorization and the experience of knowledge of how and when to deploy emo- emotion. Personality and Social Psychology tion regulatory strategies; and (6) the capac- Review, 10(1), 20–46. ity and ability to deploy top-down control systems to implement these strategies. One Barrett, L. F., Mesquita, B., Ochsner, K. N., & important direction for future research is Gross, J. J. (2007). The experience of emotion. examining how each of these differences—­ Annual Review of Psychology, 58, 373–403. and others—­may interact to produce vari- ous forms of psychopathology. Barrett, L. F., Ochsner, K. N., & Gross, J. J. (2007). Automaticity and emotion. In J. Our goal in presenting this valuation A. Bargh (Ed.), Social Psychology and the framework is to provide a common plat- Unconscious (pp. 173–218). New York: Psy- form for analyzing the neural systems that chology Press. are important for many different types of valuation. The impetus for this framework Bishop, S. J. (2007). Neurocognitive mechanisms came from the observation that neural sys- of anxiety: An integrative account. Trends in tems implicated in emotion generation and Cognitive Sciences, 11(7), 307–316. emotion regulation overlapped in important ways with neural systems implicated in other Bouton, M. E. (2004). Context and behavioral literatures that typically are not considered processes in extinction. Learning and Mem- side by side (see Figure 2.3). Across all these ory, 11(5), 485–494. research domains an organism’s adaptive capacity crucially hinges on the coordina- Buhle, J. T., Kober, H., Ochsner, K. N., Mende-­ tion of multiple valuation systems in real Siedlecki, P., Weber, J., Hughes, B. L., et al. time, and a key challenge for future research (2012). Common representation of pain and is delineating these PVA interactions, ideally negative emotion in the midbrain periaque- with adequate specificity to permit more ductal gray. Social Cognitive and Affective explicitly computational approaches. We Neuroscience. [E-publication ahead of print] believe that an explicitly integrative valu- ation framework represents a step in this Butler, E. A., Egloff, B., Wilhelm, F. H., Smith, direction, and holds out the possibility of N. C., Erickson, E. A., & Gross, J. J. (2003). better coordinating hitherto unconnected The social consequences of expressive suppres- research literatures, while simultaneously sion. Emotion, 3(1), 48–67. deepening our understanding of each one. Cato, M. A., Crosson, B., Gokcay, D., Soltysik, D., Wierenga, C., Gopinath, K., et al. (2004). Processing words with emotional connotation: An fMRI study of time course and laterality in rostral frontal and retrosplenial cortices. Journal of Cognitive Neuroscience, 16(2), 167–177. Corcoran, K. A., & Quirk, G. J. (2007). Recall- ing safety: Cooperative functions of the ven- tromedial prefrontal cortex and the hippo-

38 BIOLOGICAL BASES campus in extinction. CNS Spectrums, 12(3), judgments reveals a spatial gradient for men- 200–206. talizing in medial prefrontal cortex. Journal Craig, A. D. (2003). Interoception: The sense of of Cognitive Neuroscience, 24(8), 1742–1752. the physiological condition of the body. Cur- Desimone, R., & Duncan, J. (1995). Neural rent Opinion in Neurobiology, 13(4), 500– mechanisms of selective visual attention. 505. Annual Review of Neuroscience, 18, 193– Craig, A. D. (2009). How do you feel—now?: 222. The anterior insula and human awareness. Dum, R. P., Levinthal, D. J., & Strick, P. L. Nature Reviews Neuroscience, 10(1), 59–70. (2009). The spinothalamic system targets Critchley, H. D. (2005). Neural mechanisms of motor and sensory areas in the cerebral cortex autonomic, affective, and cognitive integra- of monkeys. Journal of Neuroscience, 29(45), tion. Journal of Comparative Neurology, 14223–14235. 493(1), 154–166. Egner, T., Etkin, A., Gale, S., & Hirsch, J. (2008). Cunningham, W. A., Raye, C. L., & Johnson, Dissociable neural systems resolve conflict M. K. (2004). Implicit and explicit evaluation: from emotional versus nonemotional distract- fMRI correlates of valence, emotional inten- ers. Cerebral Cortex, 18(6), 1475–1484. sity, and control in the processing of attitudes. Etkin, A., Egner, T., Peraza, D. M., Kandel, E. Journal of Cognitive Neuroscience, 16(10), R., & Hirsch, J. (2006). Resolving emotional 1717–1729. conflict: A role for the rostral anterior cingu- Cunningham, W. A., & Zelazo, P. D. (2007). late cortex in modulating activity in the amyg- Attitudes and evaluations: A social cognitive dala. Neuron, 51(6), 871–882. neuroscience perspective. Trends in Cognitive Everitt, B. J., Parkinson, J. A., Olmstead, M. C., Sciences, 11(3), 97–104. Arroyo, M., Robledo, P., & Robbins, T. W. Damasio, A., Damasio, H., & Tranel, D. (2013). (1999). Associative processes in addiction and Persistence of feelings and sentience after reward: The role of amygdala–v­ entral striatal bilateral damage of the insula. Cerebral Cor- subsystems. Annals of the New York Acad- tex, 23(4), 833–846. emy of Sciences, 877, 412–438. Davachi, L. (2006). Item, context and relational Ferstl, E. C., & von Cramon, D. Y. (2002). What episodic encoding in humans. Current Opin- does the frontomedian cortex contribute to ion in Neurobiology, 16(6), 693–700. language processing: coherence or theory of Davis, J. I., Gross, J. J., & Ochsner, K. N. (2011). mind? NeuroImage, 17(3), 1599–1612. Psychological distance and emotional experi- Figner, B., Knoch, D., Johnson, E. J., Krosch, A. ence: What you see is what you get. Emotion, R., Lisanby, S. H., Fehr, E., et al. (2011). Lat- 11(2), 438–444. eral prefrontal cortex and self-­control in inter- Delgado, M. R., Gillis, M. M., & Phelps, E. A. temporal choice. Nature Neuroscience, 13(5), (2008). Regulating the expectation of reward 538–539. via cognitive strategies. Nature Neuroscience, Frankenstein, U. N., Richter, W., McIntyre, M. 11(8), 880–881. C., & Remy, F. (2001). Distraction modu- Delgado, M. R., Jou, R. L., LeDoux, J. E., & lates anterior cingulate gyrus activations dur- Phelps, E. A. (2009). Avoiding negative out- ing the cold pressor test. NeuroImage, 14(4), comes: Tracking the mechanisms of avoidance 827–836. learning in humans during fear conditioning. Gallagher, H. L., & Frith, C. D. (2003). Func- Frontiers in Behavioral Neuroscience, 3, 33. tional imaging of “theory of mind.” Trends in Delgado, M. R., Nearing, K. I., Ledoux, J. E., Cognitive Sciences, 7(2), 77–83. & Phelps, E. A. (2008). Neural circuitry Gianotti, L. R., Knoch, D., Faber, P. L., Lehm- underlying the regulation of conditioned fear ann, D., Pascual-M­ arqui, R. D., Diezi, C., et and its relation to extinction. Neuron, 59(5), al. (2009). Tonic activity level in the right pre- 829–838. frontal cortex predicts individuals’ risk tak- Delgado, M. R., Olsson, A., & Phelps, E. A. ing. Psychological Science, 20(1), 33–38. (2006). Extending animal models of fear con- Goldin, P. R., McRae, K., Ramel, W., & Gross, ditioning to humans. Biological Psychology, J. J. (2008). The neural bases of emotion regu- 73(1), 39–48. lation: Reappraisal and suppression of nega- Denny, B. T., Kober, H., Wager, T. D., & Och- tive emotion. Biological Psychiatry, 63(6), sner, K. N. (2012). A meta-­analysis of func- 577–586. tional neuroimaging studies of self- and other Gray, M. A., Beacher, F. D., Minati, L., Nagai,

Valuation and Emotion Regulation 39 Y., Kemp, A. H., Harrison, N. A., et al. interaction with transcranial direct current (2012). Emotional appraisal is influenced by stimulation—­the example of punishing unfair- cardiac afferent information. Emotion, 12(1), ness. Cerebral Cortex, 18(9), 1987–1990. 180 –191. Knoch, D., Pascual-­Leone, A., Meyer, K., Treyer, Gross, J. J. (1998). Antecedent- and response-­ V., & Fehr, E. (2006). Diminishing reciprocal focused emotion regulation: Divergent con- fairness by disrupting the right prefrontal cor- sequences for experience, expression, and tex. Science, 314(5800), 829–832. physiology. Journal of Personality and Social Kober, H., Barrett, L. F., Joseph, J., Bliss-­ Psychology, 74(1), 224–237. Moreau, E., Lindquist, K., & Wager, T. D. Gross, J. J. (2007). The handbook of emotion (2008). Functional grouping and cortical–­ regulation. New York: Guilford Press. subcortical interactions in emotion: A meta-­ Hamann, S. B., Ely, T. D., Hoffman, J. M., & analysis of neuroimaging studies. NeuroIm- Kilts, C. D. (2002). Ecstasy and agony: Acti- age, 42(2), 998–1031. vation of the human amygdala in positive Kober, H., Mende-S­iedlecki, P., Kross, E. F., and negative emotion. Psychological Science, Weber, J., Mischel, W., Hart, C. L., et al. 13(2), 135–141. (2010). Prefrontal–s­triatal pathway underlies Hare, T. A., Camerer, C. F., & Rangel, A. (2009). cognitive regulation of craving. Proceedings Self-c­ ontrol in decision-m­ aking involves mod- ulation of the vmPFC valuation system. Sci- of the National Academy of Sciences USA, ence, 324(5927), 646–648. 107(33), 14811–14816. Harrison, N. A., Gray, M. A., Gianaros, P. J., Kravitz, D. J., Saleem, K. S., Baker, C. I., & & Critchley, H. D. (2010). The embodiment Mishkin, M. (2011). A new neural framework of emotional feelings in the brain. Journal of for visuospatial processing. Nature Reviews Neuroscience, 30(38), 12878–12884. Neuroscience, 12(4), 217–230. Hartley, C. A., & Phelps, E. A. (2010). Chang- Kurth, F., Zilles, K., Fox, P. T., Laird, A. R., ing fear: The neurocircuitry of emotion regu- & Eickhoff, S. B. (2010). A link between the lation. Neuropsychopharmacology, 35(1), systems: Functional differentiation and inte- 136–146. gration within the human insula revealed by Holland, P. C., & Gallagher, M. (2004). meta-a­ nalysis. Brain Structure and Function, Amygdala–f­rontal interactions and reward 214(5–6), 519–534. expectancy. Current Opinion in Neurobiol- Lane, R. D., Reiman, E. M., Axelrod, B., Yun, ogy, 14(2), 148–155. L. S., Holmes, A., & Schwartz, G. E. (1998). Johnstone, T., van Reekum, C. M., Urry, H. L., Neural correlates of levels of emotional aware- Kalin, N. H., & Davidson, R. J. (2007). Failure ness: Evidence of an interaction between emo- to regulate: Counterproductive recruitment of tion and attention in the anterior cingulate top-down prefrontal–­subcortical circuitry in cortex. Journal of Cognitive Neuroscience, major depression. Journal of Neuroscience, 10(4), 525–535. 27(33), 8877–8884. LeDoux, J. E. (1993). Emotional memory: In Kalisch, R. (2009). The functional neuroanat- search of systems and synapses. Annals of the omy of reappraisal: Time matters. Neurosci- New York Academy of Sciences, 702, 149– ence and Biobehavioral Reviews, 33(8), 1215– 157. 1226. LeDoux, J. E. (2000). Emotion circuits in the Kalisch, R., Wiech, K., Critchley, H. D., & brain. Annual Review of Neuroscience, 23, Dolan, R. J. (2006). Levels of appraisal: A 155–184. medial prefrontal role in high-level appraisal LeDoux, J. E., & Gorman, J. M. (2001). A call of emotional material. NeuroImage, 30(4), to action: Overcoming anxiety through active 1458–1466. coping. American Journal of Psychiatry, Keay, K. A., & Bandler, R. (2001). Parallel cir- 158(12), 1953–1955. cuits mediating distinct emotional coping Levenson, R. W. (1999). The intrapersonal func- reactions to different types of stress. Neuro- tions of emotion [Special issue]. Cognition and science and Biobehavioral Reviews, 25(7–8), Emotion, 13(5), 481–504. 669 – 678. Leventhal, H. (1984). A perceptual–­motor the- Knoch, D., Nitsche, M. A., Fischbacher, U., ory of emotion. Advances in Experimental Eisenegger, C., Pascual-­Leone, A., & Fehr, E. Social Psychology, 17, 117–182. (2008). Studying the neurobiology of social Levy, D. J., & Glimcher, P. W. (2011). Compar- ing apples and oranges: Using reward-s­pecific

40 BIOLOGICAL BASES and reward-g­ eneral subjective value represen- Miller, E. K. (2000). The prefrontal cortex and tation in the brain. Journal of Neuroscience, cognitive control. Nature Reviews Neurosci- 31(41), 14693–14707. ence, 1(1), 59–65. Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M., Pfeifer, J. H., & Way, B. Miller, E. K., & Cohen, J. D. (2001). An inte- M. (2007). Putting feelings into words: Affect grative theory of prefrontal cortex function. labeling disrupts amygdala activity in response Annual Review of Neuroscience, 24, 167– to affective stimuli. Psychological Science, 202. 18(5), 421–428. Lieberman, M. D., Jarcho, J. M., Berman, S., Mischel, W., Shoda, Y., & Rodriguez, M. L. Naliboff, B. D., Suyenobu, B. Y., Mandelkern, (1989). Delay of gratification in children. Sci- M., et al. (2004). The neural correlates of pla- ence, 244(4907), 933–938. cebo effects: A disruption account. NeuroIm- age, 22(1), 447–455. Mitchell, J. P. (2009). Inferences about men- Lieberman, M. D., Jarcho, J. M., & Satpute, A. tal states. Philosophical Transactions of the B. (2004). Evidence-b­ ased and intuition-­based self-k­nowledge: An fMRI study. Journal of Royal Society of London B: Biological Sci- Personality and Social Psychology, 87(4), ences, 364(1521), 1309–1316. 421– 435. Mobbs, D., Hassabis, D., Seymour, B., March- Lieberman, M. D., Ochsner, K. N., Gilbert, D. ant, J. L., Weiskopf, N., Dolan, R. J., et al. T., & Schacter, D. L. (2001). Do amnesics (2009). Choking on the money: Reward-based exhibit cognitive dissonance reduction?: The performance decrements are associated with role of explicit memory and attention in atti- midbrain activity. Psychological Science, tude change. Psychological Science, 12(2), 20(8), 955–962. 135–140. Mobbs, D., Petrovic, P., Marchant, J. L., Has- Lindquist, K. A., & Barrett, L. F. (2008). Con- sabis, D., Weiskopf, N., Seymour, B., et al. structing emotion: The experience of fear as a (2007). When fear is near: Threat imminence conceptual act. Psychological Science, 19(9), elicits prefrontal–­periaqueductal gray shifts in 898–903. humans. Science, 317(5841), 1079–1083. Lindquist, K. A., Wager, T. D., Kober, H., Bliss-­ Muhlberger, A., Neumann, R., Wieser, M. J., Moreau, E., & Barrett, L. F. (2012). The brain & Pauli, P. (2008). The impact of changes in basis of emotion: A meta-­analytic review. spatial distance on emotional responses. Emo- Behavioral and Brain Sciences, 35(3), 121– tion, 8(2), 192–198. 143. Murray, E. A., O’Doherty, J. P., & Schoenbaum, Maas, W. (2000). On the computational power G. (2007). What we know and do not know of winner-­take-all. Neural Computation, about the functions of the orbitofrontal cortex 12(11), 2519–2535. after 20 years of cross-s­ pecies studies. Journal Mauss, I. B., Levenson, R. W., McCarter, L., of Neuroscience, 27(31), 8166–8169. Wilhelm, F. H., & Gross, J. J. (2005). The tie Neely, J. H. (1991). Semantic priming effects in that binds?: Coherence among emotion experi- visual word recognition: A selective review of ence, behavior, and physiology. Emotion, 5(2), current findings and theories. In D. Besner & 175–190. G. Humphreys (Eds.), Basic processes in read- McClure, S. M., Ericson, K. M., Laibson, D. I., ing: Visual word recognition (pp. 264–336). Loewenstein, G., & Cohen, J. D. (2007). Time Hillsdale, NJ: Erlbaum. discounting for primary rewards. Journal of Ochsner, K. N., & Barrett, L. F. (2001). A mul- Neuroscience, 27(21), 5796–5804. tiprocess perspective on the neuroscience of McClure, S. M., Laibson, D. I., Loewenstein, G., emotion. In T. J. Mayne & G. A. Bonanno & Cohen, J. D. (2004). Separate neural sys- (Eds.), Emotions: Currrent issues and future tems value immediate and delayed monetary directions (pp. 38–81). New York: Guilford rewards. Science, 306(5695), 503–507. Press. McRae, K., Hughes, B., Chopra, S., Gabrieli, J. Ochsner, K. N., & Gross, J. J. (2005). The cogni- D., Gross, J. J., & Ochsner, K. N. (2010). The tive control of emotion. Trends in Cognitive neural bases of distraction and reappraisal. Sciences, 9(5), 242–249. Journal of Cognitive Neuroscience, 22(2), Ochsner, K. N., & Gross, J. J. (2008). Cogni- 248–262. tive emotion regulation: Insights from social cognitive and affective neuroscience. Currents Directions in Psychological Science, 17(1), 153–158. Ochsner, K. N., Hughes, B., Robertson, E. R.,

Valuation and Emotion Regulation 41 Cooper, J. C., & Gabrieli, J. D. (2009). Neural Ploghaus, A., Becerra, L., Borras, C., & Bor- systems supporting the control of affective and sook, D. (2003). Neural circuitry underly- cognitive conflicts. Journal of Cognitive Neu- ing pain modulation: Expectation, hypnosis, roscience, 21(9), 1842–1855. placebo. Trends in Cognitive Sciences, 7(5), Ochsner, K. N., Ray, R. D., Cooper, J. C., Rob- 197–200. ertson, E. R., Chopra, S., Gabrieli, J. D. E., et al. (2004). For better or for worse: Neural Polk, T. A., Drake, R. M., Jonides, J. J., Smith, M. systems supporting the cognitive down- and R., & Smith, E. E. (2008). Attention enhances up-­regulation of negative emotion. NeuroIm- the neural processing of relevant features and age, 23(2), 483–499. suppresses the processing of irrelevant features Ochsner, K. N., Ray, R. R., Hughes, B., McRae, in humans: A functional magnetic resonance K., Cooper, J. C., Weber, J., et al. (2009). imaging study of the Stroop task. Journal of Bottom-up and top-down processes in emo- Neuroscience, 28(51), 13786–13792. tion generation: Common and distinct neural mechanisms. Psychological Science, 20(11), Price, J. L. (1999). Prefrontal cortical networks 1322–1331. related to visceral function and mood. Annals Ochsner, K. N., Silvers, J. A., & Buhle, J. T. of the New York Academy of Sciences, 877, (2012). Functional imaging studies of emo- 383–396. tion regulation: A synthetic review and evolv- ing model of the cognitive control of emotion. Quirk, G. J., & Beer, J. S. (2006). Prefron- tal involvement in the regulation of emo- Annals of the New York Academy of Sciences, tion: Convergence of rat and human studies. 1251, E1–E24. Curentr Opinion in Neurobiology, 16(6), Olsson, A., & Ochsner, K. N. (2008). The role of 723–727. social cognition in emotion. Trends in Cogni- tive Sciences, 12(2), 65–71. Rangel, A., Camerer, C., & Montague, P. R. Ongur, D., Ferry, A. T., & Price, J. L. (2003). (2008). A framework for studying the neurobi- Architectonic subdivision of the human orbital ology of value-based decision making. Nature and medial prefrontal cortex. Journal of Com- Reviews Neuroscience, 9(7), 545–556. parative Neurology, 460(3), 425–449. Ortony, A., Clore, G. L., & Collins, A. (1988). Richards, J. M., & Gross, J. J. (2000). Emotion The cognitive structure of emotions. New regulation and memory: The cognitive costs of York: Cambridge University Press. keeping one’s cool. Journal of Personality and Packard, M. G. (2009). Anxiety, cognition, and Social Psychology, 79(3), 410–424. habit: A multiple memory systems perspective. Brain Research, 1293, 121–128. Rolls, E. T. (1999). The brain and emotion. Pessoa, L. (2008). On the relationship between Oxford, UK: Oxford University Press. emotion and cognition. Nature Reviews Neu- roscience, 9(2), 148–158. Russell, J. A., & Barrett, L. F. (1999). Core Pessoa, L. (2009). How do emotion and motiva- affect, prototypical emotional episodes, and tion direct executive control? Trends in Cogni- other things called emotion: Dissecting the tive Sciences, 13(4), 160–166. elephant. Journal of Personality and Social Pessoa, L., Kastner, S., & Ungerleider, L. G. Psychology, 76(5), 805–819. (2003). Neuroimaging studies of attention: From modulation of sensory processing to Saxe, R. (2006). Uniquely human social cog- top-down control. Journal of Neuroscience, nition. Current Opinion in Neurobiology, 23(10), 3990–3998. 16(2), 235–239. Phelps, E. A. (2006). Emotion and cognition: Insights from studies of the human amygdala. Scherer, K. R. (2001). Appraisal considered as a Annual Review of Psychology, 57, 27–53. process of multilevel sequential checking. In K. Phillips, M. L., Ladouceur, C. D., & Drevets, W. R. Scherer & A. Schorr (Eds.), Appraisal pro- C. (2008). A neural model of voluntary and automatic emotion regulation: Implications cesses in emotion: Theory, methods, research for understanding the pathophysiology and (pp. 92–120). New York: Oxford University neurodevelopment of bipolar disorder. Molec- Press. ular Psychiatry, 13(9), 829, 833–857. Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.). (2001). Appraisal processes in emo- tion: Theory, methods, research. New York: Oxford University Press. Schoenbaum, G., Saddoris, M. P., & Stalnaker, T. A. (2007). Reconciling the roles of orbi- tofrontal cortex in reversal learning and the encoding of outcome expectancies. Annals of the New York Academy of Sciences, 1121, 320 –335.

42 BIOLOGICAL BASES Schultz, W., Dayan, P., & Montague, P. R. Wager, T. D., Davidson, M. L., Hughes, B. L., (1997). A neural substrate of prediction and Lindquist, M. A., & Ochsner, K. N. (2008). reward. Science, 275(5306), 1593–1599. Prefrontal–s­ubcortical pathways mediating successful emotion regulation. Neuron, 59(6), Sharot, T., De Martino, B., & Dolan, R. J. 1037–1050. (2009). How choice reveals and shapes expected hedonic outcome. Journal of Neuro- Wager, T. D., Jonides, J., & Reading, S. (2004). science, 29(12), 3760–3765. Neuroimaging studies of shifting attention: A meta-­analysis. NeuroImage, 22(4), 1679– Sharot, T., Shiner, T., & Dolan, R. J. (2010). 1693. Experience and choice shape expected aversive outcomes. Journal of Neuroscience, 30(27), Wager, T. D., & Smith, E. E. (2003). Neuroim- 9209–9215. aging studies of working memory: A meta-­ analysis. Cognitive, Affective, and Behavioral Thompson-S­ chill, S. L., Bedny, M., & Goldberg, Neuroscience, 3(4), 255–274. R. F. (2005). The frontal lobes and the regu- lation of mental activity. Current Opinion in Williams, L. E., & Bargh, J. A. (2008). Keeping Neurobiology, 15(2), 219–224. one’s distance: The influence of spatial dis- tance cues on affect and evaluation. Psycho- Tracey, I., Ploghaus, A., Gati, J. S., Clare, S., logical Science, 19(3), 302–308. Smith, S., Menon, R. S., et al. (2002). Imaging attentional modulation of pain in the periaq- Willis, W. D., & Westlund, K. N. (1997). Neu- ueductal gray in humans. Journal of Neurosci- roanatomy of the pain system and of the path- ence, 22(7), 2748–2752. ways that modulate pain. Journal of Clinical Neurophysiology, 14(1), 2–31. Urry, H. L., van Reekum, C. M., Johnstone, T., Kalin, N. H., Thurow, M. E., Schaefer, H. Wunderlich, K., Rangel, A., & O’Doherty, J. S., et al. (2006). Amygdala and ventromedial P. (2009). Neural computations underlying prefrontal cortex are inversely coupled during action-b­ased decision making in the human regulation of negative affect and predict the brain. Proceedings of the National Academy diurnal pattern of cortisol secretion among of Sciences USA, 106(40), 17199–17204. older adults. Journal of Neuroscience, 26(16), 4415 – 4425. Young, L., Camprodon, J. A., Hauser, M., Pascual-­Leone, A., & Saxe, R. (2010). Disrup- van Veen, V., Krug, M. K., Schooler, J. W., & tion of the right temporoparietal junction with Carter, C. S. (2009). Neural activity pre- transcranial magnetic stimulation reduces the dicts attitude change in cognitive dissonance. role of beliefs in moral judgments. Proceedings Nature Neuroscience, 12(11), 1469–1474. of the National Academy of Sciences USA, 107(15), 6753–6758. Wager, T. D. (2005). The neural bases of placebo effects in pain. Current Directions in Psycho- Zaki, J., Davis, J. I., & Ochsner, K. N. (2012). logical Science, 14(4), 175–179. Overlapping activity in anterior insula during interoception and emotional experience. Neu- Wager, T. D., Atlas, L. Y., Leotti, L. A., & Rill- roImage, 62(1), 493–499. ing, J. K. (2011). Predicting individual differ- ences in placebo analgesia: Contributions of Zysset, S., Huber, O., Ferstl, E., & von Cramon, brain activity during anticipation and pain D. Y. (2002). The anterior frontomedian cor- experience. Journal of Neuroscience, 31(2), tex and evaluative judgment: An fMRI study. 439 – 452. NeuroImage, 15(4), 983–991. Wager, T. D., Barrett, L. F., Bliss-­Moreau, E., Zysset, S., Huber, O., Samson, A., Ferstl, E. C., Lindquist, K., Duncan, S., Kober, H., et al. & von Cramon, D. Y. (2003). Functional spe- (2008). The neuroimaging of emotion. In M. cialization within the anterior medial prefron- Lewis, J. M. Haviland-J­ones, & L. F. Barrett tal cortex: A functional magnetic resonance (Eds.), The handbook of emotion (3rd ed., imaging study with human subjects. Neurosci- pp. 249–271). New York: Guilford Press. ence Letters, 335(3), 183–186.

Chapter 3 Temporal Dynamics of Emotion Regulation Greg€Hajcak€Proudfit Jonathan€P.€Dunning Daniel€Foti Anna€Weinberg Gross and colleagues have framed emotion in frontal activation precede decreases in generation and regulation as a recursive amygdala activity. process involving attention and appraisal, whereby an emotional response unfolds over This chapter focuses on a different neu- time as a function of both how attention is roimaging method that complements the allocated to an emotion-eÂ

44 BIOLOGICAL BASES ErPs and the lPP presentation of a visual stimulus, frequently What Is an ERP? images from the International Affective ERPs are a direct measure of neural activ- Picture System (IAPS; Lang, Bradley, & ity. Scalp-recorded ERPs reflect the summed Cuthbert, 2008), a standardized database activity of excitatory and inhibitory post- containing hundreds of pictures that have synaptic potentials (PSPs) generated by large been normed on ratings of hedonic valence populations of pyramidal cortical neurons and arousal. Neural activity elicited by emo- aligned perpendicular to the cortical sur- tional compared to neutral pictures, and as a face. Because these neurons are oriented in function of emotion regulation instructions, the same direction, simultaneously occur- can then be averaged and compared within ring or closely occurring signals typically and between subjects. do not cancel each other out; moreover, because PSPs tend to be rather sustained in The LPP nature (compared to, for example, action The LPP is a sustained positive-going ERP potentials), it is possible for this activity to with a central–parietal scalp distribution; summate and propagate to the surface of the the amplitude of the LPP is modulated by scalp (Luck, 2005). This summated activity emotional content, becoming more positive is what is recorded in the ongoing EEG (see as early as 200 milliseconds following the Figure 3.1). ERPs then reflect the activity presentation of both pleasant and unpleas- of the EEG, time-locked to specific events ant compared to neutral stimuli (see Figure and averaged across many trials to increase 3.1; Foti, Hajcak, & Dien, 2009; Hajcak signal and reduce noise. In the studies we & Olvet, 2008; Hajcak, Weinberg, Mac- discuss, the event of interest is typically the Namara, & Foti, 2011). The LPP is larger FIGurE 3.1. A participant views images as the ongoing electroencephalogram (EEG) is recorded via sensors affixed to the scalp (left). Event-related potentials (ERPs) reflect the activity of the EEG time- locked to an event of interest, in this case, the onset of an image. On the right, the LPP is presented as a positive-going deflection in the waveform, which is enhanced for both pleasant and unpleasant images compared to neutral (data from Weinberg & Hajcak, 2010). Additionally, while the early portion of the LPP (400–800 milliseconds) has a parietal distribution—represented here by the positive difference between unpleasant and neutral images, as indicated by the dark area on the scalp—the later LPP (after 1,000 milliseconds) has a broader and more frontal distribution.

Temporal Dynamics of Emotion Regulation 45 following the presentation of emotional emotional stimuli (Hajcak et al., 2010; Thi- images (Cuthbert, Schupp, Bradley, Birbau- ruchselvam, Hajcak, & Gross, 2012). mer, & Lang, 2000; Foti et al., 2009), words (Kissler, Herbert, Winkler, & Jung- The modulation of the LPP is thought to hofer, 2009; Tacikowski & Nowicka, 2010), reflect the motivational salience of the stim- and even hand gestures (Flaisch, Häcker, ulus being viewed. In the case of emotional Renner, & Schupp, 2011). stimuli, salience is thought to be determined by the content of the stimuli—­and how rele- It has been argued that the LPP reflects vant the content is to basic biological imper- sustained attention to visual stimuli (Haj- atives (e.g., survival themes that include cak, MacNamara, & Olvet, 2010; Hajcak defense and reproduction). Indeed, evidence & Olvet, 2008; Hajcak et al., 2011; Wein- suggests that the LPP is most enhanced by berg & Hajcak, 2011b). Furthermore, there image content more directly related to these may be functional differences between imperatives, such as images of mutilation early compared to late portions of the LPP, and gore, or erotic images (Weinberg & highlighting the utility of this component Hajcak, 2010). Additionally, food depriva- in tracking multiple processes over time. tion compared to satiated states enhances Relatively early portions of the LPP (i.e., the magnitude of the LPP elicited by images 300–600 milliseconds) appear to reflect of food—a motivationally salient category more obligatory processing of emotional of images when one is hungry (Stock- stimuli, whereas later time windows (i.e., burger, Schmälzle, Flaisch, Bublatzky, & 600 milliseconds and beyond) might reflect Schupp, 2009). The LPP is also enhanced more sustained and elaborative engagement by personally relevant stimuli (e.g., photo- with stimuli (Olofsson, Nordin, Sequeira, & graphs of relatives, or one’s own name and Polich, 2008; Weinberg & Hajcak, 2011b; face; Grasso & Simons, 2010; Tacikowski Weinberg, Hilgard, Bartholow, & Hajcak, & Nowicka, 2010) compared to even very 2012). For instance, in one study, larger LPPs familiar famous faces. in response to task-­irrelevant pictures pre- dicted greater behavioral interference—i­n There is also evidence that extrinsic both within- and between-­subjects analyses manipulations of salience can impact the (Weinberg & Hajcak, 2011b). Moreover, LPP. It is possible to designate specific stim- though the early LPP was modulated by uli as “targets” by asking participants to emotional stimuli, it was the later portion count or otherwise respond to them when of the LPP (i.e., just preceding target onset) they are presented among other nontarget that uniquely predicted behavioral interfer- stimuli. The LPP is enhanced for arbitrarily ence: Only increased sustained attention just designated target stimuli compared to non- prior to target presentation related to slower target stimuli (Ferrari, Bradley, Codispoti, reaction times. & Lang, 2010; Weinberg et al., 2012), even when the same images serve as both targets During passive picture-­viewing para- and nontargets (Weinberg et al., 2012). The digms, the emotional modulation of the LPP LPP is largest when emotional images are is sustained throughout stimulus presenta- also targets, compared to both neutral tar- tion (i.e., up to several seconds; Hajcak et gets and emotional nontargets (Ferrari et al., al., 2010), and can even be observed for as 2010; Weinberg et al., 2012). This suggests long as 1,000 milliseconds after stimulus that intrinsic and extrinsic manipulations of offset. These data are consistent with the salience are additive in their impact on the notion that emotional stimuli not only cap- magnitude of the LPP. ture but also hold attention (Vuilleumier, 2005)—and we have suggested that the LPP Work that combines EEG and fMRI sug- is sensitive to the continued allocation of gests that the LPP reflects widespread and attention to emotional stimuli. Consistent concurrent activity across the visual system with this view, it is possible to manipulate (Sabatinelli, Keil, Frank, & Lang, 2013), the duration of the LPP by instructions including the visual cortex (Bradley et al., that encourage maintainance of previously 2003; Keil et al., 2002) as well as occipital, viewed images in working memory, again parietal, and inferotemporal regions of the suggesting that this later component may brain (Sabatinelli, Lang, Keil, & Bradley, reflect more volitional engagement with 2007). Rather than the activation of a sin- gle anatomical node, it is possible that the

46 BIOLOGICAL BASES LPP reflects fairly widespread neuromodu- of emotional processing across the lifespan; latory activity (de Rover et al., 2012; Haj- affective modulation of the LPP has been cak et al., 2010), or communication among observed in both very young children (Den- diverse brain regions. This latter possibility nis & Hajcak, 2009; Kujawa, Hajcak, Tor- is supported by a recent study indicating pey, Kim, & Klein, 2012) and in adults as that the magnetic equivalent of the LPP was old as 81 (Kisley, Wood, & Burrows, 2007; generated in occipitoparietal and prefrontal Langeslag & Van Strien, 2010). Further- cortices—­and that the emotional modula- more, the LPP has been used to test theo- tion of the LPP might index the coordina- ries of aging, which posit that younger com- tion of frontoparietal attention networks pared to older participants are more reactive (Moratti, Saugar, & Strange, 2011). to unpleasant stimuli, and less reactive to pleasant stimuli (Kisley et al., 2007; Lang- The involvement of frontal areas in the eslag & Van Strien, 2010). generation of the LPP is consistent with stud- ies indicating that both physical stimulation Additionally, the LPP has been used effec- (Hajcak, Anderson, et al., 2010) and func- tively to characterize the abnormal process- tional activation (based on working memory ing of affective stimuli in anxiety, mood, and load; MacNamara, Ferri, & Hajcak, 2011) substance use disorders (e.g., Dunning et al., of the dorsolateral prefrontal cortex reduces 2011; Foti, Olvet, Klein, & Hajcak, 2010; the amplitude of the LPP. Moreover, fron- Weinberg & Hajcak, 2011a). For instance, tal involvement in the LPP is also consistent consistent with the notion that depression with our observation that the scalp distri- is characterized by motivational disengage- bution of the emotional modulation of the ment from the environment, depressed indi- LPP during passive picture viewing becomes viduals tend to show decreased emotional evident at central and even frontal recording modulation of the LPP compared to healthy sites after approximately 1,000 milliseconds controls (Foti et al., 2010). Likewise, while (see Figure 3.1; Foti et al., 2009; MacNa- cocaine use disorders are associated with a mara, Foti, & Hajcak, 2009). decreased LPP in response to normative emo- tional stimuli, individuals with this diagno- A substantial literature indicates that the sis also show an increased LPP in response LPP is a robust measure of neural activity to cocaine-­related images compared to associated with emotional processing. In healthy controls (Dunning et al., 2011). The contrast to the majority of peripheral and relationship of the LPP to anxiety appears central measures that are sensitive to emo- to be somewhat context-­dependent. For tional stimuli, emotional modulation of the instance, anxiety has been associated with LPP is highly stable. Whereas skin conduc- an increased LPP when pictures are task-­ tance, heart rate, facial muscle activity, and irrelevant (MacNamara, Ferri, et al., 2011), neural activation indexed by fMRI habitu- suggesting deficits in disengaging attention ate over repeated presentations of stimuli from emotional stimuli. However, there is (Codispoti & De Cesarei, 2007; Codispoti, also evidence from passive-­viewing tasks— Ferrari, & Bradley, 2006; Phan, Liberzon, in which the pictures are task-­relevant—t­ hat Welsh, Britton, & Taylor, 2003), emotional anxiety is instead characterized by initial modulation of the LPP does not (Codispoti vigilance for threat images, as indicated by et al., 2006; Codispoti, Ferrari, & Bradley, enhancement of early ERP components, fol- 2007; Olofsson & Polich, 2007). lowed by a failure to engage in sustained processing of these images, as indicated by The LPP and Individual Differences an attenuation of the LPP (Weinberg & Haj- The LPP has proven to be an effective cak, 2011a). These data indicate that the measure of attentional allocation to emo- LPP is a useful tool for quantifying abnor- tional stimuli across multiple populations. malities in the temporal dynamics of emo- For example, enhancement of LPP ampli- tional processing across different forms of tude by emotional content appears cross-­ psychopathology. Below, we review findings culturally (Hot, Saito, Mandai, Kobayashi, that have extended this literature to emotion & Sequeira, 2006; Yen, Chen, & Liu, 2010). regulation, focusing in particular on atten- The LPP has also been applied to the study tional deployment and reappraisal.

Temporal Dynamics of Emotion Regulation 47 The LPP and Emotion Regulation tive method of down-r­egulating emotional Directed Attention response. An emotional response to a stimulus unfolds over time as a function of multiple pro- Online Regulation cesses, including the allocation of attention Attempts to regulate an emotional response to the content of the stimulus, as well as the after a stimulus has been encountered are meaning that is assigned to that stimulus considered “online” regulation. Recent (Sheppes & Gross, 2011). With regard to work has examined whether the amplitude emotion regulation, distraction and other of the LPP can be manipulated by explic- attentional deployment strategies alter the itly directing participants’ visual attention emotional response by directing attention to various areas within unpleasant images. away from affective content. Spatial atten- This approach is considered online regula- tion, in particular, plays a large role in tion insofar as visual attention is manipu- the electrocortical response to emotional lated during the generation of the emotional images (Eimer, Holmes, & McGlone, 2003; response. In one study, participants were Holmes, Vuilleumier, & Eimer, 2003; Keil, shown an unpleasant picture for 6 sec- Moratti, Sabatinelli, Bradley, & Lang, 2005) onds; after passively viewing the image for and, consequently, emotion regulation pro- 3 seconds, participants’ attention was then cesses. Indeed, a growing body of work directed to either an arousing or nonarous- demonstrates that directing attention, either ing area of the picture for the last 3 seconds implicitly or explicitly, away from emotion- of each trial. As indicated in Figure 3.2, ally salient features of a stimulus is an effec- directing attention to nonarousing com- LPP to unpleasant Reduction in LPP when pictures during focusing on nonarousing passive viewing areas of unpleasant pictures FIGURE 3.2.  Grand averaged ERPs at electrode site Pz elicited by neutral pictures (solid black line; always associated with an instruction to focus on a nonarousing area of the image) and unpleasant pic- tures (associated with an instruction to attend to either a nonarousing [dotted line] or arousing [dashed line area of the image]). Picture onset occurred at 0 milliseconds and the instruction tone occurred at 3,000 milliseconds. The figure is based on data collected by Hajcak, Dunning, and Foti (2009).

48 BIOLOGICAL BASES pared to highly arousing areas of unpleasant neutral flickering stimuli, and modulation of images resulted in a decreased LPP (Dun- this activity is linked to facilitated process- ning & Hajcak, 2009). This finding was ing in the visual cortex (Keil et al., 2003). subsequently replicated with a focus on the In a study combining the LPP and ssVEP time course of LPP modulation: Sequential within a single paradigm, directing attention significance testing indicated that reduction to nonarousing areas of unpleasant pictures of the LPP occurred 620 milliseconds after reduced both of these electrocortical indices, signaling participants to attend to nonar- indicating that each neural measure tracks ousing areas of unpleasant images (Hajcak, the effective down-r­egulation of the emo- Dunning, & Foti, 2009). tional response via attentional deployment (Hajcak et al., 2013). Moreover, modula- Directed attention also appears to impact tions of the LPP and ssVEP were unrelated, the emotional response to representations suggesting that these measures provide dis- in working memory during a post-s­timulus tinct sources of information regarding emo- period. Thiruchselvam and colleagues tional processing. (2012) had participants view unpleasant and neutral pictures, then maintain representa- Cognitive Reappraisal tions of each image in working memory Unlike distraction and other attentional after picture offset. Participants were then deployment techniques, where attention is instructed to focus on arousing or nonarous- directed away from the emotional content ing aspects of their mental representation. of a stimulus, reappraisal involves attend- Results showed that focusing on nonarous- ing directly to emotional content and alter- ing compared to arousing aspects of pic- ing the emotional response by reinterpret- tures held in working memory was related ing the meaning of the stimulus (Sheppes & to a reduction in both the LPP and self-­ Gross, 2011). In this way, emotion regula- reported ratings of unpleasantness. These tion is achieved while engaging directly with results suggest that the emotional response emotionally arousing aspects of stimuli, an can be modulated, and indexed by the LPP, approach that has been shown to be effec- even after images have been fully encoded in tive for reducing both subjective and periph- working memory. eral physiological indicators of emotional arousal (Urry, 2010). Of interest is how LPP Convergence of Directed Attention amplitude may complement these data as a Findings across Neural Measures neurobiological index of effective emotion This same pattern of LPP modulation via regulation, clarifying the impact of cognitive directed spatial attention has also been rep- reappraisal on the time course of emotional licated and extended to steady-­state visual processing. evoked potentials (ssVEP; Hajcak, MacNa- mara, Foti, Ferri, & Keil, 2013). The ssVEP According to the process-s­pecific tim- can be used to index visual processing of ing hypothesis, online emotion regulation stimuli flickering at a particular frequency. It likely requires significant effort (i.e., cogni- is not an ERP component, but rather a mea- tive resources) in order to modify existing sure of the degree to which the EEG signal and incoming emotional information—a­ nd over occipital cortex is being driven at the that effort would be directly proportional flicker frequency. For instance, suppose two to the strength and intensity of the emotion faces are simultaneously presented to the left being experienced (Sheppes & Gross, 2011). and right of fixation, flickered at 8 and 18 On the other hand, several studies have Hz, respectively. The ssVEP would be charac- presented emotion regulation instructions terized by increased activity at both frequen- to participants prior to the presentation of cies, which is typically maximal at occipital emotional stimuli. In this way, regulation electrodes. Furthermore, if participants were goals might be established while emotional instructed to attend to the left face, power at intensity is low, in anticipation of the emo- 8 Hz in the ssVEP would be increased (Hill- tional response to come. Anticipatory regu- yard et al., 1997). Related to this chapter, the lation ought to impact the initial processing ssVEP is larger for emotional compared to of an affective stimulus, indicated by the

Temporal Dynamics of Emotion Regulation 49 affective modulation of the LPP elicited by In three additional studies, the impact of the first presentation of the stimulus. cognitive reappraisal within an anticipatory emotion regulation context was examined, In one of the first investigations of cogni- testing whether alteration of the manner tive reappraisal using the LPP, Hajcak and in which participants appraise stimuli dur- Nieuwenhuis (2006) had participants view ing the initial presentation would modulate a series of unpleasant IAPS images for 1 LPP amplitude. Participants were instructed second; an instruction to “reinterpret” (i.e., to use effortful cognitive reappraisal of reappraise the picture in order to reduce emotional scenes (Krompinger, Moser, one’s negative response) or “attend” (i.e., & Simons, 2008; Moser, Hajcak, Bukay, focus on one’s natural feelings about the & Simons, 2006) and angry faces (Blech- picture) was then presented for 4,500 mil- ert, Sheppes, Di Tella, Williams, & Gross, liseconds, after which the same picture was 2012). Prior to viewing each stimulus, par- again presented for 2,000 milliseconds. ticipants were instructed to decrease the Compared to unpleasant pictures presented intensity of their emotional response by rein- after the attend instructions, those presented terpreting the image. Compared to a passive after reappraisal instructions were associ- viewing condition, the LPP elicited by pleas- ated with a reduced LPP, beginning 200 mil- ant and unpleasant images was blunted for liseconds after picture onset—an effect that the reappraisal condition at an early latency lasted for the duration of picture presenta- range, beginning approximately 300–600 tion. Furthermore, greater reduction in the milliseconds following picture presentation. LPP was related to greater reduction in self-­ Together, these lines of research demonstrate reported arousal ratings of the images. This that LPP amplitude tracks the successful study was the first to demonstrate that the down-r­egulation of an emotional response magnitude of the LPP could be modulated using cognitive reappraisal, whether reap- by cognitive reappraisal, down-­regulating praisal is enacted online or in anticipation of the LPP by reinterpreting the meaning of an affective stimulus. emotional stimuli. Building on these lines of research, Thi- Anticipatory Regulation ruchselvam, Blechert, Sheppes, Rydstrom, In the Hajcak and Nieuwenhuis (2006) and Gross (2011) compared the effects of study, regulation instructions were pre- different anticipatory regulation strategies sented after the first encounter with the on LPP amplitude at both the initial pre- emotional stimulus, indicating that the LPP sentation of stimuli and a later reexposure. is sensitive to cognitive reappraisal when it Whereas the aforementioned studies com- is done online. Separate from this study, and pared reappraisal to a passive viewing con- consistent with the notion of anticipatory dition, here the impact of reappraisal on the regulation, experimental manipulations that temporal dynamics of LPP amplitude was vary how participants attend to affective also contrasted with the impact of distrac- stimuli during the initial presentation have tion instructions. Reappraisal was achieved also been shown to modulate the LPP. For by instructing participants to reinterpret example, Hajcak, Moser, and Simons (2006) affective stimuli in a more neutral man- had participants view pleasant and unpleas- ner. Distraction, on the other hand, was ant pictures, and either categorize the pic- achieved by asking participants to generate tures affectively (i.e., indicating whether neutral thoughts unrelated to the stimulus, the picture was pleasant or unpleasant) or such as imagining complex geometric pat- nonaffectively (i.e., indicating how many terns. During a baseline session, the LPP in people were present in the picture). Results response to unpleasant images was recorded indicated that the LPP elicited by both pleas- under conditions of passive viewing, effort- ant and unpleasant pictures was reduced ful reappraisal, and distraction; 30 minutes when participants evaluated images—­and, later, participants passively viewed the same presumably, interpreted the meaning of the images with no regulation instructions. images—­along a nonaffective compared to Effortful reappraisal, compared to passive an affective dimension. viewing, yielded a blunted LPP only within a relatively later latency range (i.e., 1,500–

50 BIOLOGICAL BASES 1,700 milliseconds). The effect of distrac- regulate the initial emotional response elic- tion was stronger and apparent earlier, with ited by the image. On each trial, the descrip- a significant and sustained reduction in LPP tion precedes the presentation of the image, amplitude as early as 300 milliseconds. A thereby shaping the first appraisal of the very different pattern emerged during the stimulus during the initial iteration of the reexposure session: Images that had previ- emotion-­generative cycle. Guiding the ini- ously been processed under the reappraisal tial stimulus appraisal with verbal descrip- condition again yielded a blunted LPP com- tions removes the potential confound of task pared to the passive viewing condition, from difficulty, such that each condition simply 800 to 1,400 milliseconds. Images that had involves listening to descriptions and then previously been processed under the dis- viewing pictures. traction condition, however, yielded a sus- tained increase in LPP amplitude compared In an initial study using this paradigm with to the passive viewing condition, beginning an adult sample, modulation of LPP ampli- at approximately 1,200 milliseconds. This tude by description type was evident as early study highlights the fact that emotion reg- as 400 milliseconds following picture pre- ulation strategies likely have a differential sentation, with neutral descriptions reducing impact on initial compared to subsequent the subsequent LPP elicited by unpleasant encounters with emotional stimuli. Unlike images (Foti & Hajcak, 2008). Compared reappraisal, distraction may only be an to neutral images—­which in this study were effective emotion regulation strategy in the always preceded by a neutral description—­ short-term, reducing LPP amplitude during the LPP in response to unpleasant images the initial stimulus exposure but resulting in preceded by unpleasant descriptions was an enhanced LPP during reexposure. increased throughout stimulus presentation, from 400 to 3,000 milliseconds. The LPP in Preappraisal response to unpleasant images was increased The previous studies indicate that LPP only from 400 to 1,000 milliseconds when amplitude is sensitive to the implementation preceded by neutral descriptions, and to a of anticipatory emotion regulation through lesser extent than when preceded by nega- cognitive reappraisal, but they do not fully tive descriptions. This temporal pattern sug- address whether the down-­regulation of the gests that unpleasant images preappraised LPP is due to a shift in stimulus appraisal/ by a neutral description were associated with meaning per se. An alternate explanation reduced early, obligatory attentional capture is that the reduction in LPP amplitude is a associated with emotion generation; more- result of reappraisal being more difficult and over, the sustained, elaborative processing more cognitively taxing than passive view- that is typically observed for highly arousing ing, an important possibility to rule out in affective stimuli was nearly absent. Unpleas- light of evidence that cognitive load results ant images preceded by neutral descriptions in a reduced LPP (MacNamara, Ferri, et al., were also rated as less unpleasant and less 2011). To pursue this issue further, several emotionally arousing than those preceded studies have used a guided version of cog- by negative descriptions, providing fur- nitive reappraisal, in which images are pre- ther evidence that manipulating the initial ceded by verbal descriptions that frame the appraisal of the stimuli effectively regulated upcoming stimuli in either a more negative the subsequent emotional response. or more neutral manner (Foti & Hajcak, 2008; MacNamara et al., 2009). In other Because neutral images were always pre- words, reappraisal frames were provided to ceded by neutral descriptions in this initial participants, a manipulation that might be study, an important alternative explanation best described as preappraisal. For example, was that the reduced LPP observed for neu- an unpleasant image of a man pointing a trally described unpleasant images was due gun at his head might be preceded by either to the mismatch between the valence of the “This man is about to commit suicide” or description and the image. To address this “This man ends up not committing suicide”; confound in a follow-­up study on preap- the latter description is designed to down-­ praisal, description and image type were fully crossed: Neutral and negative descrip- tions preceded both neutral and unpleas- ant images (MacNamara et al., 2009). For

Temporal Dynamics of Emotion Regulation 51 instance, a neutral image of a drill might be Although the aforementioned studies used described as “This drill is used by a repair- paradigms in which the stimuli were task-­ man” or “This drill was used in a grisly mur- relevant and the primary focus of attention, der,” with the latter description designed modifying the initial stimulus appraisal to up-r­egulate the emotional response elic- also appears to impact the LPP elicited by ited by the image. As before, the LPP was task-­irrelevant, distracting images. For reduced in response to unpleasant images instance, during a task in which participants preceded by neutral descriptions compared were required to judge the spatial orienta- to those preceded by negative descriptions. tion of geometric shapes, mutilation and Importantly, though, the LPP was increased neutral scenes were presented as distrac- in response to neutral pictures that were tors (Mocaiber et al., 2010). These images preceded by negative compared to neutral were described beforehand as being either descriptions. Framing neutral images in a real or fictitious, thereby shaping the initial more negative manner elicited affective mod- appraisal. In the “real” context, unpleasant ulation of the LPP that would not normally images elicited an increased LPP compared be observed for low-­arousal, neutral stimuli. to neutral images in an early latency range These data demonstrate that the LPP tracks (i.e., 300–600 milliseconds) and were asso- the emotionality of the preappraisal and not ciated with significant reaction time slow- the match between description and image. In ing, indicating attentional capture that inter- fact, at late latencies (>1,500 milliseconds), fered with performance on the primary task. affective modulation of LPP amplitude was In the “fictitious” context, however, there apparent only for images preceded by nega- was no affective modulation of the LPP and tive descriptions, regardless of whether the no reaction time slowing, indicating that image itself was unpleasant or neutral. Thus, the distractor images were less salient and the later portion of the LPP was only sensi- produced less behavioral interference. This tive to meaning imbued by the preappraisals. study is consistent with other research link- ing the LPP to attentional capture and task Altering the context in which stimuli are interference (Weinberg & Hajcak, 2011b), initially appraised has a robust impact on and it demonstrates that the LPP elicited by LPP amplitude during the first viewing of a task-i­rrelevant stimuli can be modulated by stimulus. A further question is how modify- preappraisal. ing the initial appraisal may influence sub- sequent iterations of the emotion-­generative Convergence of Cognitive Reappraisal process when the stimuli are encountered Findings across Neural Measures again. To examine this, MacNamara, Och- The work reviewed thus far focused on iden- sner, and Hajcak (2011) first presented tifying the consequences of regulation on images paired with negative or neutral the emotion-g­enerative process, as indexed descriptions, then conducted a second view- by electrocortical measures. In other words, ing session 30 minutes later without the pre- electrocortical measures were used to index ceding descriptions. During the reexposure the downstream impact of emotion regu- session, images that had previously been pre- lation on neural activity. In addition to appraised in a more neutral manner elicited this, it may be possible to measure neural a blunted LPP in a middle latency range (i.e., activation responsible for these regulation 700–1,400 milliseconds), and the images effects. For instance, Parvaz, MacNamara, were rated as being less unpleasant and less Goldstein, and Hajcak (2012) examined emotionally arousing. This is a less pro- the LPP and alpha bandwidth power while tracted effect of preappraisal on LPP than participants either passively viewed or reap- observed previously (Foti & Hajcak, 2008; praised unpleasant pictures. Consistent with MacNamara et al., 2009), suggesting that previous work, the amplitude of the LPP preappraisal effects may diminish over time. was reduced when reappraising unpleas- However, modifying the initial appraisal of ant pictures. However, reappraisal was also an affective stimulus—w­ hether by effortful associated with increased left prefrontal reappraisal (Thiruchselvam et al., 2011) or cortex activation, as evidenced by reduced preappraisal (MacNamara, Ochsner, et al., left frontal alpha band power (alpha power 2011)—appears to yield down-r­egulation of the LPP upon reexposure.

52 BIOLOGICAL BASES is inversely related to brain activity). LPP over time through active attentional avoid- amplitude and alpha power were not corre- ance. In one study of individuals with a lated in this study, though, suggesting that spider phobia, the LPP elicited by spider both of these measures contributed unique stimuli was increased compared to other information about cognitive reappraisal negative stimuli during an early time win- processes. These data are consistent with dow (340–770 milliseconds) but not a later fMRI research indicating that reappraisal time window (800–1,500 milliseconds; is associated with increased activity in the Leutgeb, Schafer, & Schienle, 2009)—con- ventromedial and lateral prefrontal cor- sistent with an attentional pattern of vigi- tex (Johnstone, van Reekum, Urry, Kalin, lance followed by avoidance. Participants & Davidson, 2007; Ochsner et al., 2004; were then treated with exposure therapy in Wager, Davidson, Hughes, Lindquist, & which they were trained to approach and Ochsner, 2008), and suggests that EEG can engage with a spider until their anxiety be used not only to index the downstream diminished, rather than avoid the spider. consequences of emotion regulation but also Compared to a wait-list control group, indi- the neural activity that may be responsible viduals in the exposure therapy condition for regulatory effects. exhibited an increased and sustained LPP in response to spider stimuli from 800 to 1,500 Clinical Applications milliseconds at a follow-u­ p testing session. Researchers have begun to apply the LPP to This effect of treatment on LPP amplitude the study of emotion regulation in clinical was specific to spider stimuli and was not contexts. For example, the effects of preap- observed for other unpleasant stimuli. These praisal have been studied in children, with data suggest that successful treatment led to the purpose of identifying risk factors that increased attention and engagement with the might aid in detection and treatment of emo- phobic object. tion regulation difficulties early in life. In one study of children ages 7–10, the LPP was Other researchers have used the LPP to reduced for unpleasant images preceded by a clarify patterns of emotion regulation defi- neutral description, compared to unpleasant cits associated with the treatment of sub- images preceded by a negative description stance use disorders. Attentional biases (Dennis & Hajcak, 2009). This modulation toward drug-r­elated stimuli play a sig- of the LPP by description type was associ- nificant role in drug abuse and addiction ated with reduced symptoms of depression (Field & Cox, 2008), and several studies and anxiety, suggesting that modulation of have found that larger LPPs are elicited by LPP amplitude may serve as an early marker drug-r­elated compared to neutral pictures for emotion regulation difficulties or mood in alcoholics (Herrmann et al., 2000; Nam- disruptions. In younger children ages 5–7, koong, Lee, Lee, Lee, & An, 2004), cocaine-­ however, preappraisal does not appear to addicted individuals (Dunning et al., 2011; modulate LPP amplitude (Decicco, Solo- Franken et al., 2008), and smokers (Littel mon, & Dennis, 2012; Dennis & Hajcak, & Franken, 2007, 2011a). Furthermore, 2009), suggesting important developmental attentional bias toward drug cues and away differences that warrant further investiga- from other, intrinsically pleasant stimuli tion. has been related to risk for relapse. In one smoking cessation study, the LPP elicited by The LPP has also been used to track cigarette-­related cues was increased among emotion regulation processes related to the all smokers, but a subgroup of individuals treatment of psychological disorders. For also exhibited a blunted LPP to normative example, individuals with phobias are char- pleasant stimuli that was associated with acterized by an attentional pattern of vigi- higher rates of relapse over 6 months (Ver- lance followed by avoidance: They rapidly sace et al., 2012). This suggests that the LPP detect and attend to phobic objects initially could serve as a biomarker for identifying but quickly direct attention away from the smokers with a higher risk of relapse (Ver- phobic object. This is a maladaptive emo- sace et al., 2012). tion regulation strategy that prevents habit- uation and reinforces the anxiety response Given that drug craving has been shown to be modulated by cognitive regulation (Kober, Kross, Mischel, Hart, & Ochsner,

Temporal Dynamics of Emotion Regulation 53 2010), Littel and Franken (2011b) sought exciting questions that can be answered in to examine whether the LPP in response to the years ahead. smoking-­related images could be modulated by emotion regulation strategies in smok- Initial research indicated that the LPP ers. Indeed, both reappraisal (i.e., viewing is reduced via reappraisal, and subsequent the scenes from an uninvolved, rational, and studies confirm that changes in stimulus detached perspective) and distraction (i.e., meaning are sufficient to modulate the LPP. identifying the dominant color in the scene) Moreover, meaning-b­ased changes that strategies reduced the amplitude of the LPP characterize reappraisal appear to alter the to smoking-r­elated pictures; this effect was LPP elicited by subsequent encounters with prevalent in both light and heavy smokers, emotional stimuli—­and more so than other suggesting that regulation of drug-r­elated emotion regulation strategies (i.e., distrac- stimuli can be accomplished by smokers tion). Both the LPP and ssVEPs are modu- with differing levels of dependency (Littel & lated by attentional deployment: When Franken, 2011b). participants focus on less arousing aspects of emotional stimuli, both metrics of neu- Future Research Directions ral activity are down-­regulated—a­nd these Although electrocortical activity has been effects become evident within about 200 used for decades to study cognitive pro- milliseconds. Because of this temporal reso- cesses that include attention, it has more lution, electrocortical indices of emotion and recently been used to measure attention to its regulation can be used to index the itera- emotional stimuli (Hajcak et al., 2011). We tive and recursive processes that determine have argued that the LPP is modulated by an emotional response. Recent work also sustained attention to motivationally salient highlights the possibility that frontal acti- stimuli, possibly indexing the coordinated vation in reappraisal can be indexed using increased activity of frontoparietal attention time-­frequency decompositions of EEG data networks. However, the precise mechanisms (Parvaz et al., 2012). of this sustained attention remain unclear. Although evidence suggests that the LPP Electrocortical measures of emotion and reflects enhanced attention to emotional its regulation have been fruitful, yet there stimuli, some have suggested that the LPP are many open avenues for further explora- might instead reflect the suppressed pro- tion, particularly with regard to the study cessing of competing stimuli (Brown, van of individual differences in emotion regula- Steenbergen, Band, de Rover, & Nieuwen- tion. What emotion strategies work best for huis, 2012; MacNamara, Ferri, et al., 2011; whom? In what ways does emotion regula- Mocaiber et al., 2010; Weinberg & Hajcak, tion go awry across psychological disorders? 2011b). Future work using fMRI might help There is an emerging clinical literature uti- to clarify whether the LPP indexes facili- lizing the LPP to shed light on the temporal tated attention to emotional stimuli. For dynamics of information processing abnor- instance, given that the LPP is modulated malities associated with psychopathology. by both attentional deployment and cogni- As discussed elsewhere in this volume (Parts tive reappraisal, it would be informative VII and VIII, Chapters 24–32), research on to know what neural circuits are similarly emotion regulation has important implica- increased or decreased during these emotion tions for the treatment of a range of psy- regulation strategies. Moreover, it is impera- chological disorders, and bringing the LPP tive to combine EEG and fMRI technologies to bear on this topic may be useful for both in future research. What neural regions and quantifying impairment in emotion regula- networks support the emotional modula- tion and clarifying mechanisms of change tion of the LPP and its reduction through over the course of treatment. As a neuro- reappraisal and attentional deployment? Do biological complement to self-r­eport and fMRI and EEG provide complementary or behavioral indices of emotion regulation redundant information within and between difficulties, LPP data may provide unique subjects? In our view, these are crucial and information regarding the time course of abnormal emotional information processing within a specific patient population, which may then become a target for subsequent intervention.

54 BIOLOGICAL BASES As evidenced by the chapters in this vol- humans. Psychopharmacology, 219(4), 971– ume, emotion regulation is a topic that 979. figures heavily in both developmental and Decicco, J. M., Solomon, B., & Dennis, T. A. clinical literatures. Although electrocortical (2012). Neural Correlates of Cognitive Reap- activity has been relatively underutilized in praisal in Children: An ERP study. Develop- these areas, EEG-based measures of neural mental Cognitive Neuroscience, 2(1), 79–80. function might be particularly suitable for Dennis, T., & Hajcak, G. (2009). The late posi- emotion regulation studies in children and tive potential: A neurophysiological marker clinical populations. The few and emerging for emotion regulation in children. Journal studies in these areas suggest a large and of Child Psychology and Psychiatry, 50(11), positive potential. 1373–1383. Dunning, J. P., & Hajcak, G. (2009). See no evil: References Directing visual attention within unpleasant images modulates the electrocortical response. Blechert, J., Sheppes, G., Di Tella, C., Williams, Psychophysiology, 46(1), 28–33. H., & Gross, J. J. (2012). See what you think: Dunning, J. P., Parvaz, M. A., Hajcak, G., Reappraisal modulates behavioral and neural Maloney, T., Alia-Klein, N., Woicik, P. A., responses to social stimuli. Psychological Sci- et al. (2011). Motivated attention to cocaine ence, 23(4), 346–353. and emotional cues in abstinent and current cocaine users—An ERP study. European Jour- Bradley, M., Sabatinelli, D., Lang, P., Fitzsim- nal of Neuroscience, 33, 1716–1723. mons, J., King, W., & Desai, P. (2003). Activa- Eimer, M., Holmes, A., & McGlone, F. (2003). tion of the visual cortex in motivated attention. The role of spatial attention in the processing Behavioral Neuroscience, 117(2), 369–380. of facial expression: An ERP study of rapid brain responses to six basic emotions. Cogni- Brown, S. B. R. E., van Steenbergen, H., Band, tive, Affective, and Behavioral Neuroscience, G. P. H., de Rover, M., & Nieuwenhuis, 3(2), 97–110. S. (2012). Functional significance of the Ferrari, V., Bradley, M., Codispoti, M., & Lang, emotion-­related late positive potential. Fron- P. (2010). Detecting novelty and significance. tiers in Human Neuroscience, 6. [E-publica- Journal of Cognitive Neuroscience, 22(2), tion ahead of print] 404 – 411. Field, M., & Cox, W. M. (2008). Attentional bias Codispoti, M., & De Cesarei, A. (2007). Arousal in addictive behaviors: A review of its devel- and attention: Picture size and emotional reac- opment, causes, and consequences. Drug and tions. Psychophysiology, 44(5), 680–686. Alcohol Dependence, 97(1–2), 1–20. Flaisch, T., Häcker, F., Renner, B., & Schupp, H. Codispoti, M., Ferrari, V., & Bradley, M. (2006). (2011). Emotion and the processing of sym- Repetitive picture processing: Autonomic and bolic gestures: An event-­related brain poten- cortical correlates. Brain Research, 1068(1), tial study. Social Cognitive and Affective Neu- 213–220. roscience, 6(1), 109–118. Foti, D., & Hajcak, G. (2008). Deconstructing Codispoti, M., Ferrari, V., & Bradley, M. (2007). reappraisal: Descriptions preceding arous- Repetition and event-­related potentials: Dis- ing pictures modulate the subsequent neural tinguishing early and late processes in affec- response. Journal of Cognitive Neuroscience, tive picture perception. Journal of Cognitive 20(6), 977–988. Neuroscience, 19(4), 577–586. Foti, D., Hajcak, G., & Dien, J. (2009). Differen- tiating neural responses to emotional pictures: Codispoti, M., Mazzetti, M., & Bradley, M. Evidence from temporal–­spatial PCA. Psycho- (2009). Unmasking emotion: Exposure dura- physiology, 46, 521–530. tion and emotional engagement. Psychophysi- Foti, D., Olvet, D., Klein, D., & Hajcak, G. ology, 46(4), 731–738. (2010). Reduced electrocortical response to threatening faces in major depressive disorder. Cuthbert, B., Schupp, H., Bradley, M., Birbau- Depression and Anxiety, 27(9), 813–820. mer, N., & Lang, P. (2000). Brain potentials in Franken, I. H. A., Dietvorst, R. C., Hesselmans, affective picture processing: Covariation with M., Franzek, E. J., Van De Wetering, B. J. M., autonomic arousal and affective report. Bio- logical Psychology, 52(2), 95–111. de Rover, M., Brown, S., Boot, N., Hajcak, G., van Noorden, M., van der Wee, N., & Nieu- wenhuis, S. (2012). Beta receptor–­mediated modulation of the late positive potential in

Temporal Dynamics of Emotion Regulation 55 & Van Strien, J. W. (2008). Cocaine craving J. (2000). Event-­related potentials and cue-­ is associated with electrophysiological brain reactivity in alcoholism. Alcoholism: Clinical responses to cocaine-r­elated stimuli. Addic- and Experimental Research, 24(11), 1724– tion Biology, 13, 386–392. 1729. Grasso, D. J., & Simons, R. F. (2011). Perceived Hillyard, S. A., Hinrichs, H., Tempelmann, C., parental support predicts enhanced late posi- Morgan, S. T., Hansen, J. C., Scheich, H., tive event-­related brain potentials to parent et al. (1997). Combining steady-­state visual faces. Biological Psychology, 86(1), 26–30. evoked potentials and fMRI to localize brain Gross, J. J. (this volume, 2014). Emotion regula- activity during selective attention. Human tion: Conceptual and empirical foundations. Brain Mapping, 5(4), 287–292. In J. J. Gross (Ed.), Handbook of emotion Holmes, A., Vuilleumier, P., & Eimer, M. (2003). regulation (2nd ed., pp. 3–20). New York: The processing of emotional facial expression Guilford Press. is gated by spatial attention: Evidence from Hajcak, G., Anderson, B., Arana, A., Borckardt, event-­related brain potentials. Cognitive Brain J., Takacs, I., George, M., et al. (2010). Dorso- Research, 16(2), 174–184. lateral prefrontal cortex stimulation modulates Hot, P., Saito, Y., Mandai, O., Kobayashi, T., electrocortical measures of visual attention: & Sequeira, H. (2006). An ERP investiga- Evidence from direct bilateral epidural cortical tion of emotional processing in European and stimulation in treatment-r­ esistant mood disor- Japanese individuals. Brain Research, 1122(1), der. Neuroscience, 170(1), 281–288. 171–178. Hajcak, G., Dunning, J. P., & Foti, D. (2009). Johnstone, T., van Reekum, C. M., Urry, H. L., Motivated and controlled attention to emo- Kalin, N. H., & Davidson, R. J. (2007). Failure tion: Time-­course of the late positive poten- to regulate: Counterproductive recruitment of tial. Clinical Neurophysiology, 120, 505–510. top-down prefrontal–s­ubcortical circuitry in Hajcak, G., MacNamara, A., Foti, D., Ferri, J., major depression. Journal of Neuroscience, & Keil, A. (2013). The dynamic allocation of 27(33), 8877–8884. attention to emotion: Simultaneous and inde- Keil, A., Bradley, M., Hauk, O., Rockstroh, B., pendent evidence from the late positive poten- Elbert, T., & Lang, P. (2002). Large-scale neu- tial and steady state visual evoked potentials. ral correlates of affective picture processing. Biological Psychology, 92(3), 447–455. Psychophysiology, 39(5), 641–649. Hajcak, G., MacNamara, A., & Olvet, D. M. Keil, A., Gruber, T., Muller, M., Moratti, S., Sto- (2010). Event-­related potentials, emotion, and larova, M., Bradley, M., & Lang, P. (2003). emotion regulation: An integrative review. Early modulation of visual perception by emo- Developmental Neuropsychology, 35(2), tional arousal: Evidence from steady-­state 129–155. visual evoked brain potentials. Cognitive, Hajcak, G., Moser, J. S., & Simons, R. F. (2006). Affective, and Behavioral Neuroscience, 3(3), Attending to affect: Appraisal strategies mod- 195–206. ulate the electrocortical response to arousing Keil, A., Moratti, S., Sabatinelli, D., Bradley, M., pictures. Emotion, 6(3), 517–522. & Lang, P. (2005). Additive effects of emo- Hajcak, G., & Nieuwenhuis, S. (2006). Reap- tional content and spatial selective attention praisal modulates the electrocortical response on electrocortical facilitation. Cerebral Cor- to unpleasant pictures. Cognitive, Affective, tex, 15(8), 1187–1197. and Behavioral Neuroscience, 6, 291–297. Kisley, M. A., Wood, S., & Burrows, C. L. Hajcak, G., & Olvet, D. M. (2008). The persis- (2007). Looking at the sunny side of life. Psy- tence of attention to emotion: Brain potentials chological Science, 18(9), 838–843. during and after picture presentation. Emo- Kissler, J., Herbert, C., Winkler, I., & Junghofer, tion, 8(2), 250–255. M. (2009). Emotion and attention in visual Hajcak, G., Weinberg, A., MacNamara, A., & word processing—­An ERP study. Biological Foti, D. (2011). ERPs and the study of emo- Psychology, 80(1), 75–83. tion. In S. Luck & E. Kappenman (Eds.), The Kober, H., Kross, E. F., Mischel, W., Hart, C. L., & Ochsner, K. N. (2010). Regulation of crav- Oxford handbook of event-r­elated poten- ing by cognitive strategies in cigarette smok- tial components (pp. 441–474). New York: ers. Drug and Alcohol Dependence, 106(1), Oxford University Press. 52–55. Herrmann, M. J., Weijers, H., Wiesbeck, G. Krompinger, J., Moser, J., & Simons, R. (2008). A., Aranda, D., Böning, J., & Fallgatter, A.

56 BIOLOGICAL BASES Modulations of the electrophysiological Mocaiber, I., Pereira, M. G., Erthal, F. S., response to pleasant stimuli by cognitive reap- Machado-P­inheiro, W., David, I. A., Cagy, praisal. Emotion, 8(1), 132–137. M., et al. (2010). Fact or fiction?: An event-­ Kujawa, A., Hajcak, G., Torpey, D., Kim, J., & related potential study of implicit emotion Klein, D. (2012). Electrocortical reactivity to regulation. Neuroscience Letters, 476(2), emotional faces in young children and associa- 84–88. tions with maternal and paternal depression. Moratti, S., Saugar, C., & Strange, B. A. (2011). Journal of Child Psychology and Psychiatry, Prefrontal–­occipitoparietal coupling under- 53, 207–215. lies late latency human neuronal responses Lang, P. J., Bradley, M. M., & Cuthbert, B. N. to emotion. Journal of Neuroscience, 31(47), (2008). International Affective Picture Sys- 17278–17286. tem (IAPS): Affective ratings of pictures and Moser, J., Hajcak, G., Bukay, E., & Simons, R. instructional manual (Technical Report A-8). (2006). Intentional modulation of emotional Gainesville: University of Florida. responding to unpleasant pictures: An ERP Langeslag, S. J. E., & Van Strien, J. W. (2010). study. Psychophysiology, 43(3), 292–296. Comparable modulation of the late positive potential by emotion regulation in younger Namkoong, K., Lee, E., Lee, C. H., Lee, B. O., and older adults. Journal of Psychophysiol- & An, S. K. (2004). Increased P3 amplitudes ogy, 24(3), 186–197. induced by alcohol-r­ elated pictures in patients Leutgeb, V., Schafer, A., & Schienle, A. (2009). with alcohol dependence. Alcoholism: Clini- An event-r­elated potential study on exposure cal and Experimental Research, 28(9), 1317– therapy for patients suffering from spider pho- 1323. bia. Biological Psychology, 82(3), 293–300. Littel, M., & Franken, I. H. A. (2007). The Ochsner, K. N., & Gross, J. J. (this volume, effects of prolonged abstinence on the process- 2014). The neural bases of emotion and emo- ing of smoking cues: An ERP study among tion regulation: A valuation perspective. In J. smokers, ex-­smokers and never-s­ mokers. Jour- J. Gross (Ed.), Handbook of emotion regula- nal of Psychopharmacology, 21(8), 873–882. tion (2nd ed., pp.23–42). New York: Guilford Littel, M., & Franken, I. H. A. (2011a). Implicit Press. and explicit selective attention to smoking cues in smokers indexed by brain potentials. Ochsner, K. N., Ray, R. D., Cooper, J. C., Rob- Journal of Psychopharmacology, 25(4), 503– ertson, E. R., Chopra, S., Gabrieli, J. D. E., et 513. al. (2004). For better or for worse: Neural sys- Littel, M., & Franken, I. H. A. (2011b). Inten- tems supporting the cognitive down-and up-­ tional modulation of the late positive potential regulation of negative emotion. NeuroImage, in response to smoking cues by cognitive strat- 23(2), 483–499. egies in smokers. PloS ONE, 6(11), e27519. Luck, S. (2005). An introduction to the event-­ Olofsson, J., Nordin, S., Sequeira, H., & Polich, related potential technique. Cambridge, MA: J. (2008). Affective picture processing: An MIT Press. integrative review of ERP findings. Biological MacNamara, A., Ferri, J., & Hajcak, G. (2011). Psychology, 77(3), 247–265. Working memory load reduces the late posi- tive potential and this effect is attenuated with Olofsson, J., & Polich, J. (2007). Affective visual increasing anxiety. Cognitive, Affective, and event-­related potentials: Arousal, repetition, Behavioral Neuroscience, 11(3), 321–331. and time-on-task. Biological Psychology, MacNamara, A., Foti, D., & Hajcak, G. (2009). 75(1), 101–108. Tell me about it: Neural activity elicited by emotional stimuli and preceding descriptions. Parvaz, M., MacNamara, A., Goldstein, R., & Emotion, 9(4), 531–543. Hajcak, G. (2012). Event-­related frontal alpha MacNamara, A., Ochsner, K. N., & Hajcak, G. as a marker of lateral prefrontal cortex acti- (2011). Previously reappraised: The lasting vation during cognitive reappraisal. Cogni- effect of description type on picture-­elicited electrocortical activity. Social Cognitive and tive, Affective, and Behavioral Neuroscience, Affective Neuroscience, 6(3), 348–358. 12(4), 730–740. Phan, K., Liberzon, I., Welsh, R., Britton, J., & Taylor, S. (2003). Habituation of rostral ante- rior cingulate cortex to repeated emotionally salient pictures. Neuropsychopharmacology, 28(7), 1344–1350. Sabatinelli, D., Keil, A., Frank, D., & Lang, P. (2013). Emotional perception: Correspon-

Temporal Dynamics of Emotion Regulation 57 dence of early and late event-r­ elated potentials inson, J. D., Minnix, J. A., Brown, V. L., et al. with cortical and subcortical functional MRI. (2012). Beyond cue reactivity: Blunted brain Biological Psychology, 92(3), 513–519. responses to pleasant stimuli predict long-term Sabatinelli, D., Lang, P., Keil, A., & Bradley, M. smoking abstinence. Addiction Biology, 17(6), (2007). Emotional perception: Correlation of 991–1000. functional MRI and event-r­elated potentials. Vuilleumier, P. (2005). How brains beware: Neu- Cerebral Cortex, 17(5), 1085–1091. ral mechanisms of emotional attention. Trends Sheppes, G., & Gross, J. J. (2011). Is timing in Cognitive Sciences, 9(12), 585–594. everything?: Temporal considerations in emo- Wager, T. D., Davidson, M. L., Hughes, B. L., tion regulation. Personality and Social Psy- Lindquist, M. A., & Ochsner, K. N. (2008). chology Review, 15(4), 319–331. Prefrontal-­subcortical pathways mediating Stockburger, J., Schmälzle, R., Flaisch, T., Bub- successful emotion regulation. Neuron, 59(6), latzky, F., & Schupp, H. T. (2009). The impact 1037–1050. of hunger on food cue processing: An event-­ Weinberg, A., & Hajcak, G. (2010). Beyond related brain potential study. NeuroImage, good and evil: The time-c­ ourse of neural activ- 47(4), 1819–1829. ity elicited by specific picture content. Emo- Tacikowski, P., & Nowicka, A. (2010). Alloca- tion, 10, 767–782. tion of attention to self-name and self-face: Weinberg, A., & Hajcak, G. (2011a). Electrocor- An ERP study. Biological Psychology, 84(2), tical evidence for vigilance-a­ voidance in Gen- 318–324. eralized Anxiety Disorder. Psychophysiology, Thiruchselvam, R., Blechert, J., Sheppes, G., 48, 842–851. Rydstrom, A., & Gross, J. J. (2011). The tem- Weinberg, A., & Hajcak, G. (2011b). The Late poral dynamics of emotion regulation: An Positive Potential predicts subsequent interfer- EEG study of distraction and reappraisal. Bio- ence with target processing. Journal of Cogni- logical Psychology, 87(1), 84–92. tive Neuroscience, 23, 2995–3007. Thiruchselvam, R., Hajcak, G., & Gross, J. J. Weinberg, A., Hilgard, J., Bartholow, B., & Haj- (2012). Looking inwards: Shifting attention cak, G. (2012). Emotional targets: Evaluative within working memory representations alters categorization as a function of context and emotional responses. Psychological Science, content. International Journal of Psychophys- 23(12), 1461–1466. iology, 84, 149–154. Urry, H. L. (2010). Seeing, thinking, and feeling: Yen, N. S., Chen, K. H., & Liu, E. H. (2010). Emotion-­regulating effects of gaze-d­irected Emotional modulation of the late positive cognitive reappraisal. Emotion, 10(1), 125– potential (LPP) generalizes to Chinese individ- 135. uals. International Journal of Psychophysiol- Versace, F., Lam, C. Y., Engelmann, J. M., Rob- ogy, 75(3), 319–325.

Chapter 4 The Neural Basis of Emotion Dysregulation Tom Johnstone Henrik Walter Humans have long had an ambivalent rela- adaptive responses in the face of complexity tionship with their emotions. Emotions have that rules out a rational calculation of all been celebrated as the inspiration for pin- our options but afford us far more flexibility nacles of human art, literature, and music. than simple stimulus–r­ esponse rules. Equally, however, emotions have often been seen as “evil spirits” that misguide us and Emotion Regulation tempt us to do things we should not—­as and Dysregulation “primal” instincts that need to be conquered Emotions are not always “correct,” based as by the rational human mind. Indeed, history they are on probabilistic systems that have is replete with fables of the terrible conse- evolved to ensure our survival across a wide quences of unchecked emotion, from the range of circumstances. One of the func- tragedies of Shakespeare and his recurring tions emotions serve is to interrupt ongo- theme of “the fatalism of overmastering pas- ing behavior when an encountered event sion” (Corson, 1890, Preface) to the appall- or stimulus is highly personally significant ing crowd violence that can occur anywhere, and deserves our attention or requires us from the streets of war-t­orn Baghdad to the to prepare for action (Frijda, 1986; Gross, neighborhoods of London. this volume). Yet we cannot afford to be constantly interrupted; emotions exist in The uncertainty with which we view a balance with other, ongoing cognitive, our emotions reflects their function, both attentional, and behavioral processes. In through our evolutionary past and in day-­ order for us to benefit, emotions need to be to-­day life. Emotions allow us to respond to appropriately managed depending on the important environmental changes quickly specific contexts in which they occur. Given and with a minimum of deliberation, and the importance of emotion regulation, it is provide us with the drive to seek out things unsurprising that neural systems have been that are good for us and avoid things that identified that serve to manage emotions are harmful. Emotion and mood1 can be and moods in such a way that, on the bal- seen as arising from heuristic neural systems ance, they contribute to, rather than inter- that act upon a vast quantity of incoming fere with, our daily lives (Ochsner & Gross, information and make speedy decisions, this volume). guiding behavior in a way that is robust against incomplete information or computa- tional overload. Emotions are vital for quick, 58

The Neural Basis of Emotion Dysregulation 59 In this chapter we consider how the dys- this volume). One important question is the function of these neural systems might extent to which different manifestations of underlie the inability to regulate emotions emotion dysregulation can be understood in appropriately. The outcome of such a fail- terms of specific patterns of dysfunction in ure to regulate can range from a short-­ one or more of these neural systems. term inability to stop getting angry during a debate with a colleague, all the way to Prefrontal Cortex long-­term psychopathology such as mood and Emotion Regulation and anxiety disorders. We limit discussion Given the broad sweep of processes we to types of emotion dysregulation that are just outlined, an exhaustive review of the in some sense not normative. Thus, we do role of different neural systems in emotion not discuss mechanisms that give rise to regulation and dysregulation would require panic behavior when a bomb explodes, even an entire book. Instead, we focus here on though remaining calm might be the most research demonstrating the involvement of adaptive thing to do in the context, since the prefrontal cortex (PFC) in the regulation panic in such a situation would be considered of “bottom-u­p” emotion-g­enerating neural quite normal. We also take a look at how circuits, which has provided the clearest knowledge of the neural systems involved in empirical distinction between generation emotion regulation, and their dysfunction in and regulation of emotions, and also poten- psychopathology, is leading to novel, brain-­ tially has the greatest implications for emo- focused clinical treatments. tion regulation–b­ ased therapies for a variety of affective disorders (see Figure 4.1; also see Emotion regulation processes can be Ochsner & Gross, this volume). Research on examined at multiple levels, from neu- prefrontal cortical regulation of emotion rotransmitters and inhibitory interneurons has largely followed one of two approaches. to cortical and subcortical network feed- Studies of cognitive reappraisal have focused back loops, to socially mediated regulation on how the explicit cognitive reinterpreta- of moods and emotional behaviors. All of tion of the emotional meaning or possible these types of regulation may be at least par- outcome of a stimulus can be used to mod- tially automatic and subconscious (Gyurak ify the resulting emotional response. On a & Etkin, this volume), so limiting a discus- more automatic level, a range of studies has sion of emotion regulation to that which is examined the more automatic engagement “deliberate” or “conscious” is not particu- of prefrontal mechanisms that serve to regu- larly useful. Gross (this volume) makes the late attention and the contents of working distinction between antecedent-f­ocused and memory, and to shield these processes from response-­focused emotion regulation. Neu- emotional interference. robiologically, this conceptual distinction can be elaborated by considering the neural Cognitive Reappraisal circuits involved in detecting and appraising Based on cognitive appraisal theories of emo- situations and contexts that call for certain tion (Scherer, Schorr, & Johnstone, 2001), emotional responses (e.g., threat detection, the study of emotion regulation through relevance detection, recognition of socially reappraisal has formed the core of cognitive rewarding stimuli, evaluation of possible neuroscience research on emotion regulation outcomes and their probabilities), as well (Gross, this volume). Reappraising the affec- as those responsible for generating the neu- tive meaning of negative emotional stimuli ral, physiological, and behavioral responses can reduce the magnitude of felt negative that make up emotions. Many of the brain emotion, as well as physiological indicators, processes involved in emotion regulation such as potentiated eyeblink startle (Jackson, have thus been extensively studied in other Malmstadt, Larson, & Davidson, 2000). domains of psychology and cognitive neuro- Eyeblink startle increases in magnitude dur- science, including (but not limited to) selec- ing presentation of negative pictures, rela- tive attention, cognitive control, working tive to neutral pictures, an effect thought to memory, and response inhibition. Effective emotion regulation involves the coordinated recruitment of these different neural mecha- nisms in specific contexts (Ochsner & Gross,

60 BIOLOGICAL BASES FIGURE 4.1.  Depiction of prefrontal cortical and subcortical regions commonly linked to emotion regulation. dmPFC, dorsomedial PFC; dACC, dorsal anterior cingulate cortex; vmPFC, ventromedial PFC; NAc: nucleus accumbens; Amyg, amygdala; vlPFC, ventrolateral PFC; dlPFC, dorsolateral PFC. be mediated by projections from the central Spontaneous Regulation nucleus of the amygdala to the brainstem and Cognitive Control (Davis, 2006). Evidence that reappraisal can In daily life we need to regulate our emo- directly influence this amygdala circuitry tions continuously and automatically in comes from consistent findings in positron order to remain focused on current thoughts emission tomographic (PET) and functional and actions; we do not always have the magnetic resonance imaging (fMRI) studies luxury of being able to reappraise con- of healthy individuals showing reappraisal-­ sciously the emotion-­inducing events around dependent decreases in amygdala activation us. A number of studies have used tasks in response to negative stimuli. Increased in which regulating emotion is required dorsolateral and ventrolateral PFC activa- but not explicitly instructed. For example, tion is commonly measured in such reap- when cognitive tasks, such as those that praisal studies (Ochsner & Gross, this tax working memory, are performed under volume), though discrepancies such as the an anxiety induction or in the presence of hemispheric lateralization of PFC activation emotional distractors, a drop in task perfor- still exist, with some studies finding predom- mance is observed. There is some evidence inantly right prefrontal activation and others that this interference is reduced when the finding left prefrontal activation.2

The Neural Basis of Emotion Dysregulation 61 cognitive load of the task increases, possi- plays an important role in the creation and bly through the automatic engagement of maintenance and/or recall of extinction of top-down control mechanisms (Blair et al., conditioned fear responses (Milad & Quirk, 2007; Clarke & Johnstone, 2013; Erthal et 2002, 2012). Consistent with these stud- al., 2005; Van Dillen, Heslenfeld, & Koole, ies, activation in the vmPFC correlates with 2009). The specific subregions of the PFC recall of extinction in humans (Kalisch et that are involved in such automatic emo- al., 2006; Milad et al., 2007; Phelps, Del- tion regulation may depend on properties gado, Nearing, & LeDoux, 2004). The role of the distractors, for example, the relative of vmPFC–­amygdala connectivity seems to involvement of ventrolateral or dorsal PFC be under genetic control, pointing to a role in regulating negative or positive distrac- of dynorphins in human extinction learning tors, respectively (Erk, Kleczar, & Walter, (Bilkei-­Gorzo et al., 2012). Thus, based on 2007). A number of the prefrontal regions its role in fear conditioning and extinction implicated in either spontaneous or explicit paradigms, the vmPFC would appear to be emotion regulation, including the ventrolat- a promising candidate as a region that can eral PFC, dorsolateral PFC, and the dorsal mediate more lateral PFC regulatory effects anterior cingulate, overlap with those com- on the amygdala. monly identified in studies of cognitive and attentional control (Mitchell, 2011). The Evidence that this might be the case comes dorsal anterior cingulate cortex (dACC) is from studies of spontaneous regulation, as posited to be involved in performance moni- well as reappraisal-b­ased down-­regulation toring and detecting when control is neces- of emotional responses to aversive pictures. sary (Ridderinkhof, Ullsperger, Crone, & For example, Etkin, Egner, Peraza, Kandel, Nieuwenhuis, 2004). Lateral regions of the and Hirsch (2006) used a variant of an emo- PFC, including the ventrolateral and dor- tional Stroop task, in which participants solateral PFC, are involved in maintaining responded to the words fear and happy and manipulating information in working superimposed on faces expressing either fear memory, and in implementing attentional, or happiness. In this task, there is a need cognitive, or behavioral adjustments (Rid- to down-r­egulate automatic responses to derinkhof et al., 2004). the emotional facial expressions when they conflict with the word (e.g., the word happy Prefrontal–Subcortical Connectivity superimposed on a fearful face). Their There are few direct neural connections results indicated that while dorsal and lat- between dorsal and lateral PFC and the eral PFC detect the need for top-down regu- amygdala (Ghashghaei, Hilgetag, & Barbas, lation, rostral ACC (rACC) implements that 2007), raising the question of how these regulation. Specifically, rACC activation lateral PFC brain regions might exert their was higher and amygdala activation was top-down regulatory influence. One possi- lower during the down-­regulation of emo- bility is that the ventromedial PFC (vmPFC), tionally conflicting information, with a neg- which has neural connections with lateral ative correlation between the two regions. PFC and the amygdala, might provide the Two studies of reappraisal-­based emotion missing link. Ventral parts of the medial PFC regulation, (Johnstone, van Reekum, Urry, have been implicated in down-­regulation Kalin, & Davidson, 2007; Urry et al., 2006) of amygdala function in nonhuman and found that individual differences in activa- human studies of fear conditioning and tion of the vmPFC during down-­regulation extinction. Extinction of conditioned fear correlated negatively with amygdala acti- responses can be considered a form of emo- vation; the more activated the vmPFC dur- tion regulation, since the fear association ing down-­regulation, the less activated the itself is not destroyed (Bouton, 2004), but amygdala. Furthermore, vmPFC activation rather a new association indicating safety is was found statistically to mediate the associ- created and inhibits or overcomes the origi- ation between dorsomedial PFC (Urry et al., nal conditioned response in specific con- 2006) ventrolateral PFC (Johnstone et al., texts. In rodent studies, the infralimbic cor- 2007) or dorsolateral PFC (Erk et al., 2010) tex (homologous to the vmPFC in humans) and amygdala activation during the down-­ regulation condition. It should be noted, however, that these studies did not find cat-

62 BIOLOGICAL BASES egorically increased activation of vmPFC robiological causes of these conditions and during down-­regulation. potential avenues for treatment are there- fore of enormous importance to the general One study (Delgado, Nearing, LeDoux, population. Most brain research into mood & Phelps, 2008) directly tested the con- and anxiety disorders has focused on the nection between reappraisal-­based emo- “bottom-up” determinants of emotions, tion regulation and extinction. Participants with a focus on subcortical and early sen- were instructed simply to attend to a colored sory processing that generates emotional square that might give them a shock (the responses and biases our behavior in a fairly conditioned stimulus) or to down-­regulate automatic way. For example, much of the their conditioned response to the square by focus in anxiety disorders has been on atten- reappraising the meaning of the color (e.g., tional biases toward anxiety-­relevant stimuli by thinking of contexts in which the color (Campbell-S­ ills, Ellard, & Barlow, this vol- had a more calming influence, such as the ume; MacLeod, Mathews, & Tata, 1986), sky for a blue square). Greater activation in generated either through low-­level sensory the down-r­egulation relative to the attend cortex or subcortical structures such as the condition was measured in dorsolateral amygdala, known to play a role in early PFC and vmPFC, with less activation in the threat detection. In depression, an extensive amygdala. Prefrontal activation correlated literature exists on the role of the monoami- negatively with skin conductance responses, nergic neurotransmitters serotonin, norepi- providing evidence of its role in regulating nephrine, and dopamine in subcortical and emotional responses. Notably, the region of subcortical–­cortical networks involved in activation in vmPFC overlapped with that mood, though for all that research, there found in prior studies of extinction using the have been no definitive breakthroughs in same stimuli. understanding the causes of depression. In summary, evidence points to a pre- Although negative emotions such as anxi- frontal regulatory network in which dorsal ety and sadness are a normal occurrence in and lateral regions of the PFC exert a top- response to adversity in all individuals, one down influence on subcortical structures feature that distinguishes those with mood involved in generating emotional responses. and anxiety disorders is their inability to Some evidence suggests that this regula- regulate certain negative emotions effec- tory effect is via the same region of vmPFC tively when they arise. Indeed, it might well involved in extinction, though whether this be argued that the defining feature of many is the primary pathway, or other important mood and anxiety disorders is the extended regulatory pathways exist (Wager, David- duration of emotional episodes rather than son, Hughes, Lindquist, & Ochsner, 2008) their intensity (phobias and panic disor- is the topic of continuing investigation (e.g. ders are obvious exceptions, though even see Clarke & Johnstone [2013] for an exam- in these cases, feelings of panic and anxi- ination of the role of dACC in inhibition of ety typically last well beyond the immediate amygdala output to PFC during spontane- trigger of the episode). Cognitive neurosci- ous threat regulation). Identifying the exact entists have argued that depression should pathways involved in effective emotion regu- not be equated with the “down” state itself, lation will be vital in determining the extent but rather with the tendency to “enter and to which the dysregulation of emotion can get stuck in this state. Thus, the neurobiol- be explained in terms of a malfunction in ogy of depression should be that of mood the prefrontal emotion regulation network. reaction and regulation rather than the mood state per se” (Holtzheimer & May- Emotion Dysregulation berg, 2011, p. 1). Importantly, it is not just and Affective Disorders negative emotions that are dysregulated in Emotional disorders such as depression and mood disorders. “Anhedonia,” a markedly anxiety are estimated by the World Health reduced interest and pleasure in normally Organization to be the greatest cause of dis- rewarding activities, is one of the three core ability worldwide (Mathers, Lopez, & Mur- features of major depression that may be ray, 2006). Efforts to understand the neu- partly explained by an inability to engage cognitively in rewarding situations and sus-

The Neural Basis of Emotion Dysregulation 63 tain positive emotions, a different type of as distinct from transient, forms of vigilance emotion regulation. (Somerville et al., 2012; Walker, Toufexis, & Davis, 2003). Although abnormalities in the neural circuitry supporting adaptive regulation of Nonetheless, it seems likely that anxiety certain emotions may play a decisive role in disorders involve more than just hypersen- determining vulnerability to mood disorders sitivity in bottom-up threat detection sys- (Davidson, Pizzagalli, Nitschke, & Putnam, tems. There is now a large body of evidence 2002), only recently have researchers explic- that highly anxious individuals, in par- itly examined this possibility. Below we ticular those with anxiety disorders, show discuss the findings separately for anxiety, deficits in prefrontal circuits that regulate which is symptomatic of a variety of disor- attention to and responding to threaten- ders and phobias; the more general negative ing stimuli. Over and above an attentional mood characteristic of depression; and the bias toward threatening information, highly dysregulation of positive emotions in anhe- anxious individuals have greater difficulty donia. in disengaging from threat-Âr

64 BIOLOGICAL BASES PFC activation in high-­anxious than in low-­ with PTSD, neuroimaging studies of other anxious individuals, and the opposite pat- anxiety disorders have shown inconsistent tern in the amygdala. This result is consis- results, with both hyper- and hypoactivation tent with a model in which rACC detects the in vmPFC, though there are fewer such stud- need for control over incoming information ies than those examining PTSD, and they that conflicts with task demands, and ven- have used a variety of different experimental trolateral PFC is subsequently involved in paradigms (Etkin & Wager, 2007). regulating attention toward task-­relevant information and away from threatening, Dysregulation of Emotion task-­irrelevant information. The lower rACC in Negative Mood and Depression and ventrolateral PFC activation and higher Brain imaging studies of depressed individu- amygdala activation seen in high-­anxious als have demonstrated abnormal patterns of individuals presumably reflects reduced activation in many of the same prefrontal prefrontal control. This deficit in top-down brain regions thought to contribute to the attentional control of task-­irrelevant distrac- regulation of mood and emotion, though the tors might not be specific to threatening or findings have not been consistent. A num- even emotional distractors: In a later study, ber of PET studies have indicated reduced Bishop (2008) reported that high trait anx- metabolism in dorsolateral PFC and dACC ious individuals showed greater interference in both the resting state and during expo- from nonemotional distractors, along with sure to negative stimuli (Fitzgerald, Laird, reduced activation in dorsolateral PFC. Thus Maller, & Daskalakis, 2008). These results elevated trait anxiety is associated with gen- are consistent with the notion that dysregu- erally reduced lateral PFC-m­ ediated top- lation of negative mood in depression might down control of attention, resulting in less be due to underrecruitment of dorsal pre- efficient attentional regulation of distract- frontal brain regions important for the regu- ing emotional information. Coupled with a lation of negative emotions, yet the fact that hypersensitive bottom-up threat detection these studies did not directly measure emo- system, the consequence would be hypervig- tion regulation limits the inferences that can ilance to and an inability to disengage from be made. potentially threatening stimuli. Some PET resting state studies have also The picture of dysfunctional prefrontal reported elevated metabolism in depressed regulation in anxiety disorders (as opposed versus nondepressed individuals in ventral to high trait anxiety) is not as clear-­cut as regions of the PFC, including ventrolateral one might presume from this model, how- and orbitofrontal cortex (Savitz & Drevets, ever. The patterns of observed prefron- 2009), which is more difficult to recon- tal hypoactivity and hyperactivity differ cile with a deficit in emotion regulation. It across anxiety disorders (Etkin & Wager, should be noted though that this region of 2007; Shin & Liberzon, 2010). Of these, the PFC serves a more general role in emo- the most consistent neuroimaging findings tion processing than its posited role in top- have been reported for posttraumatic stress down emotion regulation, including flexible disorder (PTSD). A number of studies have updating of stimulus value, as well as moni- shown impaired extinction of conditioned toring of the autonomic state (Rolls, 2004). responses to aversive stimuli in PTSD, and In addition to top-down projections to the hypoactivation in vmPFC/rACC has been amygdala, it also receives excitatory input reported in response to both trauma-r­elated from both the amygdala and sensory corti- and non-t­rauma-­related aversive stimuli, as ces. In the absence of top-down regulatory well as during extinction, other emotional signals, then, hyperactivity in ventral PFC and cognitive tasks, and resting baseline in depressed individuals might reflect more (Shin & Liberzon, 2010). Activation in this bottom-up-­driven elaborative or ruminative region also negatively correlates with PTSD processing of emotional information. symptom severity. All of these results suggest that PTSD is characterized by a disrupted A number of fMRI studies have mea- ability to regulate previously conditioned sured prefrontal cortical activation dur- fear responses through the engagement of ing instructed emotion regulation tasks. vmPFC extinction circuitry. In contrast Beauregard, Paquette, and Levesque (2006)

The Neural Basis of Emotion Dysregulation 65 reported increased dorsal ACC activation in in sustained effects of emotion regulation. depressed compared with healthy individu- In healthy controls, regulation effects on als when they were decreasing emotional the amygdala could still be demonstrated responses to sad films. No group differences after the emotion regulation task; negative were found in the more anterior and lateral pictures that had been paired with a down-­ PFC regions found previously in studies of regulation instruction invoked not only reappraisal in nondepressed samples. In a reduced amygdala activation compared to study of reappraisal of negatively valenced an attend condition during the initial emo- emotional pictures, Johnstone et al. (2007) tion regulation task but also reduced amyg- found that healthy individuals showed left-­ dala activation in a simple picture-w­ atching lateralized activation of ventrolateral PFC, task 15 minutes later. Furthermore, the whereas depressed individuals showed bilat- degree of reduction in the second task was eral activation. In the healthy group, vmPFC correlated with dorsolateral PFC activa- activation was negatively correlated with tion in the first task. In contrast, depressed amygdala activation and found statistically patients had no such lasting effect of emo- to mediate a negative association between tion regulation; activation in response to ventrolateral PFC and the amygdala, con- previously down-­regulated pictures was no sistent with a top-down regulation path- different than that in response to nonregu- way from ventrolateral PFC via vmPFC to lated pictures. Together, these findings sug- the amygdala. However, for the depressed gest that depression is characterized not by group, a positive correlation was found the negative mood itself but by the tendency between vmPFC and amygdala activation, to get stuck in that mood (Holtzheimer & with no correlation between lateral PFC and Mayberg, 2011). amygdala. One possible explanation for this finding is that in depressed individuals, the Kanske, Heissler, Schönfelder, and Wessa lack of regulatory input from left dorsal or (2012) recently described how remitted lateral prefrontal regions results in sustained patients with a diagnosis of depression vmPFC activation, under either greater show a deficit in down-­regulation of the ­bottom-up or right lateral PFC influence. amygdala when reappraising socially nega- Both explanations would be consistent with tive pictures but not when using a distrac- the elevated vmPFC metabolism in depres- tion strategy. Additionally, they found that sion seen in some PET studies, perhaps down-­regulation of the amygdala was bet- reflecting greater self-­referential processing. ter in those patients who habitually used reappraisal strategies, as measured with the Results consistent with a model of Emotion Regulation Questionnaire (ERQ; reduced lateral PFC top-down regulation of Gross & John, 2003). This finding extends amygdala in depression have now been seen the work of Abler, Erk, Herwig, and Walter in a number of emotion regulation stud- (2007), whose examination of patients with ies, although lateralization of this reduced concurrent depression demonstrated that PFC involvement has not been consistently those who had higher reappraisal scores, as found across studies. In a study with par- measured with the ERQ, showed less antici- tially remitted depressed patients, Erk et patory amygdala activation to negative stim- al. (2010) demonstrated that although uli. Kanske et al.’s study (2012) shows that depressed patients are able to down-r­ egulate neurobiological deficits during reappraisal their amygdala using distancing, a form of extend into remission, pointing to the pos- reappraisal, the degree of down-r­egulation sibility that impaired emotion regulation of was diminished with increasing scores of negative stimuli might be a trait marker of depression (i.e., the more depressed the depression, as suggested previously in behav- patients were, the less they were able to ioral studies (Ehring, Fischer, Schnülle, down-r­ egulate the amygdala). This could be Bösterling, & Tuschen-C­ affier, 2008). explained by reduced activation of the right dorsolateral PFC, which also showed signif- Dysregulation of Positive Emotion: icantly reduced coupling with the amygdala Anhedonia in depressed patients. The dysregulation of emotion in depression is not limited to negative mood and emo- Erk et al. (2010) also showed differences between healthy and depressed individuals

66 BIOLOGICAL BASES tions. One of the core features of depres- positive emotions were more pronounced sion is anhedonia, a deficit in the experi- than those for negative emotions. In a recent ence of positive emotions and the ability study of up-r­ egulation of positive emotion in to take pleasure from previously rewarding response to pleasant pictures, Heller et al. stimuli. Anhedonia is a particularly debili- (2009) found that, compared with controls, tating condition, because it removes the patients with depression were unable to sus- pleasurable experiences that motivate much tain activation in the ventral striatum across of what we do in life. The neural basis of an experimental session, and that the degree anhedonia is not well understood, but it is to which ventral striatal activity was sus- likely that it involves the frontostriatal net- tained correlated with self-­reported positive work, which includes connected regions of emotion, as well as functional connectivity the striatum, particularly the ventral stria- between the ventral striatum and a region tum,3 and the PFC. Based on extensive ani- in lateral PFC. In a follow-­up study on the mal and human research, the frontostriatal same patients after 8 weeks of antidepres- network is regarded as a core system for the sant treatment, Heller et al. (2013) found learning of reward signals, the experience that those who demonstrated the greatest of pleasure, and the motivation of reward-­ increase in their ability to sustain ventral seeking behaviors. Despite its role in positive striatal activation and frontostriatal con- emotion, however, research linking the fron- nectivity while up-r­egulating positive emo- tostriatal network to anhedonia has been tion also showed the largest increases in mixed, with some studies reporting reduced positive emotion. It is worth noting that this activity in the striatum (Epstein et al., 2006; study only involved antidepressant treat- Keedwell, Andrew, Williams, Brammer, & ment, and it is unclear whether those show- Phillips, 2005) and others showing no such ing greatest symptom improvement did so effect (Knutson, Bhanji, Cooney, Atlas, & as a consequence of increased frontostriatal Gotlib, 2008; Schaefer, Putnam, Benca, & engagement. Future studies that examine the Davidson, 2006). neural impact of cognitive-­behavioral thera- pies on regulation of positive emotions will One explanation for these inconsisten- shed further light on the causes and possible cies might be the model of anhedonia being treatments for anhedonia. In an example of tested. Anhedonia has often been consid- neurally guided emotion regulation therapy, ered to reflect a tonically dampened posi- a recently published proof of concept study, tive emotion system, in which the general Linden et al. (2012) used neurofeedback capacity to experience pleasure is reduced with fMRI to target this problem in depres- (Meehl, 1975). An alternative hypothesis is sion. Eight patients with depression learned that an inability to prolong the duration of to increase activation in brain regions positive emotion, or increase the intensity thought to be related to positive emotions, of a rewarding experience—­both aspects of such as the insula and the vMPFC, during emotion regulation—m­ ight underlie at least four neurofeedback sessions. Their depres- some anhedonic symptoms (Tomarken & sive symptoms were reduced significantly, Keener, 1998). The ventral striatum has bidi- whereas a control group that underwent a rectional connections with a number of pre- training procedure with the same cognitive frontal cortical regions, and brain imaging strategies, but without neurofeedback, did studies have shown top-down modulation of not show clinical improvement. ventral striatal activation in humans (Del- gado, Gillis, & Phelps, 2008; Staudinger, Genes, Development, Erk, & Walter, 2011). and Emotion Regulation Although much has been learned in recent Compared with the regulation of emo- years about the neural underpinnings of tional responses to negative stimuli, there disorders of emotion regulation, we are a have been very few studies of top-down long way from understanding the enormous regulation of positive emotions. Kim and individual differences in our ability to regu- Hamann (2007) showed that regulation of late our moods and emotions. What are the positive emotion engaged prefrontal regions partially overlapping those involved in regu- lating negative emotions, and that subcor- tical (e.g., amygdala) effects of regulating

The Neural Basis of Emotion Dysregulation 67 causes of these individual differences? To active emotion regulation there were no dif- what extent do they reflect innate biologi- ferences in amygdala activation. Moreover, cal differences? Are there aspects of develop- it was shown that subjects with an s-aÂ

68 BIOLOGICAL BASES lescents compared to adults when required state connectivity was inversely correlated to modulate attention to emotional stimuli with anxiety symptoms but positively cor- (Monk et al., 2003; Nelson et al., 2003) related with depressive symptoms in female or when using explicit strategies to regu- adolescents, pointing to one possible factor late emotions (Levesque et al., 2004). Some in the development of PFC-s­ ubcortical emo- researchers point to the mismatch between tion regulation circuitry. The developmental development of subcortical brain regions neurobiological profile of individual differ- involved in emotion generation and prefron- ences in emotion regulation has barely been tal regions exerting top-down control as key studied, yet it is likely to have far-­reaching to understanding emotion dysregulation in consequences for our understanding of emo- children and adolescents (Casey, Totten- tion dysregulation throughout life. ham, Liston, & Durston, 2005). Because the PFC and its connections with subcortical limbic structures continue to mature from Cortical Brain Stimulation childhood through adolescence into early as a Treatment for Depression adulthood, there is a particular risk that maladaptive prefrontal–­limbic connectivity Up to now we have considered psychologi- and associated dysregulation of emotion will cal self-r­egulation of emotion. However, it develop, rendering individuals vulnerable to is also possible to regulate emotions using psychopathology later in life. technical devices that alter activity in brain Very few researchers have addressed the regions that are involved in emotion regu- neurobiology of childhood or adolescent lation. These are particularly important emotion regulation. Perlman et al. (2012) when dysregulation of emotion is present as examined emotion regulation in 14 depressed a pathological condition. Three techniques adolescents (ages 13–17 years). Compared of brain stimulation—r­epetitive transcra- to an age-m­ atched healthy control group, nial magnetic stimulation (rTMS), a closely the depressed group showed increased related technique transcranial direct current amygdala activation and less connectivity stimulation (tDCS), and deep brain stimula- between amygdala and medial PFC/insula tion (DBS)—have recently been proposed as during the control condition. Frontal–­ methods offering a viable treatment for oth- amygdala connectivity was correlated with erwise untreatable depression. psychosocial function, hinting that those depressed adolescents with less intact PFC–­ amygdala regulatory circuitry were more Repetitive Transcranial severely affected. No differences between Magnetic Stimulation the depressed and nondepressed groups were rTMS involves generating trains of brief seen during the active regulation condition, bursts of a strong magnetic field close to however, which might indicate that active the scalp. When the magnetic field is ori- emotion regulation via cognitive reappraisal ented perpendicular to the cortical surface, is able to overcome differences in baseline electric currents are induced in the underly- emotion reactivity, at least in the short term, ing neural tissue to a depth of several mil- when explicitly prompted. limeters (a consequence of Faraday’s law of In a study of adult patients with depres- electromagnetic induction). While the exact sion, Frodl, Reinhold, Koutsouleris, Rei- mechanism of action of rTMS on networks ser, and Meisenzahl (2010) have shown of neurons is not yet fully understood, dif- that previous childhood stress is associated ferent frequencies of rTMS can have either with reductions in prefrontal gray matter, short- or long-­term facilitatory or disruptive and that prefrontal gray matter volume and effects on cortical processing (Fitzgerald, childhood neglect interact to predict adult Fountain, & Daskalakis, 2006). In general, depression duration. A recent study (Burghy low-f­requency (< 1 Hz) rTMS is thought to et al., 2012) showed that elevated cortisol inhibit cortical processing, whereas high-­ levels associated with early life stress was frequency (> 5 Hz) is thought to enhance predictive of reduced vmPFC-a­ mygdala rest- cortical processing. Supporting evidence ing state functional connectivity in females. for this model of rTMS effects on cortical Furthermore, vmPFC-­amygdala resting function comes from PET studies following

The Neural Basis of Emotion Dysregulation 69 rTMS to lateral PFC, which show greater passing a small direct electrical current of blood flow and glucose metabolism fol- 1–2 milliamps (mA) between two electrodes lowing high-Âf

70 BIOLOGICAL BASES for tDCS). Although the exact mechanism of sal DBS is achieving these promising results, action for both brain stimulation treatments although Holtzheimer and Mayberg (2011) remains unknown, it is noteworthy that the suggest that modulation of a mood regula- dorsolateral PFC is consistently activated in tion network may be the underlying mecha- brain imaging studies of reappraisal, and nism. plays an important role in working mem- ory and executive control. The implication Conclusions is that lack of engagement of dorsolateral Research on how the human brain creates PFC, specifically left dorsolateral PFC, in and regulates our emotions is still in its depressed patients compromises their ability infancy, but it has already given us remark- to regulate their emotions, and that stimu- able insights into the complexity of the brain lating these areas leads to an improvement systems responsible for shaping our moods in emotion regulation through reappraisal and emotions. In many ways human brain or executive control. Until researchers spe- imaging studies have validated the notion cifically test the effects of dorsolateral PFC that the neural systems supporting flexible stimulation (either tDCS or rTMS) on regu- emotion regulation are a key component of lation of emotion, this hypothesis remains mental health. Knowledge of the role of spe- speculative. One possibility to enhance the cific corticolimbic networks in emotion dys- effect of rTMS or tDCS is to use neuroim- regulation and psychopathology is now lead- aging to determine which hemisphere of the ing to new experimental treatments involving PFC is dysfunctional to direct the stimula- brain stimulation and neurofeedback. This tion protocol according to these findings progress notwithstanding, there are many (Schönfeldt-L­ ecuona et al., 2010), possibly outstanding questions to be resolved. One using an emotion regulation task. Given that central question is the exact role of different prefrontal tDCS has been shown to improve subregions of PFC in different types of emo- learning in working memory tasks (Brasil-­ tion (dys)regulation. For example, although Neto, 2012) it is conceivable that tDCS could there is consistent evidence for vmPFC be applied during the training of psychologi- involvement in extinction recall, the exact cal emotion regulation in order to boost the role of vmPFC in more explicit, cognitive learning of emotion regulation strategies. types of emotion regulation such as reap- praisal remains unclear. It is possible that Deep Brain Stimulation this question will be answered not by rela- DBS involves the delivery of a small electric tively crude fMRI measures of the amount current directly to parts of the brain through of vmPFC activation but with more nuanced implanted wire electrodes, with stimulation techniques that can capture changes to the delivered via an implantable pulse genera- way vmPFC encodes the hedonic value of tor. As an invasive surgical procedure, it has stimuli, perhaps under “direction” from lat- so far been investigated as a treatment for eral PFC. Multivariate pattern analysis (cf. the most intractable and debilitating cases Davis & Poldrack, 2013), for example, can of major depressive disorder in a number probe not only how much a brain region is of small, open-­label (i.e., nonblind) stud- engaged during a task but also the extent to ies. The most common site for stimula- which it represents different types of infor- tion is the white matter in the subcallosal mation. Combining such methods with mea- cingulate region, chosen on the basis of its sures of whole-­brain effective and functional connectivity with brain regions thought to connectivity should lead to testable models be important for antidepressant response of emotion regulation networks. (Holtzheimer & Mayberg, 2011; Lozano et al., 2008; Mayberg et al., 2005). These stud- Another unresolved question is how dif- ies have shown remission rates of 40–60% ferent emotion regulation disturbances map over an extended period (up to several years) onto different symptoms. In this chapter, we in patients with severe depression who have have rather simplistically discussed anxiety failed to respond to any other treatments. disorders and depression separately, even Given the relatively small sample sizes, it is though there is a large amount of comorbid- difficult to pinpoint exactly how subcallo- ity between these and other disorders. Are

The Neural Basis of Emotion Dysregulation 71 there common or distinct emotion regula- nucleus, and the former refers to a subregion tion processes involved? To what extent is of the striatum, which might contain a more emotion dysregulation distinct, or does it varied collection of neurons. overlap with more general disturbances in top-down control? To address these fun- References damental questions will require larger scale studies (including meta-a­ nalyses) that Abler, B., Erk, S., Herwig, U., & Walter, H. combine data from different samples and (2007). Anticipation of aversive stimuli acti- experimental tasks (e.g., nonemotional tasks vates extended amygdala in unipolar depres- involving attentional control, working mem- sion. Journal of Psychiatric Research, 41(6), ory, and response inhibition). 511–522. Further research on the impact of genetic Allan, C. L., Herrmann, L. L., & Ebmeier, K. variation and early life environment on the P. (2011). Transcranial magnetic stimulation development of the brain systems that allow in the management of mood disorders. Neuro- us to regulate our emotions will also be psychobiology, 64(3), 163–169. important if we are to develop useful predic- tors of vulnerability to disorders of emotion Bar-­Haim, Y., Lamy, D., Pergamin, L., regulation. Along similar lines, we know Bakermans-K­ ranenburg, M. J., & van IJzen- almost nothing about how age-r­elated neu- doorn, M. H. (2007). Threat-­related atten- ral decline affects emotion regulation. Some tional bias in anxious and nonanxious indi- of the brain regions showing the greatest viduals: A meta-a­ nalytic study. Psychological age-­related atrophy, such as lateral PFC, are Bulletin, 133(1), 1–24. a key component of emotion regulation cir- cuitry. Do older individuals show the same Beauregard, M., Paquette, V., & Levesque, J. sorts of age-r­elated decline in emotion regu- (2006). Dysfunction in the neural circuitry of lation as they do in cognitive function? Early emotional self-r­egulation in major depressive evidence suggests that the picture might not disorder. NeuroReport, 17(8), 843–846. be so straightforward (van Reekum et al., 2011; Winecoff, LaBar, Madden, Cabeza, & Bilkei-G­ orzo, A., Erk, S., Schurmann, B., Mauer, Huettel, 2011). Far more cross-­sectional and D., Michel, K., Boecker, H., et al. (2012). Dyn- longitudinal research relating age, measures orphins regulate fear memory: From mice to of cortical integrity, cognitive function, and men. Journal of Neuroscience, 32(27), 9335– emotion regulation capability is required to 9343. answer these important questions. Bishop, S. J. (2008). Trait anxiety and impover- Notes ished prefrontal control of attention. Nature Neuroscience, 12(1), 92–98. 1. In this chapter we have adopted a liberal con- sideration of emotion and mood, using the Bishop, S. J., Duncan, J., Brett, M., & Lawrence, terms interchangeably. Although these terms A. D. (2004). Prefrontal cortical function might be considered more or less distinct by and anxiety: Controlling attention to threat-­ emotion theorists, the evidence for a clear dis- related stimuli. Nature Neuroscience, 7(2), tinction in terms of neurobiology is lacking. 184–188. In addition, from a clinical standpoint, mood dysregulation is at least as important as emo- Blair, K. S., Smith, B. W., Mitchell, D. G. V., tion dysregulation. Morton, J., Vythilingam, M., Pessoa, L., et al. (2007). Modulation of emotion by cognition 2. Most studies have not explicitly tested the and cognition by emotion. NeuroImage, 35(1), hemispheric lateralization of prefrontal acti- 430 – 440. vation, so caution must be exercised in infer- ring preferential involvement of one hemi- Bouton, M. E. (2004). Context and behavioral sphere in reappraisal of emotional stimuli. processes in extinction. Learning and Mem- ory, 11(5), 485–494. 3. Although the term ventral striatum is some- times used interchangeably with nucleus Brasil-­Neto, J. P. (2012). Learning, memory, and accumbens, the latter refers to a specific transcranial direct current stimulation. Fron- tiers in Neuropsychiatric Imaging and Stimu- lation, 3, 80. Burghy, C. A., Stodola, D. E., Ruttle, P. L., Molloy, E. K., Armstrong, J. M., Oler, J. A., et al. (2012). Developmental pathways to amygdala–p­ refrontal function and internaliz-

72 BIOLOGICAL BASES ing symptoms in adolescence. Nature Neuro- Personality and Individual Differences, 44(7), science, 15(12), 1736–1741. 1574–1584. Campbell-S­ ills, L., Ellard, K., & Barlow, D. (this Epstein, J., Pan, H., Kocsis, J. H., Yang, Y., volume, 2014). Emotion regulation in anxiety Butler, T., Chusid, J., et al. (2006). Lack of disorders. In J. J. Gross (Ed.), Handbook of ventral striatal response to positive stimuli in emotion regulation (2nd ed., pp. 393–412). depressed versus normal subjects. American New York: Guilford Press. Journal of Psychiatry, 163(10), 1784–1790. Casey, B. J., Tottenham, N., Liston, C., & Dur- Erk, S., Kleczar, A., & Walter, H. (2007). ston, S. (2005). Imaging the developing brain: Valence-­specific regulation effects in a work- What have we learned about cognitive devel- ing memory task with emotional context. opment? Trends in Cognitive Sciences, 9(3), NeuroImage, 37(2), 623–632. 104–110. Erk, S., Mikschl, A., Stier, S., Ciaramidaro, A., Caspi, A., Hariri, A. R., Holmes, A., Uher, R., & Gapp, V., Weber, B., et al. (2010). Acute and Moffitt, T. E. (2010). Genetic sensitivity to the sustained effects of cognitive emotion regula- environment: The case of the serotonin trans- tion in major depression. Journal of Neurosci- porter gene and its implications for studying ence, 30(47), 15726–15734. complex diseases and traits. American Journal Erthal, F., De Oliveira, L., Mocaiber, I., Pereira, of Psychiatry, 167(5), 509–527. M., Machado-­Pinheiro, W., Volchan, E., et al. Cisler, J. M., & Koster, E. H. W. (2010). Mecha- (2005). Load-d­ ependent modulation of affec- nisms of attentional biases towards threat in tive picture processing. Cognitive, Affective, anxiety disorders: An integrative review. Clin- and Behavioral Neuroscience, 5(4), 388–395. ical Psychology Review, 30(2), 203–216. Etkin, A., Egner, T., Peraza, D. M., Kandel, E. Clarke, R. J., & Johnstone, T. (2013). Prefrontal R., & Hirsch, J. (2006). Resolving emotional inhibition of threat processing protects work- conflict: A role for the rostral anterior cingu- ing memory from interference. Frontiers in late cortex in modulating activity in the amyg- Human Neuroscience, 7, 228. dala. Neuron, 51(6), 871–882. Corson, H. (1890). An introduction to the study Etkin, A., Klemenhagen, K. C., Dudman, J. T., of Shakespeare. Boston: Heath. Retrieved Rogan, M. T., Hen, R., Kandel, E. R., et al. from http://archive.org/details/introduction- (2004). Individual differences in trait anxiety tost01cors. predict the response of the basolateral amyg- Davidson, R. J., Pizzagalli, D., Nitschke, J. B., & dala to unconsciously processed fearful faces. Putnam, K. (2002). Depression: Perspectives Neuron, 44(6), 1043–1055. from affective neuroscience. Annual Review Etkin, A., & Wager, T. D. (2007). Functional of Psychology, 53(1), 545–574. neuroimaging of anxiety: A meta-a­ nalysis of Davis, M. (2006). Neural systems involved emotional processing in PTSD, social anxiety in fear and anxiety measured with fear-­ disorder, and specific phobia. American Jour- potentiated startle. American Psychologist, nal of Psychiatry, 164(10), 1476–1488. 61(8), 741–756. Fitzgerald, P. B., Fountain, S., & Daskalakis, Davis, T., & Poldrack, R. A. (2013). Measuring Z. J. (2006). A comprehensive review of the neural representations with fMRI: Practices effects of rTMS on motor cortical excitabil- and pitfalls. Annals of the New York Acad- ity and inhibition. Clinical Neurophysiology, emy of Sciences. 117(12), 2584–2596. Delgado, M. R., Gillis, M. M., & Phelps, E. A. Fitzgerald, P. B., Laird, A. R., Maller, J., & Das- (2008). Regulating the expectation of reward kalakis, Z. J. (2008). A meta-­analytic study via cognitive strategies. Nature Neuroscience, of changes in brain activation in depression. 11(8), 880–881. Human Brain Mapping, 29(6), 683–695. Delgado, M. R., Nearing, K. I., LeDoux, J. Friedel, E., Schlagenhauf, F., Sterzer, P., Park, S., E., & Phelps, E. A. (2008). Neural circuitry Bermpohl, F., Ströhle, A., et al. (2009). 5-HTT underlying the regulation of conditioned fear genotype effect on prefrontal–a­ mygdala cou- and its relation to extinction. Neuron, 59(5), pling differs between major depression and 829–838. controls. Psychopharmacology, 205(2), 261– Ehring, T., Fischer, S., Schnülle, J., Bösterling, 271. A., & Tuschen-­Caffier, B. (2008). Charac- Frijda, N. H. (1986). The emotions. Cambridge, teristics of emotion regulation in recovered UK: Cambridge University Press. depressed versus never depressed individuals. Frodl, T., Reinhold, E., Koutsouleris, N., Reiser,

The Neural Basis of Emotion Dysregulation 73 M., & Meisenzahl, E. M. (2010). Interaction diminished maintenance of fronto-­striatal of childhood stress with hippocampus and brain activation. Proceedings of the National prefrontal cortex volume reduction in major Academy of Sciences, 106(52), 22445–22450. depression. Journal of Psychiatric Research, Holtzheimer, P. E., & Mayberg, H. S. (2011). 44(13), 799–807. Stuck in a rut: Rethinking depression and its Ghashghaei, H. T., Hilgetag, C. C., & Barbas, treatment. Trends in Neurosciences, 34(1), H. (2007). Sequence of information process- 1–9. ing for emotions based on the anatomic dia- Jackson, D. C., Malmstadt, J. R., Larson, C. logue between prefrontal cortex and amyg- L., & Davidson, R. J. (2000). Suppression dala. NeuroImage, 34(3), 905–923. and enhancement of emotional responses to Gilbert, K. E. (2012). The neglected role of posi- unpleasant pictures. Psychophysiology, 37(4), tive emotion in adolescent psychopathology. 515–522. Clinical Psychology Review, 32(6), 467–481. Johnstone, T., van Reekum, C. M., Urry, H. L., Gillihan, S. J., Rao, H., Wang, J., Detre, J. A., Kalin, N. H., & Davidson, R. J. (2007). Failure Breland, J., Sankoorikal, G. M. V., et al. to regulate: Counterproductive recruitment of (2010). Serotonin transporter genotype modu- top-down prefrontal–s­ubcortical circuitry in lates amygdala activity during mood regula- major depression. Journal of Neuroscience, tion. Social Cognitive and Affective Neurosci- 27(33), 8877–8884. ence, 5(1), 1–10. Kalisch, R., Korenfeld, E., Stephan, K. E., Weis- Gross, J. J. (this volume, 2014). Emotion regula- kopf, N., Seymour, B., & Dolan, R. J. (2006). tion: Conceptual and empirical foundations. Context-d­ ependent human extinction memory In J. J. Gross, (Ed.), Handbook of emotion is mediated by a ventromedial prefrontal and regulation (2nd ed., pp. 3–20). New York: hippocampal network. Journal of Neurosci- Guilford Press. ence, 26(37), 9503–9511. Gross, J. J., & John, O. P. (2003). Individual dif- Kanske, P., Heissler, J., Schönfelder, S., & Wessa, ferences in two emotion regulation processes: M. (2012). Neural correlates of emotion reg- Implications for affect, relationships, and ulation deficits in remitted depression: The well-­being. Journal of Personality and Social influence of regulation strategy, habitual regu- Psychology, 85(2), 348–362. lation use, and emotional valence. NeuroIm- Gyurak, A., & Etkin, A. (this volume, 2014). A age, 61(3), 686–693. neurobiological model of implicit and explicit Karg, K., Burmeister, M., Shedden, K., & Sen, emotion regulation. In J. J. Gross (Ed.), S. (2011). The serotonin transporter promoter Handbook of emotion regulation (2nd ed., variant (5-HTTLPR), stress, and depression pp. 76–90). New York: Guilford Press. meta-­analysis revisited: Evidence of genetic Hariri, A. R., Mattay, V. S., Tessitore, A., Fera, moderation. Archives of General Psychiatry, F., & Weinberger, D. R. (2003). Neocorti- 68(5), 444–454. cal modulation of the amygdala response to Keedwell, P. A., Andrew, C., Williams, S. C. R., fearful stimuli. Biological Psychiatry, 53(6), Brammer, M. J., & Phillips, M. L. (2005). 494–501. The neural correlates of anhedonia in major Heinz, A., Braus, D. F., Smolka, M. N., Wrase, depressive disorder. Biological Psychiatry, J., Puls, I., Hermann, D., et al. (2005). 58(11), 843–853. Amygdala–p­refrontal coupling depends on a Kim, S. H., & Hamann, S. (2007). Neural cor- genetic variation of the serotonin transporter. relates of positive and negative emotion regu- Nature Neuroscience, 8(1), 20–21. lation. Journal of Cognitive Neuroscience, Heller, A. S., Johnstone, T., Light, S. N., Peter- 19(5), 776–798. son, M. J., Kolden, G. G., Kalin, N. H., et al. Knutson, B., Bhanji, J. P., Cooney, R. E., Atlas, L. (2013). Relationships between changes in sus- Y., & Gotlib, I. H. (2008). Neural responses to tained fronto-s­triatal connectivity and posi- monetary incentives in major depression. Bio- tive affect in major depression resulting from logical Psychiatry, 63(7), 686–692. antidepressant treatment. American Journal Lemogne, C., Gorwood, P., Boni, C., Pessiglione, of Psychiatry, 170(2), 197–206. M., Lehéricy, S., & Fossati, P. (2011). Cogni- Heller, A. S., Johnstone, T., Shackman, A. J., tive appraisal and life stress moderate the Light, S. N., Peterson, M. J., Kolden, G. G., effects of the 5-HTTLPR polymorphism on et al. (2009). Reduced capacity to sustain amygdala reactivity. Human Brain Mapping, positive emotion in major depression reflects 32(11), 1856–1867.

74 BIOLOGICAL BASES Levesque, J., Joanette, Y., Mensour, B., Beaudoin, strates. Behavioural Brain Research, 217(1), G., Leroux, J. M., Bourgouin, P., et al. (2004). 215–231. Neural basis of emotional self-r­egulation in Monk, C. S., McClure, E. B., Nelson, E. E., childhood. Neuroscience, 129(2), 361–369. Zarahn, E., Bilder, R. M., Leibenluft, E., et al. (2003). Adolescent immaturity in attention-­ Linden, D. E. J., Habes, I., Johnston, S. J., Lin- related brain engagement to emotional facial den, S., Tatineni, R., Subramanian, L., et al. expressions. NeuroImage, 20(1), 420–428. (2012). Real-­time self-r­egulation of emotion Nelson, E. E., McClure, E. B., Monk, C. S., networks in patients with depression. PLoS Zarahn, E., Leibenluft, E., Pine, D. S., et al. ONE, 7(6), e38115. (2003). Developmental differences in neuro- nal engagement during implicit encoding of Loo, C. K., Alonzo, A., Martin, D., Mitchell, P. emotional faces: An event-­related fMRI study. B., Galvez, V., & Sachdev, P. (2012). Transcra- Journal of Child Psychology and Psychiatry nial direct current stimulation for depression: and Allied Disciplines, 44(7), 1015–1024. 3-week, randomised, sham-c­ontrolled trial. Ochsner, K. N., & Gross, J. J. (this volume, British Journal of Psychiatry, 200(1), 52–59. 2014). The neural bases of emotion and emo- tion regulation: A valuation perspective. In J. Lozano, A. M., Mayberg, H. S., Giacobbe, P., J. Gross (Ed.), Handbook of emotion regula- Hamani, C., Craddock, R. C., & Kennedy, tion (2nd ed., pp. 23–42). New York: Guilford S. H. (2008). Subcallosal cingulate gyrus Press. deep brain stimulation for treatment-­resistant Padberg, F., & George, M. S. (2009). Repeti- depression. Biological Psychiatry, 64(6), 461– tive transcranial magnetic stimulation of the 467. prefrontal cortex in depression. Experimental Neurology, 219(1), 2–13. MacLeod, C., Mathews, A., & Tata, P. (1986). Perlman, G., Simmons, A. N., Wu, J., Hahn, Attentional bias in emotional disorders. Jour- K. S., Tapert, S. F., Max, J. E., et al. (2012). nal of Abnormal Psychology, 95(1), 15–20. Amygdala response and functional connectiv- ity during emotion regulation: A study of 14 Mathers, C. D., Lopez, A. D., & Murray, C. J. depressed adolescents. Journal of Affective L. (2006). The burden of disease and mortal- Disorders, 139(1), 75–84. ity by condition: Data, methods, and results Pezawas, L., Meyer-L­indenberg, A., Drab- for 2001. In A. D. Lopez, C. D. Mathers, ant, E. M., Verchinski, B. A., Munoz, K. E., M. Ezzati, D. T. Jamison, & C. J. L. Murray Kolachana, B. S., et al. (2005). 5-HTTLPR (Eds.), Global burden of disease and risk fac- polymorphism impacts human cingulate–­ tors (pp. 45–241). New York: World Bank/ amygdala interactions: A genetic susceptibility Oxford University Press. mechanism for depression. Nature Neurosci- ence, 8(6), 828–834. Mayberg, H. S., Lozano, A. M., Voon, V., Phelps, E. A., Delgado, M. R., Nearing, K. I., & McNeely, H. E., Seminowicz, D., Hamani, LeDoux, J. E. (2004). Extinction learning in C., et al. (2005). Deep brain stimulation for humans: Role of the amygdala and vmPFC. treatment-­resistant depression. Neuron, 45(5), Neuron, 43(6), 897–905. 651– 660. Posner, M. I., Snyder, C. R., & Davidson, B. J. (1980). Attention and the detection of signals. Meehl, P. E. (1975). Hedonic capacity: Some Journal of Experimental Psychology, 109(2), conjectures. Bulletin of the Menninger Clinic, 160 –174. 39(4), 295–307. Ridderinkhof, K. R., Ullsperger, M., Crone, E. A., & Nieuwenhuis, S. (2004). The role of the Milad, M. R., & Quirk, G. J. (2002). Neurons medial frontal cortex in cognitive control. Sci- in medial prefrontal cortex signal memory for ence, 306(5695), 443–447. fear extinction. Nature, 420(6911), 70–74. Rolls, E. T. (2004). The functions of the orbi- tofrontal cortex. Brain and cognition, 55(1), Milad, M. R., & Quirk, G. J. (2012). Fear extinc- 11–29. tion as a model for translational neuroscience: Savitz, J., & Drevets, W. C. (2009). Bipolar and Ten years of progress. Annual Review of Psy- major depressive disorder: Neuroimaging the chology, 63(1), 129–151. developmental-­degenerative divide. Neuro- Milad, M. R., Wright, C. I., Orr, S. P., Pitman, R. K., Quirk, G. J., & Rauch, S. L. (2007). Recall of fear extinction in humans activates the ventromedial prefrontal cortex and hip- pocampus in concert. Biological Psychiatry, 62(5), 446–454. Mitchell, D. G. V. (2011). The nexus between decision making and emotion regulation: A review of convergent neurocognitive sub-

The Neural Basis of Emotion Dysregulation 75 science and Biobehavioral Reviews, 33(5), (2012). Interactions between transient and 699–771. sustained neural signals support the genera- Schaefer, H. S., Putnam, K. M., Benca, R. M., tion and regulation of anxious emotion. Cere- & Davidson, R. J. (2006). Event-­related func- bral Cortex, 23(1), 49–60. tional magnetic resonance imaging measures Speer, A. M., Kimbrell, T. A., Wassermann, E. of neural activity to positive social stimuli in M., Repella, J., Willis, M. W., Herscovitch, pre- and post-­treatment depression. Biological P., et al. (2000). Opposite effects of high and Psychiatry, 60(9), 974–986. low frequency rTMS on regional brain activity Schardt, D. M., Erk, S., Nüsser, C., Nöthen, M. in depressed patients. Biological Psychiatry, M., Cichon, S., Rietschel, M., et al. (2010). 48(12), 1133–1141. Volition diminishes genetically mediated Staudinger, M. R., Erk, S., & Walter, H. (2011). amygdala hyperreactivity. NeuroImage, 53(3), Dorsolateral prefrontal cortex modulates 943–951. striatal reward encoding during reappraisal of Scherer, K. R., Schorr, A., & Johnstone, T. reward anticipation. Cerebral Cortex, 21(11), (2001). Appraisal processes in emotion: The- 2578–2588. ory, methods, research. New York: Oxford Tomarken, A. J., & Keener, A. D. (1998). Fron- University Press. tal brain asymmetry and depression: A self-­ Schönfeldt-­Lecuona, C., Lefaucheur, J.-P., regulatory perspective. Cognition and Emo- Cardenas-­Morales, L., Wolf, R. C., Kam- tion, 12(3), 387–420. mer, T., & Herwig, U. (2010). The value of Urry, H. L., van Reekum, C. M., Johnstone, T., neuronavigated rTMS for the treatment of Kalin, N. H., Thurow, M. E., Schaefer, H. depression. Neurophysiologie Clinique, 40(1), S., et al. (2006). Amygdala and ventromedial 37– 43. prefrontal cortex are inversely coupled during Schutter, D. J. L. G. (2009). Antidepressant effi- regulation of negative affect and predict the cacy of high-f­requency transcranial magnetic diurnal pattern of cortisol secretion among stimulation over the left dorsolateral prefron- older adults. Journal of Neuroscience, 26(16), tal cortex in double-­blind sham-­controlled 4415 – 4425. designs: A meta-­analysis. Psychological Medi- Van Dillen, L. F., Heslenfeld, D. J., & Koole, S. cine, 39(1), 65–75. L. (2009). Tuning down the emotional brain: Schwartz, C. E., & Rauch, S. L. (2004). Temper- An fMRI study of the effects of cognitive load ament and its implications for neuroimaging on the processing of affective images. Neuro- of anxiety disorders. CNS Spectrums, 9(4), Image, 45(4), 1212–1219. 284–291. van Reekum, C. M., Schaefer, S. M., Lapate, R. Schwartz, C. E., Wright, C. I., Shin, L. M., C., Norris, C. J., Greischar, L. L., & David- Kagan, J., & Rauch, S. L. (2003). Inhibited son, R. J. (2011). Aging is associated with and uninhibited infants “grown up”: Adult positive responding to neutral information but amygdalar response to novelty. Science, reduced recovery from negative information. 300(5627), 1952–1953. Shackman, A. J., Sarinopoulos, I., Maxwell, Social Cognitive and Affective Neuroscience, J. S., Pizzagalli, D. A., Lavric, A., & David- 6(2), 177–185. son, R. J. (2006). Anxiety selectively disrupts Wager, T. D., Davidson, M. L., Hughes, B. L., visuospatial working memory. Emotion, 6(1), Lindquist, M. A., & Ochsner, K. N. (2008). 40 – 61. Prefrontal-s­ubcortical pathways mediating Shin, L. M., & Liberzon, I. (2010). The neuro- successful emotion regulation. Neuron, 59(6), circuitry of fear, stress, and anxiety disorders. 1037–1050. Neuropsychopharmacology, 35(1), 169–191. Walker, D. L., Toufexis, D. J., & Davis, M. Slotema, C. W., Blom, J. D., Hoek, H. W., & (2003). Role of the bed nucleus of the stria Sommer, I. E. C. (2010). Should we expand terminalis versus the amygdala in fear, stress, the toolbox of psychiatric treatment methods and anxiety. European Journal of Pharmacol- to include repetitive transcranial magnetic ogy, 463(1–3), 199–216. stimulation (rTMS)? Journal of Clinical Psy- Winecoff, A., LaBar, K. S., Madden, D. J., chiatry, 71(7), 873–884. Cabeza, R., & Huettel, S. A. (2011). Cognitive Somerville, L. H., Wagner, D. D., Wig, G. S., and neural contributors to emotion regulation Moran, J. M., Whalen, P. J., & Kelley, W. M. in aging. Social Cognitive and Affective Neu- roscience, 6(2), 165–176.

Chapter 5 A Neurobiological Model of Implicit and Explicit Emotion Regulation Anett Gyurak Amit Etkin Emotions take center stage in our lives tory contexts. By anchoring this conceptual through the ways they influence how we framework in neurobiological findings, this think and behave. Yet emotions are not model of emotion regulation brings together deterministic; just as much as they influ- basic, clinical, neuroscientific, and trans- ence us, we have the ability to regulate and lational research. Our goal in this chapter change the way our emotional responses is to demonstrate how an understanding unfold. Most research on emotion regula- anchored at the implicit–e­xplicit spectrum tion has focused on how people explicitly informs an understanding of psychopathol- (using deliberate, effortful means) accom- ogy that cuts across traditional diagnostic plish the regulation of emotion (e.g., Och- boundaries. sner & Gross, 2005). We have previously argued that it is useful to anchor emotion Component Processes regulatory processes along an implicit (non- and Definition of Emotion Regulation conscious, automatic) to explicit (deliberate, The Modal Model of Emotions effortful) dimension within a dual-­process Using previous definitions (Gross & Thomp- framework (Gyurak, Gross, & Etkin, 2011). son, 2007; Gross, this volume) we define In this chapter, we demonstrate the utility of “emotions” as interpersonal and transac- the implicit–e­ xplicit framework to organize tional (Keltner & Haidt, 2001) processes a wealth of neurobiological findings from that play out in the rich fabric of real or basic affective and clinical neuroscience. imagined social relationships, for example, We utilize this new expanded framework feelings of embarrassment over a social faux to illustrate its benefits for understanding pas, or anger when one is slighted. Emotions healthy and disordered emotion regulatory arise when individuals attend to a situation processes. and see it as relevant to their goals, either in the short or long term (Levenson, 1999). We present paradigmatic cases of explicit As emotions unfold, they engage loosely and implicit regulation to highlight the par- coupled changes in the domains of subjec- allels and differences between implicit and tive experience, behavior, and central and explicit forms of regulation. We then explore peripheral physiology (Mauss, Levenson, cases in which implicit and explicit regula- McCarter, Wilhelm, & Gross, 2005). Many tory processes blend together, to illustrate the fluid boundaries between implicit and explicit regulation in many emotion regula- 76

Neurobiological Model of Implicit and Explicit Emotion Regulation 77 of the central and peripheral physiologi- tions (Bargh, 1994; Moors & De Houwer, cal reactions have phylogenetic continuity 2006). In general, attempts to create a clear across species (LeDoux, 2012). Moreover, dichotomy capable of completely dividing these reactions simultaneously include pro- implicit and explicit processes have been cesses generally seen as “emotional” (e.g., unsuccessful. Drawing on this body of lit- a withdrawal response or a feeling), as well erature, for the purposes of this review, we as “cognitive” (e.g., an attentional focus on define a spectrum, rather than discrete cat- a relevant stimulus; Pessoa, 2010). Finally, egories, and characterize explicit emotion emotions can be viewed as “push” forces regulation as those processes that require that predispose us to act in certain ways conscious effort for initiation, demand some (Frijda, 1987; Keltner & Gross, 1999; Lev- level of monitoring during implementation, enson, 2003). and are associated with some level of insight and awareness. Implicit regulation, is evoked Together, these features of emotion con- automatically, run to completion with- stitute the modal model of emotion: a goal-­ out monitoring, and can happen without driven person–­situation transaction that insight and awareness. As outlined below, structures attention, has particular meaning specific examples of types of emotion regu- to an individual, and gives rise to an evo- lation may vary with respect to the degree lutionarily rooted coordinated yet flexible of explicitness or implicitness and, as such, multisystem response (Gross & Thompson, this definition most clearly describes pro- 2007; Gross, this volume). However, just cesses at the extreme ends of this spectrum. as emotions have many compulsory fea- Our view is that there is scientific utility in tures that are well-c­ onserved evolutionarily, using the explicit–­implicit spectrum: It is emotions have also been understood to be more parsimonious to understand different malleable since the work of William James emotion regulatory paradigms currently in (1884). It is this latter aspect of emotion the literature within this framework rather that is most crucial for an analysis of emo- than characterize each paradigm at the level tion regulation, because it is this feature that of constituent processes (level of monitoring, makes any form of regulation possible. awareness, etc.). Emotion Regulation We view both implicit and explicit Following past work, we define “emotion forms of regulation as functional—a­s hav- regulation” as functional processes that ing arisen to help navigate the rich social influence the intensity, duration, and type world in which humans exist. The ability of emotion experienced (Gross & Thomp- to deploy either form of regulation flexibly son, 2007). Emotion regulation permits in a context-­sensitive manner might be a flexibility in emotional responding in accord marker of mental health. Additionally, the with one’s momentary as well as longer-­ conscious, effortful operations of explicit term goals. Emotion regulatory processes emotion regulatory processes are useful in a may have their effects at one or more points variety of contexts for adjusting emotional in the emotion generative process, and may reactions. Because of its cognitively demand- dampen, intensify, or simply maintain emo- ing nature, however, explicit regulation is tions, depending on an individual’s current not something in which one can effectively implicit or explicit goals. Emotion regula- engage all of the time. Rather, use of effi- tory processes may also change the degree to cient implicit emotion regulation processes, which emotion response components cohere which may run without awareness, is likely a as the emotion unfolds. critical component to well-being and every- day emotional functioning. We conceptualize emotion regulation as occurring along a spectrum from con- Most contexts in which emotions are reg- scious, effortful, and controlled regulation ulated involve a variable mixture of implicit (which we call explicit) to unconscious, and and explicit processes. The degree to which possibly effortless, or automatic regulation ongoing regulation is implicit or explicit (implicit). The history of explicit–i­mplicit may also change depending on the individ- process description in psychology has been ual, context, or time course of the emotion marked by arguments over terms and defini- or its regulation. For example, a particular regulatory process may be explicit in healthy

78 BIOLOGICAL BASES individuals (suppressing a testy response to frontal regions and might be more directly a colleague) but implicit in patients (expres- relevant to fear- and anxiety-­related regula- sive suppression), in whom it may have been tion (Gyurak et al., 2011). automatized through overrepetition, to the point that it has become habitual. Addition- The Medial Prefrontal Cortex ally, explicit and implicit regulation change Regions of the medial frontal lobes, includ- in quick succession within one person over ing the ventral portions of the anterior cin- time. For example, one might go from con- gulate cortex (vACC) and ventromedial sciously labeling one’s feelings of frustration prefrontal cortex (vmPFC) have long been at work to reflexively looking at the positive implicated in emotional processes, specifi- outcome that might result from putting in cally in regulation of emotions. We review the extra time. relevant evidence in turn below. In addi- tion to emotion regulation, these regions Taken together, our proposal is that it is are important in fear extinction and recall- most fruitful to view emotion regulation as ing inhibitory extinction memories a day reflecting a spectrum of processes, rang- or more after training (Milad et al., 2009; ing from the mostly explicit to the mostly Phelps, Delgado, Nearing, & LeDoux, implicit. In applying this framework and 2004). Similarly, the vACC/vmPFC is acti- systematically testing its components we vated when exposure to a distant versus can gain important insights into the neu- close threat occurs, suggesting that it may ral circuits that are critical for the dynamic be involved in planning adaptive responses, process of emotion regulation, and into the a regulatory function (Mobbs et al., 2007). utility of implicit and explicit processes and The vACC/vmPFC shows negative func- their relevance for well-being and, by exten- tional connectivity with the amygdala in a sion, psychopathology. range of emotional tasks and has anatomical connectivity to the amygdala (Myers-S­ chulz Core Neural Structures & Koenigs, 2012). Finally, during explicit of Emotion Regulation emotion regulation, the vACC may also act Our focus in this chapter is on neurobiologi- as an intermediary between dlPFC and the cally investigated paradigms of emotion reg- amygdala (Ochsner & Gross, 2005). ulation. This section provides a brief over- view of the core structures in the prefrontal Selected Empirical Findings cortex that are thought to be critical for across the Spectrum the monitoring and regulation of emotion. of Emotion Regulation Figure 5.1 introduces the lateral–m­edial Below we review a number of neurobiologi- gradient along which explicit and implicit cally investigated examples of emotion regu- processes can be usefully lined up along. latory processes along the explicit–­implicit We then expand on these as we review each spectrum. Though these examples are paradigm below. ordered roughly in the degree to which they involve implicit or explicit processes, we The Lateral Prefrontal Cortex remind the reader of the fluid and relative The dorsolateral (dlPFC) and ventrolateral nature of implicit and explicit regulatory (vlPFC) prefrontal cortices are commonly processes. We review the relevance of each associated with emotion regulation (e.g., example for increasing our understanding Kalisch, 2009). Along with their role in of psychopathology. We start on the implicit cognitive control and executive function- end. ing, these regions are typically described in the context of deliberate, effortful, and con- Regulation of Emotional Conflict scious regulation of emotion (Gyurak et al., Our first case of implicit emotion regula- 2011). However, as described earlier, emo- tion is based on the emotional conflict task tion regulation can occur more reflexively (Egner, Etkin, Gale, & Hirsch, 2008; Etkin, and outside of our awareness, and this type of regulation implicates more medial pre-

Neurobiological Model of Implicit and Explicit Emotion Regulation 79 Emotional conflict adaptation, x = –2, y = 44, z = 2; from Egner et al. (2008) Habituation, x = –2, y = 44, z = 2; from Hare et al. (2008) Extinction, x = –2, y = 23, z = –8; from Milad et al. (2009) Anticipation, x = –2, y = 40, z = 7; from Straube et al. (2009) Verbal labeling (triangles in the medial and lateral prefrontal cortices), x = –2, y = 48, z = 16 and, x = 24, y = –10, z = 44, respectively; from Lieberman et al. (2007) Reappraisal, calculated as the mean of the absolute values of all vLPFC and dlPFC activation in Study 2a plotted on the right hemisphere for illustrative purposes, x = 42, y = 27, z = 21; from Kalisch (2009) FIGURE 5.1.  Putative foci for implicit and explicit forms of emotional regulation on a spectrum. Foci plotted are approximate; x coordinate = –2 for visualization purposes on the medial slice. Egner, Peraza, Kandel, & Hirsch, 2006). over it, or fearful face with the word happy The task is the emotional analogue of the written over it). Because reading is automa- classic Stroop paradigm (Stroop, 1935), and tized (Stroop, 1935), during conflict tri- takes advantage of a variant of the congru- als, participants need to enact control over ency sequence effects originally reported by reading the word in favor of labeling the Gratton, Coles, and Donchin (1992) in non- emotional expression. Consistent with this, emotional conflict tasks (Botvinick, Nys- it takes longer to respond to conflict trials trom, Fissell, Carter, & Cohen, 1999). In the than to no-­conflict trials; this slow-down in emotional conflict task, participants are pre- response time due to emotional conflict has sented with photographs of emotional faces been termed the congruency effect. Implicit (fearful or happy) with the words fear or emotion regulation in this task is indexed happy written over them. The task is to indi- by trial-to-trial changes in the congruency cate whether the facial expression is happy effect as a function of the congruency effect or fearful by pressing a button and to do this on the previous trial (trial n – 1). Specifically, fast and accurately. The word written on the conflict on trial n – 1 triggers an increase in photo either matches the facial expression regulation, thereby reducing susceptibility (no-­conflict trials: happy face with the word to emotional conflict on trial n. Consistent happy written over it, or fearful face with with this, response times to incongruent tri- the word fear written over it) or is incongru- als are usually faster after incongruent tri- ent with the facial expression (conflict tri- als (iI = current Incongruent preceded by an als: happy face with the word fear written incongruent trial) than after congruent trials

80 BIOLOGICAL BASES (cI = current Incongruent preceded by a con- more, analysis of functional magnetic reso- gruent trial). Implicit emotion regulation is nance imaging (fMRI) data acquired during therefore quantified by contrasting response the emotional conflict task revealed that times on iI with cI trials. This behavioral patients with GAD or depression failed to slowdown effect was shown to occur outside engage the circuitry normally implicated of awareness (Etkin, Prater, Hoeft, Menon, in this task. Specifically, patients failed to & Schatzberg, 2010). The task qualifies as activate the vACC during regulation of emo- an emotion regulation paradigm, because tional conflict and to dampen emotional the stimulus itself evokes behavioral adjust- conflict evaluation-­related activity in the ment in the context of the goal of responding amygdala. These data indicate that regula- fast and accurately to emotionally conflict- tion of emotional conflict, as a prototypi- ing stimuli. This task fits our definition of cal exemplar of implicit emotion regulation, implicit regulation because the process is captures the critical inability of patients to uninstructed, likely results from the stimu- manage emotional processing and engage in lus properties itself, and is effortless and task-­related behavior. proceeds without awareness; despite care- ful probing, participants do not report any Extinction and Habituation awareness of the key processes of the task Both extinction and habituation are basic (Etkin et al., 2010). forms of learning that are important for adaptive behavioral responding to emotional Parallel imaging (Egner et al., 2008) and challenges. Both of these qualify as emotion lesion (Maier & di Pellegrino, 2012) studies regulation because they underlie the ability of regulation of nonemotional conflict sug- to change emotional responding to fit situa- gest that key elements of the neural circuitry tional demands. Both of these processes have involved in implicit regulation of emotional aspects that happen outside of conscious conflict is specific to the emotional content. awareness and, as such, share implicit emo- Specifically, neuroimaging studies suggest a tion regulatory elements but are likely to be regulatory interplay between anterior cin- a blend of implicit and explicit processes in gulate cortex (ACC) and emotion reactivity humans, as detailed below (Delgado, Near- regions in the limbic system in the emotional ing, LeDoux, & Phelps, 2008; Dijksterhuis conflict task: increased activation in the & Smith, 2002; Lovibond, 2004). vACC, and dampened reactivity in the amyg- dala (Etkin et al., 2006; Etkin et al., 2010; Extinction involves repeated presenta- see Figure 5.1). In the nonemotional ana- tion of an innocuous stimulus (e.g., a blue logue of the task, in which subjects judged cross) that was previously associated with the gender of emotional faces, while trying an aversive stimulus (e.g., a loud noise) and to ignore congruent or incongruent gender had acquired the ability to elicit a defensive labels, regulation was achieved through response (e.g., startle eyeblink) in the absence a dissociable neural pathway, involving of the aversive stimulus. Through repeated increased activation in the dlPFC, and posi- presentation without the aversive outcome, tive coupling to target-­specific visual corti- a new inhibitory memory is formed between cal areas in the fusiform face area, and as the innocuous stimulus and absence of the such, activation of the vACC and dampen- aversive outcome, such that the conditioned ing of amygdalar reactivity was specific to defensive reaction is no longer elicited. Even implicit emotion regulation (Egner et al., though extinction can happen without the 2008). individual being aware of the change in contingency, consciously registering the Applications of this task to clinical disor- contingency changes during extinction has ders revealed a marked deficit both behav- been shown to facilitate extinction (for a iorally and in neural network engagement as review, see Lovibond, 2004). This evidence a function of anxiety and depression (Etkin implicates the presence of both implicit and et al., 2010; Etkin & Schatzberg, 2011). Spe- explicit processes in extinction. cifically, nonmedicated patients with either generalized anxiety disorder (GAD) or Importantly, the neural circuitry support- comorbid depression and GAD were unable ing extinction shares common elements with to regulate the effect of emotional conflict the circuitry mapped out for the regulation as compared to healthy controls. Further-