Neural substrates of the interaction of emotional stimulus processing and motor inhibitory control: An emotional linguistic go/no-go fMRI study
Introduction
Approach and withdrawal represent fundamental behavioral tendencies. Successful adaptation requires that they be selectively invoked by a context-appropriate balance of agency and inhibition. Inhibitory control is a multi-domain executive function critical for flexible responsivity to changing task demands, and thereby an essential component of adaptive behavioral regulation. A phylogenetically and ontogenetically later-appearing function (Booth et al., 2003, Casey et al., 1997, Williams et al., 1999), inhibitory control afforded by evolutionary and maturational development may underlie human capacity for future-related thought (e.g., capacity for response inhibition enables transcendence of the “default mode” of human behavior (Mesulam 2002), permitting representation of alternative outcomes, thereby enabling meaningful behavioral decision-making) (Fuster, 2000, Goldman-Rakic, 1988). Inhibitory control is susceptible to impairment in a variety of developmental (Casey et al., 1997), degenerative (Neary et al., 2005, Royall et al., 2002, Shulman, 1997), and acquired (Damasio, 1999) processes, with potentially serious maladaptive consequences.
Neural substrates of response inhibition have been probed in animal model (Iverson and Mishkin, 1970, Roberts and Wallis, 2000), behavioral (Drewe, 1975), physiologic (e.g., Mathalon et al., 2003), and imaging studies (e.g., Fassbender et al., 2004, Garavan et al., 2002, Horn et al., 2003, Kelly et al., 2004, Konishi et al., 1999, Rubia et al., 2001). Convergent findings indicate that response inhibition is significantly mediated by prefrontal cortex (PFC) circuits (e.g., Mesulam, 2000), both dorsal (Braver et al., 2001, Miller and Cohen, 2001) and ventral. Of the latter, orbital frontal cortex (OFC) dysfunction has been long associated with behavioral disinhibition (e.g., Anderson et al., 1999, Rolls, 1996). An “acquired sociopathic” syndrome (e.g., impulsivity, context-dysproportionate aggression) can follow medial OFC (mOFC) damage (Bigelow, 1850, Paradiso et al., 1999, Price et al., 1990), prompting the hypothesis that mOFC dysfunction underlies behavioral dyscontrol in antisocial personality disorder (Damasio, 2000). Patients with personality disorders marked by behavioral disinhibition (e.g., borderline personality disorder) have been shown to behave similarly to OFC lesion patients on impulse control measures (Berlin et al., 2005). Neural networks combining OFC with key limbic structures (e.g., amygdala) have emerged as fundamental mediators of decision-making requiring cognitive–emotional integration (Bechara et al., 2000a).
The go/no-go task (Donders, 1868) has been variably adapted to neuropsychologically probe response inhibition. Go/no-go tasks involve execution or inhibition of a motor response, triggered by a go- or no-go stimulus, respectively. Demand to respond quickly creates a pre-potent response tone which must be inhibited when cued by a “no-go” stimulus. Although seemingly behaviorally simple, task performance involves multiple sub-processes, including stimulus discrimination, response selection, motor preparation, response inhibition, and error monitoring.
Imaging studies of go/no-go type response inhibition tasks have used a variety of methodologies (Rubia et al., 2001). Frontal regions variably identified include OFC, dorsolateral PFC (DLPFC), ventrolateral PFC (VLPFC), and anterior cingulate cortex (ACC) (e.g., Casey et al., 1997, De Zubicaray et al., 2000, Horn et al., 2003, Kawashima et al., 1996, Liddle et al., 2001, Watanabe et al., 2002). Non-frontal regions include parietal, temporal, and striatal (e.g., Garavan et al., 1999, Horn et al., 2003, Watanabe et al., 2002). Go/no-go imaging studies have been used to probe mechanisms of select psychiatric disorders. For example, Vollm et al. (2004) demonstrated aberrant neural activation patterns during go/no-go task performance by individuals with personality disorders marked by behavioral dyscontrol (e.g., antisocial personality disorder).
Inconsistencies of activation patterns across response inhibition studies have prompted some investigators to consider a “multi-domain model” of inhibitory control (Mostofsky et al., 2003) where different functions (e.g., motor versus cognitive) are at least in part regulated by different inhibitory mechanisms, correspondingly mediated by different brain regions. Accordingly, activation patterns associated with inhibitory control in part depend on the specific cognitive/behavioral inhibitory process invoked. In contrast, certain regional activations are conserved across response inhibition studies, implying relatively task-independent neural substrates of response inhibition (Wager et al., 2005). For example, Rubia et al. (2001) identified a middle-infero-mesial frontal and inferior parietal network. Others have demonstrated a supra-modal (i.e., stimulus sensory modality-independent) paralimbic/neocortical network (Laurens et al., 2005).
A successful behavioral repertoire involves not only balancing agency and inhibition, but healthy development of adaptive bi-directional modulation of emotion and cognitive control (Critchley, 2003, Davidson, 2000, Gehring and Willoughby, 2002, Lewis et al., 2006). Many psychiatric disorders involve behavioral dyscontrol which becomes prominent within certain emotional contexts (e.g., Posner et al., 2002). Integrative neurocognitive models are being applied to behavioral disorders characterized by dysfunctional interactions of emotion and behavioral control (Blair, 2005). Although the neurocircuitry of emotion has been long studied (e.g., Britton et al., 2006, Phan et al., 2002), mechanisms of the interaction of emotion and cognitive control are only recently being explored (Dolan, 2002, Zald et al., 2002), revealing complex neural interactions (e.g., Hare et al., 2005, Liberzon et al., 2000, Pessoa and Ungerleider, 2004, Phan et al., 2005, Shafritz et al., 2006), and disease-specific abnormalities (e.g., Elliott et al., 2004). Imaging studies probing neural mediation of emotion-modulated behavioral control use neuropsychological probes which join emotional manipulation with a behavioral control task (Yamasaki et al., 2002). For example, Elliott et al. (2000a) used a linguistic go/no-go task in which emotional valence was used to define stimulus targets, theorizing that observed neural activations reflected modulation of behavioral control by emotional valence. Others have employed similar tasks using emotional facial stimuli. Hare et al. (2005) observed slower response (interpreted as decreased approach) to negative facial targets and reduced inhibitory performance (interpreted as decreased avoidance) in response to positive non-target facial expressions. Shafritz et al. (2006) observed inferior frontal and insular activation associated with response inhibition during emotional facial expression go/no-go task performance, while a comparison non-emotional letter–symbol response inhibition task did not activate these regions. The investigators concluded that inhibition within an emotional context recruited neural substrates beyond those activated by non-emotional response inhibition (Shafritz et al., 2006).
When interpreting results of experimental designs that probe the interaction of emotion and behavioral control by using explicit emotion as motor response (“target”) signifier, it is relevant to note that the presumed emotional component of such a task may represent more a categorization (i.e. deciding if a stimulus is “happy” or “sad”) than an emotional task (Elliott et al., 2000a, Elliott et al., 2000b). Evidence suggests that explicit labeling of affect has distinct neural correlates (Crosson et al., 2002, Hariri et al., 2003, Teasdale et al., 1999, Taylor et al., 2003).
We developed an emotional linguistic go/no-go fMRI paradigm with a factorial block design to specifically investigate neural substrates of the interaction of emotional stimulus processing and inhibitory control in both normal subjects and patient populations. This paradigm introduces stimulus emotional valence incidentally relative to the explicit behavioral task demand (i.e., go or no-go). It is thought that such interaction better approximates that operating in many real-world socio-behavioral contexts. Sufficiency of incidental emotional stimuli to modulate neural activity is supported by diverse data (e.g., Isenberg et al., 1999, Perlstein et al., 2002, Whalen et al., 1998). Rather than requiring subjects explicitly use emotional stimulus content to guide performance on a behavioral task, this paradigm joins task demand (i.e., go or no-go) with concomitant task-unrelated emotional stimulus valence manipulation, to probe the modulation of the former by the latter.
Building on coalescing knowledge of the functional neuroanatomy of cognitive control, and accumulating evidence of the interaction of emotion and response inhibition, we hypothesized functional changes in prefrontal (e.g., mOFC) and anterior limbic (e.g., amygdala) sites reflecting the modulation of behavioral inhibitory control by emotional stimulus processing. Further, we hypothesized valence-distinct activity variations enabling identification of negative valence-specific neural substrates of the interaction of emotional stimulus processing and response inhibition.
Section snippets
Subjects
Fourteen healthy subjects (10 females; mean age 23.9 years, range 18–31; 12 right-handed (Edinburgh Handedness Inventory; Oldfield, 1971)) participated in the study. All were screened using the Structured Clinical Interview for DSM-IV (First et al., 1996); exclusion criteria included any psychiatric, neurological, or medical condition, either current or past. No subjects were taking psychotropic medications; two subjects were using oral contraceptives. All subjects met acceptable limits of head
Behavioral
Valence ratings (Fig. 2) revealed significant differences among negative, neutral, and positive word ratings (p < 0.001, respectively). Response times (Fig. 3) were significantly slower in no-go versus go blocks within negative and positive valence conditions (p < 0.01), with a trend toward such difference within neutral valence (p = 0.066). There were no significant valence-dependent differences in total mean response times (i.e., mean RT of all responses within all Go- and No-Go blocks of a
Discussion
Results overall indicate support for hypotheses regarding fronto-limbic activity changes associated with, and valence-distinct variation in the neural network engaged by, the interaction of emotional stimulus processing and behavioral inhibitory task demand. Because of its relevance to pathologic disturbances of the human behavioral repertoire, we contextualize results with special attention to findings revealed by those contrasts highlighting negative valence-specific behavioral inhibition,
Conclusion
Building on emerging functional neuroanatomic understanding of response inhibition, and complementing recent studies of emotional–cognitive interaction using explicit emotion, this study aimed to elucidate neural mechanisms by which incidental emotional stimulus processing modulates response inhibition. It is thought that such interaction better approximates that operating in many real-world socio-behavioral contexts. The conjunction of emotional stimulus processing with behavioral inhibitory
References (134)
- et al.
A global estimator unbiased by local changes
NeuroImage
(2001) - et al.
Neural development of selective attention and response inhibition
NeuroImage
(2003) - et al.
Direct comparison of prefrontal cortex regions engaged by working and long-term memory tasks
NeuroImage
(2001) - et al.
Cognitive and emotional influences in anterior cingulate cortex
Trends Cog. Sci.
(2000) Is schizophrenia the price that Homo sapiens pays for language?
Schizophr. Res.
(1997)Cognitive neuroscience needs affective neuroscience (and vice versa)
Brain Cogn.
(2000)- et al.
Removal of confounding effects of global signal in functional MRI analyses
NeuroImage
(2001) - et al.
Motor response suppression and the prepotent tendency to respond: a parametric fMRI study
Neuropsychologia
(2000) Go/no-go learning after frontal lobe lesions in humans
Cortex
(1975)- et al.
Abnormal ventral frontal response during performance of an affective go/no-go task in patients with mania
Biol. Psychiatry
(2004)