Review articleConflict detection and resolution rely on a combination of common and distinct cognitive control networks
Introduction
Cognitive control is the ability to orchestrate thought and action in accordance with internal goals (Miller and Cohen, 2001). It has been conceptualized as a set of control functions that may include working memory, response selection, response inhibition, and task switching (Lenartowicz et al., 2010, Sabb et al., 2008). Its core system, the fronto-parietal network (FPN), meaningfully contributes to a variety of task contexts. The FPN allows rapid reconfiguration of information flow across multiple task-relevant brain networks, such as the visual network, auditory network, and default mode network (Cole et al., 2013). Alterations of this control system might contribute to a striking range of mental diseases (Cole et al., 2014). In the laboratory, various stimulus-response compatibility (SRC) tasks, such as the Stroop task (Stroop, 1935), the Eriksen flanker task (Gratton et al., 1992), and the Simon task (Simon and Small, 1969), have been employed to study cognitive control functionality. The SRC effect is the phenomenon in which performance is worse (i.e., slower and more erroneous) when mappings of stimuli to responses are incongruent than when they are congruent (Fitts and Seeger, 1953, Proctor and Vu, 2006).
Based on the distinct SRC tasks, several researchers have put forward brain network models of cognitive control from an attention perspective. Fan et al. (2005) proposed three separable anatomical networks related to the components of attention. The alerting network, the orienting network, and the executive control network activate the thalamic, parietal, and anterior cingulate cortex, respectively. Corbetta and Shulman (2002) identified two partially segregated attentional systems. The top-down system, which includes parts of the intraparietal cortex and superior frontal cortex, is involved in preparing and applying goal-directed selection. The bottom-up system, which includes the temporoparietal cortex and inferior frontal cortex, is specialized for the detection of behaviorally relevant, salient or unexpected stimuli.
Different from two attention networks involved in cognitive control, Botvinick et al. (2001) proposed the conflict-monitoring (CM) model of cognitive control. This model describes a single, “all-purpose” conflict-control loop that can be recruited to generally handle different types of conflicting representations; the loop comprises the anterior cingulate cortex (ACC) for conflict detection and the prefrontal cortex for executive control (Botvinick et al., 2001). According to the CM model, many types of conflicts will yield highly similar patterns of brain activation because they share a centralized module of cognitive control. The expanded parallel distributed processing (PDP) model further suggests that the mechanisms of cognitive control are adaptive and self-regulating (Botvinick and Cohen, 2014). The model proposes that the anterior cingulate cortex implements the conflict monitoring by modulating the activity of control representations, and dopamine assists the adaptive gating by regulating the updating of control representations in the prefrontal cortex. However, according to the dimensional overlap (DO) model proposed by Kornblum et al., SRC effects can occur independently when at least two of three dimensions (task-relevant stimulus dimension, task-irrelevant stimulus dimension, and response dimension) overlap (Kornblum, 1994). In a typical Stroop task, the SRC effect involves stimulus-based processing (S-S conflict) as the conflict stems from incongruence between task-relevant (SR, e.g., ink color) and task-irrelevant (SI, e.g., word meaning) stimulus features (Egner et al., 2007, Liu et al., 2010). In a typical Simon task, the SRC effect involves response-based processing (S-R conflict) as the conflict results from incongruence between a task-irrelevant stimulus feature (SI, e.g., the location of the stimuli) and a response feature (R, e.g., button press) (Egner et al., 2007). Under this definition, S-S and S-R conflicts belong to distinct DO types and are resolved by distinct control mechanisms. Supporting the DO model, the domain-specific model further proposes that specific conflict-control loops are involved in processing S-S and S-R conflicts (Egner, 2008). The model suggests that the SRC effects that stem from S-S and S-R conflicts uniquely activate specific brain regions because distinct cognitive control mechanisms are engaged in parallel by S-S and S-R conflicts.
Functional magnetic resonance imaging (fMRI) studies might provide insight into these theoretical debates and issues because they have the potential to demonstrate whether S-S and S-R conflict processing engage common or distinct brain mechanisms. Previous fMRI studies manipulating S-S and S-R conflicts have found that the brain uses distinct but parallel cognitive control mechanisms to resolve these different forms of cognitive interference (Egner et al., 2007, Liston et al., 2006, van Veen and Carter, 2005). In contrast, some studies have found that although the specific brain activation patterns are not identical across conflict domains, S-S and S-R conflicts share a common neural mechanism of attentional control and top-down modulation (Fan et al., 2003, Jiang and Egner, 2014, Kim et al., 2010, Kim et al., 2011, Liu et al., 2004, Milham et al., 2001). Some studies have even found completely overlapping activations across conflict domains (Peterson et al., 2002).
The differing results in conjunction with confounding factors make it difficult to obtain a clear understanding of the conflict-control processes in the human brain. First, the heterogeneity of the results is partly due to diverse experimental paradigms developed by various research groups that have aimed to address different aspects of cognitive control, such as motivation (Soutschek et al., 2014), attentional switching (Kim et al., 2012), and anticipatory control processes (Aarts et al., 2008). Second, it is unknown whether activation patterns reflect information processing relevant to the cognitive control process itself or serve incidental functions. Although an fMRI study harnessed multivoxel pattern analysis (MVPA) decoding S-S and S-R conflicts to overcome these limitations of traditional studies and showed a hybrid architecture of conflict processing entailing both domain-specific and domain-general components (Jiang and Egner, 2014), a single study is unlikely to provide decisive results regarding cognitive control processing.
Therefore, it is crucial to pool prior studies together and examine the core common and distinct conflict-processing networks in the human brain by combining theory-driven and data-driven approaches. One method of meta-analysis, activation likelihood estimation (ALE) (Turkeltaub et al., 2002), allows statistically verifiable concurrence across functional neuroimaging studies, revealing regions with the highest “likelihood” of activation, i.e., regions in which concurrence is highest.
The main goal of the current study is to assess whether cognitive control mechanisms underlying DO conflicts are general or distinct by performing a meta-analysis of the results of 111 recent neuroimaging studies. Three different patterns of results that relate to different cognitive control models are possible. 1) Domain-general activation. According to the CM model and expanded PDP model, which initially insisted on an all-purpose control module, cognitive control areas associated with S-S and S-R conflict processing would be activated completely consistently. 2) Domain-specific activation. Based on the DO model and the domain-specific model, S-S and S-R conflicts would show separate neural activation patterns because of their conflict-specific processing strategies. 3) Mixed activation. However, considering the inefficiency of conflict processing by a unitary control process and the impossibility of endless control mechanisms for each potential source of conflict, the combination of domain-general and domain-specific models is a more reasonable explanation. Specifically, we expected a hybrid neural architecture of conflict-control involving both specific and general brain areas to process S-S and S-R conflicts.
Section snippets
Study identification
Four independent researchers conducted a thorough search of the literature for fMRI studies examining S-S and S-R conflict processing in humans. The terms used to search the online citation indexing service PUBMED (through July 2017) were “fMRI” and “Stroop/Flanker/SNARC/Simon/Navon/Global-Local” by the first researcher, and “functional magnetic/resonance imaging/fMRI” in the abstract and “Stroop/Flanker/SNARC/Simon” in all fields by the second researcher. The terms used to search the online
Results
The all-inclusive analysis of 141 experiments showed significant activation of a large cluster, including the left supplementary motor area (SMA), right ACC, bilateral inferior frontal cortex/dorso lateral prefrontal cortex (IFC/DLPFC), bilateral inferior parietal cortex (IPC), bilateral superior parietal cortex (SPC), bilateral insula, bilateral thalamus, and right caudate (see Fig. 2 and Table 1).
S-S conflicts activated a subset of the aforementioned networks, including the bilateral
Discussion
A fundamental challenge to goal-directed behavior is that multitudes of stimuli compete for control over our actions. Conflict control, i.e., the ability to resolve this competition, is consistent with an organism’s current goals. From a theory-driven perspective, we discovered common neural networks in S-S and S-R conflict processing, including the left SMA/dorsal ACC, bilateral insula, and right SPC. In addition, distinct neural substrates and specific functions subserved the processing of
Contributions
XL designed and supervised the study. QL and GY made equal contributions to this study, performing the literature search, data extraction and organization and preparing the manuscript. MWC revised the manuscript and provided technical support for data visualization. ZL and YQ participated in performing the literature search and data extraction.
Conflict of interest
None of the authors has a conflict of interest to declare.
Acknowledgments
We thank Carrisa Cocuzza from Rutgers University for help with language editing and for proofreading the manuscript, as well as Kai Wang, Weizhi Nan, and Di Fu for conducting literature searches and screenings, as well as data input. We thank the two anonymous referees for their comments, which significantly enhanced the revised version of this paper. This research was supported by the National Natural Science Foundation of China [Grants 31070987, 31571161, 31200782, and 61621136008/DFG TRR-169
References (81)
- et al.
Spatial normalization of brain images with focal lesions using cost function masking
Neuroimage
(2001) - et al.
Interference and facilitation effects during selective attention: an H215O PET study of Stroop task performance
Neuroimage
(1995) - et al.
Processing conflicting information: facilitation, interference, and functional connectivity
Neuropsychologia
(2008) - et al.
Differential activation of right superior parietal cortex and intraparietal sulcus by spatial and nonspatial attention
Neuroimage
(1998) - et al.
Striatal and thalamic GABA level concentrations play differential roles for the modulation of response selection processes by proprioceptive information
Neuroimage
(2015) - et al.
A core system for the implementation of task sets
Neuron
(2006) - et al.
Separate conflict-specific cognitive control mechanisms in the human brain
Neuroimage
(2007) Multiple conflict-driven control mechanisms in the human brain
Trends Cogn. Sci.
(2008)- et al.
Co-activation patterns distinguish cortical modules, their connectivity and functional differentiation
Neuroimage
(2011) - et al.
Activation likelihood estimation meta-analysis revisited
Neuroimage
(2012)
Behavior, sensitivity, and power of activation likelihood estimation characterized by massive empirical simulation
Neuroimage
Cognitive and brain consequences of conflict
Neuroimage
The activation of attentional networks
Neuroimage
Contrasting network and modular perspectives on inhibitory control
Trends Cogn. Sci.
Multiple cognitive control mechanisms associated with the nature of conflict
Neurosci. Lett.
Common and distinct neural mechanisms of attentional switching and response conflict
Brain Res.
Effect of neurofeedback training on the neural substrates of selective attention in children with attention-deficit/hyperactivity disorder: a functional magnetic resonance imaging study
Neurosci. Lett.
Neural correlates of the reverse Simon effect in the Hedge and Marsh task
Neuropsychologia
Anterior cingulate and posterior parietal cortices are sensitive to dissociable forms of conflict in a task-switching paradigm
Neuron
Common and distinct neural substrates of attentional control in an integrated Simon and spatial Stroop task as assessed by event-related fMRI
Neuroimage
The relative involvement of anterior cingulate and prefrontal cortex in attentional control depends on nature of conflict
Brain Res. Cogn. Brain Res.
Attentional control in the aging brain: insights from an fMRI study of the stroop task
Brain Cogn.
The BOLD response during Stroop task-like inhibition paradigms: effects of task difficulty and task-relevant modality
Brain Cogn.
The unity and diversity of executive functions and their contributions to complex frontal lobe tasks: a latent variable analysis
Cogn Psychol
An event-related functional MRI study comparing interference effects in the Simon and Stroop tasks
Brain Res. Cogn. Brain Res.
Subcortical processes of motor response inhibition during a stop signal task
Neuroimage
Switching attention and resolving interference: fMRI measures of executive functions
Neuropsychologia
Meta-analysis of the functional neuroanatomy of single-word reading: method and validation
Neuroimage
Right-hemispheric cortical contributions to language ability in healthy adults
Brain Lang.
Common and unique components of response inhibition revealed by fMRI
Neuroimage
The neural correlates of perceptual load induced attentional selection: an fMRI study
Neuroscience
Separating semantic conflict and response conflict in the Stroop task: a functional MRI study
Neuroimage
Anticipatory activity in anterio cingulate cortex can be independent of conflict and error likelihood
J. Neurosci.
Brain regions underlying response inhibition and interference monitoring and suppression
Eur. J. Neurosci.
Salience network integrity predicts default mode network function after traumatic brain injury
Proc. Natl. Acad. Sci. U. S. A.
The computational and neural basis of cognitive control: charted territory and new frontiers
Cogn. Sci.
Conflict monitoring and cognitive control
Psychol. Rev.
The problem of functional localization in the human brain
Nat. Rev.
Right anterior cingulate cortex infarction and transient speech aspontaneity
Arch. Neurol.
Distinct neural correlates for resolving stroop conflict at inhibited and noninhibited locations in inhibition of return
J. Cogn. Neurosci.
Cited by (47)
Cross-modal conflict deficit in children with attention-deficit/hyperactivity disorder
2024, Journal of Experimental Child PsychologyEffective connectivity analysis reveals the time course of the Stroop effect in manual responding
2023, Biological PsychologyHearts, flowers, and fruits: All children need to reveal their post-error slowing
2023, Journal of Experimental Child Psychology“Antagonistic” cooperation of control regions in bilingual language production: An effective connectivity study
2022, NeuropsychologiaCitation Excerpt :Another view, however, believes that the DLPFC and ACC cooperate with each other in a different way. The ACC monitors cognitive conflicts and sends signals to the DLPFC, while the DLPFC resolves conflicts and adjusts behaviours (Heidlmayr et al., 2020; Kerns et al., 2004; Li et al., 2017; Pestalozzi et al., 2020). Hence, the roles of the ACC and DLPFC in bilingual language control are likely conflict monitoring and resolution, respectively (Abutalebi et al., 2012; Abutalebi and Green, 2016; De Bruin et al., 2014; Reverberi et al., 2018; J. Wu et al., 2019).
Neural biomarkers of suicidal behavior: From cognition and circuits to cells (and back)
2022, Biomarkers in Bipolar DisordersAffective Dysregulation in Children Is Associated With Difficulties in Response Control in Emotional Ambiguous Situations
2022, Biological Psychiatry: Cognitive Neuroscience and NeuroimagingCitation Excerpt :In TD controls, as opposed to children with AD, modulations of activity in the right middle frontal gyrus (BA 10), the right inferior/orbitofrontal gyrus (BA 47), the left inferior parietal cortex (BA 40), and the precuneus (BA 7) are associated with the differential C-cluster amplitude modulations between congruent and incongruent trials. Meta-analytic imaging data show that conflicts in Stroop tasks, requiring the suppression of irrelevant semantic information (43,67), modulate activity in middle and inferior frontal regions, as well as in the anterior cingulate cortex (43). Imaging findings in AD suggest that the MFC, the amygdala, the nucleus accumbens, and the orbitofrontal cortex are dysfunctional (1,14,15); the current data show that these regions—particularly the orbitofrontal regions—are important to resolving cognitive-emotional conflicts in TD participants.
- 1
These authors contributed equally.