Choong-Wan Woo

Cluster-extent based thresholding in fMRI analyses: Pitfalls and recommendations

Cluster-extent based thresholding is currently the most popular method for multiple comparisons correction of statistical maps in neuroimaging studies, due to its high sensitivity to weak and diffuse signals. However, cluster-extent based thresholding provides low spatial specificity; researchers can only infer that there is signal somewhere within a significant cluster and cannot make inferences about the statistical significance of specific locations within the cluster. This poses a particular problem when one uses a liberal cluster-defining primary threshold (i.e., higher p-values), which often produces large clusters spanning multiple anatomical regions. In such cases, it is impossible to reliably infer which anatomical regions show true effects. From a survey of 814 functional magnetic resonance imaging (fMRI) studies published in 2010 and 2011, we show that the use of liberal primary thresholds (e.g., p<.01) is endemic, and that the largest determinant of the primary threshold level is the default option in the software used. We illustrate the problems with liberal primary thresholds using an fMRI dataset from our laboratory (N=33), and present simulations demonstrating the detrimental effects of liberal primary thresholds on false positives, localization, and interpretation of fMRI findings. To avoid these pitfalls, we recommend several analysis and reporting procedures, including 1) setting primary p<.001 as a default lower limit; 2) using more stringent primary thresholds or voxel-wise correction methods for highly powered studies; and 3) adopting reporting practices that make the level of spatial precision transparent to readers. We also suggest alternative and supplementary analysis methods.

Distinct Brain Systems Mediate the Effects of Nociceptive Input and Self-Regulation on Pain

Two distinct parallel neural systems independently contribute to our overall experience of pain – separately modulated by noxious input and by cognitive self-regulation.

Somatic and vicarious pain are represented by dissociable multivariate brain patterns

Understanding how humans represent others’ pain is critical for understanding pro-social behavior. ‘Shared experience’ theories propose common brain representations for somatic and vicarious pain, but other evidence suggests that specialized circuits are required to experience others’ suffering. Combining functional neuroimaging with multivariate pattern analyses, we identified dissociable patterns that predicted somatic (high versus low: 100%) and vicarious (high versus low: 100%) pain intensity in out-of-sample individuals. Critically, each pattern was at chance in predicting the other experience, demonstrating separate modifiability of both patterns. Somatotopy (upper versus lower limb: 93% accuracy for both conditions) was also distinct, located in somatosensory versus mentalizing-related circuits for somatic and vicarious pain, respectively. Two additional studies demonstrated the generalizability of the somatic pain pattern (which was originally developed on thermal pain) to mechanical and electrical pain, and also demonstrated the replicability of the somatic/vicarious dissociation. These findings suggest possible mechanisms underlying limitations in feeling others’ pain, and present new, more specific, brain targets for studying pain empathy. DOI: http://dx.doi.org/10.7554/eLife.15166.001

Towards a neurophysiological signature for fibromyalgia

Abstract Patients with fibromyalgia (FM) show characteristically enhanced unpleasantness to painful and nonpainful sensations accompanied by altered neural responses. The diagnostic potential of such neural alterations, including their sensitivity and specificity to FM (vs healthy controls) is unknown. We identify a brain signature that characterizes FM central pathophysiology at the neural systems level. We included 37 patients with FM and 35 matched healthy controls, and analyzed functional magnetic resonance imaging responses to (1) painful pressure and (2) nonpainful multisensory (visual–auditory–tactile) stimulation. We used machine-learning techniques to identify a brain-based FM signature. When exposed to the same painful stimuli, patients with FM showed greater neurologic pain signature (NPS; Wager et al., 2013. An fMRI-based neurologic signature of physical pain. N Engl J Med 2013;368:1388–97) responses. In addition, a new pain-related classifier (“FM-pain”) revealed augmented responses in sensory integration (insula/operculum) and self-referential (eg, medial prefrontal) regions in FM and reduced responses in the lateral frontal cortex. A “multisensory” classifier trained on nonpainful sensory stimulation revealed augmented responses in the insula/operculum, posterior cingulate, and medial prefrontal regions and reduced responses in the primary/secondary sensory cortices, basal ganglia, and cerebellum. Combined activity in the NPS, FM pain, and multisensory patterns classified patients vs controls with 92% sensitivity and 94% specificity in out-of-sample individuals. Enhanced NPS responses partly mediated mechanical hypersensitivity and correlated with depression and disability (Puncorrected < 0.05); FM-pain and multisensory responses correlated with clinical pain (Puncorrected < 0.05). The study provides initial characterization of individual patients with FM based on pathophysiological, symptom-related brain features. If replicated, these brain features may constitute objective neural targets for therapeutic interventions. The results establish a framework for assessing therapeutic mechanisms and predicting treatment response at the individual level.

Neuroimaging-based biomarker discovery and validation

Though developing biological markers for chronic pain has been a major goal of the field for decades, such biomarkers have not yet made their way into clinical practice. However, given the potential uses of biomarkers in multiple aspects of prevention and treatment—such as pain and risk factor assessment, diagnosis, prognosis, treatment selection, drug discovery, and more—efforts to discover new pain biomarkers have been expanding [5; 6; 8; 30]. Recent advances in human neuroimaging, including functional and structural Magnetic Resonance Imaging (fMRI/sMRI) combined with machine learning techniques, are bringing us closer to the goal of developing objective, brain-based markers of the neural functions and neuropathology that underlie chronic pain [2; 7; 25; 33]. These brain measures are particularly promising as biomarkers for chronic pain. Though pain is reliably induced by peripheral nociceptive input, many forms of chronic pain may arise from neuropathology in the supra-spinal circuits that govern the construction of pain experience and long-term motivation [1; 14; 26; 32]. Particularly, structural neuroimaging measures could provide more stable markers of neuropathology of chronic pain, including stable features underlying pain risk and resilience [2; 3; 11; 19; 28; 29]. Gray-matter changes have also been associated with a number of conditions that are often co-morbid with chronic pain, including depression [4; 22; 24], stress [10; 12; 20], post-traumatic stress disorder [17; 21; 27], and early-life adversity [13; 18; 23; 31]. Therefore, structural measures may provide important clues about supra-spinal contributions to both pain and related risk factors (Fig. 1). Figure 1 Key common brain regions that show structural changes across different conditions related to chronic pain, including depression, stress, post-traumatic stress disorder (PTSD), and early-life adversity. In this issue, Labus et al. [16] developed a new neuroimaging biomarker for irritable bowel syndrome (IBS) using structural MRI data, based on a relatively large sample of 80 IBS patients and 80 healthy controls. They used sparse Partial Least Squares-Discriminant Analysis (sPLS-DA), a method that allowed them to both develop a classification model based on brain structure and identify the regions that make the most important contributions to the classification. They subsequently tested the predictive model on a “holdout” sample of 26 IBS patients and 26 healthy controls. The model discriminated patients from controls with 70% accuracy (compared to a chance accuracy of 50%), providing a moderate but reliable morphological brain signature for IBS. Rather than being the end of the story, this study serves as a starting point for biomarker discovery and validation. Like other brain ‘signatures’ [30], the signature they identified can become a ‘research product’ that can be tested on multiple samples from different laboratories, and validated or challenged in various ways. The more the marker for IBS status or IBS risk holds up to the scrutiny of being characterized across samples and populations, the more useful it will become. Importantly, there is a set of desirable characteristics that a useful neuroimaging biomarker should demonstrate throughout the biomarker development process. We briefly describe several such characteristics (summarized in Table 1), and then relate them to the findings of Labus et al. [16]. Table 1 Desirable characteristics of neuroimaging biomarkers

Influence of dorsolateral prefrontal cortex and ventral striatum on risk avoidance in addiction: a mediation analysis

BACKGROUND Alterations in frontal and striatal function are hypothesized to underlie risky decision making in drug users, but how these regions interact to affect behavior is incompletely understood. We used mediation analysis to investigate how prefrontal cortex and ventral striatum together influence risk avoidance in abstinent drug users. METHOD Thirty-seven abstinent substance-dependent individuals (SDI) and 43 controls underwent fMRI while performing a decision-making task involving risk and reward. Analyses of a priori regions-of-interest tested whether activity in dorsolateral prefrontal cortex (DLPFC) and ventral striatum (VST) explained group differences in risk avoidance. Whole-brain analysis was conducted to identify brain regions influencing the negative VST-risk avoidance relationship. RESULTS Right DLPFC (RDLPFC) positively mediated the group-risk avoidance relationship (p < 0.05); RDLPFC activity was higher in SDI and predicted higher risk avoidance across groups, controlling for SDI vs. CONTROLS Conversely, VST activity negatively influenced risk avoidance (p < 0.05); it was higher in SDI, and predicted lower risk avoidance. Whole-brain analysis revealed that, across group, RDLPFC and left temporal-parietal junction positively (p ≤ 0.001) while right thalamus and left middle frontal gyrus negatively (p < 0.005) mediated the VST activity-risk avoidance relationship. CONCLUSION RDLPFC activity mediated less risky decision making while VST mediated more risky decision making across drug users and controls. These results suggest a dual pathway underlying decision making, which, if imbalanced, may adversely influence choices involving risk. Modeling contributions of multiple brain systems to behavior through mediation analysis could lead to a better understanding of mechanisms of behavior and suggest neuromodulatory treatments for addiction.

What reliability can and cannot tell us about pain report and pain neuroimaging.

In this issue of PAIN , Letzen et al. examine the test–retest reliability of functional magnetic resonance imaging (fMRI) connectivity (fcMRI) measures and self-reported pain during pain stimulation and show that fcMRI measures are less reliable than self-reported pain. This study is a valuable endeavor, and as we move toward developing and using neuroimaging-based biomarkers for clinical purposes (eg, making predictions and decisions about an individual), efforts to establish reliability and reproducibility of candidate biomarkers will become more and more critical. Based on the results, Letzen et al. concluded, and we agree, that fMRI measures are noisier than self-reported pain. This might be a big issue if we view fMRI measures as a substitute for pain ratings. However, the real value of using brain measures to study pain is not in replacing pain ratings but in serving to (1) provide a better understanding of how the brain generates and regulates pain and (2) provide ways to see and measure its component neurobiological processes. The reason that we need brain markers for pain is not that pain ratings are “flawed” in their reliability or that they are “flawed” at all, but that pain ratings reflect a complex mix of brain and psychological processes. For example, one person can report more pain than another because of differences in nociception, emotion, decision making, self-awareness, social cognition, and communicative tendencies. Because there is no single process that causes people to report more or less pain, self-reported pain provides only limited clues on what the underlying causes and what the best course of treatment might be. Many other disorders are similarly heterogeneous, and symptoms alone have not proven to be sufficient to guide effective treatment. In cancer, for example, diagnosis and treatment have progressively shifted from symptoms and overt signs to molecular subtypes that respond to tailored molecular treatments. Reliability is an important measurement property, though it is just one piece of the puzzle. Principally, reliability places constraints on the utility of a measure for assessing individual differences. However, those constraints are more subtle than it first appears. Below, we briefly elaborate on what reliability is and what constraints it does and does not place on the use of fMRI in assessment and personalized medicine. 1. Reliability: more is not always better

Multiple faces of pain: effects of chronic pain on the brain regulation of facial expression

Abstract Pain behaviors are shaped by social demands and learning processes, and chronic pain has been previously suggested to affect their meaning. In this study, we combined functional magnetic resonance imaging with in-scanner video recording during thermal pain stimulations and use multilevel mediation analyses to study the brain mediators of pain facial expressions and the perception of pain intensity (self-reports) in healthy individuals and patients with chronic back pain (CBP). Behavioral data showed that the relation between pain expression and pain report was disrupted in CBP. In both patients with CBP and healthy controls, brain activity varying on a trial-by-trial basis with pain facial expressions was mainly located in the primary motor cortex and completely dissociated from the pattern of brain activity varying with pain intensity ratings. Stronger activity was observed in CBP specifically during pain facial expressions in several nonmotor brain regions such as the medial prefrontal cortex, the precuneus, and the medial temporal lobe. In sharp contrast, no moderating effect of chronic pain was observed on brain activity associated with pain intensity ratings. Our results demonstrate that pain facial expressions and pain intensity ratings reflect different aspects of pain processing and support psychosocial models of pain suggesting that distinctive mechanisms are involved in the regulation of pain behaviors in chronic pain.

fMRI in analgesic drug discovery

Functional magnetic resonance imaging offers an unprecedented opportunity to evaluate and compare drug effects on human brain activity and to provide a systems-level prediction of how drugs for chronic pain affect the brain, thus accelerating drug discovery and repurposing. Functional magnetic resonance imaging offers an unprecedented opportunity to evaluate and compare drug effects on human brain activity and to provide systems-level predictions for how new drugs for chronic pain will affect the brain, thus accelerating drug discovery and repurposing (Duff et al., this issue).

Cognitive self-regulation influences pain-related physiology

Cognitive self-regulation can shape pain experience, but little is known about whether it affects autonomic responses to painful events. In this study, participants (N = 41) deployed a cognitive strategy based on reappraisal and imagination to regulate pain up or down on different trials while skin conductance responses (SCR) and electrocardiogram (ECG) activity were recorded. Using a machine learning approach, we developed stimulus-locked SCR and ECG physiological markers predictive of pain ratings. The markers demonstrated high sensitivity when predicting pain ratings, r = 0.55-0.83. In an independent dataset (N = 84), they discriminated different levels of painful heat with 74-93% accuracy and showed some specificity relative to discriminating levels of vicarious pain (50-71% accuracy; chance is 50%). Cognitive self-regulation increased and decreased both pain ratings and physiology in accordance with regulatory goals. These findings suggest that self-regulation can impact autonomic nervous system responses to painful stimuli and provide pain-selective autonomic profiles for future studies. Author Summary It is well known that cognitive self-regulation can modulate pain perception in humans, yet its physiological consequences are difficult to quantify, as the pain- and task-related physiological responses are intricately intertwined. Here, we developed physiological markers predictive of pain report from skin conductance response (SCR) and electrocardiogram (ECG) data collected from 41 participants while they experienced painful thermal stimulations without regulating their pain. These markers were validated on an independent dataset, and then tested for effects of reappraisal-based cognitive self-regulation. When participants were instructed to use this strategy to increase the amount of pain they experienced, expression of the pain-predictive physiological markers increased, and when participants were instructed to reduce the amount of pain they felt, expression of the physiological markers decreased. These results demonstrate that cognitive pain regulation using a conscious, reappraisal-based strategy not only impacts the way participants report pain, but also the way their autonomic physiology responds to pain.