Experimental and methodological factors affecting test-retest reliability of amygdala BOLD responses

Amygdala fMRI-A Critical Appraisal of the Extant Literature

Even before the advent of fMRI, the amygdala occupied a central space in the affective neurosciences. Yet this amygdala-centred view on emotion processing gained even wider acceptance after the inception of fMRI in the early 1990s, a landmark that triggered a goldrush of fMRI studies targeting the amygdala in vivo. Initially, this amygdala fMRI research was mostly confined to task-activation studies measuring the magnitude of the amygdala's response to emotional stimuli. Later, interest began to shift more towards the study of the amygdala's resting-state functional connectivity and task-based psychophysiological interactions. Later still, the test-retest reliability of amygdala fMRI came under closer scrutiny, while at the same time, amygdala-based real-time fMRI neurofeedback gained widespread popularity. Each of these major subdomains of amygdala fMRI research has left its marks on the field of affective neuroscience at large. The purpose of this review is to provide a critical assessment of this literature. By integrating the insights garnered by these research branches, we aim to answer the question: What part (if any) can amygdala fMRI still play within the current landscape of affective neuroscience? Our findings show that serious questions can be raised with regard to both the reliability and validity of amygdala fMRI. These conclusions force us to cast doubt on the continued viability of amygdala fMRI as a core pilar of the affective neurosciences.

Hippocampus, amygdala, and insula activation in response to romantic relationship dissolution stimuli: A case-case-control fMRI study on emerging adult students

BACKGROUND

Romantic relationship dissolutions (RRDs) are associated with posttraumatic stress symptoms (PTSS). Functional magnetic resonance imaging in RRD studies indicate overlapping neural activation similar to posttraumatic stress disorder. These studies combine real and hypothetical rejection, and lack contextual information and control and/or comparison groups exposed to non-RRD or DSM-5 defined traumatic events.

AIM

We investigated blood oxygen level dependent (BOLD) activation in the hippocampus, amygdala, and insula of participants with RRDs compared with other traumatic or non-trauma stressors.

METHODS

Emerging adults (mean age = 21.54 years; female = 74.7 %) who experienced an RRD (n = 36), DSM-5 defined trauma (physical and/or sexual assault: n = 15), or a non-RRD or DSM-5 stressor (n = 28) completed PTSS, depression, childhood trauma, lifetime trauma exposure, and attachment measures. We used a general and customised version of the International Affective Picture System to investigate responses to index-trauma-related stimuli. We used mixed linear models to assess between-group differences, and ANOVAs and Spearman's correlations to analyse factors associated with BOLD activation.

RESULTS

BOLD activity increased between index-trauma stimuli as compared to neutral stimuli in the hippocampus and amygdala, with no significant difference between the DSM-5 Trauma and RRD groups. Childhood adversity, sexual orientation, and attachment style were associated with BOLD activation changes. Breakup characteristics (e.g., initiator status) were associated with increased BOLD activation in the hippocampus and amygdala, in the RRD group.

CONCLUSION

RRDs should be considered as potentially traumatic events. Breakup characteristics are risk factors for experiencing RRDs as traumatic.

LIMITATION

Future studies should consider more diverse representation across sex, ethnicity, and sexual orientation.

Neural correlates of individual differences in multimodal emotion recognition ability

Studies have reported substantial variability in emotion recognition ability (ERA) - an important social skill - but possible neural underpinnings for such individual differences are not well understood. This functional magnetic resonance imaging (fMRI) study investigated neural responses during emotion recognition in young adults (N = 49) who were selected for inclusion based on their performance (high or low) during previous testing of ERA. Participants were asked to judge brief video recordings in a forced-choice emotion recognition task, wherein stimuli were presented in visual, auditory and multimodal (audiovisual) blocks. Emotion recognition rates during brain scanning confirmed that individuals with high (vs low) ERA received higher accuracy for all presentation blocks. fMRI-analyses focused on key regions of interest (ROIs) involved in the processing of multimodal emotion expressions, based on previous meta-analyses. In neural response to emotional stimuli contrasted with neutral stimuli, individuals with high (vs low) ERA showed higher activation in the following ROIs during the multimodal condition: right middle superior temporal gyrus (mSTG), right posterior superior temporal sulcus (PSTS), and right inferior frontal cortex (IFC). Overall, results suggest that individual variability in ERA may be reflected across several stages of decisional processing, including extraction (mSTG), integration (PSTS) and evaluation (IFC) of emotional information.

Frontal-midline theta and posterior alpha oscillations index early processing of spatial representations during active navigation

Previous research has demonstrated that humans combine multiple sources of spatial information such as self-motion and landmark cues, while navigating through an environment. However, it is unclear whether this involves comparing multiple representations obtained from different sources during navigation (parallel hypothesis) or building a representation first based on self-motion cues and then combining with landmarks later (serial hypothesis). We tested these two hypotheses (parallel vs. serial) in an active navigation task using wireless mobile scalp EEG recordings. Participants walked through an immersive virtual hallway with or without conflicts between self-motion and landmarks (i.e., intersections) and pointed toward the starting position of the hallway. We employed the oscillatory signals recorded during mobile wireless scalp EEG as means of identifying when participant representations based on self-motion vs. landmark cues might have first emerged. We found that path segments, including intersections present early during navigation, were more strongly associated with later pointing error, regardless of when they appeared during encoding. We also found that there was sufficient information contained within the frontal-midline theta and posterior alpha oscillatory signals in the earliest segments of navigation involving intersections to decode condition (i.e., conflicting vs. not conflicting). Together, these findings suggest that intersections play a pivotal role in the early development of spatial representations, suggesting that memory representations for the geometry of walked paths likely develop early during navigation, in support of the parallel hypothesis.

Exaggerated amygdala response to threat and association with immune hyperactivity in depression

BACKGROUND

Depression is characterized by altered neurobiological responses to threat and inflammation may be involved in the development and maintenance of symptoms. However, the mechanistic pathways underlying the relationship between the neural underpinnings of threat, inflammation and depressive symptoms remain unknown.

METHODS

Twenty participants with major depressive disorder (MDD) and 17 healthy controls (HCs) completed this study. Peripheral blood mononuclear cells (PBMCs) were collected and stimulated ex vivo with lipopolysaccharide (LPS). We then measured a broad array of secreted proteins and performed principal component analysis to compute an aggregated immune reactivity score. Subjects completed a well-validated emotional face processing task during functional magnetic resonance imaging (fMRI). Amygdala activation was measured during perception of threat for the main contrast of interest: fear > happy face. Participants completed the Mood and Anxiety Symptom Questionnaire (MASQ) and the Perceived Stress Scale (PSS). Correlation analyses between amygdala activation, the aggregate immune score, and symptom were computed across groups. A mediation analysis was also performed across groups to further explore the relationship between these three variables.

RESULTS

In line with our hypotheses and with prior work, the MDD group showed greater amygdala activation in response to threat compared to the HC group [t35 = -2.038, p = 0.049]. Internal consistency of amygdala activation to threat was found to be moderate. Response to an ex vivo immune challenge was greater in MDD than HC based on the computed immune reactivity score (PC1; t35 = 2.674, p = 0.011). Amygdala activation was positively correlated with the immune score (r = 0.331, p = 0.045). Moreover, higher amygdala activation was associated with greater anxious arousal measured by the MASQ (r = 0.390, p = 0.017). Exploring the role of stress, we found that higher perceived stress was positively associated with both inflammatory response (r = 0.367, p = 0.026) and amygdala response to threat (r = 0.325, p = 0.050). Mediation analyses showed that perceived stress predicted anxious arousal, but neither inflammation nor amygdala activation fully accounted for the effect of perceived stress on anxious arousal.

CONCLUSION

These data highlight the potential importance of threat circuitry hyperactivation in MDD, consistent with prior reports. We found that higher levels of inflammatory biomarkers were associated with higher amygdala activation, which in turn was associated with anxious arousal. Future research utilizing larger sample sizes are needed to replicate these preliminary results.

Reliability and stability challenges in ABCD task fMRI data

Trait stability of measures is an essential requirement for individual differences research. Functional MRI has been increasingly used in studies that rely on the assumption of trait stability, such as attempts to relate task related brain activation to individual differences in behavior and psychopathology. However, recent research using adult samples has questioned the trait stability of task-fMRI measures, as assessed by test-retest correlations. To date, little is known about trait stability of task fMRI in children. Here, we examined within-session reliability and long-term stability of individual differences in task-fMRI measures using fMRI measures of brain activation provided by the adolescent brain cognitive development (ABCD) Study Release v4.0 as an individual's average regional activity, using its tasks focused on reward processing, response inhibition, and working memory. We also evaluated the effects of factors potentially affecting reliability and stability. Reliability and stability (quantified as the ratio of non-scanner related stable variance to all variances) was poor in virtually all brain regions, with an average value of 0.088 and 0.072 for short term (within-session) reliability and long-term (between-session) stability, respectively, in regions of interest (ROIs) historically-recruited by the tasks. Only one reliability or stability value in ROIs exceeded the 'poor' cut-off of 0.4, and in fact rarely exceeded 0.2 (only 4.9%). Motion had a pronounced effect on estimated reliability/stability, with the lowest motion quartile of participants having a mean reliability/stability 2.5 times higher (albeit still 'poor') than the highest motion quartile. Poor reliability and stability of task-fMRI, particularly in children, diminishes potential utility of fMRI data due to a drastic reduction of effect sizes and, consequently, statistical power for the detection of brain-behavior associations. This essential issue urgently needs to be addressed through optimization of task design, scanning parameters, data acquisition protocols, preprocessing pipelines, and data denoising methods.

Importance of test-retest reliability for promoting fMRI based screening and interventions in major depressive disorder

Proponents of personalized medicine have promoted neuroimaging in three areas of clinical application for major depression: clinical prediction, outcome evaluation, and treatment, via neurofeedback. Whereas psychometric considerations such as test-retest reliability are basic precursors to clinical adoption for most clinical instruments, we show, in this article, that basic psychometrics have not been regularly attended to in fMRI of depression. For instance, no fMRI neurofeedback study has included measures of test-retest reliability, despite the implicit assumption that brain signals are stable enough to train. We consider several factors that could be useful to aid clinical translation, including (1) attending to how the BOLD response is parameterized, (2) identifying and promoting regions or voxels with stronger psychometric properties, (3) accounting for within-individual changes (e.g., in symptomatology) across time, and (4) focusing on tasks and clinical populations that are relevant for the intended clinical application. We apply these principles to published prognostic and neurofeedback data sets. The broad implication of this work is that attention to psychometrics is important for clinical adoption of mechanistic assessment, is feasible, and may improve the underlying science.

Measurement of stretch-evoked brainstem function using fMRI

Knowledge on the organization of motor function in the reticulospinal tract (RST) is limited by the lack of methods for measuring RST function in humans. Behavioral studies suggest the involvement of the RST in long latency responses (LLRs). LLRs, elicited by precisely controlled perturbations, can therefore act as a viable paradigm to measure motor-related RST activity using functional Magnetic Resonance Imaging (fMRI). Here we present StretchfMRI, a novel technique developed to study RST function associated with LLRs. StretchfMRI combines robotic perturbations with electromyography and fMRI to simultaneously quantify muscular and neural activity during stretch-evoked LLRs without loss of reliability. Using StretchfMRI, we established the muscle-specific organization of LLR activity in the brainstem. The observed organization is partially consistent with animal models, with activity primarily in the ipsilateral medulla for flexors and in the contralateral pons for extensors, but also includes other areas, such as the midbrain and bilateral pontomedullary contributions.

Test-retest reliability of approach-avoidance conflict decision-making during functional magnetic resonance imaging in healthy adults

Neural and behavioral mechanisms during approach-avoidance conflict decision-making are relevant across various psychiatric disorders, particularly anxiety disorders. Studies using approach-avoidance conflict paradigms in healthy adults have identified preliminary neural mechanisms, but findings must be replicated and demonstrated as reliable before further application. This study sought to replicate previous findings and examine test-retest reliability of behavioral (approach behavior, reaction time) and neural (regions of interest [ROIs]) responses during an approach-avoidance conflict task conducted during functional magnetic resonance imaging (fMRI). Thirty healthy adults completed an approach-avoidance conflict task during fMRI on two occasions (mean interval: 17 days; range: 11-32). Effects of task condition during three task phases (decision-making, affective outcome and monetary reward) and intraclass correlation coefficients (ICCs) were calculated across time points. Results replicated that approach behavior was modulated by conflict during decision-making. ROI activations were replicated such that dorsal anterior cingulate cortex (dACC) was modulated by conflict during decision-making, and dACC, striatum, and anterior insula were modulated by valence during affective outcomes (p's <.0083). Approach behavior during conflict demonstrated excellent reliability (ICCs ≥.77). Activation of dACC during conflict decision-making and anterior insula during negative outcomes demonstrated fair reliability (ICCs = .51 and .54), and dACC and striatum activation demonstrated good reliability during negative outcomes (ICCs = .63 and .69). Two additional ROIs (amygdala, left dorsolateral prefrontal cortex) showed good reliability during negative outcomes (ICCs ≥.60). These results characterize several specific behavioral and neuroimaging responses that are replicable and sufficiently reliable during approach-avoidance conflict decision-making to support future utility.

Test-retest reliability of emotion regulation networks using fMRI at ultra-high magnetic field

Given the importance of emotion regulation in affective disorders, emotion regulation is at the focus of attempts to identify brain biomarkers of disease risk, treatment response, and brain development. However, to be useful as an indicator for individual characteristics of brain functions - particularly as a biomarker in a clinical context - ensuring reliability is a key challenge. Here, we systematically evaluated test-retest reliability of task-based functional magnetic resonance imaging (fMRI) activity within neural networks associated with emotion generation and regulation across three sessions. Acquiring fMRI data at ultra-high field (7T), we examined region- and voxel-wise test-retest reliability of brain activity in response to a well-established emotion regulation task for predefined region-of-interests (ROIs) implicated in four neural networks. Test-retest reliability varied considerably across the emotion regulation networks and respective ROIs. However, core emotion regulation regions, including the ventrolateral and dorsolateral prefrontal cortex (vlPFC and dlPFC) as well as the middle temporal gyrus (MTG) showed high reliability. Our findings thus support the role of these prefrontal and temporal regions as promising candidates for the study of individual differences in emotion regulation as well as for neurobiological biomarkers in clinical neuroscience research.

Novel polygenic risk score as a translational tool linking depression-related changes in the corticolimbic transcriptome with neural face processing and anhedonic symptoms

Convergent data from imaging and postmortem brain transcriptome studies implicate corticolimbic circuit (CLC) dysregulation in the pathophysiology of depression. To more directly bridge these lines of work, we generated a novel transcriptome-based polygenic risk score (T-PRS), capturing subtle shifts toward depression-like gene expression patterns in key CLC regions, and mapped this T-PRS onto brain function and related depressive symptoms in a nonclinical sample of 478 young adults (225 men; age 19.79 +/- 1.24) from the Duke Neurogenetics Study. First, T-PRS was generated based on common functional SNPs shifting CLC gene expression toward a depression-like state. Next, we used multivariate partial least squares regression to map T-PRS onto whole-brain activity patterns during perceptual processing of social stimuli (i.e., human faces). For validation, we conducted a comparative analysis with a PRS summarizing depression risk variants identified by the Psychiatric Genomics Consortium (PGC-PRS). Sex was modeled as moderating factor. We showed that T-PRS was associated with widespread reductions in neural response to neutral faces in women and to emotional faces and shapes in men (multivariate p < 0.01). This female-specific reductions in neural response to neutral faces was also associated with PGC-PRS (multivariate p < 0.03). Reduced reactivity to neutral faces was further associated with increased self-reported anhedonia. We conclude that women with functional alleles mimicking the postmortem transcriptomic CLC signature of depression have blunted neural activity to social stimuli, which may be expressed as higher anhedonia.

Visual cortical regions show sufficient test-retest reliability while salience regions are unreliable during emotional face processing

Functional magnetic resonance imaging studies frequently use emotional face processing tasks to probe neural circuitry related to psychiatric disorders and treatments with an emphasis on regions within the salience network (e.g., amygdala). Findings across previous test-retest reliability studies of emotional face processing have shown high variability, potentially due to differences in data analytic approaches. The present study comprehensively examined the test-retest reliability of an emotional faces task utilizing multiple approaches to region of interest (ROI) analysis and by examining voxel-wise reliability across the entire brain for both neural activation and functional connectivity. Analyses included 42 healthy adult participants who completed an fMRI scan concurrent with an emotional faces task on two separate days with an average of 25.52 days between scans. Intraclass correlation coefficients (ICCs) were calculated for the 'FACES-SHAPES' and 'FACES' (compared to implicit baseline) contrasts across the following: anatomical ROIs identified from a publicly available brain atlas (i.e., Brainnetome), functional ROIs consisting of 5-mm spheres centered on peak voxels from a publicly available meta-analytic database (i.e., Neurosynth), and whole-brain, voxel-wise analysis. Whole-brain, voxel-wise analyses of functional connectivity were also conducted using both anatomical and functional seed ROIs. While group-averaged neural activation maps were consistent across time, only one anatomical ROI and two functional ROIs showed good or excellent individual-level reliability for neural activation. The anatomical ROI was the right medioventral fusiform gyrus for the FACES contrast (ICC ​= ​0.60). The functional ROIs were the left and the right fusiform face area (FFA) for both FACES-SHAPES and FACES (Left FFA ICCs ​= ​0.69 & 0.79; Right FFA ICCs ​= ​0.68 & 0.66). Poor reliability (ICCs ​< ​0.4) was identified for almost all other anatomical and functional ROIs, with some exceptions showing fair reliability (ICCs ​= ​0.4-0.59). Whole-brain voxel-wise analysis of neural activation identified voxels with good (ICCs ​= ​0.6-0.74) to excellent reliability (ICCs ​> ​0.75) that were primarily located in visual cortex, with several clusters in bilateral dorsal lateral prefrontal cortex (DLPFC). Whole-brain voxel-wise analyses of functional connectivity for amygdala and fusiform gyrus identified very few voxels with good to excellent reliability using both anatomical and functional seed ROIs. Exceptions included clusters in right cerebellum and right DLPFC that showed reliable connectivity with left amygdala (ICCs ​> ​0.6). In conclusion, results indicate that visual cortical regions demonstrate good reliability at the individual level for neural activation, but reliability is generally poor for salience regions often focused on within psychiatric research (e.g., amygdala). Given these findings, future clinical neuroimaging studies using emotional faces tasks to examine individual differences might instead focus on visual regions and their role in psychiatric disorders.

Maternal overprotection in childhood is associated with amygdala reactivity and structural connectivity in adulthood

Recently, we reported that variability in early-life caregiving experiences maps onto individual differences in threat-related brain function. Here, we extend this work to provide further evidence that subtle variability in specific features of early caregiving shapes structural and functional connectivity between the amygdala and medial prefrontal cortex (mPFC) in a cohort of 312 young adult volunteers. Multiple regression analyses revealed that participants who reported higher maternal overprotection exhibited increased amygdala reactivity to explicit signals of interpersonal threat but not implicit signals of broad environmental threat. While amygdala functional connectivity with regulatory regions of the mPFC was not significantly associated with maternal overprotection, participants who reported higher maternal overprotection exhibited relatively decreased structural integrity of the uncinate fasciculus (UF), a white matter tract connecting these same brain regions. There were no significant associations between structural or functional brain measures and either maternal or paternal care ratings. These findings suggest that an overprotective maternal parenting style during childhood is associated with later functional and structural alterations of brain regions involved in generating and regulating responses to threat.

Self-rated amygdala activity: an auto-biological index of affective distress

Auto-biological beliefs—beliefs about one’s own biology—are an understudied component of personal identity. Research participants who are led to believe they are biologically vulnerable to affective disorders report more symptoms and less ability to control their mood; however, little is known about the impact of self-originating beliefs about risk for psychopathology, and whether such beliefs correspond to empirically derived estimates of actual vulnerability. Participants in a neuroimaging study (n = 1256) completed self-report measures of affective symptoms, perceived stress, and neuroticism, and an emotional face processing task in the scanner designed to elicit threat responses from the amygdala. A subsample (n = 63) additionally rated their own perceived neural response to threat (i.e., amygdala activity) compared to peers. Self-ratings of neural threat response were uncorrelated with actual threat-related amygdala activity measured via BOLD fMRI. However, self-ratings predicted subjective distress across a variety of self-report measures. In contrast, in the full sample, threat-related amygdala activity was uncorrelated with self-report measures of affective distress. These findings suggest that beliefs about one’s own biological threat response—while unrelated to measured neural activation—may be informative indicators of psychological functioning.

Acute alcohol administration dampens central extended amygdala reactivity

Alcohol use is common, imposes a staggering burden on public health, and often resists treatment. The central extended amygdala (EAc)—including the bed nucleus of the stria terminalis (BST) and the central nucleus of the amygdala (Ce)—plays a key role in prominent neuroscientific models of alcohol drinking, but the relevance of these regions to acute alcohol consumption in humans remains poorly understood. Using a single-blind, randomized-groups design, multiband fMRI data were acquired from 49 social drinkers while they performed a well-established emotional faces paradigm after consuming either alcohol or placebo. Relative to placebo, alcohol significantly dampened reactivity to emotional faces in the BST. To rigorously assess potential regional differences in activation, data were extracted from unbiased, anatomically predefined regions of interest. Analyses revealed similar levels of dampening in the BST and Ce. In short, alcohol transiently reduces reactivity to emotional faces and it does so similarly across the two major divisions of the human EAc. These observations reinforce the translational relevance of addiction models derived from preclinical work in rodents and provide new insights into the neural systems most relevant to the consumption of alcohol and to the initial development of alcohol abuse in humans.