Costafreda SG, Brammer MJ, David AS, Fu CH. Predictors of amygdala activation during the processing of emotional stimuli: a meta-analysis of 385 PET and fMRI studies. Brain Res Rev 58: 57-70

Institute of Psychiatry, King's College London, UK.
Brain Research Reviews (Impact Factor: 5.93). 07/2008; 58(1):57-70. DOI: 10.1016/j.brainresrev.2007.10.012
Source: PubMed


Although amygdala activity has been purported to be modulated by affective and non-affective factors, considerable controversy remains on its precise functional nature. We conducted a meta-analysis of 385 functional neuroimaging studies of emotional processing, examining the effects of experimental characteristics on the probability of detecting amygdala activity. All emotional stimuli were associated with higher probability of amygdala activity than neutral stimuli. Comparable effects were observed for most negative and positive emotions, however there was a higher probability of activation for fear and disgust relative to happiness. The level of attentional processing affected amygdala activity, as passive processing was associated with a higher probability of activation than active task instructions. Gustatory-olfactory and visual stimulus modalities increased the probability of activation relative to internal stimuli. Aversive learning increased the probability of amygdala activation as well. There was some evidence of hemispheric specialization with a relative left-lateralization for stimuli containing language and a relative right-lateralization for masked stimuli. Methodological variables, such as type of analysis and magnet strength, were also independent predictors of amygdala activation.

Download full-text


Available from: Cynthia H Y Fu, Oct 04, 2015
402 Reads
  • Source
    • "No specific behavioral responses were required during scanning because of the inclusion of subliminal presentations and our aim to isolate activation elicited by emotion stimuli independent of behavioral task demands. A large meta-analysis of 385 studies has shown passive processing is associated with a higher probability of activation than an active task (Costafreda et al, 2008). Because stimuli were presented via goggles, stimuli filled participant's field of view. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Although the cost of poor treatment outcomes of depression is staggering we do not yet have clinically useful methods for selecting the most effective antidepressant for each depressed person. Emotional brain activation is altered in MDD and implicated in treatment response. Identifying which aspects of emotional brain activation are predictive of general and specific responses to antidepressants may help clinicians and patients when making treatment decisions. We examined whether amygdala activation probed by emotion stimuli is a general or differential predictor of response to three commonly prescribed antidepressants, using functional magnetic resonance imaging (fMRI). A test-retest design was used to assess patients with major depressive disorder (MDD) in an academic setting as part of the International Study to Predict Optimized Treatment in Depression. 80 MDD outpatients were scanned pre-treatment and 8 weeks after randomization to the selective serotonin reuptake inhibitors escitalopram and sertraline and the serotonin- norepinephrine reuptake inhibitor, venlafaxine-extended release (XR). 34 matched controls were scanned at the same timepoints. We quantified the blood oxygen level–dependent signal of the amygdala during subliminal and supraliminal viewing of facial expressions of emotion. Response to treatment was defined by >=50% symptom improvement on the 17-item Hamilton Depression Rating Scale. Pre-treatment amygdala hypo-reactivity to subliminal happy and threat was a general predictor of treatment response, regardless of medication type (cohen’s d effect size .63 to .77; classification accuracy, 75%). Responders showed hypo-reactivity compared to controls at baseline, and an increase toward “normalization” post-treatment. Pre-treatment amygdala reactivity to subliminal sadness was a differential moderator of non-response to venlafaxine-XR (cohen’s d effect size 1.5; classification accuracy, 81%). Non-responders to venlafaxine-XR showed pre-treatment hyper-reactivity which progressed to hypo-reactivity rather than normalization post-treatment and hypo-reactivity post-treatment was abnormal compared to controls. Impaired amygdala activation has not previously been highlighted in the general versus differential prediction of antidepressant outcomes. Amygdala hypo-reactivity to emotions signaling reward and threat predicts the general capacity to respond to antidepressants. Amygdala hyper-reactivity to sad emotion is involved in a specific non-response to a serotonin- norepinephrine reuptake inhibitor. The findings suggest amygdala probes may help inform the personal selection of antidepressant treatments.
    Neuropsychopharmacology 03/2015; DOI:10.1038 /npp.2015.89 · 7.05 Impact Factor
  • Source
    • "Moreover, in the two-dimensional affective space described by valence and arousal (Russell, 1980, 2003), both arousal and quadratic valence are measures of emotional salience (Lewis et al., 2007). The linear correlation of activation with lexical arousal and the quadratic correlation with valence ratings support the notion that it is emotional intensity and attention to (emotionally) salient stimuli – rather than a specific type of emotion, i.e., positive vs. negative – that is crucial for amygdala activation (Anderson & Sobel, 2003; Costafreda et al., 2008; Small et al., 2003; Wallentin et al., 2011). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Previous studies suggested that the emotional connotation of single words automatically recruits attention. We investigated the potential of words to induce emotional engagement when reading texts. In an fMRI experiment, we presented 120 text passages from the Harry Potter book series. Results showed significant correlations between affective word (lexical) ratings and passage ratings. Furthermore, affective lexical ratings correlated with activity in regions associated with emotion, situation model building, multi-modal semantic integration, and Theory of Mind. We distinguished differential influences of affective lexical, inter-lexical, and supra-lexical variables: differential effects of lexical valence were significant in the left amygdala, while effects of arousal-span (the dynamic range of arousal across a passage) were significant in the left amygdala and insula. However, we found no differential effect of pas- sage ratings in emotion-associated regions. Our results support the hypothesis that the emotion potential of short texts can be predicted by lexical and inter-lexical affective variables.
    Brain and Language 01/2015; 142:96-114. DOI:10.1016/j.bandl.2015.01.011 · 3.22 Impact Factor
  • Source
    • "Contrary to the wellknown right hemisphere dominance model of emotion processing which derives largely from studies of cortical lesions [Lane and Nadel, 2000], several meta-analyses on human neuroimaging data reported no significant amygdala lateralization. It is emphasized that lateralization might rather be related to age, gender, personality [Wager et al., 2003; Zald, 2003], stimulus type or other methodological variables [Costafreda et al., 2008; Fusar-Poli et al., 2009]. One recent quantitative meta-analysis examined the temporal dynamics of bilateral amygdala and pointed out that the right amygdala is under more rapid modulation and returns to baseline level faster [Sergerie et al., 2008]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent evidence suggests that early life stress (ELS) changes stress reactivity via reduced resting state functional connectivity (rs-FC) between amygdala and the prefrontal cortex. Oxytocin (OXT) modulates amygdala connectivity and attenuates responses to psychosocial stress, but its effect appears to be moderated by ELS. Here we first investigate the effect of ELS on amygdala-prefrontal rs-FC, and examine whether ELS-associated changes of rs-FC in this neural circuit predict its response to psychosocial stress. Secondly, we explore the joint effect of OXT and ELS on the amygdala-prefrontal circuit. Eighteen healthy young males participated in a resting-state fMRI study of OXT effects using a double-blind, randomized, placebo-controlled, within-subject crossover design. We measured the rs-FC to bilateral amygdalae and subsequently assessed changes of state anxiety and prefrontal responses to psychosocial stress. Multiple linear regressions showed that ELS, specifically emotional abuse, predicted reduced rs-FC between the right amygdala and pregenual anterior cingulate cortex (pgACC), which in turn predicted elevated state anxiety after psychosocial stress. In subjects with lower ELS scores, stronger pgACC-amygdala rs-FC predicted stronger pgACC deactivation during the psychosocial stress task, and this rest-task interaction was attenuated by OXT. In subjects with higher ELS scores however, the rest-task interaction was altered and OXT showed no significant effect. These findings highlight that ELS reduces pgACC-amygdala rs-FC and alters how rs-FC of this circuit predicts its stress responsiveness. Such changes in pgACC-amygdala functional dynamics may underlie the altered sensitivity to the effects of OXT after ELS. Hum Brain Mapp, 2014. © 2014 Wiley Periodicals, Inc.
    Human Brain Mapping 10/2014; 35(10). DOI:10.1002/hbm.22553 · 5.97 Impact Factor
Show more