It has been hypothesized in theories that neuroticism is associated with hyperactivity of the amygdala in response to negative stimuli (Canli, 2008; Eysenck, 1967; Gray, 1982, 1991). A combination of higher activation levels as well as lower activation thresholds was suggested to underlie this hyperactivity (Ormel et al., 2013a). Previous research has proposed a role for the amygdala in threat detection, because of its function in directing attention to salient stimuli in the environment and projections to the visual cortex (Ormel et al., 2013a). It would have been plausible to find increased amygdala activation to be associated with higher scores on neuroticism during the processing of negative stimuli. However, we did not find evidence for such a relationship in terms of consistently increased amygdala activation across studies. Furthermore, a sensitivity analysis was conducted with a kernel size of 10 mm, which is better suited to find smaller effects such as those expected in the amygdala, but no association was found between neuroticism and amygdala activation either. Notably, however, based on functional as well as structural connectivity studies, it seems more likely that disconnectivity between the amygdala and frontal regions may underlie the observed heightened emotional reactivity to negative events in high neurotic individuals (Bjørnebekk et al., 2013; Cremers et al., 2010; Ormel et al., 2013a; Xu and Potenza, 2012). An alternative explanation may be that not initial amygdala reactivity is associated with neuroticism, but the speed of amygdala recovery after a negative event, which has recently been shown by Schuyler and colleagues (2012).
Furthermore, we hypothesized that neuroticism would be related to decreased activation in brain areas associated with emotion processing in response to positive stimuli. However, no significant correlations with neuroticism survived FDR multiple comparisons correction for the contrast (positive>neutral). Although the power was lower to detect significant differences for this contrast (i.e. a lower number of whole-brain studies was found for this contrast), this null finding suggests that differences in the processing of negative stimuli play a dominant role in neuroticism, reactivity and emotional instability are the core elements of neuroticism and that alterations in abovementioned psychological processes contribute to this.
2
which corresponds with the observed negativity bias in high neurotic individuals (Canli, 2008;
Chan et al., 2007; Ormel et al., 2013a). Nonetheless, additional research is needed to confirm this conclusion.
2.5.6 Limitations
Several limitations regarding the data analysis and data selection should be taken into account.
A potential issue of concern regarding the PCM method is the selection of the width of the kernel, which is used to smooth the coordinate-based data. Previous research has suggested a kernel of 15-25 mm in order to obtain optimal sensitivity (Costafreda, 2012). Furthermore, a drawback of the selected data was the great variety in i) questionnaires used to assess neuroticism, ii) task paradigms to investigate the neural correlates of neuroticism and iii) the application of various regressors, such as negative mood, additional personality traits (e.g. extraversion and conscientiousness), age and sex. It has been recognized that neuroticism is a heterogeneous construct measured with questionnaires including different facet scales (Ormel et al., 2013a). Ormel and colleagues (2013a) proposed that future research would benefit from deconstructing neuroticism into lower-order facets, cognitive-emotional processes or innate and acquired neuroticism in order to investigate the underlying neurobiology of seemingly opposing facets (e.g. anxious distress; internalizing versus irritable distress; externalizing) and the basic genetic architecture of the trait by using an endophenotypic approach. Furthermore, the role of abovementioned variables that are of relevance for neuroticism should be investigated more thoroughly. For example, there are indications for sex-related differences in neuroticism levels (women generally score higher on neuroticism than men) and brain activity during the processing of emotion (Ormel et al., 2013a). Another example concerns the additional personality traits, including extraversion, conscientiousness, agreeableness and openness. In principle, the Big Five factors are conceptually independent dimensions, however, significant correlations have regularly been found between aforementioned traits in questionnaire research (Hoekstra et al., 1996).
2.6 Conclusion
The aim of the current meta-analysis was to provide a quantitative summary of the literature, using a parametric coordinate-based meta-analysis (PCM) approach. Data were pooled for emotion processing tasks that investigated negative as well as positive emotional stimuli (compared to neutral stimuli) to identify brain regions that are consistently associated with neuroticism across studies. Differences in brain activation were found to be associated with neuroticism during fear learning, anticipation of aversive stimuli and the processing and regulation of emotion. The meta-analytic findings were incorporated into a model of emotion processing in neuroticism, which allows concrete hypotheses to be formulated in order to direct future research. Associations between neuroticism and abovementioned psychological processes can be tested separately on a
behavioral, psychological and neurobiological level. Furthermore, causal as well as bidirectional relationships can be explored between the different psychological processes and associations with neuroticism. Moreover, it will be valuable to examine the role of these psychological processes in psychiatric disorders for which neuroticism is considered to be a risk factor (e.g. anxiety disorders and depression) to investigate how neuroticism predisposes individuals to the development of psychopathology.
2.7 Acknowledgements
The current study was supported by the Ministry of Education of the Netherlands (609022).
Furthermore, S.G.C. was supported by the National Institute of Health Research (NIHR) Academic Clinical Lectureship and P.H. was supported by the NIHR Biomedical Research Unit in Dementia at South London, Maudsley NHS Foundation Trust (SLaM) and the Institute of Psychiatry, King’s College London. The sponsors did not play a role in the collection, analysis and interpretation of the data; writing the manuscript or the decision to submit the article for publication.