Learning the predictive value of biologically relevant stimuli is fundamental to successful behavioural adaptation in humans and other animals and pathological fear learning underlies affective disorders, such as post-traumatic stress disorder or anxiety. Putative neural circuits mediating emotional learning have been extrapolated from animal models but still await experimental confirmation in humans. To test whether the neural networks underlying emotional learning are comparable between humans and other species, we used functional magnetic resonance imaging and independent component analysis to identify neural networks engaged during a repeated differential fear conditioning and extinction task in humans. The results of this study provide strong evidence for evolutionary continuity by showing that fear learning is supported by comparable networks in humans, rodents, and non-human primates. The results further reveal that in the human brain, activity in one amygdala-based network associated with repeated aversive learning is anti-correlated with activity in the striatal-hippocampal-ventromedial prefrontal network commonly associated with extinction. This novel finding is highly relevant for affective pathologies because it suggests that emotional learning might crucially depend on the balanced interaction between these two networks. Finally, the results show a dissociation between functional connectivity of the dorsal and ventral amygdala with insular and parietal cortices during initial and repeated conditioning. This result extends previous findings and suggests a potential role for two core networks during fear acquisition, i.e., the salience network involving anterior insular cortex and the attention network involving posterior parietal cortex.

Friedreich ataxia (FRDA) is a genetic disorder defined by progressive degeneration within the cerebellum and corticospinal tracks. While most readily characterized by impairments in motor coordination, deficits have also been observed in cognitive control and working memory processes. These observations motivate the hypothesis that focal changes in the cerebellum may influence function in, and functional interactions with, association cortices of the cerebrum.

Twenty-nine individuals with FRDA and 34 age/gender matched healthy controls performed a block-design verbal n-back working memory task with two levels of load (“0-Back” and “2-Back”), concurrent with whole-brain functional magnetic resonance imaging. Group differences in task-related brain activations (2-Back > 0-Back) and task-related cerebro-cerebellar connectivity (Psychophysiological Interactions) were assessed.

Alongside matched behavioural performance, individuals with FRDA had significantly reduced brain activations in lateral regions of the cerebellar cortex, maximally in lobule VI bilaterally. Within the cerebral cortex, analogous functional disturbances were observed in left-lateralized insular and rostrolateral prefrontal cortices. Convergent abnormalities in functional connectivity were also observed between these cerebellar and cerebral regions, supporting a putative link between cerebellar dysfunction and downstream cerebral consequences.

These results highlight the consequence of cerebellar pathology to distributed brain function underlying higher-order cognition, and provide evidence for the critical role that cerebro-cerebellar interactions play in healthy human cognition. More specifically, this study supports the conceptualization of FRDA as a disorder of large-scale, spatially distributed cerebral and cerebellar circuitry.

Resting-state connectivity studies have identified a network of correlated brain regions - the default mode network (DMN) - which is thought to underpin self-referential, introspective processing. Depression is a condition characterised by altered self-referential and emotional processing and thus understanding changes in DMN connectivity constitutes an important research objective. Initial resting-state studies of depression documented a pattern of hyper-connectivity between DMN regions (e.g., posterior cingulate cortex, PCC, medial prefrontal cortex, MPFC, and lateral prefrontal regions), and hypo-connectivity between DMN and other regions (insula, precuneus, thalamus, temporal and superior frontal cortices). We compared resting-state seed-to-voxel functional connectivity in a depression group with a matched control group using the CONN toolbox (employing component-based noise correction) to estimate correlations with two DMN seed regions (PCC, MPFC). Preliminary analyses found that the depression group exhibited significantly stronger negative correlations between both seeds and posterior temporal cortices (middle/superior temporal gyri) than the control group (voxel level p < .005; cluster level False Discovery Rate p <.05). Our results add to existing studies of resting-state connectivity in depression. Moreover, the results are consistent with functional data during internally-directed tasks (autobiographical memory and future thinking) from the same sample indicating decreased mean signal in temporal regions relative to controls and altered task-related connectivity with DMN seeds (bilateral hippocampus). Negative connectivity between areas known to support self-referential and emotional processing and temporal cortices may be associated with qualitative differences in mind-wandering for the depression group and/or reflective of connectivity changes occurring in response to temporal atrophy.

Brain region-specific changes have been demonstrated with a variety of cognitive training interventions. The effect of cognitive training on brain subnetworks, however, remains largely unknown, with studies limited to functional networks. Here, we used a well-established working memory training program and state-of-the art neuroimaging methods in 40 healthy adults (21 females, mean age: 26.5 years). Near and far transfer training effects were assessed using computerized working memory and executive function tasks. Adaptive working memory training led to improvement on (non)trained working memory tasks, but no generalization to tasks of reasoning, planning, or inhibition. Graph theoretical analysis of the structural (white matter) network connectivity (‘connectome’) revealed increased global integration within a fronto-parietal attention network following adaptive working memory training compared to non-adaptive control. Furthermore, the impact on the outcome of graph theoretical analyses of different white matter metrics to infer ‘connection strength’ was evaluated. Increased efficiency of the fronto-parietal network was best captured when using connection strengths derived from MR metrics that are thought to be more sensitive to differences in myelination (putatively indexed by the (quantitative) longitudinal relaxation rate, R1) than previously used diffusion MRI metrics (fractional anisotropy or fibre-tracking recovered streamlines). Our findings emphasize the critical role of the specific microstructural markers in providing important hints towards the mechanisms underpinning training-induced plasticity that may drive memory improvement in clinical populations.