Decreased Fronto-Parietal and Increased Default Mode Network Activation is Associated with Subtle Cognitive Deficits in Elderly Controls.

BACKGROUND: Cognitive functions progressively deteriorate during aging and neurodegenerative diseases. The present study aims at investigating differences in working memory performance as well as functional brain changes during the earliest stages of cognitive decline in health elderly individuals. METHODS: 62 elderly individuals (41 females), including 41 controls (35 females) and 21 middle cognitive impairment subjects (6 females), underwent neuropsychological assessment at baseline and an fMRI examination in a N-back paradigm contrasting 2-back vs. 0-back condition. Upon a 18 months follow-up, we identified stable controls (sCON) with preserved cognition and deteriorating controls (dCON) with -1SD decrease of performances in at least two neuropsychological tests. Data analyses included accuracy and reaction time (RT) for the 2-back condition and general linear model (GLM) for the fMRI sequence. RESULTS: At the behavioral level, sCON and dCON performed better than MCI in terms of accuracy and reaction time. At the brain level, functional differences in regions of the fronto-parietal network (FPN) and of the Default Mode Network (DFM) were observed. Significantly lower neural activations in the bilateral inferior and middle frontal gyri were found in MCI versus both dCON / sCON and for dCON versus sCON. Significantly increased activations in the anterior cingulate cortex and posterior cingulate cortex and bilateral insula were found in MCI versus both dCON / sCON and in dCON versus sCON. CONCLUSION: The present study suggests that brain functional changes in FPN and DMN anticipate differences in cognitive performance in healthy elderly individuals with subsequent subtle cognitive decline.

Inter-individual variability in metacognitive ability for visuomotor performance and underlying brain structures.

Metacognition refers to the ability to discriminate between one's own correct and incorrect decisions. The neurobiological underpinnings of metacognition have mainly been studied in perceptual decision-making. Here we investigated whether differences in brain structure predict individual variability in metacognitive sensitivity for visuomotor performance. Participants had to draw straight trajectories toward visual targets, which could unpredictably deviate around detection threshold, report such deviations when detected, and rate their confidence level for such reports. Structural brain MRI analyses revealed that larger gray-matter volume (GMV) in the left middle occipital gyrus, left medial parietal cortex, and right postcentral gyrus predicted higher deviation detection sensitivity. By contrast, larger GMV in the right prefrontal cortex but also right anterior insula and right fusiform gyrus predicted higher metacognitive sensitivity. These results extend past research by linking metacognitive sensitivity for visuomotor behavior to brain areas involved in action agency (insula), executive control (prefrontal cortex) and vision (fusiform).

Aversive stimuli exacerbate defensive motor behaviour in motor conversion disorder.

Conversion disorder or functional neurological symptom disorder (FND) can affect the voluntary motor system, without an organic cause. Functional symptoms are thought to be generated unconsciously, arising from underlying psychological stressors. However, attempts to demonstrate a direct relationship between the limbic system and disrupted motor function in FND are lacking. We tested whether negative affect would exacerbate alterations of motor control and corresponding brain activations in individuals with FND. Ten patients and ten healthy controls produced an isometric precision-grip contraction at 10% of maximum force while either viewing visual feedback of their force output, or unpleasant or pleasant emotional images (without feedback). Force magnitude was continuously recorded together with change in brain activity using fMRI. For controls, force output decayed from the target level while viewing pleasant and unpleasant images. Patients however, maintained force at the target level without decay while viewing unpleasant images, indicating a pronounced effect of negative affect on force output in FND. This emotional modulation of force control was associated with different brain activation patterns between groups. Contrasting the unpleasant with the pleasant condition, controls showed increased activity in the inferior frontal cortex and pre-supplementary motor area, whereas patients had greater activity in the cerebellum (vermis), posterior cingulate cortex, and hippocampus. Engagement of a cerebellar-limbic network in patients is consistent with heightened processing of emotional salience, and supports the role of the cerebellum in freezing responses in the presence of aversive events. These data highlight a possible neural circuit through which psychological stressors elicit defensive behaviour and modulate motor function in FND.

Neural underpinnings of background acoustic noise in normal aging and mild cognitive impairment.

Previous contributions in younger cohorts have revealed that reallocation of cerebral resources, a crucial mechanism for working memory (WM), may be disrupted by parallel demands of background acoustic noise suppression. To date, no study has explored the impact of such disruption on brain activation in elderly individuals with or without subtle cognitive deficits. We performed a functional Magnetic Resonance Imaging (fMRI) study in 23 cases (mean age=75.7 y.o., 16 men) with mild cognitive impairment (MCI) and 16 elderly healthy controls (HC, mean age=70.1 y.o., three men) using a 2-back WM task, under two distinct MRI background acoustic noise conditions (louder vs. lower noise echo-planar imaging). General linear models were used to assess brain activation as a function of group and noise. In both groups, lower background noise is associated with increased activation of the working memory network (WMN). A decrease of the normally observed deactivation of the default mode network (DMN) is found under louder noise in both groups. Unlike HC, MCI cases also show decreased deactivation of the DMN under both louder and lower background noise. Under louder noise, this decrease is observed in anterior parts of the DMN in HC, and in the posterior cingulate cortex in MCI cases. Our results suggest that background acoustic noise has a differential impact on WMN activation in normal aging as a function of the cognitive status. Only louder noise has a disruptive effect on the usually observed DMN deactivation during WM task performance in HC. In contrast, MCI cases show altered DMN reactivity even in the presence of lower noise.

Acute caffeine administration effect on brain activation patterns in mild cognitive impairment.

Previous studies showed that acute caffeine administration enhances task-related brain activation in elderly individuals with preserved cognition. To explore the effects of this widely used agent on cognition and brain activation in early phases of cognitive decline, we performed a double-blinded, placebo-controlled functional magnetic resonance imaging (fMRI) study during an n-back working memory task in 17 individuals with mild cognitive impairment (MCI) compared to 17 age-matched healthy controls (HC). All individuals were regular caffeine consumers with an overnight abstinence and given 200 mg caffeine versus placebo tablets 30 minutes before testing. Analyses included assessment of task-related activation (general linear model), functional connectivity (tensorial-independent component analysis, TICA), baseline perfusion (arterial spin labeling, ASL), grey matter density (voxel-based morphometry, VBM), and white matter microstructure (tract-based spatial statistics, TBSS). Acute caffeine administration induced a focal activation of the prefrontal areas in HC with a more diffuse and posteromedial activation pattern in MCI individuals. In MCI, TICA documented a significant caffeine-related enhancement in the prefrontal cortex, supplementary motor area, ventral premotor and parietal cortex as well as the basal ganglia and cerebellum. The absence of significant group differences in baseline ASL perfusion patterns supports a neuronal rather than a purely vascular origin of these differences. The VBM and TBSS analyses excluded potentially confounding differences in grey matter density and white matter microstructure between MCI and HC. The present findings suggest a posterior displacement of working memory-related brain activation patterns after caffeine administration in MCI that may represent a compensatory mechanism to counterbalance a frontal lobe dysfunction.