Gait impairment during complex walking in older adults is thought to result from a progressive failure to compensate for deteriorating peripheral inputs by central neural processes. It is the primary hypothesis of this article that failure of higher cerebral adaptations may already be present in middle-aged adults who do not present observable gait impairments. We, therefore, compared metabolic brain activity during steering of gait (ie, complex locomotion) and straight walking (ie, simple locomotion) in young and middle-aged individuals. Cerebral distribution of [18F]-fluorodeoxyglucose, a marker of brain synaptic activity, was assessed during over ground straight walking and steering of gait using positron emission tomography in seven young adults (aged 24 ± 3) and seven middle-aged adults (aged 59 ± 3). Brain regions involved in steering of gait (posterior parietal cortex, superior frontal gyrus, and cerebellum) are retained in middle age. However, despite similar walking performance, there are age-related differences in the distribution of [18F]-fluorodeoxyglucose during steering: middle-aged adults have (i) increased activation of precentral and fusiform gyri, (ii) reduced deactivation of multisensory cortices (inferior frontal, postcentral, and fusiform gyri), and (iii) reduced activation of the middle frontal gyrus and cuneus. Our results suggest that preclinical decline in central sensorimotor processing in middle age is observable during complex walking.

Neuroimaging-derived markers are used to model post-stroke impairment. Among these, lesion size, corticospinal-tract lesion-load (CST-LL) and resting-state functional-connectivity (rs-FC) have been correlated with impairment. It has been shown that the sensory cortex (S1) is associated with motor learning and is essential for post-stroke recovery; yet stroke-induced changes in S1 connectivity alone are yet to be investigated. We aim to determine whether interhemispheric rs-FC could be used to refine imaging models of stroke-related impairment. Subjects’ post-stroke and age-matched controls underwent rs-fMRI. Stroke-related disability was correlated with lesion size, CST-LL and interhemispheric S1 and M1 rs-FC as independent seeds. Regression analyses were performed to assess the contribution of these markers in stroke-related deficits. Post-stroke subjects showed an asymmetrical pattern of rs-FC in which affected hemisphere S1 and M1 were mostly connected with ipsi-lesional regions. Correlations between rs-FC and stroke-severity were found. Adding rs-FC of S1 to the regression model of impairment decreased the variance 31% compared to lesion size only. After a stroke, S1 interhemispheric connectivity is decreased, with S1 only connected with ipsi-lesional regions. This asymmetry correlates with neurological and motor impairment. Furthermore, when combined with lesion anatomical measures, S1 connectivity might be an important marker in explaining stroke outcome.

The completion of an antisaccade delays the reaction time (RT) of a subsequent prosaccade; however, the converse switch does not influence RT. In accounting for this result, the task-set inertia hypothesis contends that antisaccades engender a persistent nonstandard task-set that delays the planning of a subsequent prosaccade. In contrast, the coordinate system transformation hypothesis asserts that the transformation required to construct a mirror-symmetrical target representation persistently inhibits prosaccade planning. The authors tested the latter hypothesis by examining switch-costs for pro- and antisaccades directed to visual (i.e., the stimuli used in previous work) and auditory targets. Notably, auditory cues are specified in a head-centered frame of reference prior to their conversion into the retinocentric coordinates necessary for saccade output. Thus, if the coordinate system transformation hypothesis is correct then auditory pro- and antisaccades should elicit a bidirectional switch-cost because each requires a coordinate transformation. RTs for visual and auditory modalities showed a reliable--and equivalent magnitude--prosaccade switch-cost. Moreover, performance (e.g., movement time) and kinematic (e.g., velocity) variables indicated the switch-cost was restricted to response planning. As such, results are incompatible with the coordinate system transformation hypothesis and therefore provide convergent evidence that a task-set inertia contributes to the prosaccade switch-cost.