Reconfigurations of cortical manifold structure during reward-based motor learning.

Adaptive motor behavior depends on the coordinated activity of multiple neural systems distributed across the brain. While the role of sensorimotor cortex in motor learning has been well established, how higher-order brain systems interact with sensorimotor cortex to guide learning is less well understood. Using functional MRI, we examined human brain activity during a reward-based motor task where subjects learned to shape their hand trajectories through reinforcement feedback. We projected patterns of cortical and striatal functional connectivity onto a low-dimensional manifold space and examined how regions expanded and contracted along the manifold during learning. During early learning, we found that several sensorimotor areas in the dorsal attention network exhibited increased covariance with areas of the salience/ventral attention network and reduced covariance with areas of the default mode network (DMN). During late learning, these effects reversed, with sensorimotor areas now exhibiting increased covariance with DMN areas. However, areas in posteromedial cortex showed the opposite pattern across learning phases, with its connectivity suggesting a role in coordinating activity across different networks over time. Our results establish the neural changes that support reward-based motor learning and identify distinct transitions in the functional coupling of sensorimotor to transmodal cortex when adapting behavior.

Interaction between decision-making and motor learning when selecting reach targets in the presence of bias and noise.

Motor errors can have both bias and noise components. Bias can be compensated for by adaptation and, in tasks in which the magnitude of noise varies across the environment, noise can be reduced by identifying and then acting in less noisy regions of the environment. Here we examine how these two processes interact when participants reach under a combination of an externally imposed visuomotor bias and noise. In a center-out reaching task, participants experienced noise (zero-mean random visuomotor rotations) that was target-direction dependent with a standard deviation that increased linearly from a least-noisy direction. They also experienced a constant bias, a visuomotor rotation that varied (across groups) from 0 to 40 degrees. Critically, on each trial, participants could select one of three targets to reach to, thereby allowing them to potentially select targets close to the least-noisy direction. The group who experienced no bias (0 degrees) quickly learned to select targets close to the least-noisy direction. However, groups who experienced a bias often failed to identify the least-noisy direction, even though they did partially adapt to the bias. When noise was introduced after participants experienced and adapted to a 40 degrees bias (without noise) in all directions, they exhibited an improved ability to find the least-noisy direction. We developed two models-one for reach adaptation and one for target selection-that could explain participants' adaptation and target-selection behavior. Our data and simulations indicate that there is a trade-off between adaptation and selection. Specifically, because bias learning is local, participants can improve performance, through adaptation, by always selecting targets that are closest to a chosen direction. However, this comes at the expense of improving performance, through selection, by reaching toward targets in different directions to find the least-noisy direction.

Accounting for the last-sampling bias in perceptual decision-making.

Human decisions are replete with biases that often reflect the underlying mechanisms of the decision-making process. The current study focused on a bias observed in different modalities and decision domains; the last-sampling bias, whereby people tend to choose the option they sampled last. One common explanation for this bias is that attention causally affects choices: The option now being attended to accumulates evidence faster than its unattended alternative. An alternative explanation is that the state of the accumulated evidence influences sampling patterns, but this mechanism remains controversial. To test these explanations, we designed a sequential sampling paradigm based on a simple two-alternative perceptual decision task. In one condition subjects could freely sample the options and in another condition the last-sampled option was predetermined. Significant last-sampling biases were observed in both conditions. We then examined possible causes of the last-sampling bias under the attentional drift-diffusion model (aDDM) framework. By comparing empirical results and model behaviors, we inferred that the influence between attention and evidence accumulation is bi-directional during perceptual decision-making.

Discovery of Potential Serum Protein Biomarkers in Ankylosing Spondylitis Using Tandem Mass Tag-Based Quantitative Proteomics.

Ankylosing spondylitis (AS) is a systemic, chronic, and inflammatory rheumatic disease that affects 0.2% of the population. Current diagnostic criteria for disease activity rely on subjective Bath Ankylosing Spondylitis Disease Activity Index scores. Here, we aimed to discover a panel of serum protein biomarkers. First, tandem mass tag (TMT)-based quantitative proteomics was applied to identify differential proteins between 15 pooled active AS and 60 pooled healthy subjects. Second, cohort 1 of 328 humans, including 138 active AS and 190 healthy subjects from two independent centers, was used for biomarker discovery and validation. Finally, biomarker panels were applied to differentiate among active AS, stable AS, and healthy subjects from cohort 2, which enrolled 28 patients with stable AS, 26 with active AS, and 28 healthy subjects. From the proteomics study, a total of 762 proteins were identified and 46 proteins were up-regulated and 59 proteins were down-regulated in active AS patients compared to those in healthy persons. Among them, C-reactive protein (CRP), complement factor H-related protein 3 (CFHR3), α-1-acid glycoprotein 2 (ORM2), serum amyloid A1 (SAA1), fibrinogen γ (FG-γ), and fibrinogen ß (FG-ß) were the most significantly up-regulated inflammation-related proteins and S100A8, fatty acid-binding protein 5 (FABP5), and thrombospondin 1 (THBS1) were the most significantly down-regulated inflammation-related proteins. From the cohort 1 study, the best panel for the diagnosis of active AS vs healthy subjects is the combination of CRP and SAA1. The area under the receiver operating characteristic (ROC) curve was nearly 0.900, the sensitivity was 0.970%, and the specificity was 0.805% at a 95% confidence interval from 0.811 to 0.977. Using 0.387 as the cutoff value, the predictive values reached 92.00% in the internal validation set (62 with active AS vs 114 healthy subjects) and 97.50% in the external validation phase (40 with active AS vs 40 healthy subjects). From the cohort 2 study, a panel of CRP and SAA1 can differentiate well among active AS, stable AS, and healthy subjects. For active AS vs stable AS, the area under the ROC curve was 0.951, the sensitivity was 96.43%, the specificity was 88.46% at a 95% confidence interval from 0.891 to 1, and the coincidence rate was 92.30%. For stable AS vs healthy humans, the area under the ROC curve was 0.908, the sensitivity was 89.29%, the specificity was 78.57% at a 95% confidence interval from 0.836 to 0.980, and the coincidence rate was 83.93%. For active AS vs healthy subjects, the predictive value was 94.44%. The results indicated that the CRP and SAA1 combination can potentially diagnose disease status, especially for active or stable AS, which will be conducive to treatment recommendation for patients with AS.