利用脑成像多体素模式分析解码认知的神经表征:原理和应用

多体素模式分析(multi-voxel pattern analysis,MVPA)是一种基于机器学习理论发展出来的新的功能磁共振数据分析技术。MVPA技术通过训练分类器,对由不同认知状态引起的多体素信号模式进行分类。与传统的基于单个体素的分析方法相比,该技术可更敏感地检测脑对认知状态的表征,并使得从神经信号解码认知状态成为可能。文章介绍MVPA技术的基本原理、分析步骤以及可以用MVPA来解决的科学问题,并对应用中可能面临的问题提供了参考建议。     

Characterisation of magnesium oxide and its interface with α-Mg in Mg–Al-based alloys

Magnesium oxide (MgO) films and particles have been collected by pressurised filtration of a Mg-8.6wt%Al-0.67wt%Zn (AZ91D) alloy melt. The morphology of the oxides and their interfaces with the α-Mg phase were investigated by high-resolution transmission electron microscopy. It was found that the oxide films consisted of large numbers of sub-micrometre-sized MgO particles, and that melt shearing can effectively break up the oxide films and disperse the oxide particles. For the first time, orientation relationships (ORs) of OR I: [1?1?1]_MgO～2° from (0?0?0?1)_α-Mg and (0?1?1)_MgO?//[2?1?1?0)_α-Mg; and OR II: (1?1?1)_MgO//(1?1?0?1)_α-Mg and [0?1?1]_MgO//[1?2?1?1]_α-Mg, were observed between the MgO particles and the α-Mg matrix. The calculated Bramfitt planar disregistries were 5.5% and 2.5% for the two ORs, respectively, indicating good lattice matching between MgO and α-Mg at the interface. With the evidence of grain refinement effect observed in the sheared AZ91D magnesium alloy, the possibility of MgO particles to act as potent nucleants for heterogeneous nucleation of α-Mg grains is discussed in terms of the crystallographic criterion.     

Application of the stagnant stage concept for monitoring Mn partitioning at the austenite-ferrite interface in the intercritical region for Fe–Mn–C alloys

The length of the stagnant stage during the new ferrite growth starting from a mixture of austenite and ferrite has been investigated for a Fe-0.17Mn-0.023C (wt%) alloy. It was found that the stagnant stage depends on the thermal path followed to create the mixture, and deduce that the tie-lines for austenite to ferrite transformation are quite different from those for ferrite to austenite transformation. The length of the stagnant stage is determined by the very local partitioning effect at the interface, and it can be used as a tool to monitor the Mn partitioning.     

Detrimental effect of cellular precipitation on the creep strength of Inconel740H

Cellular precipitation (also known as discontinuous precipitation) has been observed at the grain boundaries of a newly developed nickel-based Inconel740H alloy designed for use at 700?°C in advanced ultrasupercritical coal-fired power plants. By means of element mapping and selected area diffraction, the cellular precipitates were identified as Cr-rich M₂₃C₆ carbides. The onset of cellular precipitation was found to follow a pucker mechanism in Inconel740H. The cellular precipitates at the grain boundaries, even at low volume fractions, were severely detrimental to the creep strength at 750?°C. The creep rupture life of Inconel740H containing cellular precipitates at grain boundaries was only one-tenth of that for the alloy without cellular precipitates. The reason for the drastically decreased creep rupture life is attributed to the poor resistance of cellular precipitates to crack propagation during creep.     

关注肿瘤患者避免过度诊疗——解析ASCO大会关于肿瘤诊疗的5项建议

现代肿瘤的诊疗水平已经有了长足的进步,先进的诊断技术使得很多肿瘤得以早期发现并积极治疗,很多晚期患者也不再面临无药可用的境地,但由于临床医生的认识存在偏差,存在过度诊断和过度治疗的问题,进而给患者带来不必要的负担和痛苦。美国临床肿瘤学会（ASC0）年会上发布了一篇名为“ASCO为改善肿瘤治疗质量及医疗花费的5项建议”的文章,特别提出要重视五项临床广泛应用、花费较大,但缺乏高级临床证据的肿瘤治疗或干预措施,也提示我们需要充分评估,综合思考,为合适的患者提供正确的诊治方案,才能给患者带来更大的获益。     

The Role of Sketching States in the Stimulation of Idea Generation: An Eye Movement Study

Sketching is widely used in design to generate creative ideas. Design studies present stimuli during sketching to enhance creativity. This study examines the effect of stimuli presented during different sketching states, especially of those presented during the stuck period. It conducted a sketching experiment that enrolled 41 students with an average of 5.7 years’ practice in industrial design. The sketching with stimuli presented at regular intervals, those with stimuli presented during the stuck period, and those without stimulus were compared. Eye movements during the sketching processes were recorded. The attention to stimuli, and the direct applications of stimuli to design ideas, and the idea connection and idea differentiation were analyzed. Participants sketching with stimuli presented during the stuck period had a higher sketching quality, compared with those in other experimental conditions. They fixated longer on stimuli, built more links among related ideas, and had a more differentiated and balanced exploration. This study then proposed the mechanisms of stimuli processing. The stimuli presented at regular intervals were stored and acted as additional memory. The stimuli presented during the stuck period were dissected and acted as supporter, inducing a more creative sketching.     

Effects of Electromyostimulation on Knee Extensors and Flexors Strength and Steadiness in Older Adults

It is known that electromyostimulation (EMS) alone or superimposed over voluntary contraction (EV) can effectively improve muscle strength. However, the effect of this type of training on the ability to control force production at submaximal levels is unknown. The authors examined the effects of EV training on steadiness in force production of knee extensors and flexors in older adults. Forty participants, including 20 men and 20 women, 60–77 years of age, were randomly allocated into a control group (CG) and an electromyostimulation superimposed over voluntary contraction (EVG) group. The EVG performed 30 bilateral isometric knee extension and flexion contractions per session, 3 training sessions per week, for 6 weeks. The variations in force production, expressed in absolute (standard deviation [SD]) and relative (coefficient of variation [CV]) terms, were assessed in isometric contractions at 5%, 15% and 25% maximal voluntary contraction (MVC) levels. Results indicated that MVC increased in knee extension and flexion in EVG (p < .05) after the training; steadiness CV also improved at 15% MVC in knee flexion (p < .05) but no significant changes were found in knee extension and steadiness SD. The training-induced changes in MVC were not correlated to steadiness CV that might indicate different mechanisms underlying these adaptations.     

Knowledge and economic growth—creativity,learning and knowledge utilization

The purpose of this study is to discuss the dynamic interdependence of knowledge and economic growth. We show that the traditional economics, which treats knowledge as parameters and is mainly developed within linearized, stable and static frameworks can hardly explain the modern economic complexity. We argue that it is necessary to treat creativity, learning and knowledge utilization as endogenous dynamic processes of social and economic evolution. To handle with the complicated dynamic interdependence of wealth and knowledge accumulation, we need to apply genuine nonlinear dynamic theory. In particular, we discuss possible benefits of government's intervention in education and science to the society when the social and economic evolution is characterized of nonlinearity, instability and multi‐equilibria.     

A Computationally Efficient State Space Approach to Estimating Multilevel Regression Models and Multilevel Confirmatory Factor Models

Although the state space approach for estimating multilevel regression models has been well established for decades in the time series literature, it does not receive much attention from educational and psychological researchers. In this article, we (a) introduce the state space approach for estimating multilevel regression models and (b) extend the state space approach for estimating multilevel factor models. A brief outline of the state space formulation is provided and then state space forms for univariate and multivariate multilevel regression models, and a multilevel confirmatory factor model, are illustrated. The utility of the state space approach is demonstrated with either a simulated or real example for each multilevel model. It is concluded that the results from the state space approach are essentially identical to those from specialized multilevel regression modeling and structural equation modeling software. More importantly, the state space approach offers researchers a computationally more efficient alternative to fit multilevel regression models with a large number of Level 1 units within each Level 2 unit or a large number of observations on each subject in a longitudinal study.