Cognitive and Neural Correlates of Mathematics Problem Solving Using Diagnostic Modeling and Dynamic Real-Time fMRI

这项研究由凯斯西储大学的一组研究人员领导，主要目的是进一步了解数学学习的神经和认知基础。重点将放在青少年使用的数学问题解决策略上，以及随着问题变得越来越难，这些策略是如何被采用和改变的。该项目将考察这些策略与与数学相关的认知技能、数学成绩测试的分数以及青少年在使用这些策略时表现出的大脑激活模式的关联。问题解决策略之所以令人感兴趣，是因为它们与更高水平的数学成就有关，而且它们可以为数学学习中的个体差异提供洞察力。该项目的长期目标是发现新的方法，根据个人的认知和神经能力定制数学学习策略。该项目还将采用创新的方法，收集有关数学任务期间神经功能的信息，这可能会影响未来对参与学习和行为的大脑系统的研究。该项目由教育和人力资源管理局(EHR)核心研究计划资助，该计划支持促进STEM(科学、技术、工程和数学)学习的文献的基础研究。参与者将是14-16岁的青少年，他们之前参加了一项关于早产和足月出生对学校头三年学习进步影响的纵向研究。措施将包括对空间技能、注意力、执行功能(如工作记忆、数字感觉和数学成就)的测试；一组难度不同的心算和分数问题，旨在揭示解决问题策略的个体差异；以及功能磁共振成像(FMRI)程序，以检查这些策略的神经关联。在对单独的年轻人样本进行试点之后，将对100名青少年进行认知和成就测试以及问题解决策略评估，其中50人将根据他们的问题解决策略来选择，以完成磁铁般的问题解决任务。具有更高级的问题解决策略(特征是依赖于提取解决方案)的青少年被假设比使用不那么高级的方法(特征是更费力的多步骤程序)的青少年表现出更完整的认知特征和更高的数学成就水平。使用更高级策略的青少年也被假设在专门进行高效数学处理的区域，如海马体和顶后区域，与不太专门的区域，如前额叶区域，显示出大脑激活的模式。随着问题难度的增加，这些参与者在激活模式上的变化也不那么明显，这表明在解决数学问题时，神经效率更高，对补偿系统的依赖更少。该项目将包括用于认知建模的新统计方法，以及一种令人兴奋的评估数学问题解决过程中大脑功能的新方法。这项研究将提供与成功采用解决问题的策略相关的认知技能和神经过程的洞察，重点是神经能力、效率和补偿。

The broad objective of this study, led by a team of researchers at Case Western Reserve University, is to further understanding of the neural and cognitive basis for mathematics learning. The focus will be on mathematics problem-solving strategies used by adolescents, and how these strategies are adopted and changed as problems become systematically more difficult. The project will examine the association of these strategies with mathematics-related cognitive skills, scores on tests of mathematics achievement, and patterns of brain activations displayed by adolescents while engaged in these strategies. Problem-solving strategies are of interest because of their relation to higher-level mathematic achievement and the insights they can offer into the basis of individual differences in mathematics learning. The longer-term objective of the project is to discover new ways to tailor mathematical learning strategies to individual cognitive and neural capacities. The project will also undertake innovative methods of collecting information about neural functioning during math tasks with the potential to influence future studies of the brain systems involved in learning and behavior. The project is funded by the EHR (Education and Human Resources directorate) Core Research program, which supports fundamental research that advances the literature on STEM (Science, Technology, Engineering and Mathematics) learning. Participants will be adolescents, 14-16 years of age, who were previously enrolled in a longitudinal study of the consequences of preterm versus term birth on academic progress across the first 3 years in school. Measures will include tests of spatial skills, attention, executive functions such as working memory, number sense, and mathematics achievement; sets of mental arithmetic and fractions problems of varying difficulty designed to reveal individual differences in problem-solving strategies; and functional magnetic resonance imaging (fMRI) procedures to examine the neural correlates of these strategies. Following piloting with a separate sample of young adults, cognitive and achievement tests and assessments of problem-solving strategies will be administered to 100 adolescents, 50 of whom will be selected on the basis of their problem-solving strategies to complete in-magnet problem-solving tasks. Adolescents with more advanced problem-solving strategies (characterized by reliance on retrieval of solutions) are hypothesized to display more intact cognitive profiles and higher levels of mathematics achievement than those using less advanced approaches (characterized by more effortful multi-step procedures). Adolescents using more advanced strategies are also hypothesized to show patterns of brain activation in regions specialized for efficient mathematics processing, such as the hippocampus and posterior parietal region, as compared to regions that are less specialized such as the prefrontal area. These participants are also expected to display less pronounced changes in activation patterns with increased problem difficulty, suggestive of greater neural efficiency in mathematics problem-solving and less reliance on compensatory systems. The project will include novel statistical methods for cognitive modeling and an exciting new approach to assess brain functioning during math problem solving. This research will provide insight into the cognitive skills and neural processes associated with successful strategy adoption for problem solving, with an emphasis on neural capacity, efficiency, and compensation.