NCS-FO: Integrated neurocognitive process models of individual differences in children’s math problem solving strategies, learning and development

该项目由斯坦福大学的一组研究人员领导，目标是开发新的神经认知模型，将行为和神经（fMRI，功能性磁共振成像）数据整合在一起，以了解数学问题解决和策略使用、训练效果和纵向发展方面的个体差异背后的计算、认知和大脑机制。理解符号是如何在大脑中处理的，对教育和认知困难的补救具有直接的意义。研究人员将对来自7至12岁儿童的数据集进行复杂的计算分析，这些数据集来自于一项功能磁共振成像研究、一项辅导研究和一项纵向研究。本研究结果将为定制培训提供基础，并为学习困难的诊断和干预程序提供新的平台。该项目由理解神经和认知系统的综合策略（NCS）资助，这是一个由计算机与信息科学与工程（CISE）、教育与人力资源（EHR）、工程（ENG）和社会、行为和经济科学（SBE）联合支持的多学科项目。这项研究将利用广泛的行为数据、认知评估以及脑成像数据，来模拟与数学问题解决相关的潜在认知动态的每时每刻变化。本文描述的多学科方法旨在开发无监督计算模型，以在逐个试验的基础上推断个体使用的潜在认知策略的差异。具体而言，本项目旨在(1)开发用于推断潜在问题解决动力和策略使用的计算认知模型；(2)开发新的综合神经认知模型，以识别潜在问题解决动力和策略使用背后不同且重叠的脑回路；(3)确定认知辅导对潜在问题解决动力和策略使用变化影响的综合认知和神经机制。计划中的研究将有助于将数学问题解耦为多个认知子过程，描述这些子过程中个体差异的来源，确定每个子过程与解决问题困难的关系，并评估这些认知子过程是否可以通过认知辅导计划加以补救。最终，本研究将增强我们对儿童数字认知的动态认知过程和问题解决策略的理解，并为儿童典型和非典型数学能力的潜在行为动力学提供新的见解。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The goal of this project, led by a team of researchers at Stanford University, is to develop novel neurocognitive models that integrate behavioral and neural (fMRI, functional magnetic resonance imaging) data to understand the computational, cognitive, and brain mechanisms underlying individual differences in mathematical problem solving and strategy use, the effects of training, and longitudinal development. Understanding how symbols are processed in the brain has direct implications for education and the remediation of cognitive difficulties. The researchers will perform sophisticated computational analyses on a dataset derived from children 7 to 12 years of age in an fMRI study, a tutoring study, and a longitudinal study. The findings of this study will provide a basis for customized training and will provide novel platforms for diagnostic and intervention procedures for learning difficulties. This project is funded by Integrative Strategies for Understanding Neural and Cognitive Systems (NCS), a multidisciplinary program jointly supported by the Directorates for Computer and Information Science and Engineering (CISE), Education and Human Resources (EHR), Engineering (ENG), and Social, Behavioral, and Economic Sciences (SBE).This research will leverage extensive behavioral data, cognitive assessments, as well as brain imaging data, to model moment-by-moment changes in latent cognitive dynamics associated with mathematical problem solving. The multidisciplinary approach described here seeks to develop unsupervised computational models to infer differences in the latent cognitive strategies used by an individual on a trial-by-trial basis. Specifically, this project aims to (1) develop computational cognitive models for inferring latent problem-solving dynamics and strategy use, (2) develop novel integrated neurocognitive models to identify distinct and overlapping brain circuits underlying latent problem-solving dynamics and strategy use, and (3) determine integrated cognitive and neural mechanisms underlying the impact of cognitive tutoring on changes in latent problem-solving dynamics and strategy use. The planned studies will help in dissociating mathematical problem solving into multiple cognitive sub-processes, characterizing sources of individual differences across these sub-processes, identifying how each of these might relate to difficulties in problem solving, and evaluate whether these cognitive sub-processes may be remediated by cognitive tutoring programs. Ultimately, this research will enhance our understanding of dynamic cognitive processes and problem-solving strategies in children’s numerical cognition and provide new insights into latent behavioral dynamics associated with typical and atypical math abilities in children.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.