Data for article: "Using mental computation training to improve complex mathematical performance"

Data for article "Using mental computation training to improve complex mathematical performance" in PLOS ONE (under review). In a previous study, mental computation training (extensive practice on multi-digit addition and subtraction problems) led to improvements on more complex forms of mathematics. These mathematical improvements were associated with improvements in numerical precision. In the current study, we investigated whether these transfer and precision effects would replicate with a more diverse population (recruited through Amazon Mechanical Turk), or when core learning features of the training (i.e., immediate feedback, outcome uncertainty, rewards) were removed. We found significant improvements in transfer and precision after participants completed the mental computation training, suggesting robust effects. Data includes the mental computation training task, the complex math (transfer) task, and the number comparison task.

本数据集对应投稿于《公共科学图书馆·综合》（PLOS ONE，处于审稿阶段）的论文《使用心算训练提升复杂数学能力》。既往研究表明，心算训练（即针对多位数加减问题的大量练习）可提升受试者的复杂数学能力，且该类数学能力提升与数值精度的改善存在关联。本研究旨在验证此前发现的迁移效应与精度效应是否可在两类条件下复现：一是通过亚马逊机械 Turk（Amazon Mechanical Turk）招募的更多元化受试者群体，二是移除训练的核心学习特征（即即时反馈、结果不确定性与奖励机制）。研究结果显示，受试者完成心算训练后，其迁移能力与数值精度均出现显著提升，提示该训练效应具有稳健性。本数据集包含心算训练任务、复杂数学（迁移）任务与数字比较任务。