随着我国基础物理教育迈入核心素养新时代，以书面形式表述的传统试题等传统测评工具的短板愈发突显。与此同时，一种被试与计算机间能进行信息交换的、人机交互式的新型试题（以下简称“人机交互试题”）应运而生。目前，人机交互试题已在国际科学教育测评领域得到广泛的研究与应用，其能有效弥补传统测评工具的短板，为顺应科学教育测评的新要求提供了重要途径。

为此，本研究在基础物理教育领域，从项目学习视域下对“如何合理地设计人机交互试题”这一问题展开研究。本研究对于促进我国物理教育教学测评工具的发展具有重要的理论意义和实践意义。

本研究的主要内容包括如下：

首先，本研究对人机交互试题、试题设计、大规模教育测评项目中的科学测评框架及科学教育领域中的项目学习测评进行综述，并基于文献综述结果确定本研究的研究设计，包括对“人机交互试题”“试题设计”“项目学习”的概念界定，以及围绕“如何在项目学习视域下设计人机交互物理试题”这一核心问题确定的研究任务、研究思路与研究方法。

其次，基于对相关研究成果的梳理，本研究厘清了人机交互物理试题的设计基础，提出项目学习视域下人机交互物理试题的设计原则，并建构项目学习视域下的人机交互物理试题设计模型，该试题设计模型涉及“分析阶段”“构造阶段”“试题生成阶段”等3个阶段，3个阶段下共包含13个要素。

再次，基于建构的项目学习视域下的人机交互物理试题设计模型，本研究确立了包含12个步骤的人机交互物理试题设计模型的应用流程，并以“绿豆芽的培育”人机交互物理试题为例，示例基于流程的试题设计过程，最终共设计“绿豆芽的培育”和“伪全息投影”两道人机交互物理试题，包括试题主体部分和评分标准部分。

最后，本研究对建构的人机交互物理试题设计模型的效度进行了实证评估。将设计的“绿豆芽的培育”人机交互物理试题制作为具备交互功能的程序并在此基础上搭建人机交互物理试题测试系统，应用该测试系统在中学开展测试，收集了129名九年级学生的作答数据。通过对实证数据的分析表明，“绿豆芽的培育”人机交互物理试题具有良好的质量和效度。基于以上实测证据可推论，项目学习视域下的人机交互物理试题设计模型具有良好的有效性，可对人机交互物理试题的设计作出有效指导。

Since the basic physics education in China has progressed into the new era of core literacy, the limitations of traditional assessment tools, such as traditional written items, are increasingly obvious. At the same time, a new type of human-computer interactive items (hereinafter referred to as "HCI items"), which can exchange information between subjects and computers, has emerged. At present, HCI items have been widely researched and applied in the international science assessment, which can effectively compensate the limitations of traditional assessment tools and provide an essential way to meet the new requirements of science assessment.

Therefore, the research is conducted in the area of basic physics education in the perspective of project-based learning to study the issue of "how to design HCI items rationally". The research is of great theoretical and practical significance in the development of instructional assessment tools for physics education in China. The main contents of the research include as follows:

First, the research provides an overview of HCI items, items design, science assessment frameworks in large-scale educational assessment project, and assessment of project-based learning in science education. Based on the results of the literature review, the research design of this research is determined, including the conceptual definitions of the terms "HCI items", " items design", and "project-based learning", as well as the research tasks, research ideas and research methods centered on the core issue of "how to design HCI physics items in the perspective of project-based learning".

Second, based on the relevant research results, the research clarifies the design basis of HCI physics items, proposes the design principles for HCI physics items in the perspective of project-based learning, then constructs the design model in the project-based learning perspective for HCI physics items, in which three phase are involved, including "analysis phase ", "construction phase ", " item generation phase ", and a total of 13 elements are included under the three phases.

Further, based on the constructed design model for HCI physics items in the perspective of project-based learning, the research establishes the application process of the design model for HCI physics items, which 12 steps are included, and takes the "Cultivation of Mung Bean Sprouts" HCI physics item as an example to demonstrate the process-based design process of the item. Eventually, two HCI physics items were designed, "Cultivation of Mung Bean Sprouts" and "Pseudo-Holographic Projection", which included the main body of the items and the scoring criteria.

In addition, the validity of the constructed design model for HCI physics items is evaluated. The designed "Cultivation of Mung Bean Sprouts" HCI physics items is created as a program with interactive functions and a HCI Physics Test system is constructed based on it, which is used to conduct a trial test in a middle school, and the response data are collected from 129 Grade nine students. The results of the analysis of the empirical data indicate that the "Cultivation of Mung Bean Sprouts" HCI physics items is of satisfactory quality and validity. Based on the above empirical evidence, it can be inferred that the design model for HCI physics items in the perspective of project-based learning is of great validity and may provide effective instructions for the design of HCI physics items.