Scholarlink

<p>Process models support analysis, design, implementation, and operation of information systems and must therefore remain understandable for diverse process stakeholders. Numerous metrics estimate model complexity, yet their ability to predict the mental effort (i.e., cognitive load) users actually experience is still unclear. This study addresses that gap with a controlled eye-tracking experiment involving 27 participants and process models that systematically vary in essential complexity (logic) and accidental complexity (layout). Some of these models combine simple and complex parts. Task complexity is explicitly manipulated so that questions target regions of differing complexity. A coarse-grained level analysis tests how model and task complexity relate to cognitive load. Then, a fine-grained level analysis segments eye-tracking data in space and time while investigating users’ visual behavior and cognitive load during model comprehension. Results show that complexity metrics align with cognitive load when the essential and accidental complexity of process models are well captured, and that the task characteristics affect this relationship. Moreover, they show that distinct phases emerge during process model comprehension and that non-task-relevant regions in process models contribute to cognitive load in a non-uniform way (especially early in a task). These findings strengthen measurement practice in Business Process Management (BPM) by establishing an empirical correspondence between a comprehensive metric suite and multimodeal cognitive load indicators and point to adaptive modeling tools that help focus attention on task-relevant parts of process models.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Beyond process model complexity: a multi-granular investigation across time and space based on eye-tracking

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月24日

查看全文
  • Thierry Sorg,
  • Amine Abbad-Andaloussi,
  • Ekkart Kindler,
  • Barbara Weber

摘要

Process models support analysis, design, implementation, and operation of information systems and must therefore remain understandable for diverse process stakeholders. Numerous metrics estimate model complexity, yet their ability to predict the mental effort (i.e., cognitive load) users actually experience is still unclear. This study addresses that gap with a controlled eye-tracking experiment involving 27 participants and process models that systematically vary in essential complexity (logic) and accidental complexity (layout). Some of these models combine simple and complex parts. Task complexity is explicitly manipulated so that questions target regions of differing complexity. A coarse-grained level analysis tests how model and task complexity relate to cognitive load. Then, a fine-grained level analysis segments eye-tracking data in space and time while investigating users’ visual behavior and cognitive load during model comprehension. Results show that complexity metrics align with cognitive load when the essential and accidental complexity of process models are well captured, and that the task characteristics affect this relationship. Moreover, they show that distinct phases emerge during process model comprehension and that non-task-relevant regions in process models contribute to cognitive load in a non-uniform way (especially early in a task). These findings strengthen measurement practice in Business Process Management (BPM) by establishing an empirical correspondence between a comprehensive metric suite and multimodeal cognitive load indicators and point to adaptive modeling tools that help focus attention on task-relevant parts of process models.