Scholarlink

<p>Traditional methods often struggle to objectively quantify the impact of salary incentives on employees’ productivity, leaving enterprise incentive strategies without a solid scientific foundation. To address this issue, this study innovatively combines recurrence quantification analysis (RQA) with electroencephalogram (EEG) signals, proposing a dynamic incentive evaluation model based on the analysis of brain chaos characteristics. By comparing the EEG signals of workers with and without reward and punishment incentives (control group vs. experimental group), key features such as deterministic (DET) and average diagonal line length (DLL) are extracted to reveal how incentives regulate work efficiency. The experiment shows that RQA diagrams of workers’ EEG under reward and punishment incentives exhibit significantly enhanced chaotic characteristics, with DET and DLL values decreasing by 13.3% and 10.4%, respectively. The accuracy of the twin support vector machine (TWSVM) reaches 98.71%, which is 0.79% and 14.37% higher than existing EEG-based incentive evaluation methods, such as the phase-locking value combined with convolutional neural network (accuracy: 97.92%) and spectral power features (accuracy: 84.34%). This study not only confirms the feasibility of EEG in incentive evaluation but also addresses the insufficient sensitivity of traditional cognitive load monitoring by integrating RQA features and a dynamic classification framework, providing a quantifiable neuroscientific basis for optimizing enterprise incentive mechanisms.</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

Study on incentive mechanism of reward and punishment on work efficiency of PCB welder based on recurrence quantification analysis and electroencephalogram signals

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年4月5日

查看全文
  • Zhang Qian,
  • Mingyue Guo,
  • Fuwang Wang

摘要

Traditional methods often struggle to objectively quantify the impact of salary incentives on employees’ productivity, leaving enterprise incentive strategies without a solid scientific foundation. To address this issue, this study innovatively combines recurrence quantification analysis (RQA) with electroencephalogram (EEG) signals, proposing a dynamic incentive evaluation model based on the analysis of brain chaos characteristics. By comparing the EEG signals of workers with and without reward and punishment incentives (control group vs. experimental group), key features such as deterministic (DET) and average diagonal line length (DLL) are extracted to reveal how incentives regulate work efficiency. The experiment shows that RQA diagrams of workers’ EEG under reward and punishment incentives exhibit significantly enhanced chaotic characteristics, with DET and DLL values decreasing by 13.3% and 10.4%, respectively. The accuracy of the twin support vector machine (TWSVM) reaches 98.71%, which is 0.79% and 14.37% higher than existing EEG-based incentive evaluation methods, such as the phase-locking value combined with convolutional neural network (accuracy: 97.92%) and spectral power features (accuracy: 84.34%). This study not only confirms the feasibility of EEG in incentive evaluation but also addresses the insufficient sensitivity of traditional cognitive load monitoring by integrating RQA features and a dynamic classification framework, providing a quantifiable neuroscientific basis for optimizing enterprise incentive mechanisms.