Wriessnegger S C, Leitner M, Kostoglou K
Institute of Neural Engineering, Graz University of Technology, Graz, Austria; BioTechMed-Graz, Graz, Austria.
Institute of Neural Engineering, Graz University of Technology, Graz, Austria.
Brain Cogn. 2024 Dec;182:106239. doi: 10.1016/j.bandc.2024.106239. Epub 2024 Nov 17.
Stress is an increasingly dominating part of our daily lives and higher performance requirements at work or to ourselves influence the physiological reaction of our body. Elevated stress levels can be reliably identified through electroencephalogram (EEG) and heart rate (HR) measurements. In this study, we examined how an arithmetic stress-inducing task impacted EEG and HR, establishing meaningful correlations between behavioral data and physiological recordings. Thirty-one healthy participants (15 females, 16 males, aged 20 to 37) willingly participated. Under time pressure, participants completed arithmetic calculations and filled out stress questionnaires before and after the task. Linear mixed effects (LME) allowed us to generate topographical association maps showing significant relations between EEG features (delta, theta, alpha, beta, and gamma power) and factors such as task difficulty, error rate, response time, stress scores, and HR. With task difficulty, we observed left centroparietal and parieto-occipital theta power decreases, and alpha power increases. Furthermore, frontal alpha, delta and theta activity increased with error rate and relative response time, while parieto-temporo-occipital alpha power decreased. Practice effects on EEG power included increases in temporal, parietal, and parieto-occipital theta and alpha activity. HR was positively associated with frontal delta, theta and alpha power whereas frontal gamma power decreases. Significant alpha laterality scores were observed for all factors except task difficulty and relative response time, showing overall increases in left parietal regions. Significant frontal alpha asymmetries emerged with increases in error rate, sex, run number, and HR and occipital alpha asymmetries were also found with run number and HR. Additionally we explored practice effects and noted sex-related differences in EEG features, HR, and questionnaire scores. Overall, our study enhances the understanding of EEG/ECG-based mental stress detection, crucial for early interventions, personalized treatment and objective stress assessment towards the development of a neuroadaptive system.
压力在我们的日常生活中所占比重日益增大，工作中或对自身的更高绩效要求会影响我们身体的生理反应。通过脑电图（EEG）和心率（HR）测量能够可靠地识别出升高的压力水平。在本研究中，我们考察了一项诱发算术压力的任务如何影响脑电图和心率，在行为数据与生理记录之间建立了有意义的关联。31名健康参与者（15名女性，16名男性，年龄在20至37岁之间）自愿参与。在时间压力下，参与者完成算术计算，并在任务前后填写压力问卷。线性混合效应（LME）使我们能够生成地形关联图，显示脑电图特征（δ、θ、α、β和γ功率）与任务难度、错误率、反应时间、压力分数和心率等因素之间的显著关系。随着任务难度增加，我们观察到左中央顶叶和顶枕叶的θ功率降低，α功率增加。此外，额叶α、δ和θ活动随错误率和相对反应时间增加，而颞顶枕叶α功率降低。脑电图功率的练习效应包括颞叶、顶叶以及顶枕叶的θ和α活动增加。心率与额叶δ、θ和α功率呈正相关，而额叶γ功率降低。除任务难度和相对反应时间外，所有因素均观察到显著的α偏侧性分数，表明左顶叶区域总体增加。随着错误率、性别、试验次数和心率增加出现显著的额叶α不对称，并且随着试验次数和心率增加也发现枕叶α不对称。此外，我们还探讨了练习效应，并注意到脑电图特征、心率和问卷分数方面的性别差异。总体而言，我们的研究增进了对基于脑电图/心电图的心理压力检测的理解，这对于早期干预、个性化治疗以及朝着神经自适应系统发展的客观压力评估至关重要。
