在充滿競爭的社會中人們一直在跟時間比賽，所以無論是學生、各業上班族、甚至服務業等族群，都會面臨時間造成的壓力，不但生活緊張、增加意外的風險亦對健康造成不利的影響。本研究目的在探討時間壓力對認知功能的影響，選用兩種不同時間分別為9秒（長時間）及5秒（短時間），來當作時間壓力的條件進行心算作業，受測者進行心算作業時以腦波（electroencephalogram, EEG）和NASA-TLX工作負荷評估量表（NASA-Task Load Index）進行客觀及主觀之量測。腦波量測分析包括：用以評估認知處理能力之腦波事件關聯電位（Event-Related Potential, ERP），及評估覺醒水準之腦波頻譜，分析評估時間壓力下之認知反應變化。本研究徵求自願之受測者，年齡介於20-28歲之間。心算作業前量測EEG與NASA-TLX工作負荷評估量表，實驗開始後進行兩階段的長時間和短時間心算作業同時進行腦波量測，各階段心算作業後進行旁側夾擊作業以量取事件關聯電位，並填寫NASA-TLX 工作負荷評估量表。研究結果顯示腦波比率指標β/α在長時間及短時間心算作業後皆顯著上升即覺醒水準顯著升高，但在長、短時間心算作業沒有顯著差異，而ERP P300成份波部分，在潛時部分，長時間心算作業之潛時較短，振幅亦較大，顯示有較差之處理能力。在NASA-TLX 工作負荷評估量表在長時間及短時間比較是有顯著差異的變化，受測者在主觀量測都覺的5秒比9秒之時間限制心算作業更有壓力。因此，本研究發現時間壓力下受測者除感受到心智負荷高之外認知反應亦變差。 In our competition society, people have to compete with time. No matter the students, office workers, service staff etc.,, will face the time pressure. The purpose of this study is to explore the effect on cognition response on time pressure. By using EEG, two sessions of experimental tasks with different response times including 9 seconds (long term) and 5 seconds (short term) for responding mental arithmetic were used to serve as two time pressure conditions. Twenty college students aged from twenty to twenty-eight years old recruited for subjects participated the experiment testing. Before experiment the subjects must fill in NASA-TLX rating scale and undertook their electroencephalogram (EEG) measured. The EEG data including event related potential (ERP) and indices of EEG power spectrum reflected respectively the recognition depth and arousal level were used to evaluate the recognition response. Before mental arithmetic tasks, the participants were measured their EEGs and NASA-TLX rating scales. Two sessions of mental arithmetic tasks were performed in counterbalance and the EEGs were measured at the same time during the experimental test. After each session, the participants undertook the flanker tasks to measure their ERP and filled in NASA-TLX rating scales. The results showed the EEG power ratio index β/α were increased, that is, the arousal level arose during mental arithmetic tasks. But there was no significant difference between long term and short term sessions at EEG power indices. For P300 components of ERP, the latency for long term of mental arithmetic task was shorter than that for short term, and the amplitude for long term of mental arithmetic task was higher than that for short term. It manifested that the participants had better treatment for long term of mental arithmetic task. For NASA-TLX rating scales measurement, the participants felt more pressure for short term of mental arithmetic task than that for long term. We found the participants revealed higher mental workload and deteriorated recognition response under higher time pressure.