| 摘要: | 隨著資訊科技的迅速發展，愈來愈的人在駕駛過程中會使用各種車上資訊輔助系統，因為這些系統會在開車過程中提供許多資訊給駕駛者，會需要使用到駕駛者的注意力，因此在設計的時候，如果未能考慮到人的能力與限制，那麼這些系統不但可能不能幫助駕駛者，甚至還可能會產生一些負面的影響。本研究主要的目的就是希望從人因工程的角度出發，透過8個實驗探討人類注意力選擇的機制以及資訊處理的模式。實驗結果建議，任何車上資訊系統人機界面的設計，都應該考慮資訊處理的瓶頸。 從實驗的結果，可以得到以下結論：(1)刺激的強度對反應時間有顯著的影響，刺激強度愈強，反應的時間愈短；(2)任務愈困難，反應的時間會愈長，因此需要的注意力愈多；(3)周遭環境中突然出現的刺激，會自動吸引人的注意力，這能力使人可以即時發覺突發狀況，但也可能造成無法控制的分心；(4)預告可以加速對刺激的感知過程，但無效的預告卻可能造成反應動作延遲；(5)同時執行多個任務，因為彼此之間的干擾，會造成任務執行的性能下降；(6)瓶頸模式似乎比較符合人類資訊處理模式；(7)目標移動的速度會影響追蹤的績效，但也會使人的注意力更為專注；(8)人聽覺刺激的感知快於視覺刺激，但因為對聽覺方位的判斷不如視覺容易，所以對聽覺的反應動作會慢於視覺；(9)視覺負荷對反應時間造成的損害大於空間相容所帶來的利益，因此在人機界面的設計上，應盡量減少不必要的干擾。 透過本研究我們對人機界面的設計提出以下4點建議：(1)人機界面的設計要考慮到人資訊處理的瓶頸；(2)在需要緊急處理的事件前可先用簡單、顯著的訊號吸引操作者；(3)警示訊號最好能採用多種型態刺激的冗餘設計，如果只能用一種型態，人對聲音的感知速度會比視覺快；(4)視覺負荷會使得反應變慢，因此在人機界面的設計上要盡量減少不必要的訊號。 As information technology continues to develop rapidly, drivers are increasingly using various in-vehicle assistant information systems for navigation, most of which have visual displays. Because these systems will offer a lot of information to drivers while driving, so they will need the driver's attention. If fail to consider people's ability and restriction in the design, then the efficiency of the equipment certainly will be unable to totally give play to, may even induce some negative influence. In order to explore the selective attention mechanism and human information-processing model, eight experiments were carried out basing on ergonomics consideration. Our experiments suggest that any human–machine interface design in driving-associated systems should consider this information-processing bottleneck. The result of those experiments could receive the following conclusions: (1)There was apparent influence in response time in the stimulus intensity, the intensity of stimulus was stronger, the shorter of the response time was; (2)The more difficult the task was, the response time would be longer, so more attention that needed; (3)The stimulus that suddenly appeared in the surrounding environment would capture the participant’s attention automaticaly, this ability enabled people to discover immediately that the state happen suddenly, but it might cause uncontrollable diverting one's attention; (4)The pre-cue could promoted the perception processing, but the invalid pre-cue might cause movements of reaction to postpone; (5) In a dual-task condition, the performance of each component task was worse than that in a single-task condition; (6)The human information-processing model appeared to be a serial bottleneck model; (7)The speed that the target moves would influence the performance of the tracking-task, but would make people's attention more absorbed; (8)The perception of auditory stimulus was faster than a visual stimulus, but because the difficult of judgment to the position of sound was not so easy as the vision, so the reaction for sound would be slow to the vision; (9)Our experiments generally indicated that the loss due to visual load appeared larger than the benefit that came from a compatible spatial effect. This finding implies that the human-machine interface design of in-vehicle systems should avoid irrelevant stimuli. Our study suggests that: (1) Any human machine interface design in systems should consider this information-processing bottleneck; (2) A simple and prominent signal could be used to attract operators' attention prior to the emergent events; (3) Warning signal should adopt various kinds of signal type redundant design, otherwise we should consider sound perception fast than visual; (4) Visual load will delay our response, so we should reduce unnecessary signal on human-machine interface design. |
