透過介面及流程設計降低走路時分心使用手機的風險

隨著智慧型手機的普及，手機持有率及行動上網率的提升使得民眾的生活愈來愈離不開手機，甚至在走路時也無可避免。走路分心使用手機會造成視線策略和肢體行為的改變，包含走路變慢、行進方向不穩定、對刺激的反應變慢等現象；追究其原因便會發現主要起因於多工任務下的分心，當邊走路邊使用手機時，注意力資源被手機佔用，導致無法分神注意周遭環境，進而影響自身與他人的安全。基於國內針對走路使用手機議題的研究仍在少數，且目前尚未有強制的規範或有效的方案，因此本研首先將藉由問卷蒐集民眾之相關經驗與主觀意見，並據此擬定改善對策，最後再透過實驗分析使用者在邊走路邊使用手機時的行為變化，以評估方案的有效性。本研究在第一階段先蒐集200份問卷，問卷內容包含民眾之基本背景資料、個人使用手機之習慣及相關經驗，以了解哪些因素會影響個人對「走路使用手機」議題的看法，並探討性別、年齡及不同使用習慣等族群間的差異，同時也調查民眾對10種不同改善方案的接受程度及自覺有效程度。結果顯示20~24歲的年輕人每日使用手機的時間最長，有74%的人每天使用手機超過4小時；此外，儘管大多數民眾（85%）都不贊同此行為，事實上卻有多達69%的民眾表示自己在填答問卷前的兩週內，曾經有走路使用手機之經驗，其中又以使用社群軟體的比例最高（36%），本研究據此結果，將針對民眾走路時使用社群軟體的行為進行分析。而針對問卷中提出的10種方案，最為民眾所接受之方案是「簡潔的介面及流程設計」（3.55分/ 5.00分），而民眾認為最有效之方案為「法令規範強制取締」（3.39分/ 5.00分）。參考問卷之結果，本研究之第二階段首先搭配人類訊息處理模型分析參與者使用社群軟體的流程，探討其注意力資源分配情形。任務中研究參與者首先進行中文打字測驗及英文打字測驗，再使用社群軟體與施測者聊天，並執行其所指定之任務。本研究將其工作流程拆解，分析每一步驟中，注意力資源是如何分配至察覺、工作記憶、長期記憶、反應選擇及反應執行等訊息處理階段。結果發現當操作涉及「輸入文字」的步驟時，所需的注意力會高於其他任務，探究其背後之原因，可能是行人「注意力太過專注意手機操作上」或「多重的注意力切換（switching）」所造成，當使用者必須在手機狹小的螢幕鍵盤上點擊，視線又必須來回核對文字輸入正確與否，倘若再同時聊天，除了文字輸入外，還需思考回應內容，情境又更加複雜，因此更為必需改善之重點。在評估實驗研究中，本研究招募20~24歲、持有智慧型手機且每天至少使用手機超過1小時之本國籍男女性各10位。參與者必須在11種不同手機操作情境，分別為（「畫面鎖定」、「彈出提示」、「無干擾」等三種介入方案，每種方案又各自包含「阻擋通知訊息」、「縮小通知訊息」、「正常顯示訊息通知」等三種次要方案，共九種方案，再加上「未握持手機行走」、「握持手機行走但不操作」等二種未操作手機的情境）下執行走路任務。實驗步道為由長、寬皆為30公分之巧拼組成之長方形步道（5.1公尺×3.6公尺），研究參與者在每個情境下皆需沿著步道逆時針行走4圈，且在實驗過程須穿戴動作追蹤裝置及眼球軌跡追蹤儀，分別用來蒐集參與者的步態指標（步行速度、步行方向偏移變異率）、注意力指標（視線在手機上的停留時間、視線自手機上轉移至環境中、再轉移至手機上的次數、對非預期視覺及聽覺提示的偵測率、對非預期視覺及聽覺提示的反應時間），並輔以SART量表與NASA-TLX量表分別評估參與者之情境察覺程度及工作負荷程度等；此外，實驗中也會出現非預期視覺及聽覺之刺激，並測量偵測率及反應時間，用以評估參與者殘餘的可用注意力資源。研究結果發現，走路時使用手機會顯著地使參與者的步行速度變慢、步態穩定度下降，在注意力方面，則會使參與者對非預期視覺提示的偵測率降低，視線必需分配給手機螢幕，察看環境的時間比例便會減少，且對環境中非預期提示的反應也會變慢；此外，在主觀量表方面，相較於未受手機干擾的兩情境，在分心使用手機的9個情境中，參與者的自覺情境察覺程度均較低。雖然本研究雖未發現達到預期之「找到一有效能降低走路使用手機風險之方案」，但綜合各指標之平均數結果可發現，在「彈出視窗提醒」的設計下，表現均較其他方案差。總而言之，走路使用手機的確會造成行人之生理及心理上的行為改變，也會對行人存在極大的風險，未來宜再深入探討此議題、尋求其他解決之道。

關鍵字

多工任務；注意力；行人安全；步態；視線

並列摘要

With the popularization of smartphones, many people cannot live without smartphones, even when walking. Using smartphones while walking may distract the users and affect their gaze and gait behavior such as walking more slowly and more unstably, taking a longer time to respond to the unexpected stimulus. The reason behind these behavioral changes is distraction from multi-tasking. When people operate smartphones while walking, their attention will be occupied. This may hinder people from allocating appropriate attention to the surrounding environment, which leads to the potential risks for both the user him-/her-self and other road users. However, the issue of distracted walking hasn’t been investigated much, and hence there is no compulsory regulation or effective solution. This study surveyed people’s related experience and subjective opinions first. Based on the results of the questionnaire, corresponding countermeasures were then generated. Subsequently, an objective experiment was conducted to analyze the behavioral changes due to distracted walking by smartphones in order to evaluate the effectiveness of the proposed solutions. In the first phase, a questionnaire survey completed by 200 respondents was used to collect self-reported information, experience of distracted walking and perspective toward the danger of distracted walking by smartphones. Each individual’s acceptance and perceived effectiveness toward 10 different solutions were also included. The results showed that among all age groups, people aged between 20 to 24 spent the longest time using smartphones per day (over 4 hours). Though most of the respondents (85%) disagreed with using smartphone while walking, 69% of the respondents reported the experience of using smartphone while walking within the last two weeks. For those who had the experience of distracted walking by smartphones, 36% of them were using social media applications. Hence, this study is mainly focused on the analysis of people’s behavior when distracted walking using social media applications. In addition, the most acceptable solution was “concise interface and workflow design” (scored 3.55/ 5.00), whereas “banning smartphone use while walking” was considered as the most effective solution (scored 3.90/ 5.00). Based on the results of questionnaire survey, the human information processing (HIP) model was used to analyze the process of using social media applications in the second phase. Participants were asked to operate Mandarin or English typing test, while chatting with the test giver. This study analyzed how the attention was allocated to different information processing components like perception, long-term memory, working memory, decision and response selection, and response execution. Findings showed that tasks associated with typing would lead to increased attentional demands. The possible reasons might be “over-focus on the smartphone” and “frequent gaze switching.” Simultaneously typing on the small virtual keyboard, checking typos and pondering over the incoming message make the tasks much more complex. Therefore, it requires improvement with a higher priority. In the simulation experiment, 20 young adults (10 male and 10 female) ranging from 20 to 24 years old, owning smartphones and operating the smartphones more than 1 hour per day were recruited. Participants were asked to complete the walking tasks under 11 different scenarios. There were three major scenarios (locked screen/ pop-up warning/ no interference involved), and under each program there were three minor scenarios including blocked notification, minimized notification, and displayed notification as usual. In addition to those 9 scenarios, there were 2 other scenarios without the operation of smartphones (free walking/ holding a smartphone without operation). Participants were asked to operate social media applications on smartphone while walking counter-clockwisely along a 5.1-meter by 3.6-meter rectangular track (with the width of 0.3 meters) for 4 laps. During the experiment, participants were asked to wear motion capturing suits and wireless eye tracking glasses in order to collect their gait data (walking speed, variability of medial-lateral deviation), attention allocation (the proportion of fixation time on the smartphone, number of gaze shifting away from smartphone and back to smartphone) and subjective responses (Situational Awareness Rating Technique score and NASA-Task Load Index). Besides, there were also unexpected visual and auditory stimulus given in the experiment. By measuring participant’s detection rate and reaction time to the stimulus, their residual attention can be estimated. This study has identified that when operating smartphones while walking, participants were generally found with decreased walking speed and more medial-lateral deviation. In addition, distracted walking by smartphone would lead to a higher probability of failing to notice the unexpected visual and auditory stimulus, less time spent looking over the environment, and slower reaction to the unexpected visual stimulus. Besides, when comparing with the 2 scenarios without operating smartphones, participants’ subjective situational awareness scores were lower than in other 9 scenarios involving smartphone operation. Although no certain solution was found to be able to effectively reduce the risks of distracted walking, the findings did suggest that solutions involving pop-up warning might be worse than others. Overall, distracted walking by smartphones will result in people’s behavioral change both physiologically and psychologically, leading to the serious risks for pedestrians. Further explorations on this issue are recommended to find out effective solutions for distracted walking by smartphones.