社交機器人之個性塑造及其對使用者感受及信任度之影響

隨著機器人科技的進步與普及，社交機器人服務不斷擴大其應用範疇，使用者對機器人的需求慢慢由功能提升到社交與情感的層級，機器人的信任度議題逐漸被重視。社交機器人開發團隊會運用個性、性別…等社會特徵，賦予機器人更顯明的擬人化特質。由於機器人可以透過言語或非言語不同溝通管道來傳達訊息給使用者，而過去人際互動的研究指出當一方個性表現不一致的情況下，會造成另一方的疑慮，進而減少後續互動的願意。因此本研究利用UBTECH Alpha 1S機器人作為遊戲關主的實驗設計，藉由不同語音與動作的組合的機器人與使用者互動，來探索機器人的互動元素與機器人的個性感受的關聯，並進一步調查機器人個性表現不一致時對使用者信任感受、行為及互動印象之影響。整體研究包括兩個部分，第一部分前導研究，以機器人的聲音與動作的四個變項：音頻（高、低）、語速（快、慢）、音量（大、小）、手勢（大、小），組成了2x2x2x2=16個機器人樣本，透過線上問卷進行調查，共收到85份有效樣本進行分析，來確立機器人互動元素與大五（Big Five）個性之關連。第二部分是正式實驗，以前導研究中最外向與最內向性之機器人語音與手勢作為實驗的控制變項，產生四種實驗條件的遊戲關主機器人：(A)外向語音與外向手勢、(B)內向語音與外向手勢、(C)內向語音與內向手勢、(D)外向語音與內向手勢。實驗採受試者間設計，招募92位參與者與不同實驗條件下的機器人進行沙漠生存遊戲，遊戲後再進行信任及印象量表之填寫。藉此探討使用者對內、外向個性表現機器人信任感受、行為與互動印象之差異，以及個性表現一致性與否的影響。研究結果發現：(1)參與者對個性表現一致的機器人的信任評分較高；(2)機器人個性表現一致與否對參與者信任行為影響不大；(3)機器人個性表現一致與否對參與者的互動印象亦無明顯影響；(4)參與者對內向手勢機器人在生命力、擬人化等項目都有較高評分。論文最後也對未來社交機器人個性設計之相關研究提出建議。

關鍵字

社交機器人；人機互動；個性表現；信任度；互動印象

並列摘要

With the advancement and popularization of robotics technology, social robot services continue to expand their application areas. The demand for robots has gradually increased from functional to social and emotional levels, and the trustworthiness of robots has gradually valued. Social robot developers often use social characteristics, such as personality and gender, to give robots a more explicit anthropomorphic trait. In addition, past research on interpersonal interactions pointed out that when a person expresses an inconsistent personality, it would cause the other party’s doubts, and thus reduce their willingness of subsequent interaction. Robots can express messages to users through verbal or non-verbal communication, so this study aims to explore the effect of robot’s personality consistency on users’ perception and trust by using the UBTECH Alpha 1S robot as a facilitator in “the desert survival game”. The study consists of two parts, the pilot study and the formal experiment. In the pilot study, 16 robot samples consisting of four variables, audio (high / low), speech rate (fast / slow), volume (large / small), and gesture (large / small), were made. Through an online questionnaire based on the Big Five model of personality, 85 valid subjects' ratings of each sample were collected and analyzed to establish the relationship between the interaction elements of robots and the Big Five personality traits. The second part is the formal experiment to explore the influence of the robot’s personality consistency on the users’ perception, trust, and impression toward the robot facilitator. The most extroverted and introverted robot voice and gesture in the pilot study were used as the experimental control variables. A between-groups experiment was conducted, 92 participants were recruited to interact with one of the four robots with distinct personality cues: (A) extroverted speech and extroverted gesture, (B) introverted speech and extroverted gesture, (C) introverted speech and introverted gesture, and (D) extroverted speech and introverted gesture. The results of the study showed: (1) participants gave higher trust scores for robots with a consistent personality; (2) the consistency of robots’ personality did not significantly affect participants’ trust-related behavior; (3) the consistency of robots’ personality had no influence on participants’ interactive experience; (4) the robots with the introverted gesture received higher scores on the animacy and anthropomorphism from participants. Finally, suggestions for future work in social robot personality design are provided based on this study.

並列關鍵字

Social Robots ； Human-Robot Interaction ； Personality ； Trust ； Impression Perception

參考文獻

網路文獻

Google Scholar

1. Business Insider Japan（2018年10月06日）80億円の期待を背負った家族型ロボットが12月18日発表 ── Pepperに次ぐ第2子誕生へ。線上檢索：2018年12月12日。網址：https://www.businessinsider.jp/post-180818

Google Scholar

2. EE Times Taiwan（2018年03月日01）。不斷進化　機器人可以為人類「做」更多。線上檢索：2018年12年12日。網址：https://www.eettaiwan.com/news/article/20180301NT31-robots-can-do-more-for-humans

Google Scholar

3. EE Times（2017年11月30日）。機器人十大預測：未來5年將改變什麼？線上檢索：2018年12月11日。網址：https://www.eettaiwan.com/news/article/20171130NT01-10-predictions-for-robotics-in-5-years

Google Scholar

4. LGNewsroom（2017年07月21日）LG AIRPORT ROBOTS TAKE OVER KOREA’S LARGEST AIRPORT。線上檢索：2018年12月12日，網址：http://www.lgnewsroom.com/

Google Scholar

國際替代計量

社交機器人之個性塑造及其對使用者感受及信任度之影響

全文下載

主題瀏覽