以玩家歷程評估Pokémon Go之移動注意力

Using the Pokémon GO History of Players to Evaluate Mobile Attention During Movement

本研究採用準實驗和參與式觀察二種研究方法，探討不同玩家水準在Pokémon GO遊戲採用的注意力分配。本研究招募60位受測者，依據Pokémon GO的等級區分為五種玩家水準，共同參與一項包括二種設計的注意力實驗：（1）螢幕之內，受測者根據實驗任務必須到訪七個精靈熱區，精靈捕獲數、補給站到訪數和花費時間被視為遊戲表現；（2）螢幕之外，參考Mack與Rock（1998）的不注意視盲實驗，在受測者未被告知的情況之下，安排三個未預期刺激逐地點依序施放，由刺激涵蓋數和資訊回報數交叉比對所得的刺激涵蓋率、未涵蓋率、有效偵測率、不注意視盲率、專注螢幕率和環視周遭率被視為注意力表現。結果顯示：（1）人潮可能成為注意力的干擾因素，顯著降低未預期刺激的有效偵測而強化不注意視盲的效果；（2）根據遊戲表現，五種玩家水準的遊戲策略大致可分為探索型、掠奪型和精準型三種典型；（3）根據注意力表現，非玩家和中量級玩家支持集中注意策略，輕量級、重量級和超量級玩家支持分割注意策略；（4）受測者的注意力策略並非隨著玩家水準而呈線性變動。根據本研究所得，建議AR應用開發者應該考慮使用者歷程而設計與其對應的移動限制機制，以確保使用者的移動安全。

Two types of research method have been conducted in a Quasi-experiment and Participant Observation to explore the attention distribution of players with various trainer-levels in Pokémon GO. The result of this research is expected to be used as a design reference for AR games. 60 subjects were recruited for this research. The players were divided into five groups according to trainer-levels in Pokémon GO. The subjects participated in two designs of attention experiments: (1) Inside the screen, subject is given a mission to visit seven Pokémon hot-zones. The subject's performance is determined by the number of Pokémon caught, visited Pokéstops, and spent time. (2) Outside the screen, this research adapted the Inattentional Blindness experiment of Mack & Rock (1998). Three locations with unexpected stimuli were implemented in order without informing the subjects prior to the experiment. The covered stimuli and reported data were used in comparison for finding out the stimulus covered rate, uncovered rate, effective detection rate, inattentional blindness rate, focused screen rate, and looking around rate as the subject's performance in attention. This research took observations of the subjects on their attention strategy inclined to focused attention or divided attention. The results show that: (1) Crowds may be a confounded factor and obviously reduced the effectiveness of the unexpected stimuli and strengthened the degree of Inattentional Blindness. (2) According to the game performance, the gaming strategies of five player-levels can be divided into three major types, which are Exploration, Predation and Precision. (3) According to the performance of attention, non-players and average players were inclined to the focused attention strategy, while casual, hardcore and heavily-hardcore players were inclined to the divided attention strategies. (4) The subject's attention strategy does not change linearly with the player's level. According to the results, it is recommended that AR application developers should design a corresponding mobile restriction mechanism in consideration of the user history to ensure safe movements of users.