本研究以相似性測量的角度來探討智慧型手機手勢操作與使用者認知間的差異。手勢辨識是一 種在計算機科學與程式語言技術的主題，透過數學的邏輯運算解釋人類手勢的動作，以達到運用手 勢來操作某種數位資訊及功能。現在這種技術被廣泛運用在智慧型手機上。隱喻是用來協助表達或 理解另外一種想法或事物的替代，意指此物即某物。互動設計師常以手勢符號或動作（即隱喻）來 實現他們的想法或感受，並以此概念作為人機介面。然而這樣的人機介面對使用者來說可能不易操 作；因為不同的使用者對同一件事物有不同的認知。因此本研究試圖探討怎麼樣的手勢操作才能將 使用者的認知負擔降到最低，讓使用者不必花太多時間來學習手勢所代表的意涵。研究的結果發現 (1) 智慧型手機之手勢操作越接近真實世界之操作行為，則越能降低使用者之認知負擔； (2) 設計師 在設計手勢操作時，應盡量避免使用相同手勢進行不同功能之操作，以免造成使用者混淆。

This paper explores the difference between Smartphone’s gesture control and users’ cognitive by similarity measurement. Gesture recognition is a topic in computer science and programming language with the goal of interpreting human gestures via mathematical algorithms. Now it is highly used in mobile phone. Metaphor is defined as the substitution of one idea or object with another, used to assist expression or understanding. Interactive designers usually use symbols of hand gesture (i.e., metaphor) to imply behavior about their thoughts and feelings and implement these concepts as a human-computer interface. However, this user interface may be not easy to operate for different users because of different cognitions. The conclusions are as follows: a) if the gesture control is more similar to personal behavior, the users’ cognition load will be minimized; b) Designers should avoid the metaphor of different functions with the same gesture when they design devices. It can avoid users to be confused.