視障者對觸覺數字之辨識研究

A study on the Tactile Number Recognition for Individuals with Visual Impairment

江佩蓉(Pei-Jung Chiang)

指導教授 ： 涂永祥

大同大學/設計學院/工業設計學系所/碩士(2009年)

摘要

本研究希望了解觸覺數字的辨識績效，以期幫助不認識點字或尚未學習點字的視障人士進行辨識。研究分為：「先期實驗」與「電梯按鍵實驗」等兩個階段進行。
「先期實驗」主要議題是探討手部辨識觸覺數字符號的能力，經由量測受測者的辨識時間與辨識錯誤率，了解無視覺狀態下對浮凸數字的辨識績效以及數字的誤判現象。12名明眼人以矇眼方式隨機測試4張不同排列方式（各有40個數字組合）之觸摸數字表。以ANOVA檢定分析性別、數字、數字字型、數字框線設計及數字排列方向對於辨識績效的影響，結果顯示男性在辨識時間及辨識錯誤率上皆優於女性；無框線的數字在辨識時間及辨識錯誤率上皆優於有框線者。線條等粗的字型在辨識時間上優於線條不等粗字型，但是在辨識錯誤率上無顯著差異。排列方向都無顯著影響。前50%的誤判字組合與誤判比率則為1誤判為7(佔11.7%)，9誤判為0(佔10.2%)；6誤判為0(佔8.7%)；8誤判為6(佔5.3%)；3誤判為5(佔4.9%)；7誤判為1(佔4.5%)；5誤判為3(佔3.7%)，最高誤判配對為1與7配對。受測者主觀評量認為外框設計、字型、字體大小、與浮凸高度為造成誤判的因素，其中外框設計被認為是誤判的主要因素。
「電梯按鍵實驗」以模擬電梯操作的方式，由15名視障者測試由點字與觸覺數字所組成之8種電梯按鍵，ANOVA分析結果顯示：「按鍵排列方式」(左點字右數字、左數字右點字、單點字、單數字)、以及「按鍵的排列行數」(單行、雙行)，在辨識時間及錯誤次數上皆無顯著差異。在「認知分類」，只認識點字者辨識績效最佳，只認識數字者最差；在「辨識方式」，無論在辨識時間或辨識錯誤次數，只用數字辨識者。本研究之成果將提供未來無障礙空間的電梯按鍵設計上的參考。

關鍵字

觸覺錯覺 ； 阿拉伯數字 ； 電梯按鍵 ； 觸覺數字符號 ； 視障者

並列摘要

The purpose of this study was to explore the performance of recognition for tactile numbers applied to buttons of elevator. The effect can take care of all of the visually impaired and provide favorable frame of reference for tactile marked for a variety of products.
There were two experiments in this study. The first experiment was to explore the performance of tactile number recognition not only by measuring the recognition rate and the recognition time but also asking the subjects to predict the possible reasons for this kind of tactile illusion. The 12 blindfolded sighted students were asked to recognize 4 different tactile number sheets which had 40 randomized numbers in each. The ANOVA showed that the number without a circle frame had the better recognition rate and time than the number with a frame, and also the male subjects had the better performance than the female. The font with an equal width in all strokes had the significantly lower recognition time than the font without an equal width in all strokes but not at the recognition rate. There was no different found between vertical and horizontal arrangements. The illusion pairs and error rate were: 1 mistaken for 7 (11.7%); 9 mistaken for 0 (10.2%); 6 mistaken for 0 (8.7%); 8 mistaken for 6 (5.3%); 3 mistaken for 5 (4.9%); 7 mistaken for 1 (4.5%); 5 mistaken for 3 (3.7%). The major reasons for these mistakes were subjectively considered to be the number frame, font, size, and tactile height.
The experiment 2 were「elevator control buttons experiment」which simulate operate of the elevator contro1 buttons with 15 participants who were visual impairment to recognize 8 different tactile number sheets. The ANOVA showed that 「arrangement of buttons」 and the 「rows of buttons」without significant differences in recognition time and recognition error.『Recognition time』and『Recognition error』from the levels of cognitive category factor shown significant differences, the people just know Braille had the best performance. The performance of people who just know Arabic numerals was the worst. Recognition time』 and 『Recognition error』from the levels of recognition method factor shown significant differences, The performance of people who recognition only by tactile numbers was the worst.
The results and conclusions might be referred at the design of elevator control buttons and other related research.

並列關鍵字

elevator control buttons ； tactile number ； visually impaired individual ； Arabic numerals ； tactile illusion

參考文獻

碼之人因評估。朝陽科技大學工業工程與管理系，碩士論文。

啟明學校為例。東海大學建築研究所，碩士論文。

[36] 行政院衛生署網站。上網日期98 年6 月。

Surface Area, and Volume by Blind and Sighted Children.

Enhanced in Blindness. The Journal of Neuroscience, 23（8）,