Liu L, Arditi A
The Arlene R. Gordon Research Institute, Lighthouse International, New York, New York 10022, USA.
Optom Vis Sci. 2001 Jan;78(1):50-5. doi: 10.1097/00006324-200101010-00014.
Acuity for letter recognition is known to be worse when multiple letters are presented with narrow interletter spacings than with wide spacings. How would interletter spacing affect the kind of errors made by human subjects?
Five-letter strings that were randomly drawn from the 26 uppercase letters of the English alphabet were presented foveally to the subjects. The interletter spacings were 1.0 and 0.1 letter height. Letter confusion matrices were constructed from the data collected using these spacing conditions.
Narrow- and wide-spacing letter strings produced different letter confusion matrices. Aside from the letter confusions that were shared by both wide- and narrow-spacing strings, narrow-spacing strings produced more random confusions and a set of unique letter confusions, which was not observed under the wide-spacing condition.
Increased random guessing and lateral interactions between features of neighboring letters can account for most of the acuity deterioration observed under the narrow-spacing condition.
已知当多个字母以窄字母间距呈现时，字母识别的敏锐度比宽间距时更差。字母间距会如何影响人类受试者所犯错误的类型？
从英语字母表的26个大写字母中随机抽取的五个字母的字符串向受试者中央凹呈现。字母间距为1.0和0.1个字母高度。根据使用这些间距条件收集的数据构建字母混淆矩阵。
窄间距和宽间距的字母串产生了不同的字母混淆矩阵。除了宽间距和窄间距字符串共有的字母混淆外，窄间距字符串产生了更多的随机混淆和一组独特的字母混淆，这在宽间距条件下未观察到。
随机猜测的增加以及相邻字母特征之间的横向相互作用可以解释在窄间距条件下观察到的大部分敏锐度下降。
