|
The purpose of this study is to investigate the influence of the
stroke count, the radical component, the complex component and
the componential structure on the similarity judgement of Chinese heir shape.
characters in terms of their shapes. The complex component refers eater than h
to the component of a character whose number of strokes is greater stricted to
than half of the total stroke counts, and the componential structure ilarity t
is restricted to the two most salient dimensions: left-right/up-down tions of
and enclosed/open, as revealed in Yeh et al. (1997). Two kinds of (reaction-ti
similarity tasks were adopted, one is discrimination task, with two ied multid
durations of stimulus presentation: 19-30 msec (threshold level) and ts werese
500-1300 msec (reaction-time level); the other is sorting task with e other wi
unlimited duration. We applied multidimensional scaling (MDS) to ctors have di
analyze the similarity data. Two kinds of sample characters were used,troke co
one with stroke counts ranged from 3 to 16, the other with more ct of radical
restricted range of stroke counts. Our results suggested that the four ons of
factors play different roles on the similarity judgement of Chinese but not a
characters. The stroke counts dominated on all kinds of experimantal ence at u
situations. The radical component and complex component had observable ecessit
influences when the durations were at threshold level and reaction-time
level, but had little effect using sorting task with unlimited orthographic s
duration. The componential structure had an effect when using the
sorting task with unlimited duration. Base on our study, we suggest
the necessity of considering both the experimental tasks and the
durations of stimulus presentation when selecting Chinese characters
as stimuli in terms of their similarity of shapes.
|