Perception of Dynamic Information in Static Handwritten Forms

Abstract

Handwriting recognition presents a problem for most theories of letter perception. How do people accurately discriminate letters given the variability in handwritten forms? We propose that perceivers detect and capitalize on information relating to production inherent in the static trace of the handwritten letter. In an "implicit detection" task, subjects' ability to detect stroke direction in a series of handwritten characters was found to be influenced by the particular drawing method used to generate the stimuli. Similar results were obtained in an "explicit detection" task in which the subjects were asked to speculate on the drawing method used to generate the samples. Our findings suggest that perceivers are able to extract information relating to production from the static trace. We propose that they can then use this information in conjunction with their knowledge of a common production process, shared by both producers and perceivers, to aid character recognition.