Rosa Senatore

A Neural Scheme for Procedural Motor Learning of Handwriting

Handwriting analysis, which requires the detection and examination of distinctive features within the ink traces representing the words, provides a valuable help in several researchs fields. In medical field, handwriting analysis provides an useful complement to other clinical investigations in diagnosing many movement disorders, such as Parkinsons disease. In forensics, the examination of particular characteristics of the ink trace allows the expert to evaluate the authenticity of an handwritten text. Handwriting recognition, which allows to optimize the handling of manuscript documents, requires the detection of distinguishing features for interpreting the characters the ink trace represents. Since any phenomenon can be better understood and analyzed when the generative process is known, investigating the process that underlies handwriting might give some guidelines for handwriting analysis. In this respect, we propose a neural scheme, envisaging that performing complex motor sequences, such as handwriting, requires the interaction among the Cortex, Basal Ganglia and Cerebellum.

Reading Cursive Handwriting

We present a method for off-line reading of cursive handwriting, which derives from modelling handwriting as a complex movement. The method includes a step for recovering the writing order from static images of handwriting, a segmentation algorithm that decomposes the “unfolded” ink into strokes, an ink matching step to compare the ink of the unknown handwriting with those of a set of reference words, of whom the transcripts are given, and a graph search algorithm to search for the best interpretation among the possible ones. The method does not involve any feature extraction, nor a classification stage and may benefit from a linguistic context, if available. We report the results of experiments on 8,000 samples, draw some conclusions and outline further developments.

A paradigm for emulating the early learning stage of handwriting: Performance comparison between healthy controls and Parkinson’s disease patients in drawing loop shapes

We present a novel paradigm, aimed at emulating the early stage of handwriting learning in proficient writers, by asking them to produce a familiar shape through a novel (unfamiliar) motor plan. Handwriting of beginner writers is characterized by slower movements, reduced spatial precision, lower fluency and reduced force regulation compared to those observed in the handwriting production of proficient writers. Features observed in the ink trace obtained with the novel motor plan and performance comparison of the handwriting obtained by familiar and unfamiliar motor plan suggest that the proposed paradigm is able to elicit non-automated movements in proficient writers. As that produced by beginner writers, handwriting of Parkinsons disease (PD) patients is characterized by lack of fluency, slowness and abrupt changes of direction. Furthermore, PD patients show impaired performance in learning novel motor behaviors, as well as in executing motor behaviors acquired before the onset of the disease. We used the proposed paradigm for comparing the performance achieved by healthy controls in writing a familiar shape through a novel motor plan with those obtained by PD patients performing a well-known motor plan for drawing the same shape. Our analysis points out some similarities between performance obtained by healthy controls and those obtained by PD patients, sustaining the hypothesis that the fine tuning of the motor plan parameters involved in the handwriting production is impaired by PD.