The development of a standardized, computer-based abbreviation algorithm

Abstract

Abstract Based upon three simple assumptions about the information-conveying properties of consonants, vowels, syllables and the relative positions of letters and letter groups within parent words, a prototype abbreviation-generating algorithm was constructed. It was reasoned, based on previous studies (Streeter et al. 1980, Reber 1976) and because of the invariant rules which are fundamental to the algorithm, that abbreviations produced by the algorithm should be more readily decoded than naturally produced, popular abbreviations. Experiment I revealed that popular abbreviations were more easily reconstructed than were their rule-based counterparts, although the margin was reliable only for abbreviations of parent words of relatively long letter length. Results of experiment II demonstrated that when feedback was provided to the subject following each reconstruction, accuracy in decoding the algorithm abbreviations improved with practice. No improvement was observed for subjects who were not provided kno...