Borowsky R, Owen W J, Fonos N
Department of Psychology, University of Saskatchewan, Saskatoon.
Can J Exp Psychol. 1999 Dec;53(4):294-305. doi: 10.1037/h0087317.
Current models of reading and speech perception differ widely in their assumptions regarding the interaction of orthographic and phonological information during language perception. The present experiments examined this interaction through a two-alternative, forced-choice paradigm, and explored the nature of the connections between graphemic and phonemic processing subsystems. Experiments 1 and 2 demonstrated a facilitation-dominant influence (i.e., benefits exceed costs) of graphemic contexts on phoneme discrimination, which is interpreted as a sensitivity effect. Experiments 3 and 4 demonstrated a symmetrical influence (i.e., benefits equal costs) of phonemic contexts on grapheme discrimination, which can be interpreted as either a bias effect, or an equally facilitative/inhibitory sensitivity effect. General implications for the functional architecture of language processing models are discussed, as well as specific implications for models of visual word recognition and speech perception.
当前的阅读和言语感知模型在语言感知过程中关于正字法和语音信息相互作用的假设上存在很大差异。本实验通过二选一的强制选择范式研究了这种相互作用，并探讨了字素和音素处理子系统之间联系的本质。实验1和实验2证明了字素语境对音素辨别具有促进主导作用（即益处超过成本），这被解释为一种敏感性效应。实验3和实验4证明了音素语境对字素辨别具有对称影响（即益处等于成本），这可以被解释为一种偏差效应，或者是同等促进/抑制的敏感性效应。文中讨论了对语言处理模型功能架构的一般影响，以及对视觉单词识别和言语感知模型的具体影响。
