作者: John D. Gould
DOI: 10.1016/S0020-7373(82)80054-0
关键词: Speech recognition 、 Key (music) 、 Constant (mathematics) 、 Composition (language) 、 Linguistics 、 Computer science 、 General Engineering
摘要: Twelve participants composed written and spoken letters under various conditions. In the first five experiments they were told which to write speak. last three could choose their method of composition. Results showed that speaking required only 35–75% time writing did. Written rated as about equally effective, being characterized more by similarities than differences. When a method, did not always select said would those circumstances. A key theoretical result was spent planning constant amount in both methods, but rather ratio. Planning two-thirds total composition time, regardless letter complexity.