کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
931774 | 1474634 | 2015 | 19 صفحه PDF | دانلود رایگان |
• Deletions in conversational speech question a tight phonology–orthography interface.
• Reduction costs of orthographically coded and uncoded segments were compared.
• Orthographic coding does not influence reduction costs in spoken-word recognition.
• Orthographic effects arise in an explicit judgment task with clear speech.
• Orthography may only influence speech processing in unnatural situations.
It has been claimed that learning to read changes the way we perceive speech, with detrimental effects for words with sound–spelling inconsistencies. Because conversational speech is peppered with segment deletions and alterations that lead to sound–spelling inconsistencies, such an influence would seriously hinder the perception of conversational speech. We hence tested whether the orthographic coding of a segment influences its deletion costs in perception. German glottal stop, a segment that is canonically present but not orthographically coded, allows such a test. The effects of glottal-stop deletion in German were compared to deletion of /h/ in German (grapheme: h) and deletion of glottal stop in Maltese (grapheme: q) in an implicit task with conversational speech and explicit task with careful speech. All segment deletions led to similar reduction costs in the implicit task, while an orthographic effect, with larger effects for orthographically coded segments, emerged in the explicit task. These results suggest that learning to read does not influence how we process speech but mainly how we think about it.
Journal: Journal of Memory and Language - Volume 85, November 2015, Pages 116–134