Abstract: Attempt to use gating paradigm and analyze data exploringly, the present study examined spoken word recognition of Chinese. The subjects were 47 undergraduates from Beijing Normal University, China. Each Mandarin monosyllable word was presented in duration-blocked format. The acoustic information of each word was increased cross groups with 40msec a gate. The subjects were asked to write down the word being presented and to indicate how confident they were about each guess on a 7-point scale. The results suggested that gating paradigm, as an approach for investigation of spoken word recognition of Chinese, provided rich information, such as isolation point (IP), and the relationship of IP and tone. Furthermore, error analysis showed more details of processing. First, tone can be recognized independently. Second, although tonal information is represented in mental lexicon in terms of distinctive features, these features are not the only constraints in on-line tone recognition. And finally, spoken word recognition, both for segmental and suprasegmental information, is influenced by the temporal order of tone pronunciation

Key words: Gating, spoken words, Chinese, tone