Phonological Prediction Data

The eye-tracking technique and printed-word version of the visual world paradigm were used to examine whether the phonological forms of incoming words would be predicted during speech comprehension. Participants were instructed to listen to Mandarin Chinese spoken sentences while looking at two single-character printed words on the screen, with one being the critical word and the other being an unrelated distractor word. Each of the spoken sentences includes a highly predictable target word. The phonological relationship between the predictable target word and the critical word was systematically manipulated. Four types of the printed critical words were included: predicted Target Word itself, Homophone Competitor (sharing the same pronunciation “consonant-vowel-lexical tone” with the target word), Tonal Competitor (only sharing the same lexical tone with the target word), or an Unrelated Word. One group of participants were required to perform a “word judgment” task (whether the spoken sentence mentions any of the words in the visual scene) in experiment 1a (exp1a_evs.rar), while another group of participants were required to perform a “pronunciation judgment” task (whether the spoken sentence mentions any of the pronunciation of the words in the visual scene) in experiment 1b (exp1b_evs.rar). In order to explore the underlying mechanism of phonological preactivation, particularly whether the phonological information is preactivated automatically or strategically, the third group of participants were required to perform “word judgment” task (exp2_task1_evs.rar) in one block and “pronunciation judgment” task (exp2_task2_evs.rar) in another block. The order of the task was counterbalanced across participants. In addition, only two types the critical words were included: phonological competitors and unrelated words.本研究采用眼动追踪技术与印刷词版本的视觉世界范式（visual world paradigm），旨在考察言语理解过程中是否会对输入词汇的语音形式进行预测。被试需在注视屏幕上呈现的两个单字印刷词的同时，聆听汉语普通话语句，其中一个为关键刺激词，另一个为无关干扰词。每段口语语句均包含一个高度可预测的目标词，关键刺激词与该可预测目标词之间的语音关联被系统操控。本次实验共设置四类印刷版关键刺激词：目标词本身、同音竞争者（与目标词共享完全相同的“辅音-元音-词汇声调”发音）、调型竞争者（仅与目标词共享相同的词汇声调），以及无关词。第一组被试在实验1a（exp1a_evs.rar）中完成“词汇判断任务”，即判断口语语句是否提及视觉场景中的任一词汇；第二组被试在实验1b（exp1b_evs.rar）中完成“语音判断任务”，即判断口语语句是否提及视觉场景中任一词汇的语音形式。为进一步探究语音预激活的潜在机制，尤其是语音信息的预激活是自动发生还是策略性产生，第三组被试需在一个实验区块中完成“词汇判断任务”（exp2_task1_evs.rar），在另一区块中完成“语音判断任务”（exp2_task2_evs.rar）。任务顺序在被试间进行了平衡。此外，本实验仅设置两类关键刺激词：语音竞争者与无关词。