de Zubicaray Greig I, McMahon Katie L, Eastburn Matthew M, Wilson Stephen J
Centre for Magnetic Resonance, The University of Queensland, Brisbane, Queensland 4072, Australia.
Neuroimage. 2002 Aug;16(4):1084-93. doi: 10.1006/nimg.2002.1135.
In the picture-word interference task, naming responses are facilitated when a distractor word is orthographically and phonologically related to the depicted object as compared to an unrelated word. We used event-related functional magnetic resonance imaging (fMRI) to investigate the cerebral hemodynamic responses associated with this priming effect. Serial (or independent-stage) and interactive models of word production that explicitly account for picture-word interference effects assume that the locus of the effect is at the level of retrieving phonological codes, a role attributed recently to the left posterior superior temporal cortex (Wernicke's area). This assumption was tested by randomly presenting participants with trials from orthographically related and unrelated distractor conditions and acquiring image volumes coincident with the estimated peak hemodynamic response for each trial. Overt naming responses occurred in the absence of scanner noise, allowing reaction time data to be recorded. Analysis of this data confirmed the priming effect. Analysis of the fMRI data revealed blood oxygen level-dependent signal decreases in Wernicke's area and the right anterior temporal cortex, whereas signal increases were observed in the anterior cingulate, the right orbitomedial prefrontal, somatosensory, and inferior parietal cortices, and the occipital lobe. The results are interpreted as supporting the locus for the facilitation effect as assumed by both classes of theoretical model of word production. In addition, our results raise the possibilities that, counterintuitively, picture-word interference might be increased by the presentation of orthographically related distractors, due to competition introduced by activation of phonologically related word forms, and that this competition requires inhibitory processes to be resolved. The priming effect is therefore viewed as being sufficient to offset the increased interference. We conclude that information from functional imaging studies might be useful for constraining theoretical models of word production.
在图-词干扰任务中，与无关词相比，当干扰词在正字法和语音上与所描绘的物体相关时，命名反应会得到促进。我们使用事件相关功能磁共振成像（fMRI）来研究与这种启动效应相关的脑血流动力学反应。明确解释图-词干扰效应的单词生成的串行（或独立阶段）模型和交互模型假定，该效应的位点在于语音代码检索水平，这一作用最近被归因于左后颞上皮质（韦尼克区）。通过向参与者随机呈现来自正字法相关和无关干扰条件的试验，并获取与每个试验估计的血流动力学反应峰值一致的图像体积，对这一假设进行了检验。在没有扫描仪噪音的情况下出现了明显的命名反应，从而可以记录反应时间数据。对这些数据的分析证实了启动效应。对fMRI数据的分析显示，韦尼克区和右前颞叶皮质的血氧水平依赖性信号降低，而在前扣带回、右眶额内侧前额叶、体感和顶下皮质以及枕叶观察到信号增加。结果被解释为支持单词生成的两类理论模型所假定的促进效应的位点。此外，我们的结果还提出了一些可能性，与直觉相反，由于语音相关词形的激活引入了竞争，正字法相关干扰物的呈现可能会增加图-词干扰，并且这种竞争需要抑制过程来解决。因此，启动效应被视为足以抵消增加的干扰。我们得出结论，功能成像研究的信息可能有助于限制单词生成的理论模型。
