SPEECH AND LANGUAGE PROCESSING IN APHASIA

本研究的主要目标是通过调查失语症患者的言语和语言加工缺陷来识别和描绘失语症患者说话和理解的过程和机制以及神经基础。研究的重点主要集中在语音结构和词汇，以及将声音映射为意义、将意义映射为声音的加工阶段。关于将语音映射到意义的加工阶段，有人假设失语症患者的缺陷源于语音结构在词汇中引起激活的程度。两个潜在的损害来源将被调查，一个集中在词汇本身的激活模式上，另一个集中在语音结构到词汇形式的映射上。词汇竞争对词汇加工的影响将通过语音类别“好”及其对词汇竞争者激活的影响，词汇密度和频率对词汇决定潜伏期的影响，以及韵律和队列效应对口语识别时间进程的影响来考察。从声音结构到词汇的映射将通过探索声学输入的各种变化对词汇通达的影响来进行，包括声学失真和标引特征如音调和说话人的影响。将使用各种方法，包括词汇决定、语义启动、辨别和眼球跟踪。关于言语产生的过程和机制，假设包括Broca失语症在内的前部失语症患者的言语输出缺陷是以发音执行障碍为基础的。为了进一步阐明这种损伤的性质，将检验三个假说：1.提出需要在短时间内对发音动作进行复杂整合的声音片段的产生将出现缺陷。研究发音方式的产生，包括停顿、塞音、擦音和滑音；2.研究表明，协同发音的缺陷将出现在更大、更复杂的语言环境中，特别是跨越单词边界。探讨元音协同发音对前辅音发音的影响，以及辅音发音对前元音发音的影响；3.提出由前失语症引起的音素性错语，即语音替换错误，具有语音或发音基础。在所有这些发音研究中，都对前后失语症患者的语音发音进行了声学分析，以推断导致发音模式的发音状态。

The broad goals of this research are to identify and delineate the processes and mechanisms and the neural substrates underlying speaking and understanding by investigating speech and language processing deficits in aphasic patients. The major focus of research is on sound structure and the lexicon and the processing stages that map sound to meaning and meaning to sound. With respect to the processing stages that map sound to meaning, it is hypothesized that the deficits of aphasic patients stem from the extent to which sound structure elicits activation in the lexicon. Two potential sources of such impairment will be investigate, one focusing on activation patterns within the lexicon itself and the other on the mapping from sound structure to lexical form. The influence of lexical competition on lexical processing will be investigated by exploring the effects of phonetic category 'goodness' and its influence on the activation of its lexical competitor, the effects of lexical density and frequency on lexical decision latencies, and rhyme and cohort effects on the time course of spoken word recognition. The mapping from sound structure to the lexicon will be investigated by exploring the influence of various changes to the acoustic input on lexical access including the influence of acoustic distortion and indexical features such as voice pitch and speaker. Various methods will be used including lexical decision, semantic priming, discrimination, and eye tracking. With respect to the processes and mechanisms contributing to speech production, it is hypothesized that the speech output deficits of anterior aphasics including Broca's aphasics have as their basis impairments in articulatory implementation. To further map out the nature of this impairment, three hypotheses will be tested: 1. It is proposed that deficits will emerge in the production of those sound segments that require the complex integration of articulatory movements over short periods of time. The production of manner of articulation including stops, affricates, fricatives, and glides will be investigated; 2. It is proposed that deficits in coarticulation will emerge in larger and more complex linguistic contexts, specifically across word boundaries. The influence of vowel coarticulation on a preceding consonant and consonant articulation on a preceding vowel will be explored; 3. It isproposed that phonemic paraphasias, i.e. sound substitutionerrors, produced by anterior aphasics have a phonetic orarticulatory basis. In all of these production studies,acoustic analyses are conducted of speech productions ofboth anterior and posterior aphasic patients to infer thearticulatory states giving rise to the acoustic patterns.