PET Studies of Auditory and Phonological Processing: Effects of Stimulus Characteristics and Task Demands

Abstract

AbstractPositron emission tomography (PET) was used to investigate the functional anatomy of auditory and phonological processing. Stimulus sets were designed to determine areas of the brain significantly activated during speech and nonspeech acoustic processing for stimuli with or without rapidly changing acoustic cues. Performance of auditory target detection tasks using these stimulus sets produced increased activation in superior temporal, frontal opercular, and medial frontal (SMA) cortices, relative to a visual fixation control task. While the medial frontal and superior temporal changes are best explained by motor and sensory components of the task, respectively, the frontal opercular changes were dependent upon the task performed upon the auditory input (mere presentation of the stimuli did not result in significant activation). On the left, the frontal opercular increases were larger when subjects performed an auditory detection task upon stimuli that incorporated rapid temporal changes (words, syllables, and tone sequences) than steady-state vowels. A converging study involving performance of orthographic (ascending letter) and phonological (long vowel sound) word discrimination tasks supports anatomical and behavioral evidence suggesting the left frontal opercular region is important for certain types of auditory/temporal analysis, as well as high-level articulatory coding.In addition to the activation increases associated with performance of auditory target detection tasks, decreases in activation were observed bilaterally along the intraparietal sulcus and superior parietal cortex, in the Rolandic sulcus, and the posterior cingulate; these decreases may reflect an attentional shift away from areas involved in the fixation task during the performance of a difficult auditory task.These results demonstrate that focusing more closely on basic neural processing differences (such as temporal integration rates) may lead to a better understanding of the specific neural processes that underlie complex phonological tasks.