The effect of task demands on the processing of attended and unattended auditory inputs as indexed by the mismatch negativity.

摘要

The main interest of this dissertation was to examine the role of attention in early auditory processing using event-related potential (ERP) recording techniques. The Mismatch Negativity (MMN), a component of the ERP, is an index of early auditory deviance detection. In the present research, the MMN was used to probe the extent to which auditory stimuli were processed when these stimuli were or were not within the focus of attention. Four experiments were carried out. The aim of Experiments 1 and 2 was to investigate whether the demands of common diversion tasks modulated the MMN. In these experiments, subjects were instructed to ignore the auditory sequence and engage in tasks that varied in the amount of attention required to complete them. The results of Experiment 1 indicated that the MMN elicited by small frequency deviants was sensitive to the nature of the task (although this effect could not be clearly related to attention). In contrast, the MMNs elicited by small intensity deviants in Experiment 2 did not vary based on the diversion task demands. A limitation of Experiments 1 and 2 was the absence of an independent measure of task demands. This limitation was addressed in Experiments 3 and 4. A visual discrimination task was employed that permitted quantification of task demands. The visual task was either easy or difficult to perform. Hence, in Experiments 3 and 4, the extent to which the attentional demands of the visual task affected the passively-elicited MMN could be more rigorously examined. Importantly, the MMNs elicited by small frequency and intensity deviants were not significantly modulated by task difficulty, in spite of a wide variation in performance measures between the easy and difficult visual tasks. Experiments 3 and 4 also investigated whether the direction of attention had an effect on the MMN. Subjects were subsequently asked to divide their attention between the visual and auditory channels and to detect all visual and auditory deviant stimuli. This allowed a comparison of the MMN when it was elicited by stimuli within or outside the focus of attention. The results of Experiment 3 indicated that the frequency MMN was unaffected by the direction of attention; the intensity MMN, however, was larger during active attention of the auditory sequence. The attentional modulation of the intensity but not frequency MMN may be due to specific deviant features being more sensitive to the effects of attention. Alternatively, these results could be explained by the difference in perceptual discriminability of the intensity and frequency deviants from the standard (the frequency deviants were detected more accurately and more rapidly than the intensity deviants). Experiment 4 tackled this issue by increasing the discriminability of the intensity deviant while decreasing the discriminability of the frequency deviant. The results showed that the frequency MMN was significantly affected by the direction of attention, whereas the intensity MMN was not. Together, the results suggest that the attentional modulation of the MMN may be limited to the processing of small stimulus changes. Attention may be viewed as a means to sharpen the memory trace of the attended stimuli and consequently enhance the processing of small auditory changes. This explanation would also be consistent with the present findings that the passively-elicited MMN is insensitive to variations in visual task difficulty.