The influence of stimulus intensity and inter-stimulus interval on the detection of pitch and loudness changes

Abstract

This study illuminates processes underlying change detection for different features (detection of pitch versus loudness changes) and different amounts of attentional allocation (automatic versus attentive change detection). For this reason, the influence of important stimulus characteristics (intensity and inter-stimulus interval (ISI)) on these different types of change detection was determined. By varying intensity, it should be clarified whether these processes are mainly sensitive to the informational content of the change or to the total amount of stimulus energy. By varying ISI, it should be determined whether they are differentially sensitive to manipulations of encoding time and/or state of sensory refractoriness. Automatic change detection was indexed by the mismatch negativity (MMN), which is a component of the event-related brain potential (ERP). Attentive change detection was indexed by the N2b and P3 components of the ERP and by behavioral performance. Human subjects were presented with a high-probability standard tone and a low-probability deviant-tone, which differed from the standard tone in frequency (Experiment I) or intensity (Experiment II). In separate blocks, the intensities of the standard stimuli were of 55 and 70 dB SPL and ISIs were of 350 and 950 ms. During the first part of the experiments, subjects were engaged in silent reading, whereas they tried to discriminate deviants from standards in the second part. The MMN elicited by a frequency change was invariant to variations in intensity and ISI, whereas the MMN elicited by an intensity change was significantly modulated by both intensity and ISI. This implies functional differences between the neural traces underlying the frequency-MMN and the intensity-MMN. In addition, there were larger effects of the ISI on the N2b and P3 amplitudes as compared with the effects on the MMN amplitudes, suggesting stronger capacity limitations for attentive change detection than for automatic change detection.