Abstract

The loudness dependence of the auditory evoked potential (LDAEP) is shown in a passive task in which each of a range of intensities is presented with equal, but low, probability, yielding a reliable relationship between N1/P2 and loudness. However, the impact of overlapping MMN and other salience-driven ERP components (e.g., P3a, Novelty P3) generated by large intensity disparities between consecutive tones in this paradigm has not been studied. Conversely, it is unclear to what extent the MMN produced by loud and soft tones is affected by stimulus intensity. Binaural tones (1000 Hz; 40-ms; 500 ms ISI) were presented at four intensities (70, 80, 90, 100 dB SPL) in a modified MMN task, in which one of the four intensities was used as a frequent standard (70%), and the remaining three as infrequent deviants (10% each), in each of four blocks, counterbalanced across subjects. Intensity-dependency of N1/P2 was validated against a standard LDAEP paradigm for a stimulus block of equiprobable intensities (70-100dB SPL) with the same short, fixed ISI. Healthy adults (N = 33) watched a silent film to direct the subject’s attention away from the auditory stimuli, and reported its content (underwater scenes) after each block. Using 72-channel ERPs, temporal principal components derived from reference-free current source density (CSD) waveforms characterized neuronal generator patterns underlying intensity-dependence of N1-P2. A second principal components analysis (PCA) was performed for difference CSD waveforms (deviant-minus-standard) calculated from comparable deviant and standard conditions at each intensity level (e.g., 100dB deviant-minus-100dB standard) to isolate activity related to deviance (i.e., MMN). Factors related to N1-P2 included: 1) N1 sink (113 ms peak; tangential sink/source topography); 2) temporal N1 sink (159 ms; radial temporal lobe sink); 3) P2 source (214; midline and bilateral temporal sources). These three factors revealed monotonic increases with intensity for standard and deviant stimuli. PCA factors derived from the CSD difference waveforms included: 1) relative MMN sink (136 ms peak; tangential sink/source topography); 2) relative vertex source (230 ms peak). The MMN sink was characterized by a late, N1-like tangential sink/source topography that increased with the level of stimulus disparity (+/- dB) and was most pronounced for louder tones. The vertex source was observed solely in the loudest deviants, likely reflecting P3a to these salient tones. N1/P2 LDAEP and MMN/P3a to stimulus disparity/salience may both be measured in the same task. [Supported by NIMH MH036295]