Abstract

Deficits in odor threshold sensitivity, discrimination and identification are common in schizophrenia, presumably originating from limbic brain structures (e.g., orbital prefrontal cortex, amygdala, piriform cortex) also linked to their cognitive and emotional disturbances. To study the neurophysiological processes underlying olfactory dysfunction in schizophrenia, nose-referenced 30-channel ERPs were recorded from 32 schizophrenia patients and 35 healthy adults (18/18 male) during odor detection of hydrogen sulfide stimuli presented for 200 ms to the left or right nostril at concentrations of 50% or 100% by a constant-flow olfactometer (variable 15-25 s ISI). Time of odor stimulation was not cued. Participants indicated when they perceived a low or high odor intensity. Patients’ performance (percentage of missed responses) was not different from healthy controls for high (23% ±16% vs. 23% ±18%) and low (45% ±19% vs. 41% ±20%) intensities. Patients and controls had similar olfactory ERP and CSD waveforms. As previously reported, ERPs rereferenced to linked mastoids had a small vertex N1 and pronounced mid-parietal P2, which was larger for high than low odor intensities. However, temporal principal components analysis (unresticted Varimax rotation of covariance loadings) of the CSD waveforms identified two distinctive PCA factors revealing bilateral frontotemporal N1 sinks (305 ms peak latency) and a mid-parietal P2 source (630 ms) that were greater for high than low intensities for both patients and controls. In agreement with findings by Turetsky et al (2003), N1 sink and P2 source were markedly reduced in patients for high intensity stimuli, providing further neurophysiological evidence of olfactory dysfunction in schizophrenia. The CSD-PCA methodology uniquely revealed neuronal generator patterns underlying olfactory ERPs that are obscured by volume-conducted, reference-biased surface potentials. [Supported by NIMH grants MH066428, MH066597 and MH082393.]