Statistical Quantification of 24-Hour and Monthly Variabilities of Spontaneous Otoacoustic Emission Frequency in Humans

Abstract

Previous evidence has suggested a relationship between spontaneous otoacoustic emissions (SOAEs) and established, biological cycles, although detailed statistical quantifications of the suggested relationships do not exist in the literature. In an attempt to statistically quantify the purported circadian and monthly influences on this phenomenon, two experiments were undertaken. The first experiment was conducted over eight weeks, investigating 31 SOAEs recorded from eight women and two men. Time series statistical analysis examined whether daily, weekly, and/or monthly cycles characterized SOAE frequency variability. Results yielded a significant monthly cycle for the majority of SOAEs recorded from the women but for none of the SOAEs recorded from the men. These results suggest the possibility that SOAE frequency fluctuation in women may be entrained to the monthly menstrual cycle. In the second experiment, hourly SOAE frequency stability was examined over a 24-h period to ascertain the nature of the daily frequency variation as precisely as possible. Four SOAEs from two subjects were examined, and time series analysis of these data included (1) modelling the autocorrelation structure of the measurements, (2) resolving each 24-h series of measurements into cyclical components of various periodicities, and (3) testing the statistical significance of given cycles within the spectrum of each series. Findings included a significant 24-h variability of frequency for each SOAE, suggesting the possibility of a circadian influence on frequency fluctuation. Results from the two experiments provide quantitative evidence supporting a hypothetical relationship between SOAEs and established, biological cycles