Alternative statistical method for detecting changes in spontaneous activity of single olfactory receptor cells

An statistical method for analysis of event data analysis, common in electrophysiological investigations of activities of olfactory receptor neurons, is proposed. The method is based on the analysis of the slopes of cumulative frequency distribution function. During the spontaneous activity the distribution function is linear which indicate that probability density distribution is uniform. During the chemical stimulation, the density of action potentials is either increased (excitation--E) or decreased (suppression--S), causing the observable change of the slope. The slope at each event is evaluated by piecewise linear regression. The type (E or S) and time of change of the slope can be determined, providing information for further statistical analyses. Graphically supported computer program (in S-PLUS) is available for interactive or routine use.