Signal source separation in the analysis of neural activity in brain

This research examines the spatial and temporal characteristics of the resp onses to stimulation of the barrel cortex in anesthetized rats using optica l imaging with particular emphasis on methods of analysis which reduce the effects of low-frequency oscillations on localization of the activated cort ical region. Image sequences were captured using a light source with a narr ow bandwidth of wavelength (590 +/- 2 nm). On each trial image data were co llected at 15 Hz and stored over a 12-s period starting 8 s before stimulat ion onset. Stimulation was for 1 s using an oscillating whisker vibrator (s imilar to1-mm deflection at 5 Hz). For each subject a total of 30 experimen tal trials were collected and averaged. There was an interstimulus interval of 26 s. The trial-averaged data were analyzed using two related signal so urce separation algorithms. Both algorithms use a weak model of the expecte d temporal response as a filter to exclude contributions from low-frequency baseline oscillations which we call the V-signal. We found that both algor ithms successfully separated most of the effects of the V-signal from the r esponse to the stimulation. The performance of the algorithms compared favo rably with the performance of related algorithms without weak constraints a nd the "ratio of means" strategy used by C. H. Chen-Bee ct al. (1996b, J. N eurosci. Methods 68:28-37; C. H. Chen-Bee et al., 2000, J. Neurosci. Method s 97:157-173). (C) 2001 Aoademic Press.