Why nerve signals should be measured monopolarly

A nerve signal, to which white noise was added, and monopolar and bipolar recordings of this signal were simulated. As a measure for nerve activity both signals were integrated (NA/sub mon/ and NA/sub bip/). The bipolar signal was rectified before integration. NA/sub MON/ appeared to be a linear measure for nerve activity while NA/sub bip/ underestimated it at high firing rates. Furthermore NA/sub mon/ was less sensitive to noise than NA/sub bip/. It was therefore concluded that the integrated non-rectified monopolar recording of a nerve signal is a better measure for the nerve activity than the integrated rectified bipolar recording.