On the photovoltaic effect in local field potential recordings

Abstract

Optogenetics allows light activation of genetically defined cell populations and the study of their link to specific brain functions. While it is a powerful method that has revolutionized neuroscience in the last decade, the shortcomings of directly stimulating electrodes and living tissue with light have been poorly characterized. Here, we assessed the photovoltaic effects in local field potential (LFP) recordings of the mouse hippocampus. We found that light leads to several artifacts that resemble genuine LFP features in animals with no opsin expression, such as stereotyped peaks at the power spectrum, phase shifts across different recording channels, coupling between low and high oscillation frequencies, and sharp signal deflections that are detected as spikes. Further, we tested how light stimulation affected hippocampal LFP recordings in mice expressing channelrhodopsin 2 in parvalbumin neurons (PV/ChR2 mice). Genuine oscillatory activity at the frequency of light stimulation could not be separated from light-induced artifacts. In addition, light stimulation in PV/ChR2 mice led to an overall decrease in LFP power. Thus, genuine LFP changes caused by the stimulation of specific cell populations may be intermingled with spurious changes caused by photovoltaic effects. Our data suggest that care should be taken in the interpretation of electrophysiology experiments involving light stimulation.

In Vivo Magnetic Recording of Neuronal Activity

“…This is important, because the magnetrode’s silicon substrate could be directly influenced by the light flash, i.e. by the photovoltaic effect (Mikulovic et al, 2016). However, the detected magnetic signals have a sharp deflection with a latency of 20–40 ms, which corresponds to the conduction delay from the retina to the cortex, ruling out a direct effect of the light flash on the magnetrode.…”

Advanced Electrical and Optical Microsystems for Biointerfacing

“…Those opaque metal microelectrodes not only generate optical shadows that prevent direct optical interrogation of cells under the microelectrodes but also produce severe light‐induced electrical artifacts that can be difficult to distinguish from real neuronal activity. [ 233,234 ] One approach to overcome those limitations is to design optically transparent microelectrodes, which will allow crosstalk‐free integration of electrophysiological recording/stimulation with optogenetics and/or optical imaging.…”

OPETH: Open Source Solution for Real-time Peri-event Time Histogram Based on Open Ephys

“…A caveat of optogenetic tagging studies is that light may induced different signals besides action potentials, including photoelectric (Becquerel) and photovoltaic effects (Kozai and Vazquez, 2015), or exciting too many neuronal elements summing up to population spikes that prevent proper spike sorting. The uneven dispersion of light in brain tissue may lead to artifacts that are hard to remove by offline referencing techniques, as pointed out in previous studies (Cardin et al, 2010;Mikulovic et al, 2016;Park et al, 2014). Most of these potential confounds can be efficiently eliminated by proper control of light intensities delivered into the brain, for which precise on-line feedback is immensely useful.…”