We compare fMRI-BOLD responses in anesthetized cats with local field potentials (LFPs), aggregate high-frequency responses (analog-Mua) and spiking activity recorded in primary and higher visual cortex of alert animals. The similarity of the activations in these electrophysiological signals to those in the BOLD is quantified by counting recording sites where different stimuli elicit the same relative activation as in the imaging experiments. Using artificial stimuli, a comparison of BOLD and LFP strongly depends on the frequency range used. Stimulating with complex or natural stimuli reduces this frequency dependence and yields a good match of LFP and BOLD. In general, this match is best between 20 and 50 Hz. The measures of high-frequency activity behave qualitatively different: the responses of the analog-Mua match those of the LFP; the spiking activity shows a low concordance with the BOLD signal. This dissociation of BOLD and spiking activity is most prominent upon stimulation with natural stimuli.
Bibliographical noteFunding Information:
helping with the fMRI experiments. Furthermore we thank the referees for their constructive and helpful comments. This work was financially supported by the Center for Neuroscience Zurich, the Swiss National Science Foundation (Grant No. 31–65 415.01), the EU/ BBW (IST-2000–28127, 01.0208–1), the National Institutes of Health (RR08079, MH61937), the Keck Foundation and the Human Frontiers Science Program.
- Area 21
- Firing rate
- Local field potential
- Natural stimuli