Ulrich Schridde

Alumni Department Physiology of Cognitive Processes

Main Focus

The neural basis of the BOLD signal during spontaneous activity and sensory stimulation

U. Schridde

Introduction and Scientific Aims

Blood-oxygen-level-dependent (BOLD) fMRI is used extensively to investigate functional localization of neural activity associated with cognitive tasks or sensory processing. Yet, the BOLD signal is only indirectly related to neural activity. Hence, caution is needed when inferring changes in neural activity from the spatial and temporal dynamics of the BOLD signal [1]. Evidence suggests that the BOLD signal mainly correlates with the local field potential (LFP), a mass neural signal, capturing various neural processes [1,2]. While the BOLD signal has been found to correlate with specific LFP frequency bands [3], it is still unclear whether it relates to each band independently or if, instead, it also reflects the relationships between different LFP bands. Our aim is to characterize the relationship between the activity expressed by LFPs at different frequencies and the changes occurring over space and time in the BOLD contrast, and to understand whether the BOLD signal reflects specific relationships among different LFP bands.


We recorded simultaneously LFP and BOLD fMRI with high spatial and temporal resolution in V1 and V2 of anesthetized non-human primates during spontaneous activity and visual stimulation. We apply information theory to characterize the relationship between the BOLD signal and LFP over time and space during spontaneous and visually induced activity.

Results and Preliminary Conclusions

We found that, during spontaneous activity, the power in the alpha and beta LFP bands conveyed information about the BOLD signal that was complementary to that conveyed by gamma power. In particular, while alpha and gamma power provided information about the amplitude of BOLD variations, beta power conveyed information about the latency with which changes in BOLD signal occurred following changes in gamma power. Our current analysis suggests that the BOLD signal does not simply reflect the total power of the LFP or the power of a specific LFP band but rather the overall shape of the LFP spectrum [4]. Additionally, these results lay the basis for identifying the contribution of different neural pathways to cortical processing using fMRI. We are currently investigating if our results also hold when visual stimuli are presented.


C. Magri, Y. Murayama, S. Panzeri


1. Logothetis NK (2008) What we can do and what we cannot do with fMRI, Nature 453 869-878.

2. Logothetis NK (2003) The underpinnings of the BOLD functional magnetic resonance imaging signal, The Journal of Neuroscience 23 3963-3971.

3. Goense JBM, Logothetis NK (2008) Neurophysiology of the BOLD signal in awake monkeys, Current Biology 18 631-640.

4. Magri C, Schridde U, Murayama Y, Panzeri S, Logothetis NK (2011) The amplitude and timing of the BOLD signal reflects the relationship between LFP power at different frequencies, The Journal of Neuroscience (in press).

Figure Caption

Three LFP bands are informative about BOLD. A) Mutual Information (median) between BOLD and LFP power at different frequencies for different lags ? (?>0 indicates that BOLD was shifted back with respect to neural activity). B) Mutual information (median, 40th – 60th percentile range) between BOLD and LFP or MUA, respectively: alpha- [8-12Hz], beta- [18-30Hz], gamma- [40-100Hz] and total LFP-power [0-100Hz], as well as MUA power [900-3000Hz] are shown. Inset: correlation between BOLD and LFP/MUA.

Go to Editor View