[Neuroscience] Neuronal activity in the brain, enhanced
(by aric.agmon from gmail.com)
Thu Oct 26 15:42:44 EST 2006
You are surely aware that there are several good studies out there (e.g. by Nikos Logothetis, among others) which have shed considerable light on the aspects of neuronal activity which give rise to the BOLD signal measured by fMRI. My posting, however, had nothing to do with fMRI - it was in response to a question which (at least as posed in its second version) also had nothing to do with fMRI, only with currents generated by neuronal activity in the cortex.
Regarding the distinction between currents and voltages, I will try to restate in some more detail. The signals measurable by EEG electrodes in humans, or by "field potential" electrodes in animals, are generated by temporally synchronous and spatially focal (=restricted to a small region or a narrow lamina) flow of current from the extracellular space into the intracellular milieu of many thousands of neurons. These neurons could be synchronously spiking, in which case we will record a "population spike", but more often they will be receiving a synchronous volley of synaptic excitation, in which case we will see a "population EPSP". The latter is the case when a beam of thalamocortical axons is activated electrically, or by focused sensory stimulus. Since many excitatory synaptic channels will be opened simultaneously (albeit briefly), in a restricted laminar position, current will flow into the neurons at that location ("current flow" refers to positive ions; negative ions would flow in the opposite direction). This is referred to as a "current sink", from a point of view within the extracellular space. Once the channels close (within a few ms, usually), the excessive positive charge is trapped inside the neurons and needs to dissipate back out; this will happen through "leak" channels distributed throughout the cell body and dendritic tree. In the case of neurons with upwards-elongated and parallel apical dendrites, like pyramidal cells, there will be a distributed current source superficially to the original current sink. To maintain local electroneutrality, the current sink and the current source will be linked by an axial intracellular current flowing upwards through the apical dendrites, and a parallel but opposite current flowing downwards in the extracellular space. It is the latter, extracellular current which gives rise to the minute voltage differences between the sink and the source (the source being more positive than the sink) which are measurable by field potential electrodes, or (if large enough) by EEG electrodes. By performing "current source density" (CSD) analysis, which essentially amounts to taking the second spatial derivative of the field potential, one can recover analytically the location, amplitude and time course of the original current sinks and sources, and thereby reveal the underlying synaptic activity.
A good description of this process can be found in the first few pages of Chapter 19 of the Bear, Connors and Paradiso textbook (3rd edition), or in Chapter 46 of the Kandel, Schwartz and Jessel textbook (4th edition). See especially the latter's Fig. 46-2.
I hope this helped.
More information about the Neur-sci