Probing neural events by fMRI at neural time scale

On the basis of the tight coupling of rCBF/metabolic load to neural activation, one would expect that fMRI signals reflect undergoing neural events at the activation sites in the brain. With relatively simple paradigms, one can follow task-induced neural events in the amplitude of fMR1 signals. We observed that with a paradigm of forepaw stimulation by two short pulses in a rat model. The fMRI signal in the somatosensory area and the evoked potential showed a neural refractory period lasted longer than 600 msec. Since the timing of these neural events can be manipulated by the task paradigm of our choice, we may be able to follow the temporal events in neural systems at the neural time scale. We have observed in rat model that MRI signal as well as evoked potential at one site can be influenced by the activation at another remote area with a particular temporal relation between the two events. One example was a cross hemisphere suppressive interference by preceding stimulus at one paw on the somatosensory activation by the other paw stimulation. When the inter-stimulus interval between the two was varied, the suppressive interaction appeared in the distinct time period of around 30-40 msec. In the absence of known contralateral projections of tbalamocortical sites, this suppression likely came through the callosal (cross-hemisphere) projection with disynaptic connections. We saw similar phenomena in the human brain. With two short visual stimulation pulses, there was a neural refractory suppression of the second activation in Vl area for the inter-stimulus interval below as long as 1 sec. This suppression appeared as reduced fMRI signal intensity (height times width) for the activation by two identical inputs which was smaller than the expected intensity, twice of the signal due to a single stimulus. On the other hand, when the interval was similar or shorter than the latency time of visual evoked potential (-100 msec), the suppression of the second activation was apparently absent and fMRI signal intensity for the two stimuli inputs was nearly the twice of the response by one stimulus. The signal height as well as width was larger than the single stimulus activation. Since the signal was stronger than the single stimulus, the two inputs must have reached the visual cortex from thalamic areas, but the suppressive condition to cause refractory response was not set in at the time of the second input arrival. These observations open a new approach in fMRI to study neural systems. Using preparative and sampling tasks as in the above examples, one can provide an extra temporal axis of neural time scale to measure the characteristics of the neural systems as well as the time course of their interaction among them.