The following musing was inspired by watching Sabine Kastner's keynote at the OHBM conference 2015:
Brain regions within the visual processing streams have receptive fields which, when a visual stimulus is presented in their preferred spatial location (the receptive field), increase their baseline activity. Furthermore, when a subject is visually cued to attend to a given spatial location, and then made to wait until a target appears in that same location, these brain regions will show elevated activity during the delay period, which gradually wanes back to baseline, before being activated again when the target appears.
What function does this elevated activity serve? The task (waiting for the target) requires that the subject’s brain be primed to attend and respond to stimulation in that visual location. Presumably the elevated activity is the mechanism by which this priming process is achieved. But if that is the case, what happens after the elevated activity has waned back to baseline levels? Priming is still observable behaviourally, even though the elevated activity is gone.
The first question should possibly be - how does elevated activity lead to priming of a RF location?Perhaps by keeping the receptive field neurons firing, the link between RF regions and higher-order brain regions can be maintained. it seems important to note that the purpose of priming a receptive field location is not simply to push-forward the target activity quickly, but perhaps primarily to orient higher-order cognitive capacities to that spatial location in anticipation of future stimulus. It is probably fair to assume that the same higher-order regions communicate with multiple receptive field neurons so that the same faculties can be directed to the whole visual field. In this sense, the priming of receptive field neurons could be seen as a beacon: guiding top-down signals and processes to the currently most important spatial location. The elevated activity in RF regions could thus be the brain’s way of keeping higher-order regions on point, and aimed at the correct spatial location, so that when the target is presented the system is already oriented towards that location.
If this ‘beacon’ property is indeed true, and achieved by repeated neural firing, what happens when the elevated activity wanes back to baseline? Again, the spatial location is still primed even after the activity has waned. It seems to me that a similar ‘beacon’ would still be necessary, however perhaps it is achieved through a different mechanism.
Perhaps the initial elevated firing serves two purposes: 1) to act as an immediate priming beacon for higher-order regions, and 2) to set up a more long-term priming beacon which doesn’t require continuous firing, which could be metabolically costly. Thus, the initial elevated firing might also trigger some form of rapid and transient long term potentiation (LTP) between the RF and higher-order brain regions. If so, this LTP could allow a spatial RF to be primed even after the elevated activity has waned.
In other words, neurally, there may be two mechanisms at play, immediately following the cue, repeated reentrant activation may represent keeping the gates open by keeping the neurons depolarised, or by keeping the RF in communication with higher-order regions. This is a costly, but immediate form of priming, and so meanwhile this reentrant activity may also be setting up a transient form of LTP, which lasts seconds or minutes, to keep that RF location primed while allowing the costly elevated activity to return to baseline.
Thus during the early phase, priming is achieved by repeated activation of RF and higher-order neurons. However, over time the mechanism is shifted such that LTP takes over as the priming mechanism, and the same priming effect is achieved through molecular potentiation.
If this system is indeed the case, it would be impressive if the two types of priming could not be distinguished - neurally, even if not behaviourally.