The Journal of Neuroscience, September 17, 2003, ():
Efficacy of Retinal Spikes in Driving Cortical Responses
J. Neurosci.
Kara and Reid 23 (24): 8547.
HTML Page - index.htslp
Files in this Data Supplement:
- Supplemental Figure 1 -
Receptive fields and cross-correlograms of all 14 disynaptically connected retino-cortical pairs. ON subfields shown in red, OFF subfields are blue, and brightness codes for response magnitude. Gridlines show individual stimulus pixels (0.6o for pair 13, 0.4o for pairs 7 & 14, and 0.8o for all other pairs). White circles are drawn over receptive fields with OFF centers and yellow circles are drawn over ON centers. In all 14 cases, the RGC receptive field center overlaps with a cortical subfield of the same sign (ON or OFF). Raw correlograms are shown in blue, filtered correlograms (see Materials and Methods) shown in red, and filter-subtracted correlograms shown in green. The horizontal red line in all correlograms represents the significance criterion for the fast neural peak, i.e., 3 standard deviations above the baseline. Except for pair 6, contribution and efficacy from filter-subtracted correlograms are virtually identical to results from shuffle-subtracted methods (see Results and compare correlograms shown in Supplementary Figure 1 with those shown in Figures 2 and 7 of the main paper). The fast neural peak in 12 of the 14 raw correlograms have rise times < 2 ms. The neural peak in pairs 8 and 9 have > 2 ms rise times.
- Supplemental Figure 2 -
Interspike interval (ISI) distributions from simultaneously recorded RGC, LGN, and cortical layer 4 simple cells. Data were collected during presentation of a drifting sine grating (4 Hz). Absolute refractory periods of RGC and V1 cells are ~ 4 ms. However, the absolute refractory period of the LGN cell is 1 ms, characteristic of bursting (which during 4 Hz drifting grating stimuli occur only at the onset of the visual response: see Kara et al., 2000). RGCs and simple cells in cortical layer 4 do not display such bursting in response to visual stimulation. Thus, we emphasize that bursting is not required for paired-spike enhancement for monosynaptic retinogeniculate connections (Usrey et al., 1998) or disynaptically for retino-geniculo-cortical connections (our current study – Kara and Reid, 2003). Mean, median, and modal ISIs for this triplet (in milliseconds) are RGC 36, 17, 16; LGN 27, 23, 2, V1 88, 9, 15; respectively. The relatively long refractory period of RGCs also impose a lower bound on the shortest RGC ISI we used for disynaptic paired-spike analysis (4 ms). ISI distributions shown here are unpublished data from a previous study (Kara et al., 2000).
