Table 1.

Average correlation strength within and across cell types

	ON parasol	OFF parasol	ON midget	OFF midget	Small bistratified
Photopic
ON parasol	0.18 ± 0.05	−0.04 ± 0.01	0.12 ± 0.03	−0.01 ± 0	0.03 ± 0.02
OFF parasol		0.07 ± 0.05	−0.03 ± 0.01	0.08 ± 0.04	—
ON midget			0.03 ± 0	−0.02 ± 0.01	0.02 ± 0.01
OFF midget				0.01 ± 0	—
Small bistratified					0.03 ± 0
Scotopic
ON parasol	0.18 ± 0.01	−0.06 ± 0.02	0.20 ± 0.02	−0.03 ± 0	0.10 ± 0.04
OFF parasol		0.08 ± 0.02	−0.05 ± 0.02	0.10 ± 0.03	−0.03 ± 0.01
ON midget			0.09 ± 0.04	−0.04 ± 0	0.06 ± 0.01
OFF midget				0.03 ± 0.01	−0.01 ± 0
Small bistratified					0.02 ± 0

Values are estimated from fits indicated by curves in Figs 2B and 3B. For cell pairs composed of different cell types, correlation strength is given at zero distance. For cell pairs composed of the same cell type, correlation strength is given at the normalized nearest neighbour distance (arrowhead on the abscissa). These estimates represent the amplitude of the strongest correlations typically observed between cells of the given types. The extrapolated theoretical correlation between cells of the same type at a distance of zero is approximately 2.7-fold higher that the value at a distance of 2 s.d.s, which is approximately the normalized nearest neighbour distance. Unreliable estimates were excluded (blank entries in table). Light levels: photopic, 830 (840, 440) P* (L (M, S) cone)⁻¹ s⁻¹; scotopic, 0.5–1.3 P* rod⁻¹ s⁻¹. Overall 54 ± 27% of the variance in correlation strength with distance was explained by the degree of receptive field overlap, with a range from 87% (ON parasol–ON parasol) to 27% (OFF midget–OFF midget).