Asymmetric pedestal masking of S-cone increments and decrements: Does sawtooth polarity matter? Scott H. Gabree & Rhea T. Eskew Jr. Department of Psychology, Northeastern University S-cone Tests on S-cone Pedestals - S+ pedestals (right half of each panel) mask more than S- pedestal (left half) regardless of test polarity (S+ and S-, top and bottom respectively)

-2

-4

SHG -15

-10

-5

0

5

10

15

0.5 0 -0.5 -1 -1.5 -2

20

S-cone Pedestal Intensity S-cone Pedestal Intensity

SHG -3

-2

-1

0

1

2

S-cone test on S-cone Pedestal

3

S-cone Pedestal Intensity S-cone Pedestal Intensity

S-cone Tests on (L+M) Pedestals Note: Signs of the (L+M) pedestals have been reversed 6

1.5

Threshold S-cone S+ in threshold ratio

S-

2

4

2

0

-2

-4

-6 -20

-15

-10

-5

0

- 5 - 10 - 15 - 20

(L+M) Pedestal Intensity (L+M) Pedestal Intensity

1 0.5 0 -0.5 -1 -1.5 -2

-3

-2

-1

0

-1

-2

-3

(L+M) Pedestal Intensity (L+M) Pedestal Intensity

- All points on graphs are displayed in units of their own threshold; up to 4 different scale factors area used for each graph - Same asymmetry would be found if plotted in cone contrast units

A. Same masking asymmetry is found with the polarity of the (L+M) pedestal, again with both S+ and S- tests B. Approximate subthreshold summation at low pedestal contrast for both pedestal chromaticities C. A and B suggest that S and (L+M) are being processed in the same pathway

Test Interval I

Ped

Contrast

Interval II

Which interval contained the pedestal + test?

Contrast

Position (deg)

Position (deg)

S- test

S+ test

SInput

S- test

∆R ∆R

Contrast (+)

- 2 AFC - Well-practiced observers - All test chromaticities were S-cone isolating (S+ or S-)

Differential Contrast Gain Control

ON

∆R ∆R

Time

Time

Contrast (-)

Ped

Output

OFF

Output

- Colored points and lines represent rapid start sawtooth stimuli and black points and lines represent rapid stop stimuli.

S+ test

S+ Input

- Flatter transducer (more contrast gain) in the ON pathway than the OFF - Higher test contrast required to reach threshold in the ON pathway (at high contrast pedestals) - Asymmetry is associated with the pedestal polarity, not the test polarity, because the local curvature around the pedestal is relatively small - Solomon and Lennie (2005) report greater contrast gain control in S-ON pathways than S-OFF in LGN

- Facilitation and masking were found for L+M pedestals (displayed above) - No facilitation or masking was found with L- or M-cone pedestals or (L-M) pedestals (data not shown) - Therefore the L and M signals to the S-cone pathways must be combined in a nonlinear fashion

8 6 4 2 0 -2 -4 -6

15

6

10

4

5

0

-5

-10

RTE

SHG

-8 -20

-15

-10

-5

0

5

10

15

S-cone test on (L+M) Pedestal S-coneThreshold Threshold S-cone

0

1

Rapid start vs. Rapid stop

-15 -20

20

S-cone Pedestal Threshold S-cone Pedestal Intensity

-15

2

0

-2

-4

SHG -10

-5

0

5

10

15

-6 -20

20

S-cone Pedestal Intensity S-cone Pedestal Intensity

-15

-10

-5

0

- 5 - 10 - 15 - 20

L+M Pedestal Intensity (L+M) Pedestal Intensity

15

8 6

S-cone S-coneThreshold Threshold

2

S-coneThreshold Threshold S-cone

4

S-cone S-coneThreshold Threshold

S-cone S-coneThreshold Threshold

2 1.5

-6 -20

Threshold S-cone S+ in threshold ratio

S+

Stimuli and Methods

6

S-cone Threshold Threshold S-cone

- “Purple” pedestals produce more masking with S-cone increment (S+) and decrement (S-) tests than do “yellow” pedestals - This suggests more contrast gain control in S-cone increment pathway (likely the S-ON pathway) - Assuming contrast polarity separates ON and OFF pathways, will the temporal polarity of the sawtooth pulse modulate this effect?

- Contrast polarity separates signals into ON and OFF pathways - What about the polarity of the sawtooth? - We expect no effect except possibly at high pedestal contrasts

S-cone S-coneThreshold Threshold

Main Ideas

4 2 0 -2 -4 -6

10

5

0

-5

-10

DR

RJG

-8 -20

-15

-10

-5

0

5

10

15

20

S-cone Pedestal Intensity S-cone Pedestal Intensity

-15 -20

-15

-10

-5

0

- 5 - 10 - 15 -20

L+MPedestal Pedestal Intensity (L+M) Intensity

- Rapid Stop waveform produced consistently less masking, but the differences from the Rapid Start waveform are quite small - The slightly lower masking could be consistent with some adaptation to the strong pedestal - More masking is found for “purple” pedestals using both polarities of the sawtooth.

Conclusions - The polarity of the sawtooth makes little difference even for strong pedestal conditions - Contrast polarity seems to determine which pathway (ON or OFF) is processing the stimulus - The temporal polarity has little effect, even at high contrast levels