amol gharat and curtis baker - mcgill vision...

1
0 90 180 270 360 20 30 10 15 20 25 Linear (F1) NonLinear (F2) H4405.006 Spatial Phase (Deg) Firing Rate (Hz) Firing Rate (Hz) Nonlinear, non-oriented receptive fields in early visual cortex: Possible precursors to second-order processing Neurons selective for the orientation of both luminance (1 st order) and contrast (2 nd order) boundaries form a considerable proportion in early visual cortex (cat Area 18, macaque V2). • Neural mechanism underlying receptive fields of these neurons is unknown. • Recent evidence suggests that inputs from LGN Y cells could be pooled together in cortex to build these receptive fields (Rosenberg et al., 2010; Demb et al., 2001). •But nonlinear RFs of Y cells can only signal increase in contrast but not decrease. •Given this, we hypothesize that there is an intermediate processing stage at the cortical level which provides inhibition to 2 nd order responsive neurons. 1 Integrated Program Neuroscience, McGill Univ.; 2 McGill Vision Res. Unit, Montreal, Canada Amol Gharat 1 and Curtis Baker 2

Upload: others

Post on 07-Oct-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Amol Gharat and Curtis Baker - McGill Vision Researchmvr.mcgill.ca/Curtis/posters/Amol_SfN_poster_13.pdf · Amol Gharat1and Curtis Baker2 Methods • Anesthetized, paralyzed cats

0 90 180 270 360

20

30

10

15

20

25

Linear (F1)NonLinear (F2)

H4405.006

Spatial Phase (Deg)

Firin

g R

ate

(Hz)

Firing Rate (H

z)

Nonlinear, non-oriented receptive fields in early visual cortex: Possible precursors to second-order processing

•  Neurons selective for the orientation of both luminance (1st order) and contrast (2nd order) boundaries form a considerable proportion in early visual cortex (cat Area 18, macaque V2). •  Neural mechanism underlying receptive fields of these neurons is unknown.

•  Recent evidence suggests that inputs from LGN Y cells could be pooled together in cortex to build these receptive fields (Rosenberg et al., 2010; Demb et al.,2001). • But nonlinear RFs of Y cells can only signal increase in contrast but not decrease.

• Given this, we hypothesize that there is an intermediate processing stage at the cortical level which provides inhibition to 2nd order responsive neurons.

1Integrated Program Neuroscience, McGill Univ.; 2McGill Vision Res. Unit, Montreal, Canada Amol Gharat1and Curtis Baker2

Methods •  Anesthetized, paralyzed cats.

•  Extracellular, single-unit recording in Area 18 using 32 channel linear arrays.

•  Response measured as average firing frequency for complex cells and 1st & 2nd harmonic modulations for simple cells.

•  Manual spike sorting using Plexon Offline Sorter.

Stimulus •  Sinusoidal grating.

•  Temporal Modulation: Drifting or Contrast reversing.

• .Y like receptive fields of these cortical neurons suggests that they receive input from LGN Y cells. • These neurons could provide an important intermediate stage in building cortical orientation selective second-order neurons

1.  Rosenberg et al 2010 2.  Demb et al 2001 3.  Crook et al 2008

Funded by CIHR grant MA-9685, MOP-119498 to CB. Vanier Canada Graduate Scholarship to AG.

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

100 µm

0.05

0.15

0.25

0.35

0.45

0.55

0.65

0.75

0.85

0.95

0

10

20

30

40

N = 135

Orientation Bias (OB)

Num

ber o

f neu

rons

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

OB = 0.09 OB = 0.68

H 4303.017 CH 22_1

H 4303.017 CH 31_1

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

Spat

ial F

requ

ency

(c

pd)

spontaneous activity

Firin

g R

ate

Firin

g R

ate

Firin

g R

ate

Firin

g R

ate

Firin

g R

ate

Firin

g R

ate

Firin

g R

ate

Firin

g R

ate

Time (sec)

0.03 cpd

0.06 cpd

0.13 cpd

0.27 cpd

0.53 cpd

1.10 cpd

2.13 cpd

Fundamental frequency response (4 Hz)

LINEAR (F1)

Frequency doubled response (8 Hz) NON-LINEAR (F2)

0 10.5

0.1 1

10

20

30

40

4

8

12

16

Contrast-Reversing (Nonlin)

Drifting (Linear)

H4209.007

Spatial Frequency (cpd)

Firin

g R

ate

(Hz)

Firing Rate (H

z)

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

Time (sec)0 10.5

Spa

tial P

hase

0o

45o

90o

135o

180o

225o

270o

315o

360o

Null phase

Null phase

Null phase

Linear Response (F1)

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

0

20

40

60

Spa

tial P

hase

0o

45o

90o

135o

180o

225o

270o

315o

360o

Time (sec)0 10.5

Non-Linear Response (F2)

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

H4505.030 28_130

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

30

210

60

240

90

270

120

300

150

330

180 0

H4505.030 CH28_1

H4505.030 CH21_4

Linear Non-LinearY cell signature

0.1 1

10

30

50

70

5

10

15

20

25

3707.006

drifting gratingcontrast reversing (F2)Contrast Modulation

Spatial Frequency (cpd)

Firin

g R

ate

(Hz)

Firing Rate (H

z)

Area 18

Non-Ori

Area 18

Ori

Linear RF Non-Linear RF

Excitatory Inhibitory Excitatory Inhibitory