Modeling Adult Visual Function Dr. James A. Bednar - - PowerPoint PPT Presentation

▶
modeling adult visual function
SMART_READER_LITE
LIVE PREVIEW

Modeling Adult Visual Function Dr. James A. Bednar - - PowerPoint PPT Presentation

Nov 25, 2022 590 likes •869 views

Modeling Adult Visual Function Dr. James A. Bednar jbednar@inf.ed.ac.uk http://homepages.inf.ed.ac.uk/jbednar CNV Spring 2012: Modeling adult function 1 Surround modulation (Series et al. 2003) Apparent contrast Detection facilitated

slide-1
SLIDE 1

Modeling Adult Visual Function

  • Dr. James A. Bednar

jbednar@inf.ed.ac.uk http://homepages.inf.ed.ac.uk/jbednar

CNV Spring 2012: Modeling adult function 1

slide-2
SLIDE 2

Surround modulation

Apparent contrast reduces Detection facilitated Contour pops out

(Series et al. 2003)

Many types of contextual interactions are known

CNV Spring 2012: Modeling adult function 2

slide-3
SLIDE 3

Surround modulation

(Schwabe et al. 2006)

Effects depend strongly on contrast (Hirsch &

Gilbert 1991), (Weliky et al. 1995) and on

distance Distance- related effects match both lateral and feedback connections

CNV Spring 2012: Modeling adult function 3

slide-4
SLIDE 4

Proposed model circuit

(Schwabe et al. 2006)

From Schwabe et al. (2006): High-threshold inhibitory interneurons Long-range excitatory lateral connections Long-range excitatory feedback connections

CNV Spring 2012: Modeling adult function 4

slide-5
SLIDE 5

LISSOM/GCAL SM model

V1 L2/3I V1 L2/3E V1 L4 (Antolik 2010; Antolik & Bednar 2012)

  • LISSOM/GCAL circuit for

surround modulation

  • Separate inhibitory

interneurons

  • Long-range excitatory

lateral connections

  • Separate simple and

complex cell layers

  • Feedback connections in

progress (Philipp Rudiger)

CNV Spring 2012: Modeling adult function 5

slide-6
SLIDE 6

SM model size tuning

(Antolik 2010)

Single-unit response to larger patterns typically increases, then decreases as inhibition is recruited

CNV Spring 2012: Modeling adult function 6

slide-7
SLIDE 7

Diversity in size tuning

(Antolik 2010)

Model matches both typical and unusual size tuning responses

CNV Spring 2012: Modeling adult function 7

slide-8
SLIDE 8

Diversity in OCTC tuning

(Antolik 2010)

Model matches both typical and unusual orientation-contrast tuning types

CNV Spring 2012: Modeling adult function 8

slide-9
SLIDE 9

The Tilt Aftereffect (TAE)

  • Bias in orientation perception after prolonged exposure
  • Allows model structure to be related to adult function
  • Classic explanation: “fatigue” – activated neurons get

tired, shifting the population average away

CNV Spring 2012: Modeling adult function 9

slide-10
SLIDE 10

Measuring perceived orientation

  • = −16.8
  • = +22.5
  • = 0.0
  • = −3.0

Activation OR preference

Neuron 3:

= 0.6

Neuron 1:

Activation OR preference = 0.24

Neuron 2:

Activation OR preference = 1.0 Activation OR preference

Average:

  • Assumption: perception based on population average
  • Vector average good for cyclic quantities
  • Decode perception before and after adaptation

CNV Spring 2012: Modeling adult function 10

slide-11
SLIDE 11

TAE in Humans and LISSOM

−90

  • −60
  • −30
  • 30
  • 60
  • 90
  • Angle on Retina

−4

  • −2
  • 2
  • 4
  • Aftereffect Magnitude

Mitchell & Muir 1976 HLISSOM

  • Direct effect for

small angles

  • Indirect effect for

larger angles

  • Null effect at

training angle

  • Human, model

match closely

CNV Spring 2012: Modeling adult function 11

slide-12
SLIDE 12

TAE Adaptation in LISSOM

Adaptation

− +

0◦

Direct

− +

10◦

Indirect

− +

60◦ Input pattern V1 Activity Histogram difference

  • Null at zero: More

inhibition, but no net change in perception

  • Direct effect: More

inhibition for angles <10◦ – Perception shifts from 10 to 14◦

  • Indirect effect: Less

inhibition for angles <60◦ – Perception shifts from 60 to 58◦

  • Due to synapses, not

tired neurons!

CNV Spring 2012: Modeling adult function 12

slide-13
SLIDE 13

McCollough effect test pattern

Before adaptation, this pattern should appear monochrome

CNV Spring 2012: Modeling adult function 13

slide-14
SLIDE 14

Adaptation pattern

Stare alternately at the two patterns for 3 minutes, moving your gaze to avoid developing strong afterimages

CNV Spring 2012: Modeling adult function 14

slide-15
SLIDE 15

McCollough effect

(McCollough 1965)

After adaptation:

  • Vertical bars

should be slightly magenta

  • Horizontal bars

should be slightly green

  • The effect should reverse if you tilt your head 90◦,

and disappear if you tilt 45◦.

CNV Spring 2012: Modeling adult function 15

slide-16
SLIDE 16

McCollough effect: data

(Ellis 1977) (Landisman & Ts’o 2002)

2.3×5.3mm macaque V1

  • Effect measured in

humans at each angle between adaptation and test

  • Strength falls off

smoothly with angle

  • V1 is earliest

possible substrate – first area showing OR selectivity; has color map

CNV Spring 2012: Modeling adult function 16

slide-17
SLIDE 17

LISSOM Color V1 Model

(Bednar et al. 2005) Green Channel Red Channel

V1

ON OFF Luminosity Green/Red Red/Green

Color Image Retina LGN

  • Input: RGB

images

  • Decomposed into

Red, Green channels (no blue in central fovea, Calkins 2001)

  • Processed by

color opponent retinal ganglia

CNV Spring 2012: Modeling adult function 17

slide-18
SLIDE 18

LISSOM OR + Color map

(Bednar et al. 2005)

  • Orientation map similar to animal maps
  • Color-selective cells occur in blobs
  • Preferences of neurons in each blob?

CNV Spring 2012: Modeling adult function 18

slide-19
SLIDE 19

Calculating McCollough Effect

  • Perceived color estimated as a vector average of all

units

  • Vector direction: + for red-selective units, - for

green-selective units

  • Weighted by activation level and amount of color

selectivity Result is a number from extreme red (positive) to extreme green (negative), with approximately 0 being monochrome.

CNV Spring 2012: Modeling adult function 19

slide-20
SLIDE 20

Model McCollough Effect

−45 −30 −15 15 30 45 60 75 90 105 120 135 −6 −4 −2 2 4 6

  • rientation of the test pattern

strength of the ME (in the model)

CNV Spring 2012: Modeling adult function 20

slide-21
SLIDE 21

Compared with human

−45 −30 −15 15 30 45 −0.2 0.2 0.4 0.6 0.8 1 1.2

  • rientation of the test pattern

strength of the ME simulated ME human data

CNV Spring 2012: Modeling adult function 21

slide-22
SLIDE 22

Summary

  • LISSOM/GCAL can be compatible with actual circuit
  • Reproduces surprising features of surround modulation
  • Afteffects arise from Hebbian adaptation of lateral

connections

  • The same self-organizing processes can drive both

development and adaptation: both structure and function

  • Novel prediction: Indirect effect due to weight

normalization

  • Project: exactly how does inverted Mexican Hat work?

CNV Spring 2012: Modeling adult function 22

slide-23
SLIDE 23

McCollough Effect

Is the effect still present?

CNV Spring 2012: Modeling adult function 23

slide-24
SLIDE 24

References

Antolik, J. (2010). Unified Developmental Model of Maps, Complex Cells and Surround Modulation in the Primary Visual Cortex. Doctoral Dissertation, School of Informatics, The University of Edinburgh, Edinburgh, UK. Antolik, J., & Bednar, J. A. (2012). A unified developmental model of maps, com- plex cells and surround modulation in the primary visual cortex. In prepa- ration. Bednar, J. A., De Paula, J. B., & Miikkulainen, R. (2005). Self-organization of color opponent receptive fields and laterally connected orientation maps. Neurocomputing, 65–66, 69–76. Calkins, D. J. (2001). Seeing with S cones. Progress in Retinal and Eye Research, 20 (3), 255–287.

CNV Spring 2012: Modeling adult function 23

slide-25
SLIDE 25

Ellis, S. R. (1977). Orientation selectivity of the McCollough effect: Analysis by equivalent contrast transformation. Perception and Psychophysics, 22 (6), 539–544. Hirsch, J. A., & Gilbert, C. D. (1991). Synaptic physiology of horizontal connections in the cat’s visual cortex. The Journal of Neuroscience, 11, 1800–1809. Landisman, C. E., & Ts’o, D. Y. (2002). Color processing in macaque striate cortex: Relationships to ocular dominance, cytochrome oxidase, and orientation. Journal of Neurophysiology, 87 (6), 3126–3137. Law, J. S. (2009). Modeling the Development of Organization for Orientation Pref- erence in Primary Visual Cortex. Doctoral Dissertation, School of Infor- matics, The University of Edinburgh, Edinburgh, UK. Law, J. S., & Bednar, J. A. (2006). Surround modulation by long-range lateral

CNV Spring 2012: Modeling adult function 23

slide-26
SLIDE 26

connections in an orientation map model of primary visual cortex devel-

  • pment and function. In Society for Neuroscience Abstracts. Society for

Neuroscience, www.sfn.org. Program No. 546.4. McCollough, C. (1965). Color adaptation of edge-detectors in the human visual

  • system. Science, 149 (3688), 1115–1116.

Mitchell, D. E., & Muir, D. W. (1976). Does the tilt aftereffect occur in the oblique meridian?. Vision Research, 16, 609–613. Schwabe, L., Obermayer, K., Angelucci, A., & Bressloff, P . C. (2006). The role of feedback in shaping the extra-classical receptive field of cortical neurons: A recurrent network model. The Journal of Neuroscience, 26 (36), 9117– 9129.

CNV Spring 2012: Modeling adult function 23

slide-27
SLIDE 27

Series, P ., Lorenceau, J., & Fregnac, Y. (2003). The “silent” surround of V1 recep- tive fields: Theory and experiments. Journal of Physiology (Paris), 97 (4– 6), 453–474. Weliky, M., Kandler, K., Fitzpatrick, D., & Katz, L. C. (1995). Patterns of excitation and inhibition evoked by horizontal connections in visual cortex share a common relationship to orientation columns. Neuron, 15, 541–552.

CNV Spring 2012: Modeling adult function 23