Abstract

We show that a broad class of visual illusions, including illusory motion, can be explained by the effects of negative afterimages. Two new illusions, illusory shading and illusory tilting, are devised on the basis of the proposed explanation. The general feature of these illusions is an alternation between a high-contrast (white or black) and a low-contrast (gray) local input signal, which can be caused either by eye motion over patterns of varied luminance or by a change in such patterns over time. A simple model of the local signal dynamics qualitatively reproduces the illusory effects by adding the negative afterimage to the original visual stimulus.

© 2002 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. A. V. Popple, D. M. Levi, M. J. Wright, “An anomalous motion illusion,” Invest. Ophthalmol. Visual Sci. Suppl. 42, S4665 (2001).
  2. B. Khang, E. A. Essock, “Apparent swinging motion from a 2-d sinusoidal pattern,” Perception 29, 453–459 (2000).
    [CrossRef] [PubMed]
  3. L. K. Cormack, S. B. Stevenson, “Exploring motion aliasing under steady illumination (abstract)” J. Vision 1, 248a (2001).
    [CrossRef]
  4. H. Ouchi, Japanese and Geometrical Art (Dover, New York, 1977).
  5. B. Pinna, G. J. Brelstaff, “A new visual illusion of relative motion,” Vision Res. 40, 2091–2096 (2000).
    [CrossRef] [PubMed]
  6. R. W. Bowen, H. R. Wilson, “A two-process analysis of pattern masking,” Vision Res. 34, 645–657 (1994).
    [CrossRef] [PubMed]
  7. H. R. Wilson, “A neural model of foveal light adaptation and afterimage formation,” Visual Neurosci. 14, 403–423 (1997).
    [CrossRef]
  8. D. H. Kelly, C. A. Burbeck, “Motion and vision. III. Stabilized pattern adaptation,” J. Opt. Soc. Am. 70, 1283–1289 (1980).
    [CrossRef] [PubMed]
  9. C. A. Burbeck, D. H. Kelly, “Role of local adaptation in the fading of stabilized images,” J. Opt. Soc. Am. A 1, 216–220 (1984).
    [CrossRef] [PubMed]
  10. P. Cavanagh, S. M. Anstis, “Brightness shift in drifting ramp gratings isolates a transient mechanism,” Vision Res. 26, 899–908 (1985).
    [CrossRef]

2001 (2)

A. V. Popple, D. M. Levi, M. J. Wright, “An anomalous motion illusion,” Invest. Ophthalmol. Visual Sci. Suppl. 42, S4665 (2001).

L. K. Cormack, S. B. Stevenson, “Exploring motion aliasing under steady illumination (abstract)” J. Vision 1, 248a (2001).
[CrossRef]

2000 (2)

B. Pinna, G. J. Brelstaff, “A new visual illusion of relative motion,” Vision Res. 40, 2091–2096 (2000).
[CrossRef] [PubMed]

B. Khang, E. A. Essock, “Apparent swinging motion from a 2-d sinusoidal pattern,” Perception 29, 453–459 (2000).
[CrossRef] [PubMed]

1997 (1)

H. R. Wilson, “A neural model of foveal light adaptation and afterimage formation,” Visual Neurosci. 14, 403–423 (1997).
[CrossRef]

1994 (1)

R. W. Bowen, H. R. Wilson, “A two-process analysis of pattern masking,” Vision Res. 34, 645–657 (1994).
[CrossRef] [PubMed]

1985 (1)

P. Cavanagh, S. M. Anstis, “Brightness shift in drifting ramp gratings isolates a transient mechanism,” Vision Res. 26, 899–908 (1985).
[CrossRef]

1984 (1)

1980 (1)

Anstis, S. M.

P. Cavanagh, S. M. Anstis, “Brightness shift in drifting ramp gratings isolates a transient mechanism,” Vision Res. 26, 899–908 (1985).
[CrossRef]

Bowen, R. W.

R. W. Bowen, H. R. Wilson, “A two-process analysis of pattern masking,” Vision Res. 34, 645–657 (1994).
[CrossRef] [PubMed]

Brelstaff, G. J.

B. Pinna, G. J. Brelstaff, “A new visual illusion of relative motion,” Vision Res. 40, 2091–2096 (2000).
[CrossRef] [PubMed]

Burbeck, C. A.

Cavanagh, P.

P. Cavanagh, S. M. Anstis, “Brightness shift in drifting ramp gratings isolates a transient mechanism,” Vision Res. 26, 899–908 (1985).
[CrossRef]

Cormack, L. K.

L. K. Cormack, S. B. Stevenson, “Exploring motion aliasing under steady illumination (abstract)” J. Vision 1, 248a (2001).
[CrossRef]

Essock, E. A.

B. Khang, E. A. Essock, “Apparent swinging motion from a 2-d sinusoidal pattern,” Perception 29, 453–459 (2000).
[CrossRef] [PubMed]

Kelly, D. H.

Khang, B.

B. Khang, E. A. Essock, “Apparent swinging motion from a 2-d sinusoidal pattern,” Perception 29, 453–459 (2000).
[CrossRef] [PubMed]

Levi, D. M.

A. V. Popple, D. M. Levi, M. J. Wright, “An anomalous motion illusion,” Invest. Ophthalmol. Visual Sci. Suppl. 42, S4665 (2001).

Ouchi, H.

H. Ouchi, Japanese and Geometrical Art (Dover, New York, 1977).

Pinna, B.

B. Pinna, G. J. Brelstaff, “A new visual illusion of relative motion,” Vision Res. 40, 2091–2096 (2000).
[CrossRef] [PubMed]

Popple, A. V.

A. V. Popple, D. M. Levi, M. J. Wright, “An anomalous motion illusion,” Invest. Ophthalmol. Visual Sci. Suppl. 42, S4665 (2001).

Stevenson, S. B.

L. K. Cormack, S. B. Stevenson, “Exploring motion aliasing under steady illumination (abstract)” J. Vision 1, 248a (2001).
[CrossRef]

Wilson, H. R.

H. R. Wilson, “A neural model of foveal light adaptation and afterimage formation,” Visual Neurosci. 14, 403–423 (1997).
[CrossRef]

R. W. Bowen, H. R. Wilson, “A two-process analysis of pattern masking,” Vision Res. 34, 645–657 (1994).
[CrossRef] [PubMed]

Wright, M. J.

A. V. Popple, D. M. Levi, M. J. Wright, “An anomalous motion illusion,” Invest. Ophthalmol. Visual Sci. Suppl. 42, S4665 (2001).

Invest. Ophthalmol. Visual Sci. Suppl. (1)

A. V. Popple, D. M. Levi, M. J. Wright, “An anomalous motion illusion,” Invest. Ophthalmol. Visual Sci. Suppl. 42, S4665 (2001).

J. Opt. Soc. Am. (1)

J. Opt. Soc. Am. A (1)

J. Vision (1)

L. K. Cormack, S. B. Stevenson, “Exploring motion aliasing under steady illumination (abstract)” J. Vision 1, 248a (2001).
[CrossRef]

Perception (1)

B. Khang, E. A. Essock, “Apparent swinging motion from a 2-d sinusoidal pattern,” Perception 29, 453–459 (2000).
[CrossRef] [PubMed]

Vision Res. (3)

P. Cavanagh, S. M. Anstis, “Brightness shift in drifting ramp gratings isolates a transient mechanism,” Vision Res. 26, 899–908 (1985).
[CrossRef]

B. Pinna, G. J. Brelstaff, “A new visual illusion of relative motion,” Vision Res. 40, 2091–2096 (2000).
[CrossRef] [PubMed]

R. W. Bowen, H. R. Wilson, “A two-process analysis of pattern masking,” Vision Res. 34, 645–657 (1994).
[CrossRef] [PubMed]

Visual Neurosci. (1)

H. R. Wilson, “A neural model of foveal light adaptation and afterimage formation,” Visual Neurosci. 14, 403–423 (1997).
[CrossRef]

Other (1)

H. Ouchi, Japanese and Geometrical Art (Dover, New York, 1977).

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (3)

Fig. 1
Fig. 1

Illusions produced by negative afterimage effects. (a) Illusory shading. Hold the picture close to your face, look in the center of the pattern, and move the picture forward and backward in front of your eyes. Luminance of different segments of the pattern appears to switch between lighter and darker shades as the motion of the picture reverses. (b) Illusory motion. Move the picture as described above. Dark and light “spokes” appear to shift when the direction of motion reverses. (c) “Swinging” illusion. Stare at the central dot for a few seconds and then steadily move your gaze between the leftmost and the rightmost dots to see the central column float in the opposite direction. It might help instead to move a tip of a pen over the picture and follow it with your eyes. (d) “Tilting” illusion. Move your eyes as in (c) and observe that gratings in the upper and lower half tilt in opposite directions. (e) The time course of the “drifting grating” illusion. As the sinusoidal grating drifts to the left with a constant speed, the overall contrast changes from its maximum to zero and back. During the low-contrast phase (frame 4) the grating appears first to accelerate and then to reverse its direction of motion for a moment. (f) The time course of the drifting grating illusion in (e) shown for one row of pixels.

Fig. 2
Fig. 2

Model of negative afterimage formation. The local visual input typical for the discussed illusions, i.e., a high-contrast stimulus (white-to-black edge) followed by a low-contrast stimulus (Gray area) is shown in the top panel. The negative afterimage component of the model (middle panel) is linearly combined with the input signal. The model output is shown in the bottom panel together with the output limits (saturation threshold) used (dotted–dashed lines). The out-of-limits portion of the output is shown by the dashed part of the curve.

Fig. 3
Fig. 3

Model simulation results. For (a) and (b) the left and the right images of the output correspond to moving the picture away from and toward observer, respectively. Arrows in (a) show the local motion. Dashed lines in (b) mark centroids of the dark spokes. Swinging and tilting stimuli are grouped together [(c) and (d), respectively], since they are produced by the same luminance pattern. For (c) a mismatch of the central column position between the input and output [indicated by dashed lines in (c)] creates an illusion of the column swinging in the direction opposite to the eye movements. In (d), circles mark regions of the strongest afterimages in the output. The overall effect is to reduce the apparent tilt. For the drifting grating illusion in (e), time intervals are shown in which afterimages create the illusion of acceleration and reversal.

Equations (2)

Equations on this page are rendered with MathJax. Learn more.

O(t, i, j)=I(t, i, j)+A(t, i, j).
A(t, i, j)t=-A(t, i, j)+I(t, i, j)τ,

Metrics