- Email: [email protected]
The freezing rotation illusion
C. Kennard & R.J. Leigh (Eds.) Progress in Brain Research, Vol. 171 ISSN 0079-6123 Copyright r 2008 Elsevier B.V. All rights reserved
CHAPTER 4.8
The freezing rotation illusion Max R. Du¨rsteler Department of Neurology, University Hospital Zurich, CH-8091 Zurich, Switzerland
Abstract: The ‘‘freezing rotation illusion’’ refers to decrease in perceived speed of a continuously rotating central region when a swaying surround co-rotates. We observed the following effects for rotations: First, when the centre and its surround are turning in the same direction, and their velocities are distinguishable, the perceived speed of the centre is lower than its physical speed. Second, when the surround and the centre are counter-rotating, the perceived speed of the centre is enhanced. Third, even when the surface size of the centre matches the surface size of the surround, swaying of the centre is unable to induce significant changes in the perceived speed of the surround, whereas swaying of the surround still induces changes in the perceived speed of the centre. Keywords: motion illusion; motion perception; figure-ground; induced motion
freezing of a rotating scene on either on headmounted or laptop display while we rolled our heads side-to-side or turned the display back and forth around its roll axis (Du¨rsteler, 2005). To demonstrate the purely visual origin of this freezing rotation illusion, I borrowed Duncker’s experimental design with a small central disc on a larger disc, each of which could rotate independently. Video clips of the freezing motion illusion can be found on the internet (e.g., http://www.youtube. com/watch?v=bZbVTi3Ibmw).
Introduction Perception of an object’s motion depends on the properties of its background. Thus, Duncker (1929) reported how illusionary motion of a central disc, induced by rotation of a larger background disc, could be nulled if subjects counter-rotated the central disc. Pavard and Berthoz (1977) reported a freezing motion phenomenon of presumably vestibular origin: a translating pattern projected on the virtual side windows of a cart appeared stationary when the cart moved abruptly forward. A similar visuo-visual illusion emerged when a stationary subject observed a visual pattern scrolling at a fixed velocity across a mobile monitor: when the monitor started moving, the scrolling pattern appeared to stop (Mesland and Wertheim, 1996; Wertheim and Reymond, 2007). In our laboratory, Stefan Hegemann and I perceived
Methods Three observers (ages 27, 27, and 59) with normal, or corrected-to-normal vision, looked at a 19 in Samsung SyncMaster 912 T liquid crystal display (33.71 25.91) connected to an HP DC7100 PC. In the first experiment, the test stimulus was composed of a surround with black und white random dots (diameter 16.51, luminance contrast 93.8%) rotating at 37.71/s, and a centre with blue and red dots
Corresponding author. Tel.: +41-44-2555542;
Fax: +41-44-2554429; E-mail: [email protected] DOI: 10.1016/S0079-6123(08)00640-7
283
284
Fig. 1. Influence of surround rotation on perceived speed of inner disc. The mean values and 95% confidence intervals of the difference between estimated and physical rotational speed of the inner disc are plotted as a function of the physical speed of the inner disc for conditions with equal (open symbols for the means of single observers, stippled line for the mean of all three observers) or opposing (filled symbol for the means of single observers, continuous line for the mean of all observers) rotational directions. In all trials the rotational speed of the surrounding annulus was set to 37.71/s.
(diameter 11.71, luminance contrast 41.8%) rotating at 5, 10, 20, or 301/s. The probe stimulus was a centre with the black and white Julesz type random dots and a grey surround. The task of the observers was to indicate if the probe’s centre velocity was faster or slower than the velocity of the centre in a preceding test stimulus. The accelerated stochastic approximation method (Kesten, 1958) was used. In the second experiment, the test stimulus was composed of a centre with a white and black random dot pattern turning at a speed of 37.71/s and a surround with blue and red dots turning at speeds of 5, 10, 20, or 301/s. The probe stimulus was composed of a surround painted with the black and white random dot pattern and a grey centre. In both experiments, I performed a two-way ANOVA with speed of the induced test stimulus and the relative direction of inducing and induced stimulus element (opposite or same) as factors, and their interaction. The figures show means with their 95% confidence limits adjusted by Tukey’s honestly significant difference (HSD) criterion for multiple comparisons.
Results First experiment The first experiment examined the effects of the surround rotation direction on the perceived speed
of a continuously turning disc. In all three observers the perceived speeds were on average significantly higher in trials with opposite rotation directions than in trials with the same directions (Fig. 1). In a two-way ANOVA (speed relative rotation direction), the rotation direction (opposite or same) was significant for all observers at a po0.0025. There was no significant interaction ( pW0.49). The differences in rotational speed estimation did not depend in a systematic way on the speed of the centre. The result of the first experiments shows that that perceived speed is enhanced in trials with opposing rotation directions. Second experiment The second experiment tested the effect of the inner disc rotation direction on the perceived speed of the surround (Fig. 2). Two observers did not show significant differences in their speed estimation for conditions with opposite and the same rotation direction ( pW0.06), one observer (diamonds) showed a significant overall difference together with a significant interaction of speed and relative direction ( po0.001), indicating that the differences were significant at some speeds only. The second experiment showed that for the same speeds of the induced stimulus, where we saw significant effects of the surround rotation on the perceived disc rotation, rotations of the inner disc
285
Fig. 2. Influence of figure rotation on perceived speed of surround. The mean values and 95% confidence intervals of the difference between estimated and physical rotational speed of the surrounding ring are plotted as a function of the physical speed of the surround for conditions with equal (open symbols for the means of single observers, stippled line for the mean of all three observers) or opposing (filled symbol for the means of single observers, continuous line for the mean of all observers) rotational directions. In all trials the rotational speed of the inner disc was set to 37.71/s.
did not have a significant effect on the perceived rotation of the surround.
rotation directions on the perceived centre’s rotational speed in the freezing rotation illusion.
Discussion References The freezing rotation illusion suggests that the rotation of an included centre is referenced to the rotation of the surround, but that the rotation of the surround is not referenced to the rotation of the included centre. The rotational speed of the centre is increased by an individually different amount, when its direction is opposite to the direction of the surround, and decreases by a fixed amount, when its direction is the same as the surround’s direction. By focusing attention on the stimulus element whose speed one had to estimate, the surround rotation may have been less effective as when focusing on the whole scene, when looking at freezing rotation illusion in the video. Neuronal correlates of the freezing motion may be found in area MST in the monkey (Duffy and Wurtz, 1991) or part of the MT+ complex in humans (Dukelow et al., 2001). After chemical lesions in area MST including the adjoining floor of the STS in macaques, a directional pursuit deficit was found, where the monkey underestimated the speed of a small target moving towards the lesion side and overestimated the speed of a target moving way from the lesion side (Du¨rsteler et al., 1987), reminiscent to the effects of surround
Duffy, C.J. and Wurtz, R.H. (1991) Sensitivity of MST neurons to optic flow stimuli. II. Mechanisms of response selectivity revealed by small-field stimuli. J. Neurophysiol., 65(6): 1346–1359. Dukelow, S.P., DeSouza, J.F., Culham, J.C., van den Berg, A.V., Menon, R.S. and Vilis, T. (2001) Distinguishing subregions of the human MT+ complex using visual fields and pursuit eye movements. J. Neurophysiol., 86(4): 1991–2000. Duncker, K. (1929) U¨ber induzierte Bewegung (Ein Beitrag zur Theorie optisch wahrgenommener Bewegung). Psychol. Forsch., 12: 180–259. Du¨rsteler, M.R. (2005) Eine neuartige Bewegungsta¨uschung: Duncker Illusion und ‘‘Motion Capture’’ im Widerstreit. Ophta, 3: 19–26. Du¨rsteler, M.R., Wurtz, R.H. and Newsome, W.T. (1987) Directional pursuit deficits following lesions of the foveal representation within the superior temporal sulcus of the macaque monkey. J. Neurophysiol., 57(5): 1262–1287. Kesten, H. (1958) Accelerated stochastic approximation. Ann. Math. Stat., 29: 41–59. Mesland, B.S. and Wertheim, A.H. (1996) A puzzling percept of stimulus stabilization. Vis. Res., 36(20): 3325–3328. Pavard, B. and Berthoz, A. (1977) Linear acceleration modifies the perceived velocity of a moving visual scene. Perception, 6(5): 529–540. Wertheim, A.H. and Reymond, G. (2007) Neural noise distorts perceived motion: the special case of the freezing illusion and the Pavard and Berthoz effect. Exp. Brain Res., 180(3): 569–576.
CHAPTER 4.8
The freezing rotation illusion Max R. Du¨rsteler Department of Neurology, University Hospital Zurich, CH-8091 Zurich, Switzerland
Abstract: The ‘‘freezing rotation illusion’’ refers to decrease in perceived speed of a continuously rotating central region when a swaying surround co-rotates. We observed the following effects for rotations: First, when the centre and its surround are turning in the same direction, and their velocities are distinguishable, the perceived speed of the centre is lower than its physical speed. Second, when the surround and the centre are counter-rotating, the perceived speed of the centre is enhanced. Third, even when the surface size of the centre matches the surface size of the surround, swaying of the centre is unable to induce significant changes in the perceived speed of the surround, whereas swaying of the surround still induces changes in the perceived speed of the centre. Keywords: motion illusion; motion perception; figure-ground; induced motion
freezing of a rotating scene on either on headmounted or laptop display while we rolled our heads side-to-side or turned the display back and forth around its roll axis (Du¨rsteler, 2005). To demonstrate the purely visual origin of this freezing rotation illusion, I borrowed Duncker’s experimental design with a small central disc on a larger disc, each of which could rotate independently. Video clips of the freezing motion illusion can be found on the internet (e.g., http://www.youtube. com/watch?v=bZbVTi3Ibmw).
Introduction Perception of an object’s motion depends on the properties of its background. Thus, Duncker (1929) reported how illusionary motion of a central disc, induced by rotation of a larger background disc, could be nulled if subjects counter-rotated the central disc. Pavard and Berthoz (1977) reported a freezing motion phenomenon of presumably vestibular origin: a translating pattern projected on the virtual side windows of a cart appeared stationary when the cart moved abruptly forward. A similar visuo-visual illusion emerged when a stationary subject observed a visual pattern scrolling at a fixed velocity across a mobile monitor: when the monitor started moving, the scrolling pattern appeared to stop (Mesland and Wertheim, 1996; Wertheim and Reymond, 2007). In our laboratory, Stefan Hegemann and I perceived
Methods Three observers (ages 27, 27, and 59) with normal, or corrected-to-normal vision, looked at a 19 in Samsung SyncMaster 912 T liquid crystal display (33.71 25.91) connected to an HP DC7100 PC. In the first experiment, the test stimulus was composed of a surround with black und white random dots (diameter 16.51, luminance contrast 93.8%) rotating at 37.71/s, and a centre with blue and red dots
Corresponding author. Tel.: +41-44-2555542;
Fax: +41-44-2554429; E-mail: [email protected] DOI: 10.1016/S0079-6123(08)00640-7
283
284
Fig. 1. Influence of surround rotation on perceived speed of inner disc. The mean values and 95% confidence intervals of the difference between estimated and physical rotational speed of the inner disc are plotted as a function of the physical speed of the inner disc for conditions with equal (open symbols for the means of single observers, stippled line for the mean of all three observers) or opposing (filled symbol for the means of single observers, continuous line for the mean of all observers) rotational directions. In all trials the rotational speed of the surrounding annulus was set to 37.71/s.
(diameter 11.71, luminance contrast 41.8%) rotating at 5, 10, 20, or 301/s. The probe stimulus was a centre with the black and white Julesz type random dots and a grey surround. The task of the observers was to indicate if the probe’s centre velocity was faster or slower than the velocity of the centre in a preceding test stimulus. The accelerated stochastic approximation method (Kesten, 1958) was used. In the second experiment, the test stimulus was composed of a centre with a white and black random dot pattern turning at a speed of 37.71/s and a surround with blue and red dots turning at speeds of 5, 10, 20, or 301/s. The probe stimulus was composed of a surround painted with the black and white random dot pattern and a grey centre. In both experiments, I performed a two-way ANOVA with speed of the induced test stimulus and the relative direction of inducing and induced stimulus element (opposite or same) as factors, and their interaction. The figures show means with their 95% confidence limits adjusted by Tukey’s honestly significant difference (HSD) criterion for multiple comparisons.
Results First experiment The first experiment examined the effects of the surround rotation direction on the perceived speed
of a continuously turning disc. In all three observers the perceived speeds were on average significantly higher in trials with opposite rotation directions than in trials with the same directions (Fig. 1). In a two-way ANOVA (speed relative rotation direction), the rotation direction (opposite or same) was significant for all observers at a po0.0025. There was no significant interaction ( pW0.49). The differences in rotational speed estimation did not depend in a systematic way on the speed of the centre. The result of the first experiments shows that that perceived speed is enhanced in trials with opposing rotation directions. Second experiment The second experiment tested the effect of the inner disc rotation direction on the perceived speed of the surround (Fig. 2). Two observers did not show significant differences in their speed estimation for conditions with opposite and the same rotation direction ( pW0.06), one observer (diamonds) showed a significant overall difference together with a significant interaction of speed and relative direction ( po0.001), indicating that the differences were significant at some speeds only. The second experiment showed that for the same speeds of the induced stimulus, where we saw significant effects of the surround rotation on the perceived disc rotation, rotations of the inner disc
285
Fig. 2. Influence of figure rotation on perceived speed of surround. The mean values and 95% confidence intervals of the difference between estimated and physical rotational speed of the surrounding ring are plotted as a function of the physical speed of the surround for conditions with equal (open symbols for the means of single observers, stippled line for the mean of all three observers) or opposing (filled symbol for the means of single observers, continuous line for the mean of all observers) rotational directions. In all trials the rotational speed of the inner disc was set to 37.71/s.
did not have a significant effect on the perceived rotation of the surround.
rotation directions on the perceived centre’s rotational speed in the freezing rotation illusion.
Discussion References The freezing rotation illusion suggests that the rotation of an included centre is referenced to the rotation of the surround, but that the rotation of the surround is not referenced to the rotation of the included centre. The rotational speed of the centre is increased by an individually different amount, when its direction is opposite to the direction of the surround, and decreases by a fixed amount, when its direction is the same as the surround’s direction. By focusing attention on the stimulus element whose speed one had to estimate, the surround rotation may have been less effective as when focusing on the whole scene, when looking at freezing rotation illusion in the video. Neuronal correlates of the freezing motion may be found in area MST in the monkey (Duffy and Wurtz, 1991) or part of the MT+ complex in humans (Dukelow et al., 2001). After chemical lesions in area MST including the adjoining floor of the STS in macaques, a directional pursuit deficit was found, where the monkey underestimated the speed of a small target moving towards the lesion side and overestimated the speed of a target moving way from the lesion side (Du¨rsteler et al., 1987), reminiscent to the effects of surround
Duffy, C.J. and Wurtz, R.H. (1991) Sensitivity of MST neurons to optic flow stimuli. II. Mechanisms of response selectivity revealed by small-field stimuli. J. Neurophysiol., 65(6): 1346–1359. Dukelow, S.P., DeSouza, J.F., Culham, J.C., van den Berg, A.V., Menon, R.S. and Vilis, T. (2001) Distinguishing subregions of the human MT+ complex using visual fields and pursuit eye movements. J. Neurophysiol., 86(4): 1991–2000. Duncker, K. (1929) U¨ber induzierte Bewegung (Ein Beitrag zur Theorie optisch wahrgenommener Bewegung). Psychol. Forsch., 12: 180–259. Du¨rsteler, M.R. (2005) Eine neuartige Bewegungsta¨uschung: Duncker Illusion und ‘‘Motion Capture’’ im Widerstreit. Ophta, 3: 19–26. Du¨rsteler, M.R., Wurtz, R.H. and Newsome, W.T. (1987) Directional pursuit deficits following lesions of the foveal representation within the superior temporal sulcus of the macaque monkey. J. Neurophysiol., 57(5): 1262–1287. Kesten, H. (1958) Accelerated stochastic approximation. Ann. Math. Stat., 29: 41–59. Mesland, B.S. and Wertheim, A.H. (1996) A puzzling percept of stimulus stabilization. Vis. Res., 36(20): 3325–3328. Pavard, B. and Berthoz, A. (1977) Linear acceleration modifies the perceived velocity of a moving visual scene. Perception, 6(5): 529–540. Wertheim, A.H. and Reymond, G. (2007) Neural noise distorts perceived motion: the special case of the freezing illusion and the Pavard and Berthoz effect. Exp. Brain Res., 180(3): 569–576.