Journal of Vision (2007) 7(8):16, 1–7 http://journalofvision.org/7/8/16/ 1

Backscroll illusion in far peripheral vision

Department of Psychology, Kwansei Gakuin University, Kiyoshi Fujimoto Nishinomiya, Hyogo, Japan

Department of Psychology, Kwansei Gakuin University, Akihiro Yagi Nishinomiya, Hyogo, Japan

The backscroll illusion refers to the apparent motion perceived in the background of a movie image that presents a locomotive object such as a person, an animal, or a vehicle. Here, we report that the backscroll illusion can occur in far peripheral visual fields at retinal eccentricity of more than 30-. In psychological experiments, we presented a walking person in profile against an ambiguously moving background of vertical counterphase grating. This stimulus, which subtended 30- of visual angle in width and height, was projected onto a hemispheric screen and positioned at horizontal eccentricity between 0- and 50- at intervals of 10-. The eccentricity was changed randomly trial by trial, and stimulus duration was as short as 350 ms so that observers could not effectively move their eyes to the stimulus. Six observers viewed the stimulus either monocularly or binocularly and reported their perceptual impression for the grating in a three-alternative forced-choice procedure: drifting left, drifting right, or flickering. Results showed that the grating appeared to move in the opposite direction of walking at high probabilities even in the far periphery. Additional experiments confirmed that walking action could be recognized from the far peripheral stimulation. Our findings suggest that the visual system uses high-level object-centered motion signals to disambiguate retinal motion signals in the whole visual field. Keywords: peripheral vision, apparent motion, biological movements Citation: Fujimoto, K., & Yagi, A. (2007). Backscroll illusion in far peripheral vision. Journal of Vision, 7(8):16, 1–7, http://journalofvision.org/7/8/16/, doi:10.1167/7.8.16.

Introduction Thorpe, Gegenfurtner, Fabre-Thorpe, and Bu¨lthoff (2001) reported that human observers could detect an animal in a photograph of a natural scene presented at a The backscroll illusion refers to apparent motion seen in far peripheral visual field of more than 30- eccentricity. a pattern behind a locomotive object such as a person, an This suggests that far peripheral vision can perform animal, or a vehicle. It is effectively demonstrated in a complex figure–ground segregation. However, perhaps movie clip in which a person makes a treadmill step due to the brief presentation time of 28 ms employed in against a counterphase grating background (Fujimoto & their study, the accuracy was relatively low and identi- Sato, 2006). Although the grating has physically ambig- fication was impossible. Ikeda, Blake, and Watanabe uous motion, it appears to drift in the opposite direction of (2005) reported that the recognition of the biological the gait. This illusion indicates that the visual system motion of human actions was poor even at a near evaluates retinal motion signals in relation to high-level peripheral field of less than 20- eccentricity. representations of object motion. We have hypothesized The perception of relatively simple motion is also that such object-centered motion perception plays a role in inaccurate in peripheral vision. For instance, Georgeson the stability of the visual world in everyday life, where and Harris (1978) reported a foveofugal bias for an retinal motion is inherently ambiguous due to the ambiguously moving pattern. They presented a counter- observer’s movements or the aperture problem (Fujimoto & phase grating at a parafoveal field of 5- eccentricity. Sato, 2006). Observers tended to perceive unidirectional motion away Here, we would like to raise a simple question. Can from the fovea. Such a foveofugal bias also has been peripheral vision perceive the backscroll illusion? If so, found in the perception of unidirectional stimuli (Ball & what amount of eccentricity makes it possible? Moving Sekuler, 1980; Dumoulin, Baker, & Hess, 2001). Other objects regularly appear in the whole visual field in studies, on the other hand, have recognized a centripetal everyday life. An investigation of peripheral vision will bias (Mateeff & Hohnsbein, 1988; Raymond, 1994). therefore provide implications for the above hypothesis. In Edwards and Badcock (1993) reported that anisotropy this study, we conducted psychological experiments in decreased with increasing eccentricity up to 24- of which human observers viewed a stimulus movie appear- eccentricity. By contrast, Fahle and Wehrhahn (1991) ing at various locations, including far peripheral visual found a foveofugal bias in far peripheral vision. Such fields up to 50- of retinal eccentricity. discrepancies may come from differences of stimuli.

doi: 10.1167/7.8.16 Received March 15, 2007; published June 29, 2007 ISSN 1534-7362 * ARVO

Downloaded from jov.arvojournals.org on 09/26/2021 Journal of Vision (2007) 7(8):16, 1–7 Fujimoto & Yagi 2

In addition to anisotropies, peripheral vision is poor at discriminating the direction and velocity of moving stimuli (Anderson & Hess, 1990; Orban, Van Calenbergh, De Bruyn, & Maes, 1985). However, this is mainly the result of undersampling at the early level of visual processing. Retinal cells and their projection to the primary visual cortex become sparse with increasing eccentricity (Curcio, Sloan, Kalina, & Hendrickson, 1990; Horton & Hoyt, 1991; Jonas, Schneider, & Naumann, 1992). Magnifying stimulus size is therefore predicted to compensate for such undersampling so that perceptual performance will improve (McKee & Nakayama, 1984; Virsu & Rovamo, 1979). It is likely that motion processing in peripheral vision is not essentially different from that in central vision. Our current study used large stimulus displays to minimize influences from inhomogeneities. Their large sizes were also natural in comparison with retinal images in everyday viewing conditions. Figure 1. Photograph of an experimental scene. An observer sat opposite to the hemispheric screen with his or her head supported by a head and chin rest. An LCD projector was placed on the desk, on which was also placed a keyboard for response Methods acquisition. See the text for details about this Elumens Vision- Station system. The box at the left side of the figure was the preamplifier of an EOG acquisition system, from which electrode Observers wires were coupled to the observer’s face. Illumination was off in actual experimental sessions. Although not clearly visible in this Six adults (22–35 years old) voluntarily participated in photograph, a stimulus movie appears around the observer’s right the experiment. Five saw stimuli with their naked eyes, shoulder, in which a girl wearing a red shirt, white short pants, and whereas the remaining one wore soft contact lenses. The red shoes appears against a background of vertical sinusoidal visual acuity of all participants was 1.0 or better. grating. A demonstration movie can be seen at the following web site: http://backscroll.jp/.

Apparatus of physical and retinal size, although it kept the same pixel resolution. The walker subtended 6- in width and We used an Elumens VisionStation system (Figure 1) 11.6–14.0- in height, corresponding to everyday retinal consisting of a hemispheric screen and an LCD projector sizes produced when one looks at a person of 1.5–1.8 m in (EPSON LMP-710). The hemispheric screen covered height from a distance of about 7.3 m. The step length was 160- of the horizontal visual field, and its physical 5.8-, defined by the distance between the arches of the feet dimensions were as follows: 1.7 m in width, 1.4 m in when both legs were the most outstretched. One step height, and 0.84 m in radius. The radius was the viewing lasted 500 ms, if completed, and apparent walking distance, which was fixed by a head and chin rest. The velocity was 11.6-/s. The grating subtended 30- in both LCD projector was placed on a desk in front of the screen width and height, with contrast modulation by a Gaussian and about 0.5 m below the observer’s eyes. A fish-eye window of 6- SD. The spatial frequency and temporal lens was mounted on the projector through which images frequency were 0.5 cycles/deg and 4.0 Hz, respectively, were projected onto the hemispheric screen.1,2 The which produced velocity components at 8-/s. projector had a resolution of 1,024 Â 768 pixels and a refresh rate of 60 Hz controlled by a personal computer (Apple Power Mac G4). Procedure

Observers initiated each trial at their own pace by Stimulus pressing a key after being signaled by a warning tone and the appearance of the fixation point. The stimulus was The stimuli were movie clips that presented a walker in presented after a 500-ms interval during which only the profile against a background of vertical sinusoidal coun- fixation point was presented. The stimulus presentation terphasing grating. This was a magnified version of the lasted 350 ms. Observers responded by pressing one of previously reported one (Fujimoto & Sato, 2006) in terms three designated keys when the stimulus disappeared.

Downloaded from jov.arvojournals.org on 09/26/2021 Journal of Vision (2007) 7(8):16, 1–7 Fujimoto & Yagi 3 There were 11 conditions of eccentricity, ranging Results between 0- and 50- at intervals of 10- in the left and right visual fields. These values pointed to the center of the stimulus. The order of the conditions was randomized. Binocular viewing Observers were asked not to move their eyes voluntarily, and eye movements were monitored with an electrooculo- The data were classified by relating the facing direction gram (EOG). All trials with more than 5- of eye movement of a person to the fovea to evaluate perceptual bias from were excluded from the data analysis.3 This occurred in the counterphase grating per se. We labeled the peripheral less than 0.5% of the trials. fields as foveopetal and foveofugal. Foveopetal refers to a The observers’ task was to report their perceptual condition in which a person faced the fovea, whereas impression for the grating in a three-alternative forced- foveofugal refers to a condition in which a person faced choice procedure. The alternatives were flickering, drifting away from the fovea. The foveopetal and foveofugal left, and drifting right. In the data analysis, the directional conditions are represented as negative and positive responses were classified as the opposite response or the eccentricities, respectively, on the abscissas of the graphs same response in relation to the walking direction. Response shown in Figure 2. percentages were collected from a total of 16 trials (8 human As shown in Figure 2a, an animated walker induced the models  2 facing directions) if no trial was excluded due to opposite response at probabilities of 74.0–97.9%. These excessive eye movement as described above. rates were all significantly higher than the same response To evaluate the statistical significance of the difference rates based on the directional indexes, 0.84 G DI values e between the directional responses, as in our previous 1.00, t values(5) 9 8.37, p values G .001. The probabilities study (Fujimoto & Sato, 2006), we calculated the Direc- were highest at 10–30- eccentricities of the foveopetal tional Index (DI) using the following equation: DI = conditions. A one-way ANOVA revealed a marginally (% opposite j % same)/(% opposite + % same). DI ranged significant effect, F(10, 50) = 1.79, p G .10. between j1 and 1. Positive values indicated dominance of Figure 2b illustrates the response probabilities for the the opposite response, whereas negative values indicated static walker condition. The opposite responses occurred dominance of the same response. Statistical significance 53.1–77.1%, rates that were also significantly higher than was evaluated by a single-sampled t test for the mean DI the same response rates, 0.40 G DI values G 0.94, t values versus 0. When the denominator was zero due to no (5) 9 2.70, p values G .05. Although the probabilities were directional responses, DI was designated as 0. highest at 10- eccentricity of the foveofugal condition, a In addition to the movie clip stimuli consisting of an one-way ANOVA showed no significant effect, F G 1. animated walker and a counterphase grating, we presented Under both walker conditions, an opposite response also two other types of movie stimuli. One consisted of a static means a specific directional perception of the grating. walker and a counterphase grating, and the other consisted That is, foveofugal perception occurred at negative of a counterphase grating alone. For each of these three eccentricities, whereas foveopetal perception occurred at movie conditions, 176 trials of stimuli were presented. positive eccentricities. The probabilities of corresponding The order of the conditions was randomized across responses occurring in the condition in which only the observers. counterphase grating was presented are illustrated in To confirm that observers could recognize walking Figure 2c. There was no fugal–petal bias, F G 1. action, we performed an additional experiment. The Figure 2d illustrates the recognition performance of stimuli were those used in the animated walker condition walking action. For the normal forward walk, correct and their reverse sequences. These reverse animations response rates were quite high and significantly above the looked like a person walking backward. The observers’ chance level of 50%, t values(5) 9 4.07, p values G .001. task was to judge the walking direction with a two- The backward walk also was correctly recognized, t values alternative forced choice: left or right. A correct response (5) 9 4.69, p values G .01, except for the +50- eccentricity to a backward walker required the integration of the gait condition, t(5) = 1.68, ns. direction and the body facing direction, whereas a forward walker could be recognized from the facing direction alone. Thus, performance for the backward walker more Monocular viewing adequately indicated the extent to which observers correctly recognized walking action. The data were classified according to nasal–temporal Observers viewed the stimuli either binocularly or retinal fields to investigate constraints from the input monocularly. In the monocular condition, observers used levels of the visual system. On the abscissas of the the dominant eye determined by the hole-in-card method. graphs shown in Figure 3, temporal and nasal fields are An eye patch covered the other eye. The sessions for each represented as negative and positive eccentricities, viewing condition ran on separate days. respectively.

Downloaded from jov.arvojournals.org on 09/26/2021 Journal of Vision (2007) 7(8):16, 1–7 Fujimoto & Yagi 4

Figure 2. Results of the binocular viewing conditions. (a) Mean probabilities of responses, indicating that a counterphase grating appeared to move in the opposite direction of an animated walker (open circle), in the same direction (filled circle), and to flicker (cross). (b) Mean probabilities of perceptual responses to a counterphase grating onto which a static walker was superimposed. (c) Mean probabilities indicating bias from a counterphase grating without the superimposition of a human figure. The open circles denote unidirectional perception corresponding to the opposite response for the animated and static walker conditions. The crosses denote flicker perception. (d) Correct response rates for discriminating the walking direction in forward walk (open circles) and backward walk (filled circle) conditions. In all graphs, the data were classified according to the walker’s direction in relation to the fovea. The negative eccentricities indicate conditions in which a person faced toward the fovea (foveopetal). By contrast, the positive eccentricities indicate conditions in which a person faced away from the fovea (foveofugal). Each marker and each error bar indicate the mean and standard deviation, respectively, for the six observers.

As shown in Figure 3a, an animated walker induced the significant differences between j50- and j20-, j30-, opposite response at probabilities of 59.4–92.7%. These +20-, +30-, and +40- eccentricities (p G .05). were significantly higher than the same response rates, Figure 3b illustrates the response probabilities for the 0.72 G DI values e 1.00, t values(5) 9 6.68, p values G .01. static walker condition. The opposite responses occurred There was a drop at 50- eccentricity in the temporal 40.6–68.8%, which were significantly higher than the field. A one-way ANOVA revealed a significant effect, same response rates, 0.33 G DI values G 0.94, t values(5) 9 F(10, 50) = 2.89, p G .01. Tukey’s HSD tests indicated 2.69, p values G .05. There is a small drop at the temporal

Downloaded from jov.arvojournals.org on 09/26/2021 Journal of Vision (2007) 7(8):16, 1–7 Fujimoto & Yagi 5

Figure 3. Results of the monocular viewing conditions. The graph formats are identical to those in Figure 2, except that the abscissas here indicate eccentricities on a retina. The positive and negative values represent the temporal and nasal sides, respectively.

50- eccentricity. A one-way ANOVA indicated a margin- observed in the far periphery were as high as those in ally significant effect, F(10, 50) = 1.85, p G .10. the center. Monocular viewing lowered the probability in Figure 3d illustrates the recognition performance of the far temporal retina. This result can be explained by the walking action. For the normal forward walk, correct constraints of input level. That is, the density of retinal response rates were quite high and significantly above the cells in the temporal field is smaller than that in the nasal chance level of 50%, t values(5) 9 4.02, p values G .05. field (Curcio et al., 1990; Jonas et al., 1992). Recognition The backward walk also was correctly recognized, t values of walk was also poor in the far temporal retina. However, (5) 9 3.72, p values G .05. the current data were not strongly correlated with eccentricity. This could be due to the ceiling effect caused by the large stimulus displays. The involvement of higher Discussion level processing was also probable. The results of the binocular condition suggested some anisotropy. The probabilities of the backscroll illusion The present results demonstrated that peripheral vision increased when a person appeared to walk away from the can perceive the backscroll illusion. The probabilities fovea. However, the results of the backward walk

Downloaded from jov.arvojournals.org on 09/26/2021 Journal of Vision (2007) 7(8):16, 1–7 Fujimoto & Yagi 6

conditions implied that a foveofugally facing person had its background image. Ikeda et al. (2005) reported that deteriorated recognition of walking action. This result is peripheral vision is poor at recognizing biological motion. paradoxical if the recognition of object locomotion However, this may be attributed to their use of artificial determines the backscroll illusion. However, this finding stimulus settings, in which a fragmented point-light figure of poorer recognition in the backward walk conditions can was embedded in dynamic point-light noise. By contrast, be explained by the interference caused by the facing in this study, we used realistic animations. Moreover, the direction of the person walking, which can determine the largeness of our stimuli was likely advantageous to direction of locomotion in everyday life because a back- recognition. There is no doubt that peripheral vision is ward gait is rare. It was possible, therefore, that shape inferior to central vision in various aspects. However, this information dominated motion information in walk rec- inferiority is relative. Our data indicate that peripheral ognition. This argument is also supported by the occur- vision has the ability to see a moving person at an rence of illusion from the static figures in the current and ordinary distance. It is therefore important to make previous experiments (Fujimoto & Sato, 2006). In considerations for the actual viewing conditions of every- addition, a foveofugal walker might require attention, for day life when investigating the recognition of natural instance, to prepare tracking eye movements; otherwise, objects or events. he or she will gradually fade away from the visual field in everyday situations. We therefore suggest that there are high-level factors that affect the backscroll illusion in peripheral vision. Further investigations are required to more fully under- Acknowledgments stand the backscroll illusion in peripheral vision. One possible idea for future studies is the use of a smaller This study was supported by the Grant-in-Aid for display, which should worsen the recognition of walking Scientific Research in Academic Frontier Promotion and alter the appearance of the illusion. A smaller display Project provided by the Japanese Ministry of Education, would also allow us to formulate a magnification factor Culture, Sports, and Technology. Part of this study was and make comparisons with other types of perceptual presented at the sixth annual meeting of the Vision phenomena in peripheral vision. Sciences Society in Sarasota, USA, May 5–10, 2006. Although neural mechanisms underlying the backscroll illusion are unclear, the most likely candidate is networks Commercial relationships: none. among motion-specific areas in the occipital–temporal Corresponding author: Kiyoshi Fujimoto. cortex of the brain. Several studies have consistently Email: [email protected]. reported that human bodily movements activated areas Address: Uegahara 1-1-155, Nishinomiya, 662-8501, around the superior temporal sulcus (for review, see Puce & Hyogo, Japan. Perrett, 2003). On the other hand, motion components of a counterphase grating are extracted in the primary visual (V1) area and integrated in the middle temporal (MT+, the human homologue to MT/MST in macaque monkeys) area Footnotes (Heeger, Boynton, Demb, Seidemann, & Newsome, 1999). Thus, neural connections among those areas 1 probably have a relation to the backscroll illusion. A There was a mismatch between the curvatures of the recent study using monkeys as subjects showed that fish-eye lens and the hemispheric screen, such that pixel different parts of MT receive inputs from different brain size enlarged gradually from the center to the edges of the areas according to representation of the visual field screen. The size at 50- eccentricity was about 1.4 times as (Palmer & Rosa, 2006). Central and near-peripheral large as that at 0- eccentricity. No compensation was representations receive inputs from various visual areas, applied to maintain the resolution of the stimulus images. whereas far periphery receives inputs exclusively from Such enlargement was trivial in comparison with the more V1, MST, and the retrosplenial cortex, which probably than 14 times enlargement produced by cortical magnifi- plays a role in visual information processing for rapid cation factors (Horton & Hoyt, 1991; Rousselet, Husk, reactions such as orienting or postural actions. The current Bennett, & Sekuler, 2005; Rovamo & Virsu, 1979). For results were slightly linked to such a neural organization. convenience, this article describes stimulus parameters in If further investigations find that the backscroll illusion is the2 central visual field. differently perceived between far and other visual fields, it Luminance decreased from the center to the periphery. may provide clues to the understanding of neural No compensation was applied because the reduction was mechanisms. small at the mean luminance level, which ranged from This study is the first to confirm that far peripheral 2 - - 203 cd/m at 0 eccentricity to 14 cd/m2 at 50 eccentricity. vision can recognize complex human motion quite We confirmed few eye movements from further robustly, to the extent that it induces apparent motion in analyses of EOG data.

Downloaded from jov.arvojournals.org on 09/26/2021 Journal of Vision (2007) 7(8):16, 1–7 Fujimoto & Yagi 7

Graefe’s Archive for Clinical and Experimental References Ophthalmology, 230, 505–510. [PubMed] Mateeff, S., & Hohnsbein, J. (1988). Perceptual latencies Anderson, S. J., & Hess, R. F. (1990). Post-receptoral are shorter for motion towards the fovea than for undersampling in normal human peripheral vision. motion away. Vision Research, 28, 711–719. Vision Research, 30, 1507–1515. [PubMed] [PubMed] Ball, K., & Sekuler, R. (1980). Human vision favors McKee, S. P., & Nakayama, K. (1984). The detection of centrifugal motion. Perception, 9, 317–325. motion in the peripheral visual field. Vision Research, [PubMed] 24, 25–32. [PubMed] Curcio, C. A., Sloan, K. R., Kalina, R. E., & Hendrickson, Orban, G. A., Van Calenbergh, F., De Bruyn, B., & Maes, H. A. E. (1990). Human photoreceptor topography. (1985). Velocity discrimination in central and periph- Journal of Comparative Neurology, 292, 497–523. eral visual field. Journal of the Optical Society of [PubMed] America A, Optics and image science, 11, 1836–1847. Dumoulin, S. O., Baker, C. L., Jr., & Hess, R. F. (2001). [PubMed] Centrifugal biases for second-order but not first-order Palmer, S. M., & Rosa, M. G. (2006). A distinct motion. Journal of the Optical Society of America A, anatomical network of cortical areas for analysis of Optics, image science, and vision, 18, 2179–2189. motion in far peripheral vision. European Journal of [PubMed] Neuroscience, 24, 2389–2405. [PubMed] Edwards, M., & Badcock, D. R. (1993). Asymmetries in Puce, A., & Perrett, D. (2003). Electrophysiology and the sensitivity to motion in depth: A centripetal bias. brain imaging of biological motion. Philosophical Perception, 22, 1013–1023. [PubMed] Transactions of the Royal Society of London Series B: Fahle, M., & Wehrhahn, C. (1991). Motion perception in Biological Sciences, 358, 435–445. [PubMed] the peripheral visual field. Graefe’s Archive for [Article] Clinical and Experimental Ophthalmology, 229, Raymond, J. E. (1994). Directional anisotropy of motion 430–436. [PubMed] sensitivity across the visual field. Vision Research, Fujimoto, K., & Sato, T. (2006). Backscroll illusion: 34, 1029–1037. [PubMed] Apparent motion in the background of locomotive Rousselet, G. A., Husk, J. S., Bennett, P. J., & Sekuler, A. B. objects. Vision Research, 46, 14–25. [PubMed] (2005). Spatial scaling factors explain eccentricity Georgeson, M. A., & Harris, M. G. (1978). Apparent effects on face ERPs. Journal of Vision, 5(10):1, foveofugal drift of counterphase gratings. Perception, 755–763, http://journalofvision.org/5/10/1/, 7, 527–536. [PubMed] doi:10.1167/5.10.1. [PubMed][Article] Heeger, D. J., Boynton, G. M., Demb, J. B., Seidemann, E., & Rovamo, J., & Virsu, V. (1979). An estimation and Newsome, W. T. (1999). Motion opponency in visual application of the human cortical magnification cortex. Journal of Neuroscience, 19, 7162–7174. factor. Experimental Brain Research, 37, 495–510. [PubMed][Article] [PubMed] Horton, J. C., & Hoyt, W. F. (1991). The representation of Thorpe, S. J., Gegenfurtner, K. R., Fabre-Thorpe, M., & the visual field in human striate cortex: A revision of Bu¨lthoff, H. H. (2001). Detection of animals in the classic Holmes map. Archives of Ophthalmology, natural images using far peripheral vision. European 109, 816–824. [PubMed] Journal of Neuroscience, 14, 869–876. [PubMed] Ikeda, H., Blake, R., & Watanabe, K. (2005). Eccentric Virsu, V., & Rovamo, J. (1979). Visual resolution, perception of biological motion is unscalably poor. contrast sensitivity, and the cortical magnification Vision Research, 45, 1935–1943. [PubMed] factor. Experimental Brain Research, 37, 475–494. Jonas, J. B., Schneider, U., & Naumann, G. O. (1992). [PubMed] Count and density of human retinal photoreceptors.

Downloaded from jov.arvojournals.org on 09/26/2021