- Email: [email protected]
Invisible motion contributes to simultaneous motion contrast
Consciousness and Cognition 18 (2009) 168–175
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Invisible motion contributes to simultaneous motion contrast Takahiro Kawabe *, Yuki Yamada Kyushu University, 6-19-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan
a r t i c l e
i n f o
Article history: Received 22 August 2008 Available online 29 January 2009
Keywords: Simultaneous motion contrast Visual awareness Motion perception Blindsight
a b s t r a c t The purpose of the present study was two-fold. First we examined whether visible motion appearance was altered by the spatial interaction between invisible and visible motion. We addressed this issue by means of simultaneous motion contrast, in which a horizontal test grating with a counterphase luminance modulation was seen to have the opposite motion direction to a peripheral inducer grating with unidirectional upward or downward motion. Using a mirror stereoscope, observers viewed the inducer and test gratings with one eye, and continuous flashes of colorful squares forming an annulus shape with the other eye. The continuous flashes rendered the inducer subjectively invisible. The observers’ task was to report whether the test grating moved upward or downward. Consequently, simultaneous motion contrast was observed even when the inducer was invisible (Experiment 1). Second, we examined whether the observers could correctly respond to the direction of invisible motion: It was impossible (Experiment 2). Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction ‘Invisible’ stimuli occasionally impact on the processing of ‘visible’ stimuli. For example, observers can perceive a contrast-modulation flicker of the grating with high spatial frequency of more than 60 cycles per degree (cpd), in spite of their being unable to discriminate the orientation of the grating (MacLeod & He, 1993). In a similar vein, He and MacLeod (2001) reported that orientation-specific aftereffects occurred after observers adapted to the grating, the orientation of which was unresolvable due to the high spatial frequency. The unresolvable orientation of the grating also affects apparent motion perception and serves as a spatial cue (Rajimehr, 2004). In addition, perceptual adaptation to an invisible flicker due to high temporal frequency has been reported (Shady, MacLeod, & Fisher, 2004). These results indicate that unresolvable orientation and flicker information is retained during visual processing, and can interplay with successive visible stimuli. Consistent with this idea, Jiang, Zhou, and He (2007) observed the brain activities that correlated with the invisible chromatic flicker with a high temporal frequency of more than 25 Hz. In addition to stimuli that are invisible due to the limitation of spatiotemporal resolution of visual processing, stimuli that are invisible due to subjective disappearance phenomena have been used to assess the effect of invisible visual information on conscious visual processing. For example, by employing motion-induced blindness (MIB: Bonneh, Cooperman, & Sagi, 2001), previous studies have reported an orientation-specific aftereffect of the invisible grating (Montaser-Kouhsari, Moradi, Zandvakili, & Esteky, 2004), and the robust formation of a negative afterimage (Hofstoetter, Koch, & Kiper, 2004). By using visual crowding (Bouma, 1970; Toet & Levi, 1992), studies have shown the orientation-selective aftereffect of invisible illusory lines (Montaser-Kouhsari & Rajimehr, 2005; Rajimehr, Montaser-Kouhsari, & Afraz, 2003) and the motion aftereffects of several types of motion stimuli, such as invisible apparent motion (Rajimehr, Vaziri-Pashkam, Afraz, & Esteky, 2004), * Corresponding author. Fax: +81 92 642 2416. E-mail address: [email protected] (T. Kawabe).
1053-8100/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2008.12.004
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
169
invisible first order motion1 (Whitney, 2005) and invisible second-order motion2 (Harp, Bressler, & Whitney, 2007; Whitney & Bressler, 2007). Furthermore, by employing binocular rivalry (Alais & Blake, 2005), previous studies have reported contrast aftereffects and spatial frequency aftereffects (Blake & Fox, 1975), the McCollough effect (White, Petry, Riggs, & Miller, 1978), and a motion aftereffect (Blake, Tadin, Sobel, Raissian, & Chong, 2006; Lehmkuhle & Fox, 1975; O’Shea & Crassini, 1981), but no facial aftereffect (Moradi, Koch, & Shimojo, 2005). On the basis of these findings, many studies have proposed that the ‘invisible’ visual stimuli are still processed in the visual system, and cause perceptual aftereffects. On the other hand, whether spatial interaction between invisible and visible stimuli affects the appearance of visible stimuli is still a subject of debate. Clifford and Harris (2005) showed that simultaneous orientation contrast3 occurred even when the surround orientation, as an inducer, was invisible due to backward masking. Pearson and Clifford (2005) also showed that the surround orientation suppressed by binocular rivalry contributed to simultaneous orientation contrast. On the other hand, the visual phantom perception4 in the static display (Gyoba, 1983) disappeared when surrounding inducers were invisible due to flash suppression or binocular rivalry (Meng, Ferneyhough, & Tong, 2007). Thus, it was the important task for vision researchers to specify what kinds of spatial interaction occurred between invisible and visible stimuli. The purpose of the present study was two-fold. First, we investigated whether simultaneous motion contrast occurred even when the inducer motion was invisible. Simultaneous motion contrast refers to the phenomenon by which ambiguous motion in the central patch is perceived to have the opposite direction to the motion in the surrounding inducer (see for example Nishida, Edwards, & Sato, 1997). Simultaneous motion contrast is deeply related to the center-surround spatial interaction in the motion processing mechanism (Murakami & Shimojo, 1993; Tadin, Lappin, Gilroy, & Blake, 2003). Thus, by investigating whether invisible inducer motion could alter the perceived motion direction in the test patch, we could directly confirm the effects of spatial interaction between invisible and visible motion on the appearance of visible motion. The second purpose of our study was to additionally examine whether the invisible motion could affect guess-based decisions for motion direction in a direct way. In a study by Clifford and Harris (2005), the invisible grating orientation induced simultaneous orientation contrast, but did not contribute to the direct discrimination of the invisible grating orientation. That is, the orientation of invisible gratings affected the appearance of visible grating in an indirect way, but it did not contribute to guess-based decisions for the orientation of invisible grating in a direct way. They discussed that the neural activity strength for the inducer orientation generating the simultaneous orientation contrast is insufficient to generate a conscious representation of the inducer orientation. The present study also confirmed whether the invisible inducer motion that altered the appearance of visible test motion also contributed to guess-based discriminations of invisible motion direction in a direct way. 2. Experiment 1 In Experiment 1a, we examined whether invisible inducer motion can contribute to simultaneous motion contrast. We rendered inducer motion invisible by means of continuous flash suppression (CFS: Tsuchiya & Koch, 2005). We also tested whether there was a difference in the magnitude of simultaneous motion contrast depending on the inducer’s visibility. Furthermore, in Experiment 1b, we checked whether simultaneous motion contrast occurred in the dichoptic presentation. Walker and Powell (1974) showed that spatial interaction of velocity processing was observed only in the monocular presentation. Thus, by investigating simultaneous motion contrast in monocular and dichoptic presentations we could specify the pathway (monocular vs. binocular) that was suppressed by continuous flashes. 2.1. Methods 2.1.1. Observers In Experiment 1a, five observers, including the two authors, participated in the experiment. In Experiment 1b, the two authors and three naive observers participated in this experiment. Apart from the authors, no observers participated in Experiment 1a. They reported that they had normal or corrected-to-normal visual acuity. Apart from the authors, the observers were naive as to the purpose of the experiment. The observers received ¥800 for their participation. 2.1.2. Apparatus Stimuli were presented on a 17-inch CRT monitor (HF703U; Iiyama, Japan). The resolution of the monitor was 1024 768 pixels and the refresh rate was 100 Hz. The presentation of stimuli and collection of data were controlled by a computer (Mac pro; Apple). Using a photometer (3298F; Yokogawa, Japan), we performed gamma correction for the luminance emitted from the monitor. The observers viewed stimuli through a mirror stereoscope (Screenscope, Stereoaids, Australia).
1 2 3 4
A A A A
kind of motion that is defined by luminance features, and processed in a linear spatiotemporal filter. kind of motion that is defined by non-luminance feature such as texture, contrast, and flicker, processed in a filter-rectify-filter system. perceptual phenomenon where the orientation of a central patch is perceived as tilted in the orientation opposed to a surround background. perceptual phenomenon where perceptual filling-in occurs between two separated low-contrast gratings.
170
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
2.1.3. Stimuli In Experiment 1a, stimuli were generated by MATLAB (Mathworks Inc.) with the Psychtoolbox extension (Brainard, 1997; Pelli, 1997). The stimuli consisted of two fixations, two square outlines, continuous flashes, and a simultaneous motion contrast display (Fig. 1). Each fixation was composed of two concentric rings with radius of 0.25° and 0.5°. The luminance of each ring was 7.5 cd/m2. Each of the square outlines had sides of 15.4° with borders 0.2° in width and a luminance of 7.5 cd/m2. On each frame of continuous flashes, two hundred rectangles were randomly positioned annularly. The center of each rectangle was set at the eccentricity in the spatial range between 3.15° and 6.1°. Each rectangle had sides of 1° (long axis) and 0.2° (short axis). Thus, the radius of the inscribed circle for the annulus region made from rectangles was 3.2° at the minimum. The color, position and side of each rectangle were randomly altered at 10 Hz. The simultaneous motion contrast display comprised a surrounding inducer grating (inducer) and a central test grating (test). The inducer had an annulus region with a thickness of 1.2°. The eccentricities of outer and inner edges of the inducer were 6.1° and 4.9°, respectively. The inducer had drifting horizontal gratings with 2.7 cpd of spatial frequency, 10% Michelson contrast, and a 2-Hz drift frequency. The drift direction of the inducer was either upward or downward. The test patch had a similar orientation, and spatial/drifting frequencies, but was a compound version of the two counter-drifting gratings. To measure the magnitude of simultaneous motion contrast, we altered the relative contrast ratio between the two counter-drifting gratings and calculated the cancel points of perceived drift direction (see Results section for details). When the two counter-drifting component gratings were superimposed, the consequent percept of the compound grating was biased in the direction of the component grating with the higher luminance contrast ratio. When the two component gratings had equal contrast, the drifting directions were balanced. Thus, the drifting direction of the compound grating was ambiguous. The luminance contrast ratio of the downward-drifting grating was 0.8, 0.7, 0.6, 0.5, 0.4, 0.3 or 0.2 when that of the upward drifting grating was 0.2, 0.3, 0.4, 0.5, 0.6, 0.7 or 0.8, respectively. The fixation and the square outline were dichoptically presented to both eyes. On the other hand, the simultaneous motion display was presented to one eye while continuous flashes were presented to the other eye. When fused, the test could be seen through the annulus of continuous flashes while the inducer was invisible due to CFS. The retinal position of the inducer corresponded to that of the continuous flashes presented in the other eye; hence, the inducer was selectively rendered invisible by continuous flashes. In Experiment 1b, properties of stimuli were identical to those in Experiment 1a except for the following: the simultaneous motion contrast display was dichoptically presented. That is, when the test and continuous flashes were presented to one eye, the inducer was presented to the other eye.
Fig. 1. A schematic representation of the stimulus presentation in Experiments 1 and 2. The observers fused the images presented on the left and right eyes by using a mirror stereoscope, and perceived the pattern placed in the yellowish region. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
171
2.1.4. Procedure Both experiments were performed in a darkroom. The observers sat 32 cm from the CRT display. As we used a mirror stereoscope, the visual distance was 40 cm. The square outline and the fixation were presented throughout the experiment. The observer pressed the spacebar to initiate each session. With 500-ms temporal intervals, the simultaneous motion contrast display with/without continuous flashes was presented for 500 ms. Continuous flashes were presented only in the invisible condition. After the disappearance of the simultaneous motion contrast display and continuous flashes, the observers first reported whether the inducer was visible or invisible by pressing assigned keys. Only when the inducer was invisible were the observers asked to report if the test had an upward or downward drift by pressing the assigned keys. The observers were urged to respond as accurately as possible without speed stress. After the response, and a temporal interval of 500–1000 ms, the next trial started automatically. Each observer performed 560 trials consisting of two visibility conditions of the inducer (visible vs. invisible) two inducer conditions (upward vs. downward) seven luminance contrast ratios 20 replications. The effect of the visibility of inducer was assessed in separate blocks. In each block, the trial order was randomized. It took 1 h for each observer to complete this experiment. 2.2. Results 2.2.1. Experiment 1a We calculated the proportions of trials in which an upward drift was perceived in the test as a function of the luminance contrast ratio of the upward drifting grating in the test. No observers reported the appearance of the inducer; thus, we used all data for the calculation. By using the psignifit program implemented in MATLAB (Wichmann & Hill, 2001a, 2001b), we individually fitted the Weibull function to the proportion, and computed the cancel point for the simultaneous motion contrast. The goodness of fit was assessed by deviance and cumulative probability estimate (criteria: p < .95). The cancel point here corresponds to the luminance contrast ratio of the upward grating inducing a perceptually balanced drift direction. Fig. 2a shows the data from one representative observer. We averaged the cancel points across observers for each inducer condition. The averaged data are plotted in Fig. 2b. For each of the visible and invisible conditions, we assessed the statistical difference in cancel points between two inducer conditions by means of a t-test. For the visible condition, the cancel points were significantly different between the inducer conditions, t(4) = 6.09, p < .004. For the invisible condition, the cancel points were also significantly different between the inducer conditions, t(4) = 3.27, p < .04. Next, we subtracted the cancel points for the downward inducer condition from those for the upward inducer condition, calculated the magnitudes of the simultaneous motion contrast, and plotted them in Fig. 2c. By means of a t-test, we analyzed the statistical difference in the magnitude of simultaneous motion contrast between visible and invisible conditions, revealing a significant difference, t(4) = 3.05, p < .04.
Fig. 2. The results of Experiment 1a. (a) The data from observer HA. The green and red dots/lines show the data in downward and upward inducer conditions, respectively. The upper and lower panels show data from visible and invisible inducer conditions, respectively. (b) Individual and averaged cancel points. The green and red bars show the cancel points for downward and upward inducer conditions, respectively. The upper and lower panel shows data from visible and invisible inducer conditions, respectively. Error bars denote standard errors of means. (c) The magnitude of simultaneous motion contrast. Error bars denote 95% confidence intervals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
172
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
As in Experiment 1a, we calculated the cancel points for the simultaneous motion contrast, and plotted in Fig. 3. We assessed the statistical significance of differences between the two inducer conditions. One observer (TN) reported subjective appearances of the inducer in several trials; thus, these trials were not included in the further analysis. No significant difference was observed between the two inducer conditions, t(4) = 0.84 and 0.53 for visible and invisible inducer conditions, respectively, p > .1. The magnitude of simultaneous motion contrast was also not significantly different between visible and invisible conditions, t(4) = 0.05, p > .1. 2.3. Discussion The data showed that invisible motion contributed to simultaneous motion contrast. When the inducer drift was upward, the cancel points (that is, the relative contrast ratio of upward drifting grating causing balanced motion percepts) were significantly greater than when the inducer drift was downward, in both visible and invisible inducer conditions. This indicates that simultaneous motion contrast occurred in both inducer conditions. On the other hand, in the dichoptic presentation no motion contrast was observed with visible and invisible motion stimuli. The results are consistent with Walker and Powell (1974) showing that the modulation of motion speed, which was based on spatial interaction of motion, relied on the monocular processing. 3. Experiment 2 In this experiment, we tested whether the observers could discriminate the direction of an invisible motion of an annulus grating (inducer in Experiment 1) in a direct way. It was reported that some patients with cerebral lesions, who had no visual awareness for stimuli, could perform the visual task with the stimuli. For example, patients can make a saccade towards the invisible target, and the magnitude of the saccade is dependent on the eccentricity of the invisible target (Pöppel, Held, & Frost, 1973). Furthermore, some patients can manually point to the invisible visual target in a relatively accurate manner (Weiskrantz, Warrington, Sanders, & Marschall, 1974), and detect the global motion presented in their blind field (Alexander & Cowey, 2009). These phenomena, so-called blindsight, indicate that, under certain conditions, human observers can judge the visual properties of an object without awareness of it. In this experiment, we were interested in whether the normal observers could also discriminate the motion directions of an invisible annulus grating. By examining the discriminability of invisible motion direction, we wanted to further clarify the processing of invisible annulus grating. The successful motion discrimination of invisible annulus grating would indicate that invisible motion of the annulus grating affects decisions for motion direction in a direct way. Meanwhile, unsuccessful
Fig. 3. The results of Experiment 1b. (a) Individual and averaged cancel points. The green and red bars show the cancel points for downward and upward inducer conditions, respectively. The upper and lower panel shows data from visible and invisible inducer conditions, respectively. Error bars denote standard errors of means. (b) The magnitude of simultaneous motion contrast. Error bars denote standard errors of means. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
173
discrimination would imply that invisible motion of the annulus grating affects visible motion perception only in an indirect way. 3.1. Methods 3.1.1. Observers Five observers participated in this experiment. One of them was one of the authors (TK) and the other four people were naive as to the purpose of the experiment. Aside from observers SK and TK, the observers did not participate in Experiments 1a or 1b. All observers reported that they had normal or corrected-to-normal visual acuity. 3.1.2. Stimuli Stimuli were identical to those used in Experiment 1 except for the following. In this experiment, no test was presented. Thus, the stimuli consisted of the fixations, square outlines, an annulus grating and continuous flashes. 3.1.3. Procedure The procedure was identical to that used in Experiment 1, except for the following. In this experiment, the observers performed two kinds of task in each trial. The first task was to report whether the visible or invisible annulus grating had an upward or downward drift by pressing the assigned keys. When the annulus grating was invisible, observers were asked to guess the motion direction. The second task was to rate the confidence in each response on motion direction. On each trial, the observers reported their confidence on a 10-point scale where ‘0’ represented no confidence, ’10’ represented maximum confidence, and ‘1’–‘9’ represented intermediate levels of confidence. They moved the cursor position on the scale presented in the display by pressing assigned keys, and determined the confidence rating score by pressing another key. They performed 160 trials (two visibilities of the annulus grating two directions 40 replications). On the other hand, observer MS performed 80 trials (two visibilities of inducers two directions 20 replications) since she quit the experiment due to the eyestrain. It took 20 min for each observer to complete the trials. 3.2. Results The proportions of trials in which the observers correctly judged the motion direction in the annulus grating were calculated for each of the visibility conditions, and are plotted in Fig. 4a. There was a significant difference in the proportion of correct responses between two visibility conditions, t(4) = 13.4, p < .0002. Moreover, the proportion of correct responses in the visible annulus grating condition was significantly different from a chance level (t(4) = 31.5, p < .00001), but not in the invisible annulus grating condition (t(4) = 1.55 p > .1). Fig. 4b shows the confidence ratings for each observer and group. There was a significant difference in the confidence rating between the two visibility conditions, t(4) = 16.9, p < .0001. The 95% confidence intervals showed that the confidence ratings in the invisible annulus grating condition were significantly lower than 4, and that those in the visible condition were significantly higher than 8 (p < .05). 3.3. Discussion The results indicate that the discrimination of invisible motion direction was impossible. The chancel-level performance and low confidence ratings in the invisible grating condition imply that when the observers decided the motion direction
Fig. 4. The results of Experiment 2. (a) The proportion correct for the discrimination of invisible motion direction and (b) the confidence ratings for the discrimination in visible (green-filled bars) and invisible (brown-outlined bars) conditions. Error bars denote 95% confidence intervals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
174
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
based on guessing, that decisions could not incorporate information about the motion of the invisible annulus grating. Therefore, it is likely that invisible motion is available in an indirect way through spatial interaction with visible motion. 4. General discussion The present study briefly addressed how invisible motion is processed in the visual system. In Experiment 1a, we found that invisible motion contributed to simultaneous motion contrast. Experiment 1b confirmed that simultaneous motion contrast occurred only when the inducer and test were monocularly presented. In Experiment 2, we showed that invisible motion did not contribute to guess-based decisions about the direction of invisible motion in a direct way. Thus, these results indicate that spatial interaction between invisible and visible motion can occur, and result in simultaneous motion contrast that grounds on a center-surround spatial interaction. The results of Experiment 1 indicate that spatial interaction between invisible and visible motion signals can affect the appearance of visible motion. Although previous studies have clarified the existence of processing for invisible motion signals, these investigations have focused on the adaptation to invisible motion signals (see for example Blake et al., 2006), priming (Blake, Ahström, & Alais, 2000), and learning (Watanabe, Náñez, & Sasaki, 2001). In addition to these phenomena, we showed that invisible inducer motion contributed to simultaneous motion contrast. This indicates that invisible inducer motion can serve as the input for center-surround spatial interaction of motion processing. On the other hand, the magnitude of simultaneous motion contrast was greater when the inducer was visible than when it was invisible. There are two speculative ideas that are consistent with the greater effect in the visible inducer condition. First, we suggest that the simultaneous motion contrast is boosted by conscious factors such as attentive tracking (Cavanagh, 1992) or altered appearance due to attention (Carrasco, Ling, & Read, 2004). However, it remains to be resolved why the effect of conscious processing is restricted to monocular stimuli. Second, one can argue that the weaker simultaneous motion contrast in the invisible inducer condition stems from the suppression of inducer motion itself due to CFS. However, it has not been validated whether the motion signal strength itself is diminished by CFS. There is a previous study showing that the suppressed representation due to binocular rivalry correlated with reduced activity in the V1 (Lee & Blake, 2002). On the contrary, a recent study (Cai, Zhou, & Chen, 2008) showed that continuous flashes did not affect the early component of visual processing (namely surround suppression at the lower-level). Neither conscious boosting nor signal diminishment due to CFS are deterministic; thus, future studies are required to disentangle these ideas. The underlying mechanism for the contribution of invisible motion to the simultaneous motion contrast is still unclear. The continuous flashes to one eye reportedly increased the contrast threshold for stimuli presented to the other eye (Tsuchiya, Koch, Gilroy, & Blake, 2006). These data are consistent with the idea that the activity in the V1 is strongly reduced (Lee & Blake, 2002). Nevertheless, significant simultaneous motion contrast was observed even when the inducer was invisible. The results perhaps imply that the internal signal strength required for motion information to be consciously perceived should be greater than that required for motion information to contribute to the simultaneous motion contrast. This idea has been already discussed with orientation processing for invisible grating (Clifford & Harris, 2005). However, it is unclear whether the magnitude of internal signals also influences the magnitude of simultaneous motion contrast since simultaneous motion contrast increase as stimulus contrast (Nishida et al., 1997). It is also possible that as described earlier post-conscious processing such as attention (Carrasco et al., 2004) or decision (Schneider & Komlos, 2008) boosts the magnitude of the simultaneous motion contrast. The specification of the binocular interactive mechanism underlying CFS and the variation of internal signal strength due to CFS are important issues to be challenged in future studies. References Alais, D., & Blake, R. (Eds.). (2005). Binocular rivalry. Cambridge, MA: MIT Press. Alexander, I., & Cowey, A. (2009). The cortical basis of global motion detection in blindsight. Experimental Brain Research, 192, 407–411. Blake, R., Ahström, U., & Alais, D. (2000). Perceptual priming by invisible motion. Psychological Science, 10, 145–150. Black, R., & Fox, R. (1974). Adaptation in invisible gratings and the site of binocular rivalry suppression. Nature, 249, 488–490. Blake, R., Tadin, D., Sobel, K. V., Raissian, T. A., & Chong, S. C. (2006). Strength of early visual adaptation depends on visual awareness. Proceedings of the National Academy of Sciences of the United States of America, 103, 4783–4788. Bonneh, Y. S., Cooperman, A., & Sagi, D. (2001). Motion-induced blindness in normal observers. Nature, 411, 798–801. Bouma, H. (1970). Interaction effects in parafoveal letter recognition. Nature, 226, 177–178. Brainard, D. H. (1997). The psychophysics toolbox. Spatial Vision, 10, 433–436. Cai, Y., Zhou, T., & Chen, L. (2008). Effects of binocular suppression on surround suppression. Journal of Vision 8(9):9, 1–10. Carrasco, M., Ling, S., & Read, S. (2004). Attention alters appearance. Nature Neuroscience, 7, 308–313. Cavanagh, P. (1992). Attention-based motion perception. Science, 257, 1563–1565. Clifford, C. W. G., & Harris, J. A. (2005). Contextual modulation outside of awareness. Current Biology, 15, 574–578. Gyoba, J. (1983). Stationary phantoms: A completion effect without motion and flicker. Vision Research, 23, 205–211. Harp, T., Bressler, D., & Whitney, D. (2007). Position shifts following crowded second-order motion adaptation reveal processing of local and global motion without awareness. Journal of Vision, 7((2):15), 1–13. doi:10.1167/7.2.15.. He, S., & MacLeod, D. I. A. (2001). Orientation-selective adaptation and tilt after-effect from invisible patterns. Nature, 411, 473–476. Hofstoetter, C., Koch, C., & Kiper, D. C. (2004). Motion-induced blindness does not affect the formation of negative afterimages. Consciousness and Cognition, 13, 691–708. Jiang, Y., Zhou, K., & He, S. (2007). Human visual cortex responds to invisible chromatic flicker. Nature Neuroscience, 10, 657–662. Lee, S.-H., & Blake, R. (2002). V1 activity is reduced during binocular rivalry. Journal of Vision, 2, 618–626. Lehmkuhle, S. W., & Fox, R. (1975). Effect of binocular rivalry suppression on the MAE. Vision Research, 15, 855–859. MacLeod, D. I. A., & He, S. (1993). Visible flicker from invisible patterns. Nature, 361, 256–258.
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
175
Meng, M., Ferneyhough, E., & Tong, F. (2007). Dynamics of perceptual filling-in of visual phantoms revealed by binocular rivalry. Journal of Vision, 7((13):8), 1–15. doi:10.1167/7.13.8.
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Invisible motion contributes to simultaneous motion contrast Takahiro Kawabe *, Yuki Yamada Kyushu University, 6-19-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan
a r t i c l e
i n f o
Article history: Received 22 August 2008 Available online 29 January 2009
Keywords: Simultaneous motion contrast Visual awareness Motion perception Blindsight
a b s t r a c t The purpose of the present study was two-fold. First we examined whether visible motion appearance was altered by the spatial interaction between invisible and visible motion. We addressed this issue by means of simultaneous motion contrast, in which a horizontal test grating with a counterphase luminance modulation was seen to have the opposite motion direction to a peripheral inducer grating with unidirectional upward or downward motion. Using a mirror stereoscope, observers viewed the inducer and test gratings with one eye, and continuous flashes of colorful squares forming an annulus shape with the other eye. The continuous flashes rendered the inducer subjectively invisible. The observers’ task was to report whether the test grating moved upward or downward. Consequently, simultaneous motion contrast was observed even when the inducer was invisible (Experiment 1). Second, we examined whether the observers could correctly respond to the direction of invisible motion: It was impossible (Experiment 2). Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction ‘Invisible’ stimuli occasionally impact on the processing of ‘visible’ stimuli. For example, observers can perceive a contrast-modulation flicker of the grating with high spatial frequency of more than 60 cycles per degree (cpd), in spite of their being unable to discriminate the orientation of the grating (MacLeod & He, 1993). In a similar vein, He and MacLeod (2001) reported that orientation-specific aftereffects occurred after observers adapted to the grating, the orientation of which was unresolvable due to the high spatial frequency. The unresolvable orientation of the grating also affects apparent motion perception and serves as a spatial cue (Rajimehr, 2004). In addition, perceptual adaptation to an invisible flicker due to high temporal frequency has been reported (Shady, MacLeod, & Fisher, 2004). These results indicate that unresolvable orientation and flicker information is retained during visual processing, and can interplay with successive visible stimuli. Consistent with this idea, Jiang, Zhou, and He (2007) observed the brain activities that correlated with the invisible chromatic flicker with a high temporal frequency of more than 25 Hz. In addition to stimuli that are invisible due to the limitation of spatiotemporal resolution of visual processing, stimuli that are invisible due to subjective disappearance phenomena have been used to assess the effect of invisible visual information on conscious visual processing. For example, by employing motion-induced blindness (MIB: Bonneh, Cooperman, & Sagi, 2001), previous studies have reported an orientation-specific aftereffect of the invisible grating (Montaser-Kouhsari, Moradi, Zandvakili, & Esteky, 2004), and the robust formation of a negative afterimage (Hofstoetter, Koch, & Kiper, 2004). By using visual crowding (Bouma, 1970; Toet & Levi, 1992), studies have shown the orientation-selective aftereffect of invisible illusory lines (Montaser-Kouhsari & Rajimehr, 2005; Rajimehr, Montaser-Kouhsari, & Afraz, 2003) and the motion aftereffects of several types of motion stimuli, such as invisible apparent motion (Rajimehr, Vaziri-Pashkam, Afraz, & Esteky, 2004), * Corresponding author. Fax: +81 92 642 2416. E-mail address: [email protected] (T. Kawabe).
1053-8100/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2008.12.004
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
169
invisible first order motion1 (Whitney, 2005) and invisible second-order motion2 (Harp, Bressler, & Whitney, 2007; Whitney & Bressler, 2007). Furthermore, by employing binocular rivalry (Alais & Blake, 2005), previous studies have reported contrast aftereffects and spatial frequency aftereffects (Blake & Fox, 1975), the McCollough effect (White, Petry, Riggs, & Miller, 1978), and a motion aftereffect (Blake, Tadin, Sobel, Raissian, & Chong, 2006; Lehmkuhle & Fox, 1975; O’Shea & Crassini, 1981), but no facial aftereffect (Moradi, Koch, & Shimojo, 2005). On the basis of these findings, many studies have proposed that the ‘invisible’ visual stimuli are still processed in the visual system, and cause perceptual aftereffects. On the other hand, whether spatial interaction between invisible and visible stimuli affects the appearance of visible stimuli is still a subject of debate. Clifford and Harris (2005) showed that simultaneous orientation contrast3 occurred even when the surround orientation, as an inducer, was invisible due to backward masking. Pearson and Clifford (2005) also showed that the surround orientation suppressed by binocular rivalry contributed to simultaneous orientation contrast. On the other hand, the visual phantom perception4 in the static display (Gyoba, 1983) disappeared when surrounding inducers were invisible due to flash suppression or binocular rivalry (Meng, Ferneyhough, & Tong, 2007). Thus, it was the important task for vision researchers to specify what kinds of spatial interaction occurred between invisible and visible stimuli. The purpose of the present study was two-fold. First, we investigated whether simultaneous motion contrast occurred even when the inducer motion was invisible. Simultaneous motion contrast refers to the phenomenon by which ambiguous motion in the central patch is perceived to have the opposite direction to the motion in the surrounding inducer (see for example Nishida, Edwards, & Sato, 1997). Simultaneous motion contrast is deeply related to the center-surround spatial interaction in the motion processing mechanism (Murakami & Shimojo, 1993; Tadin, Lappin, Gilroy, & Blake, 2003). Thus, by investigating whether invisible inducer motion could alter the perceived motion direction in the test patch, we could directly confirm the effects of spatial interaction between invisible and visible motion on the appearance of visible motion. The second purpose of our study was to additionally examine whether the invisible motion could affect guess-based decisions for motion direction in a direct way. In a study by Clifford and Harris (2005), the invisible grating orientation induced simultaneous orientation contrast, but did not contribute to the direct discrimination of the invisible grating orientation. That is, the orientation of invisible gratings affected the appearance of visible grating in an indirect way, but it did not contribute to guess-based decisions for the orientation of invisible grating in a direct way. They discussed that the neural activity strength for the inducer orientation generating the simultaneous orientation contrast is insufficient to generate a conscious representation of the inducer orientation. The present study also confirmed whether the invisible inducer motion that altered the appearance of visible test motion also contributed to guess-based discriminations of invisible motion direction in a direct way. 2. Experiment 1 In Experiment 1a, we examined whether invisible inducer motion can contribute to simultaneous motion contrast. We rendered inducer motion invisible by means of continuous flash suppression (CFS: Tsuchiya & Koch, 2005). We also tested whether there was a difference in the magnitude of simultaneous motion contrast depending on the inducer’s visibility. Furthermore, in Experiment 1b, we checked whether simultaneous motion contrast occurred in the dichoptic presentation. Walker and Powell (1974) showed that spatial interaction of velocity processing was observed only in the monocular presentation. Thus, by investigating simultaneous motion contrast in monocular and dichoptic presentations we could specify the pathway (monocular vs. binocular) that was suppressed by continuous flashes. 2.1. Methods 2.1.1. Observers In Experiment 1a, five observers, including the two authors, participated in the experiment. In Experiment 1b, the two authors and three naive observers participated in this experiment. Apart from the authors, no observers participated in Experiment 1a. They reported that they had normal or corrected-to-normal visual acuity. Apart from the authors, the observers were naive as to the purpose of the experiment. The observers received ¥800 for their participation. 2.1.2. Apparatus Stimuli were presented on a 17-inch CRT monitor (HF703U; Iiyama, Japan). The resolution of the monitor was 1024 768 pixels and the refresh rate was 100 Hz. The presentation of stimuli and collection of data were controlled by a computer (Mac pro; Apple). Using a photometer (3298F; Yokogawa, Japan), we performed gamma correction for the luminance emitted from the monitor. The observers viewed stimuli through a mirror stereoscope (Screenscope, Stereoaids, Australia).
1 2 3 4
A A A A
kind of motion that is defined by luminance features, and processed in a linear spatiotemporal filter. kind of motion that is defined by non-luminance feature such as texture, contrast, and flicker, processed in a filter-rectify-filter system. perceptual phenomenon where the orientation of a central patch is perceived as tilted in the orientation opposed to a surround background. perceptual phenomenon where perceptual filling-in occurs between two separated low-contrast gratings.
170
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
2.1.3. Stimuli In Experiment 1a, stimuli were generated by MATLAB (Mathworks Inc.) with the Psychtoolbox extension (Brainard, 1997; Pelli, 1997). The stimuli consisted of two fixations, two square outlines, continuous flashes, and a simultaneous motion contrast display (Fig. 1). Each fixation was composed of two concentric rings with radius of 0.25° and 0.5°. The luminance of each ring was 7.5 cd/m2. Each of the square outlines had sides of 15.4° with borders 0.2° in width and a luminance of 7.5 cd/m2. On each frame of continuous flashes, two hundred rectangles were randomly positioned annularly. The center of each rectangle was set at the eccentricity in the spatial range between 3.15° and 6.1°. Each rectangle had sides of 1° (long axis) and 0.2° (short axis). Thus, the radius of the inscribed circle for the annulus region made from rectangles was 3.2° at the minimum. The color, position and side of each rectangle were randomly altered at 10 Hz. The simultaneous motion contrast display comprised a surrounding inducer grating (inducer) and a central test grating (test). The inducer had an annulus region with a thickness of 1.2°. The eccentricities of outer and inner edges of the inducer were 6.1° and 4.9°, respectively. The inducer had drifting horizontal gratings with 2.7 cpd of spatial frequency, 10% Michelson contrast, and a 2-Hz drift frequency. The drift direction of the inducer was either upward or downward. The test patch had a similar orientation, and spatial/drifting frequencies, but was a compound version of the two counter-drifting gratings. To measure the magnitude of simultaneous motion contrast, we altered the relative contrast ratio between the two counter-drifting gratings and calculated the cancel points of perceived drift direction (see Results section for details). When the two counter-drifting component gratings were superimposed, the consequent percept of the compound grating was biased in the direction of the component grating with the higher luminance contrast ratio. When the two component gratings had equal contrast, the drifting directions were balanced. Thus, the drifting direction of the compound grating was ambiguous. The luminance contrast ratio of the downward-drifting grating was 0.8, 0.7, 0.6, 0.5, 0.4, 0.3 or 0.2 when that of the upward drifting grating was 0.2, 0.3, 0.4, 0.5, 0.6, 0.7 or 0.8, respectively. The fixation and the square outline were dichoptically presented to both eyes. On the other hand, the simultaneous motion display was presented to one eye while continuous flashes were presented to the other eye. When fused, the test could be seen through the annulus of continuous flashes while the inducer was invisible due to CFS. The retinal position of the inducer corresponded to that of the continuous flashes presented in the other eye; hence, the inducer was selectively rendered invisible by continuous flashes. In Experiment 1b, properties of stimuli were identical to those in Experiment 1a except for the following: the simultaneous motion contrast display was dichoptically presented. That is, when the test and continuous flashes were presented to one eye, the inducer was presented to the other eye.
Fig. 1. A schematic representation of the stimulus presentation in Experiments 1 and 2. The observers fused the images presented on the left and right eyes by using a mirror stereoscope, and perceived the pattern placed in the yellowish region. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
171
2.1.4. Procedure Both experiments were performed in a darkroom. The observers sat 32 cm from the CRT display. As we used a mirror stereoscope, the visual distance was 40 cm. The square outline and the fixation were presented throughout the experiment. The observer pressed the spacebar to initiate each session. With 500-ms temporal intervals, the simultaneous motion contrast display with/without continuous flashes was presented for 500 ms. Continuous flashes were presented only in the invisible condition. After the disappearance of the simultaneous motion contrast display and continuous flashes, the observers first reported whether the inducer was visible or invisible by pressing assigned keys. Only when the inducer was invisible were the observers asked to report if the test had an upward or downward drift by pressing the assigned keys. The observers were urged to respond as accurately as possible without speed stress. After the response, and a temporal interval of 500–1000 ms, the next trial started automatically. Each observer performed 560 trials consisting of two visibility conditions of the inducer (visible vs. invisible) two inducer conditions (upward vs. downward) seven luminance contrast ratios 20 replications. The effect of the visibility of inducer was assessed in separate blocks. In each block, the trial order was randomized. It took 1 h for each observer to complete this experiment. 2.2. Results 2.2.1. Experiment 1a We calculated the proportions of trials in which an upward drift was perceived in the test as a function of the luminance contrast ratio of the upward drifting grating in the test. No observers reported the appearance of the inducer; thus, we used all data for the calculation. By using the psignifit program implemented in MATLAB (Wichmann & Hill, 2001a, 2001b), we individually fitted the Weibull function to the proportion, and computed the cancel point for the simultaneous motion contrast. The goodness of fit was assessed by deviance and cumulative probability estimate (criteria: p < .95). The cancel point here corresponds to the luminance contrast ratio of the upward grating inducing a perceptually balanced drift direction. Fig. 2a shows the data from one representative observer. We averaged the cancel points across observers for each inducer condition. The averaged data are plotted in Fig. 2b. For each of the visible and invisible conditions, we assessed the statistical difference in cancel points between two inducer conditions by means of a t-test. For the visible condition, the cancel points were significantly different between the inducer conditions, t(4) = 6.09, p < .004. For the invisible condition, the cancel points were also significantly different between the inducer conditions, t(4) = 3.27, p < .04. Next, we subtracted the cancel points for the downward inducer condition from those for the upward inducer condition, calculated the magnitudes of the simultaneous motion contrast, and plotted them in Fig. 2c. By means of a t-test, we analyzed the statistical difference in the magnitude of simultaneous motion contrast between visible and invisible conditions, revealing a significant difference, t(4) = 3.05, p < .04.
Fig. 2. The results of Experiment 1a. (a) The data from observer HA. The green and red dots/lines show the data in downward and upward inducer conditions, respectively. The upper and lower panels show data from visible and invisible inducer conditions, respectively. (b) Individual and averaged cancel points. The green and red bars show the cancel points for downward and upward inducer conditions, respectively. The upper and lower panel shows data from visible and invisible inducer conditions, respectively. Error bars denote standard errors of means. (c) The magnitude of simultaneous motion contrast. Error bars denote 95% confidence intervals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
172
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
As in Experiment 1a, we calculated the cancel points for the simultaneous motion contrast, and plotted in Fig. 3. We assessed the statistical significance of differences between the two inducer conditions. One observer (TN) reported subjective appearances of the inducer in several trials; thus, these trials were not included in the further analysis. No significant difference was observed between the two inducer conditions, t(4) = 0.84 and 0.53 for visible and invisible inducer conditions, respectively, p > .1. The magnitude of simultaneous motion contrast was also not significantly different between visible and invisible conditions, t(4) = 0.05, p > .1. 2.3. Discussion The data showed that invisible motion contributed to simultaneous motion contrast. When the inducer drift was upward, the cancel points (that is, the relative contrast ratio of upward drifting grating causing balanced motion percepts) were significantly greater than when the inducer drift was downward, in both visible and invisible inducer conditions. This indicates that simultaneous motion contrast occurred in both inducer conditions. On the other hand, in the dichoptic presentation no motion contrast was observed with visible and invisible motion stimuli. The results are consistent with Walker and Powell (1974) showing that the modulation of motion speed, which was based on spatial interaction of motion, relied on the monocular processing. 3. Experiment 2 In this experiment, we tested whether the observers could discriminate the direction of an invisible motion of an annulus grating (inducer in Experiment 1) in a direct way. It was reported that some patients with cerebral lesions, who had no visual awareness for stimuli, could perform the visual task with the stimuli. For example, patients can make a saccade towards the invisible target, and the magnitude of the saccade is dependent on the eccentricity of the invisible target (Pöppel, Held, & Frost, 1973). Furthermore, some patients can manually point to the invisible visual target in a relatively accurate manner (Weiskrantz, Warrington, Sanders, & Marschall, 1974), and detect the global motion presented in their blind field (Alexander & Cowey, 2009). These phenomena, so-called blindsight, indicate that, under certain conditions, human observers can judge the visual properties of an object without awareness of it. In this experiment, we were interested in whether the normal observers could also discriminate the motion directions of an invisible annulus grating. By examining the discriminability of invisible motion direction, we wanted to further clarify the processing of invisible annulus grating. The successful motion discrimination of invisible annulus grating would indicate that invisible motion of the annulus grating affects decisions for motion direction in a direct way. Meanwhile, unsuccessful
Fig. 3. The results of Experiment 1b. (a) Individual and averaged cancel points. The green and red bars show the cancel points for downward and upward inducer conditions, respectively. The upper and lower panel shows data from visible and invisible inducer conditions, respectively. Error bars denote standard errors of means. (b) The magnitude of simultaneous motion contrast. Error bars denote standard errors of means. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
173
discrimination would imply that invisible motion of the annulus grating affects visible motion perception only in an indirect way. 3.1. Methods 3.1.1. Observers Five observers participated in this experiment. One of them was one of the authors (TK) and the other four people were naive as to the purpose of the experiment. Aside from observers SK and TK, the observers did not participate in Experiments 1a or 1b. All observers reported that they had normal or corrected-to-normal visual acuity. 3.1.2. Stimuli Stimuli were identical to those used in Experiment 1 except for the following. In this experiment, no test was presented. Thus, the stimuli consisted of the fixations, square outlines, an annulus grating and continuous flashes. 3.1.3. Procedure The procedure was identical to that used in Experiment 1, except for the following. In this experiment, the observers performed two kinds of task in each trial. The first task was to report whether the visible or invisible annulus grating had an upward or downward drift by pressing the assigned keys. When the annulus grating was invisible, observers were asked to guess the motion direction. The second task was to rate the confidence in each response on motion direction. On each trial, the observers reported their confidence on a 10-point scale where ‘0’ represented no confidence, ’10’ represented maximum confidence, and ‘1’–‘9’ represented intermediate levels of confidence. They moved the cursor position on the scale presented in the display by pressing assigned keys, and determined the confidence rating score by pressing another key. They performed 160 trials (two visibilities of the annulus grating two directions 40 replications). On the other hand, observer MS performed 80 trials (two visibilities of inducers two directions 20 replications) since she quit the experiment due to the eyestrain. It took 20 min for each observer to complete the trials. 3.2. Results The proportions of trials in which the observers correctly judged the motion direction in the annulus grating were calculated for each of the visibility conditions, and are plotted in Fig. 4a. There was a significant difference in the proportion of correct responses between two visibility conditions, t(4) = 13.4, p < .0002. Moreover, the proportion of correct responses in the visible annulus grating condition was significantly different from a chance level (t(4) = 31.5, p < .00001), but not in the invisible annulus grating condition (t(4) = 1.55 p > .1). Fig. 4b shows the confidence ratings for each observer and group. There was a significant difference in the confidence rating between the two visibility conditions, t(4) = 16.9, p < .0001. The 95% confidence intervals showed that the confidence ratings in the invisible annulus grating condition were significantly lower than 4, and that those in the visible condition were significantly higher than 8 (p < .05). 3.3. Discussion The results indicate that the discrimination of invisible motion direction was impossible. The chancel-level performance and low confidence ratings in the invisible grating condition imply that when the observers decided the motion direction
Fig. 4. The results of Experiment 2. (a) The proportion correct for the discrimination of invisible motion direction and (b) the confidence ratings for the discrimination in visible (green-filled bars) and invisible (brown-outlined bars) conditions. Error bars denote 95% confidence intervals. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
174
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
based on guessing, that decisions could not incorporate information about the motion of the invisible annulus grating. Therefore, it is likely that invisible motion is available in an indirect way through spatial interaction with visible motion. 4. General discussion The present study briefly addressed how invisible motion is processed in the visual system. In Experiment 1a, we found that invisible motion contributed to simultaneous motion contrast. Experiment 1b confirmed that simultaneous motion contrast occurred only when the inducer and test were monocularly presented. In Experiment 2, we showed that invisible motion did not contribute to guess-based decisions about the direction of invisible motion in a direct way. Thus, these results indicate that spatial interaction between invisible and visible motion can occur, and result in simultaneous motion contrast that grounds on a center-surround spatial interaction. The results of Experiment 1 indicate that spatial interaction between invisible and visible motion signals can affect the appearance of visible motion. Although previous studies have clarified the existence of processing for invisible motion signals, these investigations have focused on the adaptation to invisible motion signals (see for example Blake et al., 2006), priming (Blake, Ahström, & Alais, 2000), and learning (Watanabe, Náñez, & Sasaki, 2001). In addition to these phenomena, we showed that invisible inducer motion contributed to simultaneous motion contrast. This indicates that invisible inducer motion can serve as the input for center-surround spatial interaction of motion processing. On the other hand, the magnitude of simultaneous motion contrast was greater when the inducer was visible than when it was invisible. There are two speculative ideas that are consistent with the greater effect in the visible inducer condition. First, we suggest that the simultaneous motion contrast is boosted by conscious factors such as attentive tracking (Cavanagh, 1992) or altered appearance due to attention (Carrasco, Ling, & Read, 2004). However, it remains to be resolved why the effect of conscious processing is restricted to monocular stimuli. Second, one can argue that the weaker simultaneous motion contrast in the invisible inducer condition stems from the suppression of inducer motion itself due to CFS. However, it has not been validated whether the motion signal strength itself is diminished by CFS. There is a previous study showing that the suppressed representation due to binocular rivalry correlated with reduced activity in the V1 (Lee & Blake, 2002). On the contrary, a recent study (Cai, Zhou, & Chen, 2008) showed that continuous flashes did not affect the early component of visual processing (namely surround suppression at the lower-level). Neither conscious boosting nor signal diminishment due to CFS are deterministic; thus, future studies are required to disentangle these ideas. The underlying mechanism for the contribution of invisible motion to the simultaneous motion contrast is still unclear. The continuous flashes to one eye reportedly increased the contrast threshold for stimuli presented to the other eye (Tsuchiya, Koch, Gilroy, & Blake, 2006). These data are consistent with the idea that the activity in the V1 is strongly reduced (Lee & Blake, 2002). Nevertheless, significant simultaneous motion contrast was observed even when the inducer was invisible. The results perhaps imply that the internal signal strength required for motion information to be consciously perceived should be greater than that required for motion information to contribute to the simultaneous motion contrast. This idea has been already discussed with orientation processing for invisible grating (Clifford & Harris, 2005). However, it is unclear whether the magnitude of internal signals also influences the magnitude of simultaneous motion contrast since simultaneous motion contrast increase as stimulus contrast (Nishida et al., 1997). It is also possible that as described earlier post-conscious processing such as attention (Carrasco et al., 2004) or decision (Schneider & Komlos, 2008) boosts the magnitude of the simultaneous motion contrast. The specification of the binocular interactive mechanism underlying CFS and the variation of internal signal strength due to CFS are important issues to be challenged in future studies. References Alais, D., & Blake, R. (Eds.). (2005). Binocular rivalry. Cambridge, MA: MIT Press. Alexander, I., & Cowey, A. (2009). The cortical basis of global motion detection in blindsight. Experimental Brain Research, 192, 407–411. Blake, R., Ahström, U., & Alais, D. (2000). Perceptual priming by invisible motion. Psychological Science, 10, 145–150. Black, R., & Fox, R. (1974). Adaptation in invisible gratings and the site of binocular rivalry suppression. Nature, 249, 488–490. Blake, R., Tadin, D., Sobel, K. V., Raissian, T. A., & Chong, S. C. (2006). Strength of early visual adaptation depends on visual awareness. Proceedings of the National Academy of Sciences of the United States of America, 103, 4783–4788. Bonneh, Y. S., Cooperman, A., & Sagi, D. (2001). Motion-induced blindness in normal observers. Nature, 411, 798–801. Bouma, H. (1970). Interaction effects in parafoveal letter recognition. Nature, 226, 177–178. Brainard, D. H. (1997). The psychophysics toolbox. Spatial Vision, 10, 433–436. Cai, Y., Zhou, T., & Chen, L. (2008). Effects of binocular suppression on surround suppression. Journal of Vision 8(9):9, 1–10. Carrasco, M., Ling, S., & Read, S. (2004). Attention alters appearance. Nature Neuroscience, 7, 308–313. Cavanagh, P. (1992). Attention-based motion perception. Science, 257, 1563–1565. Clifford, C. W. G., & Harris, J. A. (2005). Contextual modulation outside of awareness. Current Biology, 15, 574–578. Gyoba, J. (1983). Stationary phantoms: A completion effect without motion and flicker. Vision Research, 23, 205–211. Harp, T., Bressler, D., & Whitney, D. (2007). Position shifts following crowded second-order motion adaptation reveal processing of local and global motion without awareness. Journal of Vision, 7((2):15), 1–13. doi:10.1167/7.2.15.
T. Kawabe, Y. Yamada / Consciousness and Cognition 18 (2009) 168–175
175
Meng, M., Ferneyhough, E., & Tong, F. (2007). Dynamics of perceptual filling-in of visual phantoms revealed by binocular rivalry. Journal of Vision, 7((13):8), 1–15. doi:10.1167/7.13.8.