Limits...
Rapid perceptual switching of a reversible biological figure.

Jackson S, Cummins F, Brady N - PLoS ONE (2008)

Bottom Line: We found evidence that some observers experience peaks in the distribution of response locations that are relatively stable across sessions.In summary, we have demonstrated that the temporal dynamics of reversal with biological motion are similar to other forms of ambiguous SFM.We conclude that perceptual switching with biological motion is a robust bistable phenomenon.

View Article: PubMed Central - PubMed

Affiliation: Cognitive Science, UCD School of Computer Science and Informatics, University College Dublin, Belfield, Dublin, Ireland, UK. stuart.jackson@ucdconnect.ie

ABSTRACT
Certain visual stimuli can give rise to contradictory perceptions. In this paper we examine the temporal dynamics of perceptual reversals experienced with biological motion, comparing these dynamics to those observed with other ambiguous structure from motion (SFM) stimuli. In our first experiment, naïve observers monitored perceptual alternations with an ambiguous rotating walker, a figure that randomly alternates between walking in clockwise (CW) and counter-clockwise (CCW) directions. While the number of reported reversals varied between observers, the observed dynamics (distribution of dominance durations, CW/CCW proportions) were comparable to those experienced with an ambiguous kinetic depth cylinder. In a second experiment, we compared reversal profiles with rotating and standard point-light walkers (i.e. non-rotating). Over multiple test repetitions, three out of four observers experienced consistently shorter mean percept durations with the rotating walker, suggesting that the added rotational component may speed up reversal rates with biomotion. For both stimuli, the drift in alternation rate across trial and across repetition was minimal. In our final experiment, we investigated whether reversals with the rotating walker and a non-biological object with similar global dimensions (rotating cuboid) occur at random phases of the rotation cycle. We found evidence that some observers experience peaks in the distribution of response locations that are relatively stable across sessions. Using control data, we discuss the role of eye movements in the development of these reversal patterns, and the related role of exogenous stimulus characteristics. In summary, we have demonstrated that the temporal dynamics of reversal with biological motion are similar to other forms of ambiguous SFM. We conclude that perceptual switching with biological motion is a robust bistable phenomenon.

Show MeSH

Related in: MedlinePlus

Eye movements and rotation speed.(A) For each individual dot making up the walker, we calculated the absolute horizontal distance traveled (in pixels) from frame to frame i.e. leftward and rightward motion are treated similarly. Combining the values for all thirteen dots at each frame, and dividing by the total motion in pixels over all frames, gives the proportion of image motion contained between any two frames in the rotation cycle. The colormap illustrates the distribution of image motion across the 60-frame sequence, with brighter values (white) indicating a greater proportion of image motion. (B) Mean saccades/min and reversals/min for each of the three observers, in the fixation (black) and eye movements encouraged (grey) conditions. Error bars represent standard deviations. To confirm that our saccade detection routines performed satisfactorily, we plotted saccade peak velocity against amplitude (i.e. the main sequence [32]). (C) For all saccades made in a particular condition, we calculated the proportion triggered during the presentation of individual frames of the rotation cycle, by noting the frame on screen during the starting sample of the saccade. The blue (fixation) and green (eye movements) lines illustrate this distribution, divided into fifteen equally-sized bins. The black line represents the distribution of image motion across the sequence. Note that the combined histogram data (row four) is not the average of individual datasets, but the pooled saccade data. (D) For each percept reversal, we compared the median fixation positions during two separate time windows prior to the button response: Mid-reversal (540ms-340ms prior to response), during which the reversal is estimated to have occurred, and Post-reversal (200ms-0ms prior to response). Each symbol corresponds to the average over all reversals for a particular observer. Horizontal and vertical lines represent the inter-quartile range along the respective axes. Note than unlike in Experiment 3, both transition types were monitored as normal in each trial. The plots separate the data for each transition type for comparison.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2601034&req=5

pone-0003982-g009: Eye movements and rotation speed.(A) For each individual dot making up the walker, we calculated the absolute horizontal distance traveled (in pixels) from frame to frame i.e. leftward and rightward motion are treated similarly. Combining the values for all thirteen dots at each frame, and dividing by the total motion in pixels over all frames, gives the proportion of image motion contained between any two frames in the rotation cycle. The colormap illustrates the distribution of image motion across the 60-frame sequence, with brighter values (white) indicating a greater proportion of image motion. (B) Mean saccades/min and reversals/min for each of the three observers, in the fixation (black) and eye movements encouraged (grey) conditions. Error bars represent standard deviations. To confirm that our saccade detection routines performed satisfactorily, we plotted saccade peak velocity against amplitude (i.e. the main sequence [32]). (C) For all saccades made in a particular condition, we calculated the proportion triggered during the presentation of individual frames of the rotation cycle, by noting the frame on screen during the starting sample of the saccade. The blue (fixation) and green (eye movements) lines illustrate this distribution, divided into fifteen equally-sized bins. The black line represents the distribution of image motion across the sequence. Note that the combined histogram data (row four) is not the average of individual datasets, but the pooled saccade data. (D) For each percept reversal, we compared the median fixation positions during two separate time windows prior to the button response: Mid-reversal (540ms-340ms prior to response), during which the reversal is estimated to have occurred, and Post-reversal (200ms-0ms prior to response). Each symbol corresponds to the average over all reversals for a particular observer. Horizontal and vertical lines represent the inter-quartile range along the respective axes. Note than unlike in Experiment 3, both transition types were monitored as normal in each trial. The plots separate the data for each transition type for comparison.

Mentions: Figure 9A depicts the distribution of walker dot motion vectors (horizontal component), normalized across the rotation cycle. We can see two obvious peaks in the distribution of absolute dot motions, at approximately one-third and two-thirds of the way through the rotation cycle (slowly cycle through Movie S1). Therefore, it is very likely that image motion may in some way influence the percept in systematic ways. Beintema, Oleksiak & van Wezel [25] have shown that interpretations of biological motion are strongly affected by the stimulus speed. These authors found that when presented at unnaturally slow speeds, biological motion is perceived to rotate in depth. This effect was particularly strong when the figure was rendered unfamiliar to observers, by scrambling or inverting the display. At higher, natural speeds, the percept was veridical with respect to the stimulus display. The authors interpreted these findings as evidence of the existence of biological motion channels tuned to higher, more natural walking speeds; channels that presumably dominate over a default assumption to perceive trajectories in depth [25]. This finding may be important in the current context. When the walker approaches angular orientations where image motion increases or decreases abruptly, changes in image motion may affect the structure/depth interpretation. For example, accelerating or decelerating movements of the hands and feet may trigger eye movements. These eye movements could then trigger perceptual alternations, with a limb moving in the opposite direction then becoming most salient, and triggering further eye movements (similar to the trapping that occurs with other ambiguous SFM, but different in the sense that it is not a ‘random’ dot that triggers eye movements). This could result in alternations congregating at systematic locations in the rotation cycle (and such patterns may be the reason observers find the ‘collapsing’ percept so frustrating).


Rapid perceptual switching of a reversible biological figure.

Jackson S, Cummins F, Brady N - PLoS ONE (2008)

Eye movements and rotation speed.(A) For each individual dot making up the walker, we calculated the absolute horizontal distance traveled (in pixels) from frame to frame i.e. leftward and rightward motion are treated similarly. Combining the values for all thirteen dots at each frame, and dividing by the total motion in pixels over all frames, gives the proportion of image motion contained between any two frames in the rotation cycle. The colormap illustrates the distribution of image motion across the 60-frame sequence, with brighter values (white) indicating a greater proportion of image motion. (B) Mean saccades/min and reversals/min for each of the three observers, in the fixation (black) and eye movements encouraged (grey) conditions. Error bars represent standard deviations. To confirm that our saccade detection routines performed satisfactorily, we plotted saccade peak velocity against amplitude (i.e. the main sequence [32]). (C) For all saccades made in a particular condition, we calculated the proportion triggered during the presentation of individual frames of the rotation cycle, by noting the frame on screen during the starting sample of the saccade. The blue (fixation) and green (eye movements) lines illustrate this distribution, divided into fifteen equally-sized bins. The black line represents the distribution of image motion across the sequence. Note that the combined histogram data (row four) is not the average of individual datasets, but the pooled saccade data. (D) For each percept reversal, we compared the median fixation positions during two separate time windows prior to the button response: Mid-reversal (540ms-340ms prior to response), during which the reversal is estimated to have occurred, and Post-reversal (200ms-0ms prior to response). Each symbol corresponds to the average over all reversals for a particular observer. Horizontal and vertical lines represent the inter-quartile range along the respective axes. Note than unlike in Experiment 3, both transition types were monitored as normal in each trial. The plots separate the data for each transition type for comparison.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2601034&req=5

pone-0003982-g009: Eye movements and rotation speed.(A) For each individual dot making up the walker, we calculated the absolute horizontal distance traveled (in pixels) from frame to frame i.e. leftward and rightward motion are treated similarly. Combining the values for all thirteen dots at each frame, and dividing by the total motion in pixels over all frames, gives the proportion of image motion contained between any two frames in the rotation cycle. The colormap illustrates the distribution of image motion across the 60-frame sequence, with brighter values (white) indicating a greater proportion of image motion. (B) Mean saccades/min and reversals/min for each of the three observers, in the fixation (black) and eye movements encouraged (grey) conditions. Error bars represent standard deviations. To confirm that our saccade detection routines performed satisfactorily, we plotted saccade peak velocity against amplitude (i.e. the main sequence [32]). (C) For all saccades made in a particular condition, we calculated the proportion triggered during the presentation of individual frames of the rotation cycle, by noting the frame on screen during the starting sample of the saccade. The blue (fixation) and green (eye movements) lines illustrate this distribution, divided into fifteen equally-sized bins. The black line represents the distribution of image motion across the sequence. Note that the combined histogram data (row four) is not the average of individual datasets, but the pooled saccade data. (D) For each percept reversal, we compared the median fixation positions during two separate time windows prior to the button response: Mid-reversal (540ms-340ms prior to response), during which the reversal is estimated to have occurred, and Post-reversal (200ms-0ms prior to response). Each symbol corresponds to the average over all reversals for a particular observer. Horizontal and vertical lines represent the inter-quartile range along the respective axes. Note than unlike in Experiment 3, both transition types were monitored as normal in each trial. The plots separate the data for each transition type for comparison.
Mentions: Figure 9A depicts the distribution of walker dot motion vectors (horizontal component), normalized across the rotation cycle. We can see two obvious peaks in the distribution of absolute dot motions, at approximately one-third and two-thirds of the way through the rotation cycle (slowly cycle through Movie S1). Therefore, it is very likely that image motion may in some way influence the percept in systematic ways. Beintema, Oleksiak & van Wezel [25] have shown that interpretations of biological motion are strongly affected by the stimulus speed. These authors found that when presented at unnaturally slow speeds, biological motion is perceived to rotate in depth. This effect was particularly strong when the figure was rendered unfamiliar to observers, by scrambling or inverting the display. At higher, natural speeds, the percept was veridical with respect to the stimulus display. The authors interpreted these findings as evidence of the existence of biological motion channels tuned to higher, more natural walking speeds; channels that presumably dominate over a default assumption to perceive trajectories in depth [25]. This finding may be important in the current context. When the walker approaches angular orientations where image motion increases or decreases abruptly, changes in image motion may affect the structure/depth interpretation. For example, accelerating or decelerating movements of the hands and feet may trigger eye movements. These eye movements could then trigger perceptual alternations, with a limb moving in the opposite direction then becoming most salient, and triggering further eye movements (similar to the trapping that occurs with other ambiguous SFM, but different in the sense that it is not a ‘random’ dot that triggers eye movements). This could result in alternations congregating at systematic locations in the rotation cycle (and such patterns may be the reason observers find the ‘collapsing’ percept so frustrating).

Bottom Line: We found evidence that some observers experience peaks in the distribution of response locations that are relatively stable across sessions.In summary, we have demonstrated that the temporal dynamics of reversal with biological motion are similar to other forms of ambiguous SFM.We conclude that perceptual switching with biological motion is a robust bistable phenomenon.

View Article: PubMed Central - PubMed

Affiliation: Cognitive Science, UCD School of Computer Science and Informatics, University College Dublin, Belfield, Dublin, Ireland, UK. stuart.jackson@ucdconnect.ie

ABSTRACT
Certain visual stimuli can give rise to contradictory perceptions. In this paper we examine the temporal dynamics of perceptual reversals experienced with biological motion, comparing these dynamics to those observed with other ambiguous structure from motion (SFM) stimuli. In our first experiment, naïve observers monitored perceptual alternations with an ambiguous rotating walker, a figure that randomly alternates between walking in clockwise (CW) and counter-clockwise (CCW) directions. While the number of reported reversals varied between observers, the observed dynamics (distribution of dominance durations, CW/CCW proportions) were comparable to those experienced with an ambiguous kinetic depth cylinder. In a second experiment, we compared reversal profiles with rotating and standard point-light walkers (i.e. non-rotating). Over multiple test repetitions, three out of four observers experienced consistently shorter mean percept durations with the rotating walker, suggesting that the added rotational component may speed up reversal rates with biomotion. For both stimuli, the drift in alternation rate across trial and across repetition was minimal. In our final experiment, we investigated whether reversals with the rotating walker and a non-biological object with similar global dimensions (rotating cuboid) occur at random phases of the rotation cycle. We found evidence that some observers experience peaks in the distribution of response locations that are relatively stable across sessions. Using control data, we discuss the role of eye movements in the development of these reversal patterns, and the related role of exogenous stimulus characteristics. In summary, we have demonstrated that the temporal dynamics of reversal with biological motion are similar to other forms of ambiguous SFM. We conclude that perceptual switching with biological motion is a robust bistable phenomenon.

Show MeSH
Related in: MedlinePlus