Backward position shift in apparent motion

Hsin-Hung Li, Won Mok Shim, Patrick CavanaghJournal of Vision 2014 · DOI

TL;DR

This paper reveals a surprising "backward" perceived shift in apparent motion, where a moving object appears to shift against its actual direction of movement, specifically for the second element in a short sequence. This effect, attributed to an "attraction" from the initial stimulus, contrasts with the more commonly observed "forward" extrapolation and highlights the complex interplay of salience and temporal dynamics in visual perception.

What Problem Does This Paper Address?

The perception of visual motion often involves shifts in perceived object location. A well-documented phenomenon is "forward extrapolation," where a moving object's perceived position is shifted slightly ahead in its direction of motion. This is often explained as a neural compensation for processing delays. However, the authors investigate a less common, and in this case, unexpected effect: a "backward" shift in apparent motion. They question how the number of discrete steps in an apparent motion sequence influences this perceived positional shift, particularly in the periphery, and whether this backward shift is a consistent phenomenon or dependent on specific stimulus configurations.

Methods

The researchers employed a series of psychophysical experiments using apparent motion displays.

  • Experiment 1: Observers viewed a three-disc sequence moving horizontally in the far periphery (20° above fixation). They were asked to judge the position of the middle disc relative to a static reference.
  • Experiment 2: This experiment systematically varied the number of discs in the apparent motion sequence (3, 5, and 7 discs) while keeping the overall spatial and temporal extents (and thus speed) constant. The goal was to see how increasing the "continuity" of the motion affected the perceived shift.
  • Experiment 3: A localization task was used to measure the position shifts of all disc locations in two-, three-, and four-disc sequences. This experiment also explored whether the response modality (perceptual judgment vs. motor response) influenced the results.

A key aspect of their methodology was the use of psychometric functions to quantify perceived positions, specifically by determining the Point of Subjective Equality (PSE).

Datasets

Not discussed in the paper. The stimuli were generated for the experiments.

Key Insights and Ideas

  • The "Backward Shift" Surprise: The most striking finding is the consistent backward shift of the second element in short apparent motion sequences. This suggests that instead of extrapolating forward, the visual system can be "pulled back" towards the initial stimulus.
  • Salience Matters: The authors propose that the initial stimulus is highly salient and "attracts" the subsequent, less salient, stimulus. This attraction effect is strongest when the motion is discrete and the steps are large.
  • Continuity is Key: As the apparent motion becomes more continuous (more discs, smaller steps), the backward shift diminishes and eventually reverses into the more typical forward shift. This suggests a transition from discrete attraction to more fluid motion extrapolation.
  • Location Specificity: The backward shift is not uniform across all elements. It is consistently observed for the second element in sequences, regardless of the total number of discs (up to four), but the effect on later elements varies.
  • Periphery Matters: The experiments were conducted in the far periphery, suggesting that these effects might be more pronounced or unique to peripheral vision, where spatial resolution and motion processing might differ from central vision.

Key Results

  • Experiment 1: Observers consistently mislocalized the middle disc of a three-disc sequence backward, opposite to the direction of motion.
  • Experiment 2: The backward shift was observed only for the three-disc sequence. With five and seven discs, the perceived shift became forward, in the direction of motion.
  • Experiment 3:
    • In two-disc motion, the first disc shifted forward, and the second backward.
  • In three-disc motion, the first disc shifted forward, the second backward, and the third showed no significant shift.
  • In four-disc motion, the first disc shifted forward, the second backward, and the third shifted strongly forward, while the fourth showed no significant shift.
  • The backward shift for the second disc increased with the number of discs in the sequence (from two to four discs).
  • The results were not significantly affected by target color (white vs. green) or motion direction (leftward vs. rightward).

Limitations and Future Work

  • Attraction Account Speculative: The authors acknowledge that their proposed "attraction" mechanism is speculative and requires further testing.
  • Other Factors: They note that other factors, such as the eccentricity of the stimuli and the specific temporal dynamics, might also play a role and need further investigation.
  • Future Directions: Future work could explore the role of motion direction (toward vs. away from the fovea) more deeply, as individual differences were observed. Differentiating between their attraction/salience proposal and other accounts like feature integration is also a key area for future research. They also suggest that the backward shift might be constrained to high-speed apparent motion with large steps in the periphery.

Summary updated

Abstract

We investigated the perceived position of visual targets in apparent motion. A disc moved horizontally through three positions from -10° to +10° in the far periphery (20° above fixation), generating a compelling impression of apparent motion. In the first experiment, observers compared the position of the middle of the three discs to a subsequently presented reference. Unexpectedly, observers judged its position to be shifted backward, in the direction opposite that of the motion. We then tested the middle disc in sequences of 3, 5, and 7 discs, each covering the same spatial and temporal extents (similar speeds). The backwards shift was only found for the three-disc sequence. With the extra discs approaching more continuous motion, the perceived shift was in the same direction as the apparent motion. Finally, using a localization task with constant static references, we measured the position shifts of all the disc locations for two-disc, three-disc and four-disc apparent motion sequences. The backward shift was found for the second location of all sequences. We suggest that the backward shift of the second element along an apparent motion path is due to an attraction effect induced by the initial point of the motion.
BibTeX
@article{li2014backward,
  author = {Hsin-Hung Li and Won Mok Shim and Patrick Cavanagh},
  title = {{Backward position shift in apparent motion}},
  journal = {Journal of Vision},
  year = {2014},
  doi = {10.1167/14.1.16},
}
Figure