Keeping Andromeda peripheral: tracking multiple targets out of the corner of your eye
Thursday, July 21, 2016
We owe most of our visual acuity to our fovea, the area of the retina that is most densely packed with photo-receptive cones. Whenever we focus on an object, we move our eyes so that the image is projected onto this area that contains nearly 150,000 cone receptors per square millimeter. Anything that is projected onto our retina outside the fovea is comparatively blurred.
The image below simulates the high acuity obtained by the tiny foveal focus together with the lesser acuity in surrounding areas:
Although foveal acuity is an indispensable attribute of human perception, it does not follow that peripheral vision is its poor cousin or just an unimportant appendage. Quite on the contrary, there are situations in which peripheral vision can get the job done whereas the fovea fails.
One such situation arises in astronomy: Many an amateur astronomer will have taken many attempts to locate and spot the Andromeda Galaxy (M31), even though it can be a spectacular sight even in a modest instrument. The solution? Do not look at the galaxy! Instead, focus your eyes on the dark sky next to it, so that your fovea is processing complete darkness, and then shift your attention—but not your eyes!—to the para-foveal area of your visual field. Presto, the Andromeda will pop out and you will be able to enjoy its rather stunning visual appearance provided you can resist the temptation to look at it. This technique is known as averted vision and is based on the fact that the part of your retina that is not dedicated to the fovea contains “rod” receptors that are particularly light sensitive (although they don’t do color vision). Once you have mastered averted vision, the night sky will reveal thousands of previously invisible nebulae and galaxies.
But even in more every-day situations, peripheral vision can assist your perception in amazing ways: for example, it allows us to detect natural objects at eccentricities of 70°. To put this into context, 90° corresponds to an object just off your left or right ear—so 70° is way out in the periphery indeed. In addition, peripheral vision features higher temporal resolution and superior motion detection than foveal vision.
A recent article in the Psychonomic Society’s journal Attention, Perception, and Psychophysics extended our understanding of this aspect of peripheral vision, namely its involvement in change detection in moving objects.
Researchers Vater, Kredel, and Hossner focused on the role of peripheral vision in the tracking and perception of multiple moving objects—a task that is of critical importance in any team sport, such as soccer or football.
The procedure used by Vater and colleagues is shown in the figure below:
At the beginning of each trial, four of the objects on the screen are arbitrarily selected as targets, and are identified to participants by being highlighted with red frames. (Panel a). The frames then disappear and all objects move in quasi-random fashion for 6 s. (Panel b). At the end of the motion phase, participants recall the targets by naming the numbers corresponding to targets that appear superimposed on all objects. (Panel c).
To ensure the involvement of peripheral vision, stimuli were projected onto a large screen (1.40 × 1.40 m, corresponding to 40° × 40° of visual angle).
The crucial manipulation was that one of the targets during the motion phase (Panel b) either changed motion or shape shortly after it had collided with another object (thereby changing its direction). On the change-motion trials, the target would briefly stop (for ½ a second) before resuming its motion. On the change-shape trials, the object would briefly turn into a diamond (for ½ second) before resuming its original shape. On no-change trials, the objects would continue on their path as per normal.
The participant had to press a response button as soon as target changed motion or shape, and they had to recall the identity of the changed target at the end of the trial (see Panel c above). On trials on which no change occurred, all four targets had to be recalled.
Because Vater and colleagues were interested in peripheral vision, they tracked participants’ eye movements throughout the task. This enabled the researchers to exclude any trial on which the participants, by chance, happened to be focusing their gaze on the target that changed shape or motion.
The results are shown in the figure below:
It is clear that people were able to identify a target that briefly stopped using their peripheral vision (gray portion of bars). Because a response was counted as correct only if both the button was pressed when the event occurred and the item had been named correctly at the end of the trial, we can be certain that people not only detected that some object had briefly stopped but also that they knew the object’s identity.
Performance on the change-shape (or form) trials was less accurate than on the change-motion trials but still quite high if one considers that those objects were outside the person’s foveal vision at the time. (The no change control trials record the accuracy of recalling all 4 objects at the end of a trial; they are of little interest here.)
One further intriguing aspect of the results is that the order of accuracy between motion and shape was reversed for a companion experiment in which the object was in the fovea: this is entirely consonant with the idea that peripheral vision is particularly good at detecting motion whereas the fovea (of course) specializes in high-acuity identification of object shapes.
Unsurprisingly, the changed target did not remain in people’s peripheral vision for long: As shown in the figure below, people moved their gaze towards the target after the change in motion or shape had been detected:
Clearly, peripheral vision can suffice to identify something we should pay attention to—and we then shift our eyes to examine the object in greater detail. This works fantastically well in team sports if you spot an opposing player reversing direction with your peripheral vision—you’d better check out why he did that by focusing on him or her. But this strategy fails if you are chasing a faint object in a telescope: there you need to resist the temptation to look at a galaxy lest it disappear in your fovea.
In the next series of experiments, Vater and colleagues will show that motion changes can be detected at the far periphery in dual-task situations. A player can therefore make the choice whether he or she needs detailed information of an opposing action by focusing on that action, or by monitoring opposing movements solely with peripheral vision and use the high resolution of foveal vision to view another player.
Article focused on in this post:
Vater, C., Kredel, R., & Hossner, E.-J. (2016). Detecting single-target changes in multiple object tracking: The case of peripheral vision. Attention, Perception, and Psychophysics. DOI: 10.3758/s13414-016-1078-7.