Cognitive Daily

Do new objects capture our attention?

Posted by Dave Munger on November 21, 2005
(6)
More »

The picture below will link you to a quick animation. The blue ring will gradually get smaller until it obscures the three “8″s, then continue to shrink until the figures are visible again. While they are obscured, the 8s will be transformed into letters (S, P, E, U, or H), and a new letter will also appear. Your job is to search for the letter U or H—it has an equal chance of appearing where any of the 8s were, or in the new spot. Click on the picture to try it out.

i-c7fae76b306b20807048b3bc85f3e7de-annulusimg.gif

Attention researchers Steven Franconeri, Andrew Hollingworth, and Daniel Simons used a similar animation to answer a key question about what attracts our attention. Recent research has led to two different hypotheses—either the appearance of a new object, or a change in the luminance (brightness) of an area of our field of view is what attracts our attention.

In many cases, both of these changes occur at the same time. For example, we’re driving down the street and a boy runs out in front of our car chasing a ball. The boy appearing in front of our car is certainly a new object, and the light reflecting off of his sweatshirt carries a different luminance value from the asphalt pavement we had been looking at.

Franconeri and his colleagues cleverly designed the animated display you just looked at to introduce a new object without a corresponding change in luminance: the object was introduced while all four objects were obscured by the blue ring, so when it appeared, there was no luminance change.

In a separate condition, the group presented the same animation, with the ring passing behind the letters, so that when the new object appeared, there was a change in luminance. They repeated the task with 2, 3, and 4 objects in the final display; in each case, participants knew that the 8s would change into letters, and that they’d be searching for a U or an H. Here are their results:

i-8515f6a67722abeadb079e6c5724a823-annulus1.gif

When the ring passes in front of the letters, there is no difference in reaction time, whether the target letter (U or H) appears in a new position or an old position. This suggests that the mere fact of a new letter appearing doesn’t attract our attention, because if it did, we’d react more quickly when the target letter appeared in a new position. Compare this to the control condition, where the letters are always in view:

i-12fb1374308867f8ae7106d4b0f04ba8-annulus2.gif

Now, reaction time is significantly faster for the new item. So when a new item actually changes the luminance of an area in our field of view, we react faster. Franconeri and his team argue that this result supports the luminance hypothesis. They suggest that when the new item corresponds to a luminance change, we direct our attention to that new item. Then we are able to complete the task more quickly—to determine if that item is an H or a U. So a change in luminance attracts our attention, but the appearance of a new object on its own does not.

Franconeri, S.L., Hollingworth, A., & Simons, D.J. (2005). Do new objects capture attention? Psychological Science, 16(4), 275-281.

(6)
More »

Comments

  1. #1 Eve
    November 21, 2005

    Umm.. the ring doesn’t disappear gradually – I just get directed to a site where there is a blue ring with four letters outside it …
    http://cognitivedaily.com/images/annulus.gif

  2. #2 Dave Munger
    November 21, 2005

    It moves pretty quickly. Try emptying your cache and refreshing your browser window. It should definitely work—I’ve tried it on several computers.

  3. #3 Dave Munger
    November 22, 2005

    “is it possible to have a change in luminance without having a new object?”

    Sure—an object changes color, for example. Or imagine a ball that is red on one half and white on the other. When the ball spins, it changes luminance.

    “What this study really seems to do is to show that new objects are detected quicker if there’s a change in luminance to highlight the new object.”

    It actually does more than that—it shows that absent a luminance change, new objects are detected at the same speed as old objects. Our attention is attracted by a luminance change, but not by a new object.

    What’s amazing to me about this study is that the animation is so rapid (and my understanding is that the original experiment was even more rapid). Take a look at the sample stimulus again: the difference between detecting versus not detecting the new object faster is a mere one-hundredth of a second while all the objects are occluded.

  4. #4 Quercus
    November 22, 2005

    Seems to me like they’re comparing two different levels of abstraction here. A change in luminance is one (out of several) character that defines a new object. So they’re not really independent things.
    In other words, while they created a new object with no change in luminance, is it possible to have a change in luminance without having a new object?

    What this study really seems to do is to show that new objects are detected quicker if there’s a change in luminance to highlight the new object. Not very surprising, really. Possibly interesting if they can determine what other characteristics help us identify new objects and compare them. I suspect that’s more of a low-level visual processing question than the investigators are interested in.

  5. #5 Brent Edwards
    November 23, 2005

    The experiment reminds me of change blindness, where one can be amazingly blind to visual changes when they occur simultaneous with the re-introduction of other visual information that had not changed. When the ring is on top of the 8s, adding new location for a letter doesn’t cause attention to register the new information and therefore has no effect on reaction time. Having the change occur on top of the ring, however, is an instantaneous change/addition in the absence of other new location changes. Visually, one can get at whether this is due to a change in luminance by changing the color that doesn’t change the luminance relative to the background. Unless I’m misunderstanding the conditions.

  6. #6 Can
    November 26, 2005

    Maybe the reason is that we’re taught to detect things with different color or luminance. We all know that new thigns have different colors. More bright, or more light, depending on the environment. This is what every designers use very often to differentiate the new comers. If all designers or maybe all websites represents new/important things by givining them an unusual location, maybe we could detect the letter more easily in the animation. Maybe the experiment should be repated after teaching people how to find the new things.