How do we find those lost keys? The color of the environment doesn't seem to matter

ResearchBlogging.orgThe other day, our car wouldn't start and Jim had to ask a neighbor over to help him jump-start it. There was much rushing in and out of the house looking for flashlights and other tools to help get the job done. After the neighbor left, Jim wanted to drive somewhere and couldn't find the keys. Clearly he had just had them because he was working on the car. Where could they be? We searched up and down throughout the house, but we couldn't find them and eventually had to use a spare set.

The next morning as I was getting ready to leave for our school carpool in our other car, I found them sitting on a workbench in the garage, just a few feet from the car! Why couldn't we find them when we needed them? Everyone's experienced a similar problem at some point, whether it's trying to find the remote for the TV or the olives in the fridge. Why are objects so hard to find sometimes and at the tip of our fingers other times?

One part of the answer is a phenomenon called "contextual cuing." Basically, this means we're good at finding things in places we've found them before, but bad at finding them elsewhere. The more often we see an object in a certain place, the quicker we are at finding it there. It doesn't take long to train people to locate new objects in this way. For example, a researcher might ask people to search for the letter "T" or "L" in an array of letters. Most of the time, the array is completely new, and the T or L is in a different place. But if occasionally the same array appears, people are quick to recognize it and find the letter much faster. It even works if only the portion of the array around the target letter stays the same and the rest of it changes.

So the question then becomes this: what are they key elements of the area you're searching? If I usually leave my keys on the kitchen counter, will I still be able to find them if Nora has cluttered it up with a baking project? In one contextual cuing study, the researchers showed viewers scenes filled with distractor objects that were either black or white, and asked them to search for a target. After they had learned to find the target object quickly among a particular pattern of distractors, the experimenters changed the distractors from black to white (or vice-versa). The advantage of contextual cuing disappeared.

Is the color of the surrounding environment really an important part of how we find objects? Krista Ehinger and James Brockmole suspected that in more realistic environments, color may not matter as much. They showed volunteers images like this and asked them to search for a tiny letter T or L:


Can you spot it? How about in this picture?


The colors here are unrealistic, but it's still obviously a shoreline scene, just like the first. Now try one more:


Again, the colors have changed, but it's clearly the same scene as the first image. Did you spot the T faster?

Ehinger and Brockmole created 16 different versions of each image, manipulating it to create different unrealistic color combinations, like this:


Viewers searched for letters in blocks of 16 images at a time. Half of the images were ones they had seen before, with the search item in the same place. The other half of the images were brand-new each time. This was repeated 16 times -- so eight the images were seen 16 times each, and 128 images were seen once each.

But the key to the study was this: Viewers were divided into groups. One group saw the identical images repeated consistently. They might have had an odd coloration, but it was the consistent each time they saw it. A second group saw a variable coloration of the repeated images. The target letter T or L was in the same place each time, but the color of the image it was in was varied each time they saw it. Here are the results:


The graph shows the cuing effect as the experiment progressed. The cuing effect is measured by subtracting the reaction time for the repeated image (whether the color of the image was variable or consistent) from the reaction time for the new images. As you can see, the effect got stronger quite quickly for both groups (as well as for a control group not shown here). Most importantly, there was no significant difference between the groups -- contextual cuing occurred whether or not the color of the repeated image changed.

In a second experiment, the colors were kept consistent until the final block, when all the colors were changed. Once again, there was no difference in the cuing effect for the last two blocks -- suggesting that the first group hadn't simply been trained to ignore colors.

So why does color not affect contextual cuing in this study when it clearly does have an impact in other studies? Ehinger and Brockmole suggest that other aspects of the scene may be more important. In the real world, we must be able to recognize scenes at different times of the day, with different weather and lighting conditions, so color changes -- even unrealistic ones -- don't faze us. Unfortunately when we put our keys in a completely unfamiliar location, no amount of coloring or other environmental cues will help us find them efficiently.

EHINGER, K., & BROCKMOLE, J. (2008). The role of color in visual search in real-world scenes: Evidence from contextual cuing Perception & Psychophysics, 70 (7), 1366-1378 DOI: 10.3758/PP.70.7.1366


More like this

This is a guest post by Daniel Griffin, one of Greta's top student writers from Spring of 2007. Does anything seem stick out about this sentence? I'm sure that if I told you to keep looking for yellow highlighted words, you would not have much trouble finding them in these first few sentences. You…
Take a look at the image below. Your job is to find the T among the sea of Ls. If you're like most people it will take just a second or two. Figure 1: If you repeat this task several dozen times, each time with a new set of Ls and T in different colors, positions, and orientations, you'll get…
Take a look at this movie (you'll need a video player like QuickTime or Windows Media Player installed in your browser to see it). You'll see four different outdoor scenes flash by, one at a time. The scene itself will only be displayed for a fraction of a second, followed immediately by a…
Take a look at these pictures. Each picture depicts four shapes -- irregular vertical columns spanning the height of the picture. It's easy to tell which letter is on a column and which is not, right? If our readers are typical, over 90 percent would agree that a is on a column and b is not. But…

Odd, I found them almost instantly in all three pictures?

By DazedNConfuzed (not verified) on 01 Oct 2009 #permalink

I think you should give color a little more credit, here. The particular color might not matter, but the contrast between the background and the search target should; that's why camouflage (and highlighters) work. If that T were dark green, it would've been much harder to find. Surely, if your keychain were hot pink or hunter orange, you would've found them right away. Unless, of course, your whole workbench is hot pink.

I'm in the habit that when I come in the house my wallet, keys and iPod are placed on my desk.

Of course my SO has 'borrowed' my keys on occasion and left them in a different place. And he's the type who doesn't remember where he leaves things.

Great fun ensues.

The fact that, even though about 7% of men are colorblind, no-one ever seems to have noticed that the condition even existed before 1794, when John Dalton first described his own deuteranopia, also suggests that color vision (full trichromacy, at any rate) serves little adaptive purpose for humans. Most other mammals, apart from primates, also get by without trichromacy. Presumably it evolved in our primate ancestors because it served some useful purpose for them (maybe finding fruit amongst leaves), but since we came down from the trees it has not been all that important for us. This experiment seems to confirm that, to a very large extent, normal human color vision is an evolutionary luxury. Nice, and occasionally useful but we do not really need it. Seeing spatial relationships (and spatio-temporal ones, such as movement) is much more important.

Took me quite a while to spot the T in the last of those three images. I think it was because my eyes were constantly drawn to the tower - which really stands out because of the contrast - and particularly those two little, bright white lines on the walls which could - at a quick glance - look like the letter l.

I agree with JRH that particular colors may not be important, after all I think we can generally find things easily enough in black and white images too. What is important is contrast, and where your eyes are naturally drawn in a given setting. In the other two images my eyes were not drawn nearly as heavily towards one part of the image, thus it was easier to scan the whole scene quickly and find the letters.

Your workbench may be in a place where your eyes don't tend to look very often, either because its not very prominent or special where it stands, or because you're not used to looking for things there. Or both. I believe, in a stressed situation you're particularly prone to focusing in certain places because you have to work quickly and focus on the places most likely to yield success. But when you're not stressed your eyes don't focus as heavily and rather scan the area broadly, so you find the keys in the unlikely place where they are.

Makes good sense to me. If you have to find something particular, you focus your eyes on small areas of the scene, while when you are simply moving around without looking for anything specific you glance around and focus less. Makes sense for an animal having to avoid predators to scan the whole scene when there's no obvious danger, to better catch unexpected movements, while focusing in on the most important areas when there is known danger around.

I also picked out the "T"s and "L" immediately in the first three photos. Not sure if this is the same thing, but I seem to do better than average on hidden items puzzles. And am good at finding Waldo. LOL.

Because I am absent-minded I've tried very hard to set up designated places to put vital things like keys. However, as with today, I still will get distracted or rushed on occasion and will lay them down in odd places. I have mislaid my cell phone yet again and still haven't run across it.

This is hair-tearingly frustrating. Particularly because when I do it, some inner voice says, "Now don't forget you put those there." But this is almost a cue to forget it. Why do I never learn? I guess that is a different psych problem altogether.

I'm in the habit that when I come in the house my wallet, keys and iPod are placed on my desk.

Of course my SO has 'borrowed' my keys on occasion and left them in a different place. And he's the type who doesn't remember where he leaves things.

Great fun ensues.

A small contribution: I have a nasty exotic weed on my land called cinquefoil. I have been slowly obliterating it from my land, but it has taken years. I have now gotten quite proficient at finding it among many other species. My vision cues first on the color of the leaves; I can key on the precise shade of green to lock onto it. After that comes shape, which is unfortunately similar to a desirable species. It takes less than a second to make a final determination. The trick, though, is the scanning process, which I can now carry out at high speed as I walk across the land. I walk at a slightly slower than normal pace and cover a swath six feet wide. But it has taken five years to learn those visual cues so deeply that the process is now so fast.

By Chris Crawford (not verified) on 04 Oct 2009 #permalink

From the annals of electrical engineering and image recognition math: the only mathematical parameter that distinctly duplicates the photo recognition of humans is image brightness differentials in space, i.e. determine the derivative in x and y of the brightness alone as a spatial pattern and other patterns with similar spatial values will "look" like the same thing. There are neural net arguments why this is likely to be the case especially given the structure of the retina as the same. All else fails to create recognition algorithms that "make sense" to the human brain.

The advantage of being a polymath...