Cognitive Daily

Boundary extension and emotion

Take a look at this picture I took last year when we went to Venice. Look at it fairly closely, because there will be a “test” at the end of this post.


We have posted on boundary extension before: it’s a simple phenomenon where our memory for a picture is consistently different from the actual picture we saw before. We remember a larger frame, or boundary for the picture, than was actually present. This effect can be measured in many different ways—for example, by asking people to sketch the picture they saw, or draw the frame on a larger picture, or, as was done in today’s experiment, by showing people several different examples of the picture they saw, and asking which one is correct.

The trick with this final method is that none of the pictures is actually correct. Usually, participants are asked to choose between pictures reduced or enlarged by 10 to 20 percent. For ordinary pictures, people will choose the picture with a 10 percent larger frame most often.

However, a significant body of research has found that people tend to narrow the focus of their attention when they are emotionally aroused—for example when they are shown violent photographs. So how does this narrowing of focus interact with boundary extension? Are we literally narrowing our focus? If so, then shouldn’t we do exactly the opposite of boundary extension—boundary restriction?

Andrew Mathews and Bundy Mackintosh of the MRC Cognition and Brain Sciences Unit of Cambridge University have been working on this problem for the past decade. They have found that negative emotional scenes do tend to have reduced boundary extension, particularly for people who tested as more anxious than average. In a new study (“Take a Closer Look: Emotion Modifies the Boundary Extension Effect,” Emotion, 2004), they explored whether other emotionally arousing images also led to boundary restriction.

They selected a set of images that included pleasant, unpleasant, and neutral images, but also images that were either relatively arousing or not arousing. Images can be both unpleasant and arousing, for example pictures of a violent assault. An unpleasant, non-arousing picture might be a photo of a body in a coffin, and a pleasant arousing picture could be a skydiver. (I imagine my picture of Venice might only be arousing to a roofing fanatic or a boating enthusiast. Maybe I ought to photoshop Godzilla storming the city or something.)

Viewers rated the images on a scale of extremely negative to extremely positive, and were tested for boundary extension by viewing each picture for 5 seconds, then at the end of the study, looking at four different pictures—ten or twenty percent larger or smaller than the original, and selecting the picture that they thought was the same. None of the participants ever suspected that the original picture was not one of there choices. Here are the results:


There are no results for extremely positive or negative, low-arousal images, because all these images were arousing. Note first that high-anxiety individuals had the lowest boundary extension ratings of all for extremely negative images. Low-anxiety individuals had normal boundary extension for these images. However, for images that were negative or neutral, both high-anxiety and low-anxiety individuals had less boundary extension for high-arousal images, and normal boundary extension for low-arousal images.

The only significant difference between high-anxiety individuals and low-anxiety individuals was for extremely negative images. Mathews and Mackintosh suspect that the difference may be due to different reactions to highly negative images: low-anxiety individuals tend to avert their eyes from highly negative images, exploring the edges of the scene, while high-anxiety individuals can’t keep themselves from focusing on the central negative image. Therefore, low-anxiety results in boundary extension, while high-anxiety does not. But regardless of anxiety levels, most people have less boundary extension for mildly negative or neutral, arousing images.

Now, remember that image I showed you at the beginning of the article? Take a look at this one.


Is it the same or different from the original picture I showed you? No peeking back at the original. Remember, you should see a larger-boundary image as the same as the original. Or I might be tricking you and showing you the same image you saw before. I’ll put the answer in the comments.


  1. #1 Dave Munger
    July 13, 2005

    The second picture has an expanded boundary compared to the first one. Since it’s a pleasant photo, it should have looked about the same to you.

  2. memory

    Boundary extension and emotion .

  3. #3 Glen Gordon
    July 13, 2005

    For some reason, I noticed in the second picture there was more of the column on the right showing. Perhaps it was because I disliked the column intruding into the image, and that part was empahsized in the second photo. By the way, great site!

  4. or some reason, I noticed in the second picture there was more of the column on the right showing. Perhaps it was because I disliked the column intruding into the image, and that part was empahsized in the second photo.

    I agree.

  5. #5 Chris Tregenza
    July 14, 2005

    I think the second picture is the same except there seemed to be more white space around it. This suggests its smaller but another part of my brain is saying its the same size.

  6. #6 Sunil Bajpai
    July 16, 2005

    I confess going back to the picture for a brief relook midway into the post. My answer: same size.

    There is something in what you report, Dave.

    Thanks and regards,


  7. […] ing Dave Munger @ 5:27 am We’ve posted on boundary extension before, here, here, and here, but we’ve never written about boundary extension and kids. Boundary extension is when […]

New comments have been disabled.