Cognitive Daily

ResearchBlogging.orgBoundary extension — misremembering the boundaries of a scene as wider than they really are — has been observed in adults as old as 84 and children as young as 6. But for kids much younger than 6, the phenomenon becomes quite difficult to study. How do you ask a 6-month-old whether the picture they’re looking at has the same borders as one they saw a few minutes ago? You can’t ask them to draw the picture for you — they can barely sit up, let alone hold a pencil.

Yet the development of boundary extension is an important aspect of the study of vision. Do babies experience the phenomenon from birth, or is it something that is learned or acquired later in life? Helene Intraub and her colleagues believe that boundary extension is a fundamental aspect of processing scenes. When we view a scene, we don’t just hold a picture of that scene in memory, we build a representation that extends beyond the boundaries of what we can see.

This makes sense as we try to navigate around the world: if we couldn’t picture what was just coming up around the next corner as we walked, then walking would become a hesitant and halting business. And since only a small portion of our visual field (corresponding to the foveal region of the retina) is actually in precise focus at any one time, even building a picture of what is right in front of us can require some active imagination.

If boundary extension is so basic to the visual system, then we might expect babies to experience it too. Paul Quinn and Intraub believe they have found a way to find out if babies might have boundary extension.

They showed two identical pictures of a teddy bear to two groups of infants — aged 3 to 4 months and 6 to 7 months. They did this four times, for fifteen seconds at a time, until the babies were bored with the pictures:


Next they showed the babies two new pictures, one showing ten percent more area, and one cropped in ten percent closer:


An experimenter who couldn’t see the pictures (and didn’t know which was which) viewed the babies through a peephole and timed how long they looked at each picture (using a separate stopwatch for each picture). Here are the results:


Both groups of babies looked at the close-up shot significantly longer than they looked at the wide shot. Quinn and Intraub say this suggests that babies experience boundary extension: if they were bored with the original picture, then if they had extended its boundary, the wide shot would have looked just like the boring original. But the close-up would look different, and would be relatively interesting by comparison.

But there’s another possibility: maybe babies just prefer close-ups to wide shots. So in a second experiment, the babies never saw the two identical mid-distance shots. They were just shown the close-up side-by-side with the wide shot, and looking time was measured as before. Here are those results:


Now there was no significant difference in looking time between the two pictures, suggesting babies had no preference for either shot. In another experiment, Quinn and Intraub verified that adults experienced boundary extension with these pictures.

So boundary extension, at least in this type of scene, seems to be experienced by babies in very much the same way adults experience it. Babies as young as three months may already be building representations of scenes just like older kids and adults do.

Quinn, P.C., Intraub, H. (2007). Perceiving “Outside the Box” Occurs Early in Development: Evidence for Boundary Extension in Three- to Seven-Month-Old Infants. Child Development, 78(1), 324-334. DOI: 10.1111/j.1467-8624.2007.01000.x


  1. #1 NoAstronomer
    June 12, 2008

    I am *definitely* suffering from ‘boundary extension’.

  2. #2 Dan
    June 18, 2008

    Not sure but would a 3-4 month old have a fully developed depth of vision? Would they find the close-up more interesting because it is more in focus than the wide shot?

  3. #3 Nigel
    June 18, 2008

    Dan, two points:

    1. These are photographs, so the baby’s depth of focus is not relevant. I think
      it is fair to assume that the researchers took the trouble to ensure that the
      bear was in focus in all the pictures, and that the photos were placed at a
      suitable distance for a baby to be able to see them clearly.
    2. In any case, experiment 2, as Dave describes it, shows that they were not more
      interested in the close-up. If anything the graph indicates that they preferred
      the wide shot (though I guess, not to a statistically significant degree).
  4. #4 Anat
    June 22, 2008

    what about the simple explanation that the zoomed in picture displayed more of something that the babies were interested in – the teddy’s features, while the zoomed out picture displayed 10% more of boring information – two empty walls? ‘Boundary extension’ does not seem to be the only reasonable explanation for these findings.

New comments have been disabled.