A functional analysis of photo-object matching skills of severely retarded adolescents.

The team worked with nonverbal teens who had severe intellectual disability. They wanted to know if trimming the background from photos would help the teens match the photo to the real object.

Each teen got two kinds of photos: regular rectangle shots and cut-out shapes that showed only the item. The staff timed how fast and how well the teens paired each photo with the real thing.

Most teens matched faster and more often when the photo was cut out. The blank background seemed to make the important part pop out.

Only a few teens did equally well with both formats, and none did better with the regular full photo.

Cordova et al. (1993) later showed that many people with moderate or severe ID can pass generalized identity-matching tests. McCann (1981) gives a simple way to make those tests easier—just trim the photo.

Delamater et al. (1986) seems to disagree. They found that teaching names with real objects created stronger generalization than teaching with picture cards. The two studies don’t clash; McCann (1981) only tested matching, while J et al. tested naming and generalization to new items. Matching and naming are different skills.

Roche et al. (1997) also looked puzzling at first. Most adults in their study could choose with objects but failed when pictures were offered. The key difference is age and task: the adults were older and had to show preference, not just match. Teens in McCann (1981) still benefited from clearer photos, so always assess before you drop pictures.

If you run matching programs or use photos for choice boards, take thirty seconds to crop the figure. Free phone apps can erase the background and save the image. Clearer photos can speed up acquisition and reduce errors for learners with severe ID, especially when you are starting a new program or probe.