Real-world examples and transfer of learning

Jennifer A. Kaminski, Vladimir M. Sloutsky and Andrew F. Heckler wrote an article that was published in Science last week. The article is called: "Learning theory: The advantage of abstract examples in learning math". A main issue discussed in the article is whether students who learn mathematics through real-world examples are able to apply this knowledge to other situations or not (the old problem about transfer of knowledge from one context to another). The article claims that their findings:

(...) cast doubt on a long-standing belief in education. The belief in using concrete examples is very deeply ingrained, and hasn't been questioned or tested.

They also discuss the issue of word problems, and they claim that:

[Word] problems could be an incredible instrument for testing what was learned. But they are bad instruments for teaching.

If, like me, you don't have full access to the articles in Science magazine, you could read a nice summary of the article with comments on Nobel Intent.