Robin Hanson at Overcoming Bias has a post on some new cognitive  learning research which contains (for me) some counterintuitive results:

In Experiment 1, students received an illustrated booklet, PowerPoint presentation, or narrated animation that explained 6 steps in how a cold virus infects the human body. The material included 6 high-interest details mainly about the role of viruses in sex or death (high group) or 6 low-interest details consisting of facts and health tips about viruses (low group). The low group outperformed the high group across all 3 media on a subsequent test of problem-solving transfer (d = .80) but not retention (d = .05). In Experiment 2, students who studied a PowerPoint lesson explaining the steps in how digestion works performed better on a problem-solving transfer test if the lesson contained 7 low-interest details rather than 7 high-interest details (d = .86), but the groups did not differ on retention (d = .26). In both experiments, as the interestingness of details was increased, student understanding decreased (as measured by transfer). Results are consistent with a cognitive theory of multimedia learning, in which highly interesting details sap processing capacity away from deeper cognitive processing of the core material during learning.

If this is borne out by other research it certainly has some far reaching consequences for how we think about classroom teaching in general. We’ve been raised in a paradigm which values the “interesting” teacher who “brings the material alive” with unexpected or fascinating details. Perhaps at some point in the future we’ll see a renaissance of dry, Ben Stein styled lecturers.

(Let’s hope not)