Confusion Aids Learning: Good News for Biased Search Results

October 3, 2011

As Google and other search engines toy with new models to increase user-friendliness, ease-of-use, and predictability of search results, a new study shows that perhaps all those qualities held so highly by search engine designers are not so hot. The article, Eric Mazur’s Keynote at ICER 2011: Observing demos hurts learning, and confusion is a sign of understanding, on Computing Education Blog, explains how just the opposite may be needed for search users to grasp new search archetypes.


Image source:

According to computer education guru, Eric Mazur, how students learn is by being taught straight facts, with no regard for their confusion. In fact, a little confusion is a good thing – it forces the individual to search for clarification, i.e. learn. This wisdom came from a study he conducted measuring the success rate of various models of teaching: demo given before direct instruction, no demo given before direct instruction, student predictions before direct instruction and discussion before direct instruction.

Contradictory to what most would assume, the direct instruction method (with nothing else) scored highest along with the discussion model. The article explains why the results were what they were:

…observing a demo is worse than having no demo at all!  The problem is that you see a demo, and remember it in terms of your misconceptions.  A week later, you think the demo showed you what you already believed…People remember models, not facts, said Mazur.

The fact that prediction and demo models failed to increase learning should be a wake-up call for search engine designers – stop coddling! I wonder if the shift to applications, not old fashioned research is accelerating certain casual and shallow learning.

Another aspect Mazur studies was the effect of confusion on the student’s part. The results showed that students who admitted confusion actually answered more correct answers than those who reported not being confused after direct instruction. Mazur surmised that the results reflect that confusion leads to one’s trying to make sense of the information.

While Bing, Blekko, and Google are tinkering with its search engine effectiveness, these firms may want to look at what the research is saying about how the general masses learn. By creating Web search engines that put training wheels on intellectual exercise, an opportunity to manipulate ideas and conceptual frameworks seems to be available.

Catherine Lamsfuss, October 3, 2011

Sponsored by


Comments are closed.

  • Archives

  • Recent Posts

  • Meta