Talking about my generation…effect

In October’s TJ, Justin Collinge writes a thought-provoking article called “Are you m*ssing a tr*ck?”. His premise is that if you leave information out of a visual, or deliberately reverse the letters in a word so that the learner has to work out what the correct message is, they are more likely to engage with it and remember it. The scientific basis for this is the “generation effect” which I remember fondly from my days of being a psychology undergraduate. I like his linking of recommendations to scientific evidence, but I think he has drawn misleading conclusions along the way.

The generation effect and the studies that he is citing are derived from experiments where people are asked to recall a random list of words. The studies usually conclude that there are two broad ways to increase the likelihood that information held in short term memory is encoded into long term memory:

1. Raise the level of attention that you pay to information as it goes in
2. Generate associations and meaning around the words (this is why the phenomenon is called the “generation effect”)

The challenge with the first of these is that there is a countervailing phenomenon “cognitive load”. The more you mix up letters, the more you increase cognitive load and hence the benefit will only be felt in the short term. The longer the course/ module the more that learners will get bored and frustrated by devices like this.

The scientific experiments are based on people’s ability to remember short bursts of words. Training courses and e-learning modules are longer events so you need to be careful about how you extrapolate what the generation effect says. My unscientifically tested belief, backed up by years of facilitating and instructional design experience, is that the more that you provide a variety of activities aimed at generating associations and meanings, the greater the impact of the learning intervention. By all means follow Justin’s advice…but in moderation is best.