Liftology 8: Timing of reactivation
We know we need reactivation, but then the question becomes “when” to reactivate. The answer is not simple, as multiple factors affect the decision. While there are multiple ways we could think about it, we suggest a useful way to to approach the problem is as a base rate, and then adjustments based upon important factors.
Research on spacing has repeatedly determined that much of what’s presented can atrophy as quickly as a day or two. We also know that the best time to reactivate is right before it’s forgotten. We also know that as the trace of memory becomes stronger, more time between reactivation helps.
That gives us a baseline, based upon memory. We really want reactivation at set intervals. We start with 2 days from the last activation, then 3 from that previous, then 4. From there, we start going every week, until we’ve reached the necessary quantities of reactivations. That latter, then, becomes the issue. How many do we need? Importantly, that changes depending on different factors. We can start with a middle number, and adjust up and down from there. 10 or 12 is probably a good starting point.
The things that adjust the total number of activations are myriad, but we believe there are four that are most important. For one, how frequently are you expected to perform the task after the learning event? If it’s frequent, you don’t need as much practice, as you’ll be getting it. If it’s not, you may need more practice to keep yourself prepared for when it does occur. Similarly, how experienced are the performers? Are they coming in with some prior knowledge, and therefore we’re just refining what they know? If so, we don’t need much practice. On the other hand, if they’re novices to the topic, we probably need more practice.
On the other side, we need to look at factors that the more they’re true, the more practice we need. For one, how complex is it, inherently? If it’s not too complex, we won’t need as much practice as if it’s really inherently complex. Then we’ll need lots of practice! And, how important is it? What are the consequences of getting it wrong? If you’re just going to have some mildly irritated folks, you probably don’t need a lot of practice, because mistakes aren’t costly. On the other hand, if it’s really costly, e.g. people die, you’re going to want a lot of practice to ensure mistakes don’t happen when it counts.
Thus, there’s a relatively fixed pattern of reactivation timings (adjusted for weekends and holidays, typically), but a more contingent number of total quantity of learning interventions fueling transformations (LIFTs). That’s our stance.