Learning is not simply the ”filling-in” of missing knowledge or skill. For any given element of knowledge or skill, a learner will be objectively competent or incompetent but may be subjectively unaware of their level of competence. When a learner is unconsciously incompetent, they believe they possess knowledge or skill when in fact they do not.
Area9’s data shows that learners are typically 15-40% unconsciously incompetent in any given subject. We hypothesize that this unconscious incompetence is the source of many errors in practice – that damage is most likely to occur when people think they know what to do, when in fact they don’t.
In addition to measuring a student’s “objective” performance on a question, we measure his/her subjective feeling of knowing. How well did they think they knew the answer to the question?
We know that it can be hard to judge the difference between different confidence levels. And we know that some students may lie. But nonetheless this information is useful, and we can leverage is to improve the precision of the system in a number of ways.
Our adaptive engine therefore works to build not only competence but confidence, increasing the likelihood that learning will transfer into behavior change.