In my last post, I wrote about memorisation as a foundational component of learning. What I am going to write about today is a system to more accurately measure memorisation than the one that is currently used.
Currently, a test setter (teacher, institution etc.) determines the content domain that a test is designed to cover, and then writes questions that sample material from the content domain and the determines how much of the content domain has been learned (memorised) by how many of the questions in the sample have been answered correctly. One of the flaws in the system is that, if the test taker misses any of the questions, they are deemed to have missed that part of the content domain the questions were designed to cover. It is an all or nothing proposition that is supposed to accurately reflect the amount of material a person has learned.
An alternative that I would like to propose is based on psychophysical measurement.
Psychophysical measurement is the mapping of physical stimuli (e.g. light) onto a psychological experience (e.g. detecting light). Because biological sensory receptors vary in their sensitivity from minute to minute, a clever way to establish a threshold for detecting the physical stimuli were devised in the late 1800s by a group of very clever scientists. These scientists acknowledged that the strength of a psychological response didn’t directly map on to the actual state of the psychical world. In other words, although no light didn’t elicit a biological response, very weak levels of light didn’t elicit a response either. Increasing the strength of the physical light signal eventually elicits a biological response, however, doing this over and over doesn’t result in the response being elicited at the same level of physical stimulus every time (some variability), and working backward (decreasing the light until it is no longer detected) leads to a different level of sensitivity.
In order to come up with a way to accurately describe what is happening, psychologist’s in the area devised a stepping procedure where the light is increased and decreased in an unpredictable manner, and the value of physical light that the person correctly detects, say 50% of the time, becomes the detection threshold for that person. This doesn’t mean that there is no detection below that level, nor that there is perfect detection above that level, but it is a number used to describe the level at which the person detects light. The same methodology is used for other physical phenomena such as sound, pressure, and heat etc.
Using the same philosophy, we could measure the level at which a person ‘knows’ (has memorised) a body of knowledge. If there were an infinite number of questions, all properly tagged with the level of knowledge (difficulty) required to answer the questions, a smart testing instrument could feed the questions at a person, increasing or decreasing the difficulty level until the person consistently answered, say, 60% of the questions correctly. This difficulty level would then accurately describe the “learning threshold” for that person in that particular content domain, at that particular point in time.
That type of system could measure the ‘learned’ (memorised) material accurately, and would be comparable between teachers and institutions. This type of testing could be a part of everyday education instead of a single point in time examination that returns a static measurement that is often used to define an individual and pigeonhole them.
Just a thought.