The Forgetting curve - Implications for assessment of vocational competency: Credit Transfer, Program duration and the definition of "Competency"

Hermann Ebbinghaus (see https://en.wikipedia.org/wiki/Hermann_Ebbinghaus) hypothesized the "Forgetting curve" in the 1890s. The forgetting curve hypothesizes the varying rate in the decline of memory over time. This curve (above) shows how information is lost over time when there is no attempt to retain it.A related concept is the strength of memory that refers to the durability that memory traces in the brain. The stronger the memory, the longer period of time that a person is able to recall it (https://en.wikipedia.org/wiki/Forgetting_curve).

"So what?" you say. The current significance of this finding has recently been championed by eminent scholars such as Robert Bjork, distinguished Professor of Psychology at the University of California, Los Angeles. His research focuses on human learning and memory and on the implications of the science of learning for instruction and training.

"So what?" you (still) say. Well the upshot of this scholarly research can be summarised as follows:

1. We forget stuff we've learned if we don't use it
2. When we learn things we learn them better if we "practice" them - This takes time (including time to "Forget")
3. Performance (in particular evidence of performance) is not evidence of learning. And, (two of my own)
4. The rate of retention declines in indirect proportion to the presence of competing demands on memory
5. We don't know what we've forgotten and we don't know that we've forgotten it

Still saying "So What"? Well, in the information-age relevant vocational skills are increasingly reliant on information (not strength or physical technique). This isn't just the ability to recall but the efficient ability to identify what information needs to be sourced and the distinct ability to source that information "just in time".

This impacts Australia's VET system in three important ways: (1) Approach to Credit Transfer, (2) Learning Program duration and (3) "Competence" as it is judged in the field. I discuss these below.

(1) Approach to Credit Transfer.

We all know Units of Competence are "mapped"; Thanks to TGA, older units can be readily compared to new units to identify differences. Sometimes a superseded unit is deemed "equivalent" to a current unit. In this case, and subject to adequate evidence of continuing practice and professional development, the holder of a superseded unit can be (and indeed is required to be) deemed competent for the equivalent, current unit.

What's wrong with this picture? Well (1) above for starters. We forget stuff, particularly if we don't use it. While two units might be equivalent to each other this says nothing about an individual. For that we have to investigate the features of the evidence for their continuing practice - Does it provide quality evidence of currency and sufficiency? Probably not. We get certified, we get a job, we use parts of our learning in the job (i.e. we practice it). We don't need to recall other parts of our training for the job, so we forget those bits.

The phrase "I've forgotten more than you know" is an epithet used to denote expertise. Today however, it can mean an experienced VET practitioner has forgotten important information through misuse.

If VET certification through CT is to be effective it needs to commit assessors to identifying knowledge gaps that arise through "forgetting". Evidence of PD and "staying up to date with advances" is not enough. And yet right now, ASQA is directing RTOs to award units through CT even if the candidate has demonstrated they have forgotten the bulk of the knowledge associated with their award of a prior (equivalent) unit. ASQA is committing RTOs (many of which know better) to ignore manifest incompetence.

(2) Learning Program Duration

The "Forgetting curve" has implications for program duration.

With "User choice" rational candidates choose the lowest "cost" RTO. Cost can be rendered in terms of time and money and effort. Understandably, and to attract students, enterprising RTOs offer the shortest course with the least cost and the least effort.

In response, ASQA has mandated minimum durations for learning programs - essentially, and quite rightly, to dissuade dodgy operators. At least for TAE, BSB, CUA, FSK and FNS training packages that I am engaged with, this has lead to a spate of what I think of as "front-loaded", short duration courses: Such courses engage the learner in intensive FtF workshops and subsequently require the learner to complete self-study activities over the remaining duration of the program. They aren't all bad - The best compel a learner through a self-reflective journey that genuinely delivers learning. Others are just a way to get around the system.

In my experience so far, such programs uniformly disregard the Forgetting curve. Learners need time to forget before they can meaningfully engage in retrieval practice (through low-stakes testing). This relates to point (2) above: When we learn things we learn them better if we "practice" them - This takes time (including time to "Forget"). If a program awards certification too soon, or it fails to compel retrieval practice then it can not validly justify learning, let alone competence. Candidates may have produced an impressive portfolio of evidence, but without "forgetting time" it is not, and can not be, evidence of learning.

(3) "Competence" as it is judged in the field

The third area of impact for the Forgetting curve is our very notion of Competence as it is judged by assessors every day. The assessment of "Competence", defined roughly as the ability to apply skills and knowledge to perform consistently to the standard required in the workplace including the ability to transfer this ability to other contexts, doesn't explicitly demand persistence. It doesn't actually demand learning.

This would just be a bit of semantic revery if it didn't have so much impact. The issue is this - How competent is someone if you observe their performance today but have no confidence in their ability to perform an equivalent task tomorrow? In other words, are we as assessors, being asked to judge a persons future competence?

This is a "can of worms". Of course we can't judge future competence; anything could happen. And yet, if a certificate is to have anything more than historical significance it is saying something about the likely competence of the individual at some point in the future.

As an aside, these days my beloved Mother passes her time in an aged-care facility. She doesn't have Alzheimers but the years are taking their toll on her cognition. When presented with a novel piece of information she understands it. An hour later, unless it's important to her, it has gone for good. We don't know how long someone will retain information, but we can make educated guesses.

Two months ago I exited a CBD Melbourne campus (where I was teaching) on a lunch break only to be confronted by an aggressive vagrant who was menacing my students and demanding money with the threat of violence. I was perplexed; I'm a teacher and an appalling but enthusiastic violinist - I have no self (or other) defence skills. Still, I adopted my best Bruce Carridine pose (recalled from the Kung Fu television series in the early 1970s), stared him down and saw him off the property. If he had seen through my bluff I have no idea what might have happened next.

I mention the above because it is a signal lesson in the "appearance" of competence but does not provide any information regarding persistence. Suppose there were three of us confronting the aggressor. One had no experience, but had just completed a one hour session in how to adopt the Kung Fu defensive position. The second is me, no training but a long memory of watching too many Kung Fu shows in the seventies. Third is an actual Kung Fu master who had years of disciplined training and just happened to be walking by.

To an assessor, without reference to knowledge of the three individuals, the performance is identical. And so the assessment must be identical. With knowledge the assessor can conclude the Kung Fu master will repeat the behaviour (because he is an expert), I will also repeat the behaviour (having demonstrated recall of a 40 year old memory). But the assessor can't ascribe persistence to the freshly trained learner - Who may or may not repeat the behaviour depending on their recall.

To judge the value of performance as evidence of competency Assessors must make an informed judgement of persistence. Persistence can not be judged without judging recall and therefore without verifying sufficient "forgetting time".

Concluding remarks

If Australian VET is to be effective it needs to take account of the "Forgetting curve". The knowledge has been around for a 100 years and most recently has been confirmed by reputable scholars using modern scientific methods. Credit Transfer protocols need to include how much a candidate has forgotten before automatically deeming competency. Learning program duration needs to reflect the need for "Forgetting time" and retesting knowledge prior to deeming competency and, finally, our concept of competency must address the notion of persistence.

Sean Kelly July 2018