Debunking myths and misconceptions about Learning

Debunking myths and misconceptions about Learning (with a capital L)

Farhad Manjoo writes in his book ‘True Enough: Learning to Live in a Post-Fact Society the real danger of living in the age of Photoshop isn’t the proliferation of fake photos  rather, it’s that true photos will be ignored as phonies.

This madness seems to be related to the fact we live in the so-called “post-truth” or “post-fact” era, in which people, in the name of democracy and freedom of speech, confuse opinion with evidence.

In his book, The Death of Expertise: The Campaign Against Established Knowledge and Why it Matters, Tom Nichols (2017) wrote: Democracy… denotes a system of government, not an actual state of equality. It means that we enjoy equal rights versus the government, and in relation to each other. Having equal rights does not mean having equal talents, equal abilities, or equal knowledge. It assuredly does not mean that everyone’s opinion about anything is as good as anyone else’s.

“Everyone is entitled to his own opinion, however, not to his own facts.” – Daniel Patrick Moynihan)

“No amount of belief makes something fact.” – James Randii

“We can, of course, be deceived in many ways. We can be deceived by believing what is untrue, however, we certainly are also deceived by not believing what is true.” – Soren Kierkegaard)

As part of the never-ending quest to improve all facets of learning, I came across an interesting  book ‘Evidence-Informed Learning Design: Creating Training to Improve Performance‘ by Mirjam Neelen and Paul A. Kirschner. Below is a summary of some of the more pertinent points raised.

The reason for the book is to debunk myths and misconceptions, learn from actual academic research and practice, and then focus on creating learning that is Effective, Efficient, and Enjoyable.

Effective:  learn what is needed to be learned in the time allotted to the learning experience.
Efficient: what is learned is learned in less time and/or with less (mental) effort
Enjoyable: the learner experiences success/a feeling of accomplishment at each stage of the learning process.

Psychology behind the prominent Learning Models has an influence on the development of learning materials:
Behaviourism: Individual demonstrates correct response or behaviour, consequence of correct or incorrect performance, a response to the performance , however not to the reasons behind that performance.
Cognitive science: Interdisciplinary study – sociology, linguistics, AI, philosophy, neuro-science, psychology, anthropology – about how people use knowledge in every day settings – perception, memory, attention, reasoning, emotion.
Constructivism: We ourselves impose meaning on the world. There is no such thing as an “objective truth”. Knowledge is fundamentally situated in the specific context that it’s learned. There are many meanings or perspectives for any event or concept and this is determined by our previous experiences.

Adaptive learning technologies (AI) – often used in e-learning environments are supplementing learning: degree of support and personalisation to learners, just-in-time help, personalised problem selection, adaptive to the extent that they take into account that learners differ in their needs, i.e. System is intelligent to the extent that it adapts to the differences and similarities between learners , continuous assessment of learners.
For complex problems, system needs to recognise and accept different solution paths.

Learning experiences: the idea that we can’t design learning per se; we can only design experiences that facilitate and support learning.

Performance problem is a discrepancy between what someone is expected to do and what they actually do.
Learning problem is about a knowledge and skills gap that needs to be filled in order to solve a performance problem. Many performance problems are not linked to a lack of knowledge or a lack of skills, but are rather linked to challenges within working environments, management styles, motivation, reward structures, etc.

The goal of training is learning (i.e. Achieving a sustainable change in behaviour and cognition) so that individuals achieve the competencies they need to perform on the job. Training is both about learning and performance.

Common problems in designing learning experiences:
Atomistic design – when complex problems are tackled by reducing all the tasks involved to simpler or smaller components. The challenge of application of new skills to effective problem solving in the workplace, is often not overcome.

Fragmentation – if people practise various skills in isolation, it can’t be expected that they’ll be able to integrate and coordinate the separately practised elements (knowledge and skills) in their actual job. The whole is mostly more than the sum of the parts. A lack of learning transfer generally follows, i.e. failure to apply what is learnt in various contexts or situations in their job.

Holistic Learning Design:
1. Start with the Performance Problem.
You need a picture of what knowledge, skills and attitudes are needed to perform the task.
2. Determine if Learning can help. (Skills development analysis)
What other issues could be limiting the performance of the individual ?
3. Clarify what success looks like. How is success/achievement measured ?
The famous saying Whatever can’t be measured cannot be managed applies in the learning environment as much as in the working environment.
4. Determine Learners’ needs. Knowledge and skills needed. Prior knowledge and skills. Gap between them and the goal. Identify learning challenges.

All four steps should focus on authentic learning tasks – as they exist in real life:
Physical context  (environment)
Social context (individual or team task)
Authentic assessment result  (deliverables that the learner needs to produce)
Criteria and standards (how the learner’s performance is measured on the job, so that we can mirror that in the learning environment).

Myths that business has about learning:
1. Training alone produces improvements in on-the-job performance
2. Information presentation is sufficient as training design.
3. Training and instructional design require no skill or competencies.
Learners know how to learn
4. Managers think learning and development is a low-priority part of their role.

5. Learning styles (visual or audio) – selecting instructional methods in line with learners’ preferences was uncorrelated or even negatively correlated with learning outcomes. Instructional method need to be selected in line with performance outcome.
It’s detrimental to simplify people to dichotomous categories and harmful to assume that what people tell us about how they learn best is true.

Instruction kills learning when instructional methods match a preferred but unproductive learning style.
It’s the instructional method that matters, depending on what needs to be learned, not the preference of the learner.

It is critical to differentiate between content that employees “need to know” versus resources that they “need to access”.
Difference between learning solutions and performance support.

Supportive information – information that helps learners perform non-routine aspects of complex tasks, such as problem-solving
Just-in-time support – or procedural information – which is information given at the moment that it will help learners perform routine aspects of a task – performance support in the workplace.

Tips for training design
– Text is more difficult to read if it is justified.
– Use figures sparingly as illustrations (interesting /funny) are seductive details that attract attention and divert it away from what is important and clutter our working memories.
– Focus on deeper conceptual understanding – factual and procedural knowledge – and know how to apply it and adapt it to new situations.
– Focus on learning and instruction.
– Create learning environments.
– Build on prior knowledge. Ascertain prior knowledge and teach accordingly. Learn best if previous knowledge is acknowledged and built on.
– Include opportunities for reflection.
– Focus on learning transfer (performance), not just on comprehension. Understand and remember  – also being able to retrieve knowledge.
– Successful training is not a one-time event – but an interactive process that considers the elements leading up to the training as well as important factors after the training event at the time of application of new skills / knowledge.
– Feedback on performance – especially specific feedback is imperative.

The finding that rewarding leads to dopamine production which in turn elicits positive feelings could have a future and help guide us in giving feedback. It is incorrect however to make the assumption that making learning “fun” leads to dopamine production and hence motivation and learning.

Learners with low prior knowledge learn more from step-by-step detailed worked-out examples while learners with higher prior knowledge learn better from solving problems.

Rote learning or inflexible learning (like multiplication tables) – is a necessary starting point for acquiring deep knowledge. It’s a prerequisite, a stepping stone.

Flexible knowledge is the ultimate desirable goal. To apply old knowledge to new situations, one must both recognise the truth in the concepts at hand and then successfully map the new problem to the familiar problem.

Importance of domain specific knowledge.
Problem-solving ability is heavily dependent on domain-specific knowledge held in the LTM (long-term memory).

Analogical thinking – transfer of knowledge from one situation to another by a process of mapping.
Problem-solving always starts with domain-specific knowledge.
Creativity is needed at a computational level, algorithmic level and implementation level.

It’s the nature and the extent of the knowledge base that determines the level of creativity.
Creativity and Innovation are an important means, not an end.
Creativity and innovation are essential for humankind to develop and move forward. We always face new problems.

Creativity equals coming up with something novel that has value, and in that word ‘value’ lies the crux: to come up with something new and of value you need knowledge.
If we want learners to be ready for the future, we need to shift the focus from domain-independent skills to domain-specific knowledge.

You need to learn the main ‘complex skills’ in an integrated manner, i.e. paying attention to the relationship between pieces. If you don’t, then you won’t be able to apply it in real life. So, when we’re dealing with complex skills, we’re dealing with complex learning.

Complex means an ambiguous query with many variables, lots of possible databases, and fuzzy features where they all interact with each other.

A whole task approach drives complex learning – able to coordinate the knowledge, skills and attitudes in the new situation.

Transfer of learning is all about flexibility and adaptive expertise – coming up with new solutions to new problems, overcome challenges.

People only start to understand basic concepts by application, by using the concepts.

Knowledge will only be retained when it’s developed in the process of actual task application.

People learn more from examples than the more abstract and general principles. After studying the examples, they suddenly understand the abstract model.

Productivity tools like spreadsheet programs, search engines etc – can be used efficiently and effectively as mind-tools to support learning.

The results showed that learning is more effective when learners take notes with pen and paper.

Instructional method – the method matters, depending on what needs to be learned and not the preference of the learner, which is usually based on what the learner ‘like more’ rather than ‘which has the better results’.
Interactivity on Multimedia – can distract from learning. Butterfly Defect = fluttering like a butterfly from one piece of seemingly interesting information to another without actually learning .

Effective feedback has been found to be one of the most powerful educational interventions to improve learning – positively affects learning outcomes and motivation to learn and can help build accurate schema (meaning a mental concept that organises categories of information and the relationship between them.)
The purpose of feedback is to reduce the gap between the learner’s current understanding/performance and a desired goal.

Feedback provided on 4 levels:
1. Task level – how well does the learner understand/perform the tasks at hand
2. Process level – How well does the learner follow the required process to achieve the goal
3. Self—regulations level – How well does the learner self-direct, monitor and regulate their actions
4. Self level – What can be said around personal evaluation and affect

Only telling learners that they’re doing something wrong can be detrimental if they don’t understand why they’re doing it wrong and what to change to do it better.

Feedback that hurts learning:
General praise and reward hinder intrinsic motivation and aren’t valuable for learning – praise needs to be specific.
Threats or discouragement can cause anxiety because they threaten self-esteem, which is not beneficial for learning.

Feedback early in the process provides a corrective concept-improving function.
Later in learning, feedback performs more of a retrieval-strengthening function.
Corrective feedback – single-loop feedback – learner provides and incorrect answer, then giving the right answer
Directive feedback – double-loop feedback – that gives direction on how things can be better or better done.
Epistemic feedback – triple-loop feedback – that helps learners learn/gather knowledge. Stimulates the learner to think about the ‘why’ in relation to carrying out a task. You don’t tell learners how they can do better; you give a hint in order to help them think about it and figure out how they can do it differently or better.
Epistemic feedback works best followed by directive feedback.

Factors to consider when giving feedback:
– the feedback relates to the critical dimensions of the goal
– specific gap between current knowledge and skill and targeted knowledge and skill by specifying how they meet the target
– the more one makes use of feedback, the more effective it is
– making feedback an explicit part of the learning process so that learners go through the feedback and do something with it, leads to significant improvement in learning
– feedbacks needs to be given frequently
– the best time to give feedback is when the learner can actually use the feedback information at that point in time.
– feedback is more effective when it provides information on how to improve the answer or solution
– this means you might need to structure your feedback better as the goal is to guide the learner in the right direction
– feedback is not a product but is a part of a process in which the outcome (output) of an operation (process) is returned (feedback) to the input.
-Feedback is meant to improve and not to judge.

Effective learning strategies:
1. Spaced (distributed ) practice  – the idea is that the ‘pause’ between two practice sessions strengthens the memory trace, as you must retrieve it in the next study session.
2. Practice tests (retrieval practice) – recalling the learning content – better transfer
3. Interleaving (variability of practice) – let practice of one topic overlap with the practicing of other topics
4. Questioning –  challenge a learner to explain why something that’s learned is actually the case. Facilitates the integration of new information into existing schemas in our memory.
5. Explain to self – let learners explain a process or procedure to themselves.
‘ What does what you have just read have to do with what you already know.’

Knowing how to do something is not enough to understand it. Learners also need to know why something is being done. And this understanding is necessary for all transfer, both near (to seemingly similar tasks and problems) and far (learning applied in real-life situations that are different from the learning context).

Spaced learning:
Repeatedly retrieving knowledge from long-term memory strengthens memory traces so that we forget less and learn more effectively. This is called retrieval strength, in contrast to storage strength.
Tackling learning in various short sessions works better than learning the same thing in one long session.

This effect was found for all types of retrieval practice; cued-recall (eg. Instructor asks learners questions to encourage them to recall definitions or concepts or this is done through technology), free-recall (learners create their own key words or questions to help them remember), recognition tasks, multiple choice questions etc .
Combining various test formats works best.

Testing as a learning strategy:
– Production – coming up with answers yourself works better than recognising answers
– Multiple practice rounds
– A mix of formats – variety in practice exercises
– Spacing practice over time /days
– Retrieval practice during and after learning sessions
– Start with retrieving facts and concepts before moving on to application

Recommendations to effectively use dual coding:
1. Slow down – take your time presenting words and images and divide them into small parts.
2. When explaining a diagram – do it verbally and not through text in or around imagery. Use signalling as well (animations or highlighting)
Don’t explain it in writing on the slide.
3. Offer images and text at the same time – not when you explain something verbally.
Temporal contiguity – the learner doesn’t have to remember the one part while processing the other.
4. Prevent working memory from becoming overloaded  – don’t offer spoken text with the same text in writing – like the Power Point example.
5. Only use ‘useful’ information – skip image or sounds used for the sake of engagement or fun. These are only distracting and increase cognitive load.

We are also dealing with the paradoxical situation that strong self-directed and self-regulated individuals already are highly knowledgeable, high-performing and high-capacity learners, however, those who probably most need to continue to learn are lower performers for whom it’s generally not so easy to direct and regulate their own learning.

Dunning-Kruger effect (unskilled and unaware of it): How difficulties in recognising one’s own incompetence lead to inflated self-assessments. Incompetent people, those with little to no knowledge of the topic, simply aren’t capable of understanding that their way of reasoning or the conclusions they draw are incorrect. People who are incompetent in an area don’t know what they don’t know and therefore overestimate their own competence.

As learners we are not particularly good at knowing our knowledge blind spots. The more knowledge we have, the easier self-regulation becomes, and the better we become at self-regulation, the more easily we increase our knowledge.

Under-appreciating the role of practice and effort in learning, leads us to underestimate our own capacity to learn. Effective learning can be fun, it can be rewarding, and it can save time – however, it is seldom easy.

The idea is to move from high levels of scaffolding to true self-regulation. Find out what people already do and what works and use that data to figure out what scaffolds or guidance to offer.

There is always a friction between short-term learning goals, focused on solving current problems, and the long-term ones, aimed at developing the required and desired skills to operate effectively and autonomously throughout one’s career.

In summary
Holistic learning experience design:
1. Starts with some kind of problem to be solved.
2. Figure out if it is a performance problem.
3. Need to clarify what success looks like
4. We need to understand the learners – what their challenges/needs are and how to support them best.
5. Complex skills – need to be acquired in an integrated manner , or they will not be able to be applied in real life.
7. Getting better at a ‘whole task’ approach to learning – authentic work tasks.
8. Fragmentation is ineffective.