Effective Learning: How we Learn and Why it Matters

Does our educational practice always rely on research findings? Apparently not; interestingly, we often rely instead on our intuition to decide what is best: ‘it feels good, hence will be effective’. This strategy, however, may be not the best one for lecturing or learning. How do we know what works and what does not?

The workshop Effective Learning, organised by Dr Lara Alcock and myself and part of the IMA Learning & Teaching series 2018, was delivered twice this year by Lara and by our world-leading team of experts from the Mathematics Education Centre, School of Science, Loughborough University. A first workshop for lecturers and students was held in November 2018 at Loughborough University and was attended by more than 150 participants. Thanks to IMA support, some came from institutions around the country, from Glasgow and Newcastle to London and Birmingham. The event was reprised at the 2019 Loughborough Learning & Teaching conference in July.

The workshop stemmed from our continual effort to inform teaching across the School of Science by up-to-date research on education, as well as from the desire to engage both staff and students in ongoing conversations around teaching and learning. It helped us to recognise the risks of relying on intuition (and standard metrics, to some extent!), and suggested starting instead from what we know about the learning process and integrating effective, research-based strategies from cognitive psychology into our classroom and learning practice. Full of audience-participation activities, the event was divided into four sessions:

Evidence in Education by Professor Matthew Inglis
How we Learn by Dr Iro Xenidou-Dervou
Effective Learning by Dr Nina Attridge & Dr Camilla Gilmore
Lecturing and Learning by Dr Lara Alcock

Below is my summary of each session, with suggestions for further reading. Slides and recordings of the event can be found on Loughborough University’s website.

Sara Lombardo, PhD, FIMA, FHEA
Professor of Mathematics
Associate Dean (T), School of Science
Loughborough University

Evidence in Education (Matthew Inglis)

What do we know about what works in education? More importantly, how do we know?

Which claims should we believe, and which should we be sceptical about? The first session of the workshop set the scene for all the others. It opened by analysing the case study of the e-proof, a resource designed at Loughborough to support students in understanding specific written proofs, and praised both by peers and students who found it useful for their learning. However, a more rigorous test of its efficacy revealed that “students who studied an e-proof did not learn more than students who had simply studied a printed proof and in fact retained their knowledge less well” [1]. The e-proof case is reviewed in a paper in the Notices of the American Mathematical Society [1], where the authors argue that “while the e-proofs made learning feel easier, it also resulted in shallower engagement and therefore poorer learning” [1].

This case study clearly poses a question: How can we evaluate teaching innovation, or teaching more generally? For e-proofs, all standard measures[1] of educational effectiveness currently used in UK Higher Education were collected, and all suggested a positive evaluation of the tool. Yet, measuring student learning revealed that all of these methods were at best misleading. Using further case studies, the session made us aware of the weakness of classroom observations and students’ evaluations: peer observation [2, 3] essentially does not distinguish between bad and good teachers, while students evaluations [4, 5], besides the well known risks of being biased, show weak positive correlations with students’ current achievement and weak negative correlations with subsequent achievement. While no studies have been done to date to assess the validity of other tools, such as lecturer portfolios, there seems to be no reasons to suppose that these perform better.

The message of the session was ultimately that there is no valid way of assessing teaching quality, beyond assessing learning gains. But assessing learning gains is extremely difficult, especially in the context of the Higher Education, where one cannot rely on control groups or standard examinations. So, should we despair? No, we need only adopt a different approach. When it comes to teaching innovations, we should evaluate proposed changes against what we know about the cognitive mechanisms that underlie learning. Fortunately, there are now good introductory texts to the science of learning [6–9], and these should be a reference for lecturers and students alike.

[1] Qualitative and quantitative student feedback, classroom observations, expert peer review.

How we Learn (Iro Xenidou-Dervou)

How do humans process information? What are their strengths and limitations? The second session was a crash course in cognitive psychology and provided the background necessary to think realistically about how learning works. It introduced a simple model of the mind, distinguishing between limited-capacity working memory and long-term memory, where the latter holds knowledge indefinitely. We learned that learning involves storing information in long-term memory, and that although we have no direct control over what it is stored, we can increase our chances of retaining a concept if we focus attention deeply on it. We came to appreciate how this works via a recall exercise based on [10] which makes use of simple words; we were told that the mechanism would be the same also for more complex concepts.

Once we have stored information, we need to be able to retrieve it. We learned that forgetting is often the result of a difficulty in retrieving memories and that cues are therefore important in the process of remembering [11]. Cues are like anchors, they help us retrieve information but they need to be specific to be effective: missing or ambiguous cues can make memories inaccessible.

Given the capacity limits of working memory, how can we increase what we can learn and remember? We learned that existing knowledge can support the leaning process in two ways [12]: it can help us think about new information in a meaningful way, and links to it can help us to remember that information. If information is meaningful, we can for example group it, or chunk it; our ability to chunk information relies on knowledge. Similarly, new information is more likely to be remembered if it is related to what is already in the memory; a rich network of associations provides more cues for retrieval.

Effective Learning (Nina Attridge & Camilla Gilmore)

Equipped with the basics of how we process information, we were ready to address questions about effective learning. What do we typically believe about how to learn effectively? Which of these beliefs are borne out by research, and which are not? The third session highlighted common erroneous beliefs and biases that influence study decisions, and reviewed evidence suggesting what we should believe and do instead. The audience was introduced to the concept of metacognition, literally thinking about thinking; metacognition is the ability to reflect and critically analyse how we think. It can, however, be very inaccurate, as the e-proof example showed: strategies that feel good for learning are often ineffective, whereas effective ones often feel difficult and fruitless. I have always tried to convey this message to my students, but have done so based on personal experience – now that I can point them to the research in cognitive psychology, perhaps the message will be more effective! A quiz run during the session reinforced the misleading nature of metacognition.

We then learned about evidence-based strategies for improving learning by improving storage of information and retrieval of information. Storage can be enhanced self-explanation and interleaving; retrieval of stored information improves using retrieval practice and spacing. Self-explanation consists in explaining to oneself the information presented; it forces the learner to think about the new information, to find links and relate new concepts to pre-existing knowledge. It thus promotes deep learning rather than memorisation [14]. By creating connections with existing knowledge, it also offers effective cues for retrieval. The idea was investigated by in a research study [14] and the related online resource can be found on the Mathematics Education Centre website. Interleaving consists in studying a new concept for long enough to understand it, then switching to a new topic, concept or idea. This strategy improves learning by helping us to identify links between different topics and distinguish between them, and it can easily be implemented by both students and lecturers [13].

When it comes to retrieve information, a common misconception is thinking that re-reading notes or re-watching lectures is an effective way to revise, while self-testing is good for identifying how much we have learned. As a matter of fact, re-studying materials is very passive and can give a false sense of security, while testing is a good way to learn in itself. Answering questions on a topic or generating test questions on material we have studied proves more effective than re-reading notes [15]. That retrieval can also be enhanced by spacing – that information is more easily learnt when it is repeated multiple times with time between the repetitions – came perhaps less as a surprise. After all, it is well known that repetita iuvant! Yet, students seem to be unaware of this, especially during revision time. The principle that content is best learnt all at once with no spacing (massing) is again a metacognitive misconception: massing feels more effective but in reality, is detrimental to learning. Spacing works because it re-shapes our forgetting curve, which empirically describes the decline of memory retention in time, and thus alters the time we retain information. The take-home message of this session was definitely that learning must be active to be effective, and if it feels difficult at times that is not necessarily a bad thing! The session included many tips for students and suggestions to lecturers on how to make learning and teaching more effective; more about these and other strategies can be found at www.learningscientists.org.

Lecturing and Learning (Lara Alcock)

In the last session, the audience was led to discuss and reflect on what the information means for lecturers and students. In particular, we explored how can we structure lectures and independent study to maximise effective learning, and who is responsible for that organisation. The discussion was facilitated by Lara Alcock, National Teaching Fellow. It reinforced a sense of a learning community and partnership between staff and students, and gave the impression that people would have gladly lingered well beyond the closing time to continue discussions. I felt that everyone left with a feeling of empowerment, ready to take learning and teaching to a more sophisticated and effective level.

If you would like to host a similar workshop at your institution, please get in touch with Lara Alcock, l.j.alcock@lboro.ac.uk.

Acknowledgements. I had many interesting and thought provoking discussions with Lara Alcock while organising the first workshop and afterwards. I would like to thank her also for reading the final version of this text.

References

Lara Alcock, Mark Hodds, Somali Roy, and Matthew Inglis. (2015) Investigating and Improving Undergraduate Proof Comprehension. Notices of the AMS Volume 62, Number 7

Michael Strong, John Gargani, and Özge Hacifazlioglu. (2011) Do We Know a Successful Teacher When We See One? Experiments in the Identification of Effective Teachers, Journal of Teacher Education, 62(4) 367?382

Brian Gill, Megan Shoji, Thomas Coen, and Kate Place. (2016) The content, predictive power, and potential bias in five widely used teacher observation instruments, Mathematica Policy Research.

Scott E. Carrell and James E. West. (2008) Does Professor Quality Matter? Evidence from Random Assignment of Students to Professors. NBER Working Paper No. 14081

Michela Braga, Marco Paccagnella, and Michele Pellizzari, (2014) Evaluating students? evaluations of professors, Economics of Education Review, 41, 71—88

Adam L. Putnam, Victor W. Sungkhasettee, and Henry L. Roediger, III. (2016) Optimizing Learning in College: Tips From Cognitive Psychology, Perspectives on Psychological Science, 11(5) 652–660

Megan Sumeracki, Oliver Caviglioli, and Yana Weinstein. (2018) Understanding How We Learn: A Visual Guide, Routledge.

The Science of Learning, https://deansforimpact.org

Daniel T. Willingham. (2009) Why Don’t Students Like School?: A Cognitive Scientist Answers Questions About How the Mind Works and What It Means for the Classroom, John Wiley & Sons.

Donald Morris, Johan D. Bransford and Jeffrey J. Franks. (1977) Levels of processing versus transfer appropriate processing, Journal of Verbal Learning and Verbal Behavior, 16(5), 519–533

Endel Tulving and Zena Pearlstone. (1966) Availability versus accessibility of information in memory for words, Journal of Verbal Learning & Verbal Behavior, 5(4), 381–391

James P. Van Overschelde and Alice F. Healy. (2001). Learning of nondomain facts in high- and low-knowledge domains, Journal of Experimental Psychology: Learning, Memory, and Cognition, 27(5), 1160–1171.

Doug Rohrer and Kelli Taylor. (2007) The shuffling of mathematics problems improves learning, Instructional Science, 35(6), 481–498

Mark Hodds, Lara Alcock, and Matthew Inglis. (2014) Self-Explanation Training Improves Proof Comprehension, Journal for Research in Mathematics Education, 45(1), 62–101

Yana Weinstein, Kathleen B. McDermott, and Henry L. Roediger III. (2010) A comparison of study strategies for passages: Rereading, answering questions, and generating questions, Journal of Experimental Psychology: Applied, 16(3), 308–316