Desirable difficulties


Initial harder learning might make for better longterm retrieval.

A lot of the discussions at my university on how to improve learning focus on how to make it easier for students to learn. That never sat quite right with me without me really having a solid basis for that feeling, so today I want to share with you an article by Adi Jaffe, who argues for “desirable difficulties in the classroom” – difficulties that make the learning process harder in the short term, but more successful in the long term.

A couple of such desirable difficulties are given in the article, some of which I want to discuss here:

Being tested on items repeatedly even after successful retrieval attempts helps long-term learning – I discuss this paper in my post titled “testing drives learning” based on The Critical Importance of Retrieval for Learning” by Karpicke and Roediger (2008), so go check it out, and for now let’s just remember that dropping an item from practice after it has been successfully recalled isn’t a good idea.

Having learners generate target material themselves rather than passively consuming it. In the paper that is referred to in the above article, students get paragraphs of text which they have to order before being able to read the whole text. Other methods might be to have students read different parts of a text and then having them reconstruct the whole text from the pieces each of them knows. Intuitively, this makes sense to me, and it is something we have been applying.

Spacing lessons on a topic out rather than massing them together (see for example Dempster (1990). “The spacing effect: A case study in the failure to apply the results of psychological research.”). This one goes against the trend of grouping instruction on specific topics together, both in terms of having “math week”, followed by “biology week”, etc, and in terms of having absolutely coherent curricula inside a specific discipline.

I have to say, I am struggling with this one. I do see the research is pretty unambiguous, but… What about all those nicely designed teaching materials that build knowledge, baby step by baby step, that we put so much effort into? Sometimes letting go of an ideal can be really hard. (On the other hand – it does make me feel a whole lot better about not properly proof-reading what I post on this blog. Desirable difficulties, people!)

Speaking of nicely designed materials, let’s get to the last point I want to discuss:

Making fonts harder to read to improve processing. This one I found really interesting: In their 2010 study “Fortune favors the bold (and the italicized): Effects of disfluency on educational outcomes”, Diemand-Yauman, Oppenheimer and Vaughan find that changing fonts from something really clear to something slightly less clear yields improvements in educational outcomes. And this holds both for laboratory as well as for classroom settings.

Thinking back to my days in school when we were often given texts that had been repeatedly photocopied of old photocopies of type-written documents, this seems intuitive to me. If I spent slightly more time reading the texts, I actually did think a little more about them, too. Deciphering did help me process, and remember. But the days of the crappy photocopies are long gone, and now we are given perfectly font-set documents on crisp white paper. However such an intervention would be really easy to implement.

I feel like I would like to read a little more on this topic before actually suggesting this at my university, but I am looking forward to digging into the literature on the last two points! How about you? Ready to go for desirable difficulties?

P.S.: There is even some research that suggests that learning in an instructional design that doesn’t cater to your preferred learning style might be one of those “desirable difficulties”. But I’ll save that one for a later date :-)

Dempster, F. (1988). The spacing effect: A case study in the failure to apply the results of psychological research. American Psychologist, 43 (8), 627-634 DOI: 10.1037//0003-066X.43.8.627

Diemand-Yauman C, Oppenheimer DM, & Vaughan EB (2011). Fortune favors the bold (and the Italicized): effects of disfluency on educational outcomes. Cognition, 118 (1), 111-5 PMID: 21040910

Karpicke, J., & Roediger, H. (2008). The Critical Importance of Retrieval for Learning Science, 319 (5865), 966-968 DOI: 10.1126/science.1152408


One thought on “Desirable difficulties

Leave a Reply