Five psychological findings every history teacher should know

This text is somewhere between what I planned to say and what I did say during my session at the Historical Association’s annual conference in Bristol yesterday, with a few reflections in italics.

I’m going to start with a couple of stories from the pillar of the local free press: Hackney Today.  As one of only a handful of local authorities still publishing a paper fortnightly, and having recently been instructed to cease doing so by central government, this is not an opportunity to be missed.  Particularly with stories like this one, a few months ago:

I particularly liked this story: not because the council are doing something about dog fouling – although I’m in favour of that too – but because of the approach they’ve taken.

The posters are waterproof and charge during the day, allowing them to glow for ten hours each night.  What I really like about this is that the approach is informed by psychology.  The image below summarises an experiment cited by Daniel Kahneman, measuring how much money academics contributed to a department’s honesty tea kitty against how much milk was drunk.  Over several weeks, without explanation, a series of pictures showing either eyes or flowers was placed above the sign requesting contributions.

The result was clear – the (scary) eyes have it:

So the logic of the council’s approach is impeccable.  Dog fouling, according to Keep Britain Tidy, increases during the winter, when more people take their dogs out under cover of darkness.  While a pair of eyes watching dog walkers may seem somewhat Orwellian, but it;s certainly logical. So, perhaps the result is unsurprising:

This would be the perfect way to introduce the intriguing and useful applications of psychology in the history classroom but for one hitch.  The research into the feeling of disgust and the discomfort this engenders is surely a significant disadvantage: this discussion could well have elicited a slight, hard-to-pin-down feeling of revulsion.  If so, I apologise and hope things will pick up from here.

To me, learning about psychology is, in many ways, like learning about history – it’s a path to seeing the world differently.  In that spirit, I’d like to summarise a few insights from contemporary psychology and suggest their possible implications in the history classroom.

I should add a couple of caveats: this is a history teacher’s view – based on a good deal of reading, but no formal training in psychology.  And it may not lead you to do anything differently – much of this validates things many teachers have been doing for years.  But I hope it will at least offer a slightly new perspective on what you’re doing.

1) Telling stories

Let me start with a story.  It’s about a new history teacher.  He thought that analysis trumped telling a story every time; he didn’t seem himself as much of a story teller, and in any case, he struggled to get his students to listen to him, so setting himself up to do so seemed to promise a fall.  With time, however, he found himself telling brief moral tales about former students – Kara always did this that or the other, and it worked brilliantly for her.  And he noticed that students would focus upon those stories, going quiet and still, in a way they never did to anything else…

I still think analysis is important – although not necessarily more important than stories; I still don’t consider myself much of a story teller.  But reading Daniel Willingham and thinking more critically about my own experience listening to others, I’ve come to realise how important stories can be though.  Willingham describes the human mind as “exquisitely tuned to understand and remember stories,” such that they can be described as “psychologically privileged.”

Willingham offers three reasons why stories are important and why people enjoy them: firstly, they are easy to comprehend – we have had the structure of narratives drummed into us from an early age.  Secondly, they are interesting: stories are consistently rated as more interesting than other formats of information, even if the content is the same.  Thirdly, they are easy to remember – partly because they require the listener or reader to make lots of inferences; partly because their causal structure helps us remember how one part of the story links to the next.

Willingham highlights four elements of a good story:

• causality
• conflict
• complications
• character

In many ways, this is a gift for history teachers.  History offers us some fantastic narratives without our having to create them afresh.  So how can we make room for the story in the lesson, how can we build lessons, units and schemes of work around stories?  There are numerous ways in which history teachers have done this – whether in the development of lessons or schemes of work from books like The Voices of Morebath or the ways in which a telling image – like that of the execution of Charles I – can be used to shape a narrative.

While there’s a good deal of interest in telling stories in historical education, it sometimes feels like they are more tacitly accepted than embraced – I certainly don’t recall any training in telling stories.  So while this is something that has a place in history teaching, it’s perhaps something we could do more with.

An aside: the conference was full of examples of teachers advocating using individuals’ stories and giving great examples of how they had done so to bring history to life.  Richard Kennett and Annie Davis‘s session was a particular highlight with numerous examples of individual lives.  Likewise, one of the best historical sessions I attended, by Robert Bicker, told the story of Shanghai during the First World War through the lens of those living there and evocative photographs of the time.  I was also reminded, in conversation with Rob Peel, that a deep knowledge of a topic is a prerequisite to identifying and using a range of stories in lessons.

2) Memory and testing

This summer, I’m returning to Sweden to do some teacher-training for the third year in a row.  To try to make the experience as interesting and worthwhile as possible, I’ve been learning Swedish each year.  In the process, I’ve come across the fabulous website Memrise.  You can first learn words, by being introduced to them and clicking on, then typing out the correct answer.

The website then retests me on the words: initially, four hours later, then a day, then two days, and so on – assuming I get them correct.  If I make a mistake, it forces me to study them again sooner.  As you can see from the list below, while I’m fine with jag reser, I frequently mix up two senses of ‘to live’ – att bo and att leva – so I’ll be retested on them sooner.

I like this because it helps me remember the language, but I also like it because it fits with the research on the power of repeated testing.  Here’s Ebbinghaus’s hundred-year old ‘forgetting curve.’

The moment we learn a thing we begin to forget it.  But every time we are retested on it, we remember it longer.  A little test at the end of a lesson is a great way both to entrench memory and to set the brain up for better learning in future.

For a start, there’s a sense of liberation here – we can turn to students and explain to them that it’s alright (in one sense) that they are forgetting things.  But we can also find ways to alleviate this issue.   Designing an interleaved curriculum – in which we return to the same topic more than once – helps us plan our way around this.  One thing I’ve tried, in a couple of different ways, has been having students redo card sorts, or face the same test again and again until they master the learning concerned – this has been to master basic chronological knowledge.  I’ve also tried running revision lessons which begin with a test, provide time to revise and then return to the same test at the end of the lesson.  Toby French has suggested the same approach, but with a different test each time – an excellent idea.  There are many ways of running these tests – you can upload your own tests onto Memrise, if you like.

What this requires is a very clear understanding of exactly what the most important things to test for – given how few things we can expect students to remember out of any one lesson.  So a tightly planned curriculum is key.

This issue was subject to some very interesting responses in the session.  Some teachers described using mini-tests each lesson, and the advantages in terms of memory (and as a way to discuss progress with students and parents).  Another mentioned the issues which interleaving had caused students with lower prior attainment – how confusing jumping around could be.  There is, presumably, a happy medium to be found between the demands of memory and those of a coherent unit of work and a feeling of progress.  Another teacher mentioned the issue of choosing exactly what matters most for students to remember, given the essentially infinite size of ‘history’.  The distinction mentioned by Jamie Byrom in his keynote later – between ‘now’ knowledge – what’s needed for the current enquiry – and ‘hereafter’ knowledge – what will be useful in future learning, is perhaps a helpful prism through which to view this question.

3) Stubborn misconceptions

It is a well-known fact that the Illuminati exist.  That they had a role in the death of Michael Jackson, and in 9/11.  Most potently, it is a known fact that the Illuminati have pulled the wool over the eyes of countless dupes of history teachers who deny their very existence.  A straw poll revealed that the vast majority in the room have found themselves facing these truths in the classroom at some time or other.  Besides the bizarre, students arrive to our lessons with all sorts of stubborn misconceptions – that humans coexisted with dinosaurs, or that Martin Luther king freed black Britons from slavery.

Psychology has a lot to tell us about how poor we are at thinking.  We are poor statistical reasoners, we substitute easy questions for harder ones at the drop of the hat, and then there’s confirmation bias.  The fact that we are not only prone to seek evidence that confirms our existing beliefs, but we are also so willing to dismiss countervailing facts, is particularly troubling in this subject.  That telling students the ‘correct’ answer can prove counter-productive: leading them to switch off or assume they were correct already, for example, is equally alarming.

Disappointingly, the literature is weaker on how we might address confirmation bias – although that may be more the fault of humans in general than researchers in particular.  But a few things which might help include:

• Using lines (and approaches) like “I’m glad no one’s fallen into the trap of…” before describing a common misconception – encouraging students both to avoid the trap and to see themselves as wary.
• Limiting the ‘air time’ of students whose views are both firmly held and entirely erroneous, saving a discussion of their beliefs for a quieter time and avoiding legitimising them to the rest of the class through extended consideration.
• Holding out on providing the ‘correct answer’ to force students to think through their reasoning (and its limitations) before we reveal correct conclusions.
• Pausing – in the face of an unexpected misconception – and saving discussion until the next lesson, allowing ourselves time to marshal the most telling evidence available.

One teacher present made two excellent points: firstly, that it was important to listen to understand exactly what students were thinking (Dylan Wiliam’s distinction between ‘evaluative’ listening – to judge – and ‘interpretive’ listening – to understand what students are thinking, came to mind here); secondly, she noted how important consciousness of our own confirmation bias – particularly our willingness to overlook our failures, could be – on this I was reminded of Doug Lemov’s line ‘the incentive is to bury the data’.  

4) Social-psychological interventions

One of my favourite examples of a social-psychological intervention was a simple ‘values affirmation’ exervise.  At the beginning of the school year, students wrote about values which were important to them.  The results were incredible: the effect was to raise black students’ grades, reducing the ‘racial achievement gap’ by 40% and maintaining a discernible effect on black students’ Grade Point Average two years later.  Social-psychological interventions are particularly attractive: they are brief, cheap and effective.  They are perhaps unusual because they:

• enlist students – the message is articulated by students to themselves – increasing its credibility in their eyes
• are stealthy – their messages are not battered around students’ heads
• and are recursive – their impact grows over time.

One study which I’ve adapted to fit the history classroom asked students to write, at the end of each topic in science, about how what they’d studied could apply to their own lives – leading to higher grades for prior poor performers in the subject.  I simply asked students how what we’d learned in history might be useful or relevant to their lives.  I ended up having to add any number of prompts, inviting students to think beyond a test, but I hope this had some effect.

Another study I liked asked university students to read about the fears and uncertainties new undergraduates often experience and then (supposedly) to write a speech addressed to younger students explaining how these anxieties evaporate.  This study showed a significant effect on students three years later.  I asked Year 9 to give advice to the next year’s Year 9 students on making a good start in history, and Year 13 to explain how best to approach coursework.  In both cases, I was seeking to modify or reinforce what I wanted students to do themselves (for example, redrafting coursework, starting early, rather than waiting until the last minute).

I’ve written about these interventions more extensively, and with detailed references, here.

5) Peaks and ends – making the conclusion matter

This graph shows the pain that two different patients experienced during their colonoscopy.  Who had a worse time?

Even though Patient B suffered for twice as long, they remembered the pain as less intense than Patient A – because it had a slightly gentler ending.  In a similar experiment, people with their hand in icy water who had a jet of slightly less cold water added at the end of a period of time remembered the pain they suffered less unpleasantly than those that didn’t – even though their hands were immersed for longer.  We are two selves, Daniel Kahneman argues – the experiencing self and the remembering self.  This is the ‘peak-end’ rule: how we recall an event is based on our emotions at the peak of our experience and at the end.

This has interesting implications for how we conclude our lessons.  At its simplest, it’s worth thinking about the difference which and encouraging final few words after a difficult lesson can modify students’ memory of what happened, mitigating the struggle or confusion they experienced.  More actively, we might wish to consider what emotion and what feeling we wish students to feel at the end of a lesson: a sense of accomplishment? an element of mystery (to be assuaged next lesson)? a single key memory?

The Cambridge Conclusions Project identified a number of facets of great conclusions: setting the lesson’s content in a fresh light, highlighting the key ideas and the conceptual core of what students have learned, placing the lesson in a broader context.  Yet it also noted that many of these possibilities are in tension: how can we avoid trivialising the complexity of all that’s been covered, while also ending on a clear note.  To what extent should lessons return to the original objectives, to what extent should they move on?  How, also can we embed memories and manage students moods.

How to end this…  by suggesting that it’s up to teachers to identify what will work best in their classrooms – based on an understanding of the psychology, as well as their strengths in the subject.

The slides from the talk can be found here.

This talk drew most heavily on:

Daniel Willingham’s Why Don’t Students Like School? (on the importance of stories and aspects of memory)

Daniel Kahneman’s Thinking, Fast and Slow (reviewed here) (on the peak-end rule and confirmation bias)

The work of Yeager and Walton (discussed here) (on social-psychological interventions)

And that of Robert Bjork (for memory and testing)

For the last of these four, and much else beside, much of my knowledge is derived from the writing of David Didau.  If you are interested in almost anything discussed here, I highly recommend his forthcoming book, What if everything you knew about education was wrong? out in June.

For large parts of the way I look to put this to use in history, I am indebted particularly to Michael Fordham and Christine Counsell.