Introducing the Behaviour Change Intervention Ontology
Techniques to help you Design for Behavior Change
Hi there, I’m Robert and welcome to my behavioral insights page. You will find posts about behavior, psychology and decision making to help grow products, innovate services and improve employee experiences.
This weeks post is about the Behavior Change Intervention Ontology (BCIO) which has been developed by the team over at the Human Behavior Change Project.
Basically we can use the Ontology to design better experiences for projects in design, innovation and change.
I wanted to take a different approach with this subject and explain it using an analogy.
My idea is to then follow this up with the practical examples in a follow up piece.
So, to kick off this story I asked someone who worked on the Ontology to explain it to us…
Let me introduce you to Ella Howes
Ella Howes is a PhD candidate at the University of Leeds. Her research is looking at what we can learn from adaptations made to complex interventions within trials impacted by the COVID-19 pandemic. Before starting her PhD, Ella worked on the Human-Behaviour Change Project and was involved in the development of the Behaviour Change Intervention Ontology and its dissemination.
Ella is also a writer. She writes a weekly newsletter about grief (with the occasional behavioural science twist!). She also co-hosts a podcast called ‘Feeling Griefy’ which you can listen to here.
Over to you Ella 😀
Introduction
I’d like to tell you about something which has the potential to really advance the field of behavioural science. However, it’s one of those things which can initially feel quite complicated, and I don’t want that to be off putting!
So, first I’m going to tell you a story. It has a love theme (because I thought that might make it a bit more exciting). I’ll then explain how this love story is like a situation that needs fixing in the field of behavioural science. Only then will I get on to the thing I want to tell you about that can help address it.
I’m going to follow this structure three times, over three ‘chapters’ to share three key insights. However, before we start the ‘love’ story, I need to give you a bit of background. We’ll call this the prologue.
Prologue
Imagine a dictionary that only includes types of things related to behaviour change interventions. However, instead of organising this dictionary into alphabetical order, it is organised into categories that all relate to a different part of a behaviour change intervention. For example, one category is for things related to who delivers an intervention (its ‘source’). Another is for where an intervention could be located (it’s ‘setting’).
In each of these categories, everything is organised from the general to the more specific. So, in the ‘who delivers the intervention’ category of things, you might have ‘health professional’ at the top of the page, and then a list of more specific health professionals (like doctor, nurse, dentist etc) underneath. It’s arranged in a hierarchy.
Like a dictionary, everything in it is clearly defined. But unlike a dictionary, the relationships between everything are shown. For example, in the ‘who delivers the intervention’ category you could clearly see that a doctor is a health professional. This means you can see how these different things fit together.
This is a very brief introduction to the hero of the love story (or at least its behavioural science equivalent). It is called the Behaviour Change Intervention Ontology, and it’s been developed at University College London as part of the Human Behaviour-Change Project (led by Professor Susan Michie and funded by the Wellcome Trust). I’ll tell you more about it along the way (but if you can’t wait that long, you can read about it here)
Now onto the first chapter of this three-part love story, which begins with a date. A date with me and you (yes you – the reader!).
Chapter 1
Finding each other
For our first date, I ask you to meet me at a café in London called Café Royal at 4pm. That’s the only information I give you, and because you’re not familiar with London, you don’t think to ask for more (rookie error!).
At 3.30pm (just half an hour before we’re meant to be meeting!) you type ‘Café Royal’ into Google maps. Up come 15 options. Which one is it?!
After wasting 10 minutes carefully crafting a text to ask me which Café Royal I’m at (wondering if it’s stupid to only be asking now!?) I tell you it’s the one in South London and send you the postcode.
SW1W 9SJ…. Oh so simple! Now you know exactly where to go!
However, because you’re in North London you’re 45 minutes away and are going to be quite late…
Luckily, I am a hugely understanding person. I send you a sweet message saying ‘oh don’t worry at all!’ (Even though it is a bit annoying to be sat waiting for so long, and it does make me wonder whether this is a bad omen for things to come…).
Despite your sweaty and rushed arrival, we end up having a wonderful time and organise to see each other again the next week!
🧠 Behavioural science equivalent:
In the field of behavioural science, papers get published at a very fast rate. While this might initially sound quite positive, it can actually get a bit messy. The more evidence that exists (of varying quality), the harder it can be to find what you’re looking for. It is also harder to know what has been done before.
This can mean we don’t make the most of what has been published, and its potential to help change behaviour is reduced. A bit like having to go to every single ‘Café Royal’ to find me … it takes longer than it should to find what is relevant for you and your study.
💡How the Behaviour Change Intervention Ontology can help:
In the Behaviour Change Intervention Ontology, every definition for something has a unique ID number attached to it. This is helpful for two reasons. Firstly, it makes it computer readable so Artificial intelligence can help us organise research papers better. Secondly, it can help us describe and report interventions more consistently so they can be found more efficiently.
For example, imagine I am trying to find out what different studies say about the effectiveness of a certain technique to change behaviour. If I have an ID number for this technique, the computer could help me find relevant studies that have tested it out much more efficiently than if I had to find them all myself (and then decide if that technique is the one I am talking about or a different one…). A bit like how the postcode helped you find me at the right Café Royal…
Chapter 2
Speaking the same language
So, it’s now our second date. We’re out for brunch and we’re talking about the tennis (because the french open is coming up). I mention a player I really admire who’s back after a lot of months of injury. You then respond by telling me you once dated them…
I nearly choke on my croissant. Dated!? Does that mean you were together together? Or is it just that you went on one single date? Or was it a few more…
I can’t really ask this in a casual way, so I spend the whole rest of our date feeling a bit uneasy (and if I’m honest, a little insecure).
In reality, you only went on two dates (a casual lunch, then a cinema trip) and didn’t see each other again because there was no spark.
But the confusion has now caused a slight tension at the table. This lack of clarity in the language we have around dating has caused a misunderstanding…
🧠 Behavioural science equivalent:
When writing up research papers and describing an intervention that took place, sometimes the same words are used to describe different things, and different words are used to describe the same thing. Like the ambiguous word ‘dating’, that means we can be unsure of what actually happened. This causes confusion.
For example, if an intervention occurred in a ‘high school’, it’s not immediately obvious for UK audiences which age group that involves? Is that a secondary school or a college?
While this example might not seem very disruptive, if it occurs again and again, the evidence base gets messier and messier. This causes big problems if we try to replicate an intervention or evaluate if it worked as planned. We also can’t always be sure if two studies are talking about the same thing when they say they are…
💡How the Behaviour Change Intervention Ontology can help:
In the Behaviour Change Intervention Ontology, every definition for something, has a unique ID number attached to it. This means that if a potentially ambiguous term is included in a paper to describe what happened in an intervention, we can look up the ID number that is referenced (like you would in a dictionary) and be clear on what is meant.
The Behaviour Change Intervention Ontology provides us with a consistent and shared language from which we can better describe interventions. This helps to reduce ambiguity and increases the clarity of our reporting (a great outcome!).
Chapter 3
Being prepared
Despite the lateness of the first date and the slight misunderstanding on the second, we end up ‘dating’. And by this, I mean we have decided we’re exclusive (but no labels just yet!).
Anyway, we decide to go away together. You suggest a hike in Scotland, and I say yes because it sounds quite romantic.
However, I’ve never gone on a proper hike before (or to Scotland), so I have no idea what to pack. You also suggested we camp along the way (!!?!?!) so I really don’t want to forget anything important.
I turn to my trusty friend Google and type in ‘what to bring on a hiking holiday to Scotland in July’. It comes up with a helpful list of things, organised into categories like ‘clothing’, ‘toiletries’, ‘outdoor items’ etc. I’m glad I checked, because it reminds me that to hike, I’ll probably need more than my flimsy plimsoles… oh and I definitely need insect repellent for the midges!!
🧠 Behavioural science equivalent:
Often, when a behaviour change intervention is written up into a published paper, key bits of information about the intervention are missing. For example, the content that was delivered is not well described. Or there is no mention of the degree to which what was planned was actually delivered (the intervention fidelity). This really undermines our ability to understand what happened and why. For example, did the intervention not work because the content was ineffective, or was it because someone forgot to deliver some key Behaviour Change Techniques? If it is not written up and reported in detail, this is very hard to know.
Missing key details out of papers also impacts our ability to reproduce a behaviour change intervention. Like trying to follow a recipe but only knowing a few of the ingredients. Again, more evidence is wasted, as well as a lot of time and money.
💡How the Behaviour Change Intervention Ontology can help:
The Behaviour Change Intervention Ontology captures all the different parts of an intervention that could be important to know about. This information is captured quite granularly. Like in the dictionary I described in the first part of this series, with categories that go from the general to the more specific.
This granular information can serve as a very detailed ‘prompt’ list for what you might want to include in the design and write up of an intervention. In the same way I used a list to see what to pack for Scotland, I could use parts of the Behaviour Change Intervention Ontology to be more detailed in the planning and write up of my intervention. This can help increase how comprehensive the reporting is. It can also prompt us to think of things we might not have ordinarily considered. For example, if we review the Behaviour Change Technique Ontology during the design of an intervention, it could encourage us to consider alternative BCTs than the ones we’re most familiar with (and that come to mind first!).
Ella’s Conclusion:
Ok, so this wasn’t the most romantic love story. And granted, being prompted to bring hiking shoes on a hike isn’t the best example of being reminded of what to pack (although finding out about midges is helpful!). But hopefully, this three-part story has given you an insight into the value and potential of the Behaviour Change Intervention Ontology for the field of behavioural science.
In summary, the Behaviour Change Intervention Ontology provides us with a clear and consistent vocabulary from which we can better describe behaviour change interventions. It is also computer readable which means AI can help us organise and find evidence more efficiently than we can otherwise.
However, it’s all very well having a glamorous and shiny tool, but if we never use it, it’s potential is limited. In other words, for the Behaviour Change Intervention Ontology to help address all of the problems I’ve highlighted, we need to people to use it! A behaviour change problem in and of itself.
P.s if you’re wondering what happened to us….Despite the midges, we had a lovely hike in Scotland. You even asked if we could make our relationship ‘official’! It was thoroughly exciting for the first few months, but then after that, your inability to ever be on time really got on my nerves. I also found it incredibly annoying that you were always so ambiguous with what you meant. So, we decided to part ways (but still remain friends!)
So how do we use this in practice? (small preview)
To use the BCIO when designing solutions you should have already selected your behaviors and done your behavioral analysis.
Then, you can follow these steps as per the BCIO website:
Identify Behaviour Change Techniques (you can get my free board with explanations and resources on BCT’s here) using the Theory and Techniques tool
Decide on:
a) mode(s) of delivery
b) schedule(s) of delivery
c) intervention source
d) style of delivery
e) tailoring
You can use the BCIO tool to fill out the above details.
Flesh out the details of the intervention content and delivery.
This may seem quite high level, and this is by design, as we wanted to just give you the first introduction to how this behavior change ontology can help you design better experiences.
If you know me by now, you know I am all about being practical, so Ella and myself will bring you some easily understandable examples and step by step instruction in our next Ontology article (coming soon…)
Thanks for reading,
Robert and Ella
Article Resources
· Here is a link to a webinar where we talk about the Behaviour Change Intervention Ontology in more depth.
· To find out more about the development of the Behaviour Change Intervention Ontology, here are the collection of papers published so far.
· You can explore the Behaviour Change Intervention Ontology in more depth here.
· There are some great training resources about the Behaviour Change Intervention Ontology which you can find here.
If you are curious learn how to design for Behavior, have a look at my courses:
https://courses.aimforbehavior.com
If your team wants to work with us on a challenge: https://www.aimforbehavior.com/consulting-in-behavior-science
 | A guest post by
|