One view of influencing user behaviour – what I’ve called the ‘errorproofing lens’ – treats a user’s interaction with a system as a set of defined target behaviour routes which the designer wants the user to follow, with deviations from those routes being treated as ‘errors’. Design can help avoid the errors, either by making it easier for users to work without making errors, or by making the errors impossible in the first place (a defensive design approach).

That’s fairly obvious, and it’s a key part of interaction design, usability and human factors practice, much of its influence in the design profession coming from Don Norman’s seminal Design of Everyday Things. It’s often the view on influencing user behaviour found in health & safety-related design, medical device design and manufacturing engineering (as poka-yoke): where, as far as possible, one really doesn’t want errors to occur at all (Shingo’s zero defects). Learning through trial-and-error exploration of the interface might be great for, say, Kai’s Power Tools, but a bad idea for a dialysis machine or the control room of a nuclear power station.

It’s worth noting a (the?) key difference between an errorproofing approach and some other views of influencing user behaviour, such as Persuasive Technology: persuasion implies attitude change leading to the target behaviour, while errorproofing doesn’t care whether or not the user’s attitude changes, as long as the target behaviour is met. Attitude change might be an effect of the errorproofing, but it doesn’t have to be. If I find I can’t start a milling machine until the guard is in place, the target behaviour (I put the guard in place before pressing the switch) is achieved regardless of whether my attitude to safety changes. It might do, though: the act of realising that the guard needs to be in place, and why, may well cause safety to be on my mind consciously. Then again, it might do the opposite: e.g. the steering wheel spike argument. The distinction between whether the behaviour change is mindful or not is something I tried to capture with the behaviour change barometer.

Making it easier for users to avoid errors – whether through warnings, choice of defaults, confirmation dialogues and so on – is slightly ’softer’ than actual forcing the user to conform, and does perhaps offer the chance to relay some information about the reasoning behind the measure. But the philosophy behind all of these is, inevitably “we know what’s best”: a dose of paternalism, the degree of constraint determining the ‘libertarian’ prefix. The fact that all of us can probably think of everyday examples where we constantly have to change a setting from its default, or a confirmation dialogue slows us down (process friction), suggests that simple errorproofing cannot stand in for an intelligent process of understanding the user.

On with the patterns, then: there’s nothing new here, but hopefully seeing the patterns side by side allows an interesting and useful comparison. Defaults and Interlock are the two best ‘inspirations’ I think, in terms of using these errorproofing patterns to innovate concepts for influencing user behaviour in other fields. There will be a lot more to say about each pattern (further classification, and what kinds of behaviour change each is especially applicable to) in the near future as I gradually progress with this project.

[column width="47%" padding="6%"]

[/column][column width="47%" padding="0%"]

[/column][column width="47%" padding="6%"]

[/column][column width="47%" padding="0%"]

[/column]
[column width="47%" padding="6%"]

[/column][column width="47%" padding="0%"]

[/column][end_columns]

Photos/screenshots by Dan Lockton except seatbelt warning image (composite of photos by Zoom Zoom and Reiver) and donor card photo by Adrienne Hart-Davis.

UPDATE: See the detailed response below from Peter of Gateshead Council, which clarifies, corrects and expands upon some of the spin given by the Mail articles. The new shakers were supplied to the chip shop staff for use behind the counter: “Our main concern was around the amount of salt put on by staff seasoning food on behalf of customers before wrapping it up… Our observations… confirmed that customers were receiving about half of the recommended daily intake of salt in this way. We piloted some reduced hole versions with local chip shops who all found that none of their customers complained about the reduced saltiness.”

A number of councils in England have given fish & chip shops replacement salt shakers with fewer holes – from the Daily Mail:

Research has suggested that slashing the holes from the traditional 17 to five could cut the amount people sprinkle on their food by more than half.

And so at least six councils have ordered five-hole shakers – at taxpayers’ expense – and begun giving them away to chip shops and takeaways in their areas. Leading the way has been Gateshead Council, which spent 15 days researching the subject of salty takeaways before declaring the new five-hole cellars the solution.

Officers collected information from businesses, obtained samples of fish and chips, measured salt content and ‘carried out experiments to determine how the problem of excessive salt being dispensed could be overcome by design’. They decided that the five-hole pots would reduce the amount of salt being used by more than 60 per cent yet give a ‘visually acceptable sprinkling’ that would satisfy the customer.

OK. This is interesting. This is where the unit bias, defaults, libertarian paternalism and industrial design come together, in the mundanity of everyday interaction. It’s Brian Wansink’s ‘mindless margin’ being employed strategically, politically – and just look at the reaction it’s got from the public (and from Littlejohn). A BBC story about a similar initiative in Norfolk also gives us the industry view:

A spokesman for the National Federation of Fish Friers called the scheme a “gimmick” and said customers would just shake the containers more.

Graham Adderson, 62, who owns the Downham Fryer, in Downham Market, said: “I think the scheme is hilarious. If you want to put salt on your fish and chips and there are only four holes, you’re just going to spend longer putting more on.”

I’m assuming Gateshead Council’s research took account of this effect, although there are so many ways that users’ habits could have been formed through prior experience that this ’solution’ won’t apply to all users. There might be some customers who always put more salt on, before even tasting their food. There might be people who almost always think the fish & chips they get are too heavily salted anyway – plenty of people, anecdotally at least, used to buy Smith’s Salt ‘n’ Shake and not use the salt at all.

And there are probably plenty of people who will, indeed, end up consuming less salt, because of the heuristic of “hold salt shaker over food for n seconds” built up over many years of experience.

Overall: I actually quite like this idea: it’s clever, simple, and non-intrusive, but I can see how the interpretation, the framing, is crucial. Clearly, when presented in the way that the councils media have done here (as a government programme to eliminate customer choice, and force us all down the road decided by health bureaucrats), the initiative’s likely to elicit an angry reaction from a public sick of a “nanny state” interfering in every area of our lives. Politicians jumping on the Nudge bandwagon need to be very, very careful that this isn’t the way their initiatives are perceived and portrayed by the press (and many of them will be, of course): it needs to be very, very clear how each such measure actually benefits the public, and that message needs to be given extremely persuasively.

Final thought: Many cafés, canteens and so on have used sachets of salt, that customers apply themselves, for many years. The decision made by the manufacturers about the size of these portions is a major determinant of how much salt is used, because of the unit bias (people assume that one portion is the ‘right’ amount), and, just as with washing machine detergent, manipulation of this portion size could well be used as part of a strategy to influence the quantity used by customers. But would a similar salt sachet strategy (perhaps driven by manufacturers rather than councils) have provoked similar reactions? I’m not sure that it would. ‘Nanny manufacturer’ is less despised than ‘nanny state’, I think, certainly in the UK.

What do you think?

British Rail’s drop-the-window- then-stick-your-hand-outside- to-use-the-handle doors puzzled over by Don Norman in The Design of Everyday Things are still very much around, though often refurbished and repainted as with this delightful/vile pink First Great Western-liveried example.

I’m assuming that this design was intended to introduce an extra step into the door-opening procedure, a speed-hump, if you like, to make it less likely that a door was opened accidentally while the train was in motion (before central door locking was introduced – which makes it less necessary). From a usability point of view, we might immediately dismiss any system which has to have such detailed instructions to inform the user about performing such a simple task, but it’s certainly interesting to consider this kind of poka-yoke. Being forced to lowering the window to get to the handle is almost like a modal ‘Are you sure you want to delete this file?’ dialogue box.

However, other concerns come into play and now need to be considered in addition: this sticker suggests keeping the window closed to cut drag and save fuel, but as I walked along the train, almost all these windows were dropped down, left in that position by the last person to close the door. The urgency of scrabbling to lower the window, stick your hand out and use the handle, with a crowd of commuters behind you probably overwrites any intentions to close the window again engendered by the ‘Make a small change’ sticker.

Our DVD player has (to me) the most irritating standby light that I have ever seen on any device. When on, the light is constantly illuminated, but when in standby the light flashes continuously (at a slow rate). This drives me mad, but results in an interesting action – it causes me to turn it off at the plug when I am not using it (which is most of the time). Suddenly one little flashing light has resulted in more energy saving than having no light.

As he notes, designing a system with an indicator which actually draws power to inform you of… ‘nothing’ … actually may not be as inefficient as a from-first-principles efficiency design process would suggest, because of that human reaction. Similarly to the Static! project’s Power-Aware Cord, you may need to use a little extra energy to make people realise how much they’re using without thinking. Although:

There is one problem with this, it only works on people who care. If I did not care about saving energy, then I would just leave the laptop plugged in and the DVD player on. That means that you have to consider how your users will handle this kind of subtle feedback and determine whether turning the light off, or encouraging unplugging, results in more energy savings.

Sometimes the most obvious design decisions may not be the ones which result in the greatest energy saving.

This is a very astute observation indeed.

Are there any other examples where this sort of effect can be usefully employed? How similar is this to the ‘useful landmine’ concept where you deliberately force/provoke/annoy yourself into taking actions you otherwise wouldn’t bother/would forget to do?

EDIT: I’ve now added the audio! Thanks everyone for the suggestions on how best to do it; the audio is hosted on this site rather than the Internet Archive as the buffering seemed to stall a bit too much. Let me know if you have any problems.

I’ve put my presentation from Persuasive 2008 on SlideShare, – because of the visual style it really needs to be listened to, or viewed alongside the text (below, or in the comments when viewing it on the SlideShare site). Alternatively, just download it [PPT, 11.6 Mb] – it comes with the notes.

P.S. The slide about defaults, with the alarm clock stuck on 12:00, is meant to show it flashing – the actual PPT file uses an animated GIF – but SlideShare’s conversion process seems to have lost this element.*

1. I’m Dan Lockton, from Brunel University in London, and I’m going to be talking about what we call ‘Design with Intent’. It’s effectively Persuasive Technology in a Wider Context.

2. Persuasive Technology is an example of design that’s intended to result in certain user behaviour.
It’s design with intent.

3. If we cast our net a bit more widely, we can see that this idea recurs across many areas of design: solutions employed in one context are often applicable to others. Our research involves developing a tool to help designers match applicable design techniques to a range of ‘target behaviours’, and we’re ultimately going to be applying this to ecodesign, guiding more sustainable product use.

4. In this presentation we’ll look at a series of Design with Intent examples across different fields not normally considered part of Persuasive Technology, then see how the ideas of PT and DwI fit together. Then I’ll quickly describe how our work’s progressed since the paper was written.

5. Before getting started, have a look at these so-called ‘anti-loitering’ benches in Oxford, England – designed to prevent users actually sitting down, as the council freely admits. The seats are too high to sit properly and curved so you slide off if you try – you can ‘perch’, but that’s it. But there’s a worthwhile lesson right here: whatever the designers’ intent might be…

6. …people will find their own ways of using things. It’s easier to bend metal than to twist arms.

7. OK. In Human-Computer Interaction, as in Product Design, the main expressions of Design with Intent relate to designing specific affordances and constraints to guide users: shaping users’ perceptions of what actions are possible, and making some actions intentionally more difficult or impossible.

8. You can ‘design out’ affordances you don’t want the user to have – constraining the options available – here, to just ‘OK’, even if the user’s not OK with that – but it doesn’t always make for the best usability.

9. Or you can be a bit cleverer, and use a forcing function (a term coined by Donald Norman) – design the system so that the ‘right’ behaviour must occur before the user can take the next step. The example here is an interlock on a Toyota: to prevent the driver starting the car while it’s in gear, the ‘Start’ button is inoperative…

10. …unless the clutch pedal is held down…

11. …while the button’s pressed. I’ll admit it took me a while to figure that one out.

12. The best-known everyday safety interlock is on the microwave oven…

13. …where it will not operate unless the door is closed. Forcing functions generally aren’t subtle. They’re tending towards the coercive side of persuasion, but because they usually help us achieve something we want, such as keeping us safe, we don’t seem to mind too much.

14. Some affordance-manipulation can be a bit more subtly persuasive. Russell Beale, a computer scientist, used the term ‘Slanty Design’ to describe design which makes certain actions slightly more difficult, to discourage them. For example, these cigarette bins are sold on the basis that they have sloping tops not for aesthetic reasons, but so people don’t just leave cigarettes or litter on top of them.

15. Another aspect of affordance/constraint thinking is the persuasive power of defaults. We all know that many users leave settings exactly how they are, or simply choose the most prominent option: as designers, we can harness this power of choice architecture – as Richard Thaler and Cass Sunstein describe it – to persuade users into making the ‘right’ choices.

16. Imagine if all washing machines simply defaulted to the most efficient cycle (maybe even sensing the load to determine this). This is, again, subtle persuasion, but could have a big impact on users’ behaviour.

17. Now, in manufacturing, it’s crucial that assembly workers follow the right procedure when building something. To a large extent these are similar problems to those we’ve just seen – we want the ‘user’ (in this case that worker) to take certain actions, probably in a certain order. Every ‘mistake’ ends up costing the company money, in one way or another. Shigeo Shingo, a Japanese engineer, believed that with clever enough ‘defensive’ design, based on observation of workers, it was possible to eliminate assembly defects altogether. He called it Poka-yoke – mistake-proofing, and many of the ideas parallel those of affordances and constraints.

18. We’re used to seeing one of the very simplest poka-yoke methods every day – the ‘snipped’ corner on SIM cards, memory cards, and so, on…

19. …which prevent ‘assembly’ errors by ensuring that they can only be inserted into devices one way.

20. This is a control poka-yoke – it actually prevents the error from occurring. These are effectively forcing functions, as discussed earlier.

21. Shingo also used warning poka-yokes extensively, where a worker (or a user) is alerted to an error condition – something’s not in the right place, or is missing, or fitted incorrectly. The seatbelt warning light here indicates to the driver that a seatbelt is not buckled. This kind of immediate feedback on user behaviour is an example of suggestion-at-the-right-moment, or kairos, as defined in Persuasive Technology. It’s the right moment to warn the driver to fasten the seat belt.

22. Volvo for many years offered a gearchange suggestion light, which (based on monitoring engine RPM and throttle position), ‘suggested’ to the driver when he or she should change gear, to ensure the best economy. That’s a simple, clever persuasive technology: it makes ‘correct’ behaviour easier by guiding the user.

23. The idea that designers might ‘inscribe’ intended behaviours into artefacts has, in various forms, been subject to some philosophical and sociological debate. Johan Redström, developing an argument by Richard Buchanan, has suggested that since all artefacts are designed with some vision or intention of how they are ultimately to be used, it may be that all design is persuasive.

24. The presence of a chair persuades me to sit down where I might not have done otherwise. Designing the chair to appear more comfortable makes it even more likely. And so on.

25. Bruno Latour and Madeleine Akrich have discussed the idea that designers can ‘script’ behaviours into artefacts. Jaap Jelsma gives the example of a dual-button toilet flush as seen here, which effectively scripts users into making a decision about their water usage. There is no default, quite deliberately; the user must make some kind of decision.

26. This discussion has many expressions in urban planning, in fact: how much does architecture control us? Langdon Winner asked ‘Do artefacts have politics?’

27. His most famous examples were these very low overpasses built over a number of parkways on Long Island, by Robert Moses – too low for buses to pass underneath, with the effect of making it more difficult for poorer people to visit the Jones Beach State Park.

28. But there’s always the danger in this area of ascribing to malice what might more reasonably be explained by other factors, and the use of Moses’ bridges as the eminent ‘artefacts with politics’ example has been challenged in recent years by a number of authors.

29. Nevertheless, it is clear that some artefacts do have politics. We saw those perch benches in Oxford earlier on. Now, rough sleeping, by the homeless or otherwise, is frowned upon by many public authorities.

30. Sometimes benches with central armrests are installed specifically to attempt to stop this behaviour, especially at airports and railway stations.

31. Some models of bench are even sold to authorities on the basis that they will ‘discourage overnight stays’.

32. Not that some users can’t find a way round this…

33. Not all such techniques are so ‘anti-user’. Spaces and seating arrangements can be designed to be sociopetal, that is, to persuade people to interact – the simplest technique is to face seats towards each other…

34. …it doesn’t always work, of course.

35. Transposing the ‘architectures of control’ concept to the digital world, Lawrence Lessig used the phrase “Code is law” to explain how the structure of the internet, and what actions are possible, effectively regulates and shapes behaviour online, regardless of what laws may actually apply. If the system makes it easy to copy music, it will happen. Simplicity is persuasive.

36. So-called technological protection measures such as digital rights management – DRM – can be seen as attempts by companies to lock down the freedom of behaviour afforded by the internet, and persuade consumers into adhering to specific business models drawn up in an offline world.

37. Some of the most prevalent efforts at designing persuasion are for purely commercial benefit. Aside from advertising itself…

38. …there are strategies such as the razor-blade model, where a product is designed to persuade the consumer into repeat purchases of consumables, by locking him or her into a particular format. Electronic authentication makes this easier to enforce: for example, some printers include a ‘handshake’ which ensures that only the original manufacturer’s (usually higher-priced) cartridges can be used. Such strategies tend towards the coercive side of persuasion.

39. So, that was a very quick run-through of examples and ideas from a range of disciplines. I hope you can see how the Design with Intent idea runs through it all. But how does the field of Persuasive Technology, as it is defined, fit with this? Much PT research focuses on persuasion with intended social benefit – such as improving health – but much persuasion in the world as a whole is about intended commercial benefit. These don’t have to be mutually exclusive, of course: a fitness equipment manufacturer or a gym persuading people to exercise fulfils both social and commercial benefit intentions.

40. So, it makes sense to think of these as two separate dimensions of the ‘Design with Intent’ space.
Another aspect is whether the impact on the immediate user is helpful or not. This is where some persuasion techniques may fall down: it might be better for society, in terms of energy saving, if you can’t put your TV on standby any more, but it’s likely to inconvenience you. This is the grey area above. So if this space represents all Design with Intent, then maybe PT, as it’s defined, is the area outlined with the dashed line: it’s centred on intended social benefit, usually (but not always) helpful to the immediate user, and possibly with intended commercial benefit too. Still, this is only one way of visualising the relationship: as the boundaries of Persuasive Technology as a field are debated and redrawn, we may find that visualisations illustrating other aspects, such as coercion vs. persuasion, and so on, become useful.

41. Going beyond what’s in the paper now, over the last few months we’ve considered and analysed many different examples from different fields, and have tried to classify these techniques to understand them better and synthesize similar ideas.

42. The techniques pretty much fall into five ‘approaches’ which, though always open to debate, are useful in defining the mind-set a designer might have in approaching the problem.

43. These techniques have then been incorporated into a ‘suggestion tool’, which, given a target behaviour, allows designers to explore applicable techniques.

44. …The target behaviours are abstract descriptions, but can be applied to many different problems; each breaks down further into more specific target behaviours.

45. The next stage of our research will be testing out this suggestion tool, both in practical workshop sessions with design students and then with design consultancies… and with an online version, too.
After that, the aim is to do user trials with prototype ‘persuasive’ products developed as a result of applying the suggestion tool to sustainable behaviour problems, comparing how well different techniques actually work in practice in terms of changing behaviour, saving energy or reducing waste.

46. To conclude, I hope this brief review of Design with Intent has been interesting, and more importantly, inspirational in terms of suggesting examples of behaviour-shaping design beyond the immediate Persuasive Technology field. Our research is only at a very early stage, but we hope in due course to be able to present some concrete results, applying ‘Design with Intent’ thinking to guiding user behaviour, specifically in sustainable design.

47. In the meantime, if you’re interested, please do have a look at the research blog – at danlockton.co.uk. Thanks for listening.

All photographs/images by Dan Lockton except:
Slide 6 – Oxford Cornmarket bench with teenagers – Stephanie Jenkins -
http://www.headington.org.uk/oxon/cornmarket/new_seat.htm
Slide 14 – two catalogue images – New Pig Corporation -
http://www.newpig.com
Slide 22 – Volvo 340/360 dashboard – Volvo 300 Mania forums -
http://www.volvo300mania.com/
Slide 27 – Wantagh Parkway overpass – Peacenic on Flickr -
http://www.flickr.com/photos/68841932@N00/73241931
Slide 28 – Jones Beach approach – New York Architecture -
http://www.nyc-architecture.com/BKN/BKN001.htm
Slide 29 – Sleeping on a Hyde Park Bench – David Basanta on Flickr -
http://www.flickr.com/photos/dbasanta/2093742562
Slide 31 – Georgetown bench – Belson Outdoors -
http://web.archive.org/web/20040417173248/http://www.belson.com/gbrec.htm
Slide 32 – ‘Happy homeless’ – Rick Abbott on Flickr -
http://www.flickr.com/photos/rickabbott/81779858

This presentation was given by Dan Lockton at Persuasive 2008, Oulu, Finland on 6 June 2008, based on the paper: Lockton, D, Harrison, D. and Stanton, N.: Design with intent: Persuasive technology in a wider context, in H. Oinas-Kukkonen et al. (eds.): Persuasive 2008, LNCS 5033. Berlin Heidelberg: Springer-Verlag, 2008. pp. 274 – 278.

A preprint version is available free from http://hdl.handle.net/2438/2138

*The clock is a Mayhem Aurora, designed by Rob Leeks and Matt Chapman, and in reality does not flash when the time isn’t set. But I didn’t have a VCR handy to photograph…

These (pretty shallow) steps in Dawlish, Devon, have been labelled as such, presumably because without this, some visitors wouldn’t notice, and would run, cycle or wheelchair down them and hurt themselves or others. Painting a white line along the edge is a common way of improving visibility of steps, but actual labelling is fairly unusual.

There is some argument that having to label an affordance in this way, rather than it being self-evident (e.g. by making the steps deeper, or putting a handrail, or something), is ‘bad design’, but I’m not sure one way or the other: from a utilitarian point of view, enormous labelling, however ‘ugly’, is probably a surer bet than providing subtle ‘cues’. Nevertheless, the poka-yoke approach would be to design out the problem entirely: make the whole thing a full-width ramp like the section at the side.

A diagram in Bill Gaver’s classic paper ‘Technology Affordances‘ [PDF, 647 kb] sets out very clearly the importance of an affordance being perceived as such by a user:

In this case we have a hidden affordance (not deliberately hidden) which has been un-hidden by the label – similar to (though not as funny as) the ‘This is a Mop Sink‘ example from Michael Darnell’s fantastic BadDesigns.com:

This series is looking at what design techniques/mechanisms are applicable to guiding a user to follow a process or path, performing actions in a specified sequence. The techniques fall roughly into three ‘approaches’. In this post, I’m going to examine the Poka-yoke approach. If you’ve been following the previous posts, you’ll probably have thought, “Well, all that’s pretty obvious.” And it is obvious – we encounter these kinds of design techniques in products and systems every day – but that’s part of the point of this bit of the research: understanding what’s out there already.

Poka-yoke approach

The mechanisms described in this approach are all based on technical (rather than explicitly human) factors, and involve designing the relationships between system elements.

Poka-yoke (Japanese: mistake-proofing) is an approach usually applied in manufacturing engineering, developed by Shigeo Shingo in the context of developing ‘zero defect’ assembly processes. The idea is to avoid slip-type errors by designing systems which prevent them occurring, prevent a user proceeding until the error condition has been rectified (control poka-yokes), or at the very least clearly warn the user of the error condition (warning poka-yokes).

Generally, when the design intent is for the user to follow a process or path in a specified sequence, a deviation from that sequence can be considered as an error, and thus the poka-yoke approach can be applicable outside its original field. Similar concepts, forcing functions, have been developed in interaction design, especially in the work of Donald Norman – the three main forcing function mechanisms, Interlock, Lock-in and Lock-out, broadly correspond to Shingo’s control poka-yoke category; all can help in assisting (or forcing) users to follow a process or sequence. In the warning poka-yoke category, the Arrangement detection mechanism is most relevant to this behaviour.

Interlock

An Interlock combines elements of both lock-ins and lock-outs (see below), and is probably the most familiar forcing function mechanism: the ability to use one function is dependent on another running or being started, another component (such as a guard) being in place, or some other condition being fulfilled.

Example: This Toyota Verso requires the clutch pedal to be depressed before the starter button will operate, to reduce the risk of starting in gear.

Car ignitions which cannot be operated unless the driver’s seat belt is fastened – a system originally promoted as ‘Interlock’ in the US – microwave ovens not operating unless the door is closed, and airline or train toilets where the lighting does not operate until the user has locked the door, are some of the highest profile everyday examples, but the principle of the interlock is extremely common in engineering and manufacturing industry, often in the context of a machine tool which will not start until a guard is in place, or where opening the case automatically cuts the power.

Interlocks are often specified when it is imperative – rather than merely desirable – that a user follow a particular sequence, or at least two steps of a sequence, in exactly the right order, but their use need not be limited to critical safety design problems. Ecodesign applications might include (for example) a car’s air conditioning system requiring the windows to be fully closed before operating, or a sink requiring the plug to be in before the tap can be left in a ‘running’ position.

Example: The ubiquitous interlock on a microwave oven ensures that the door is closed before the oven will start.

Lock-in

The Lock-in mechanism in this context (rather than an economic one) refers to a system arranged such that a process, procedure or operation is kept active – the user can’t exit the operation until a certain condition is met, or the ‘correct’ next step is taken. This can be implemented using sensors, logic processing, physical architecture, or a number of other ways.

As Norman puts it, this prevents “someone from prematurely stopping” an operation – this could mean letting some ongoing process run its course to completion before starting the next, or denying the user access to another function which might interfere with the current process. It can also prevent accidental cancelling of an operation – inadvertent deviation from a specified sequence – by introducing an extra ‘confirmation’ step.

Example: The confirmation dialogue displayed by some software when a user attempts to exit can be seen as a lock-in to prevent inadvertent ending of the application.

Lock-out

Lock-out is closely related to Lock-in: in this case, the mechanism makes it difficult or impossible for the user to start certain operations, or denies or impedes access to particular areas or functions. In the context of encouraging or forcing a user to follow a path or process in a specified sequence, a lock-out helps prevent inadvertent or mistaken steps in that sequence. It can also help prevent an operation being started too early in the sequence, and may also be implemented as an extra ‘confirmation’ step.

Example: This file backup application prevents a user modifying the properties of a scheduled backup task while it is running – ensuring that the correct sequence is followed.

Arrangement detection

Arrangement detection is a ‘warning’ rather than ‘control’ poka-yoke mechanism, and may be considered as a ‘feedback’ analogue of interlocks, lock-ins and lock-outs – providing a warning (audible, visual, tactile) when system elements are incorrectly arranged (physically or procedurally).

Arrangement detection is about warning the user that the path or process is occurring in an incorrect sequence, rather than actually forcing the user to follow the correct sequence. While there are a number of possible warning poka-yoke mechanisms alerting users to incorrect behaviour, arrangement detection is most relevant to the specific issue of sequencing.

Example: The seat belt warning on car dashboards (in this case a Fiat Punto) is an arrangement detection poka-yoke, providing a visual (and often also audible) alert that a belt is not buckled while the engine is running, or the car is moving.

In part 4, we’ll look at the Persuasive Interface approach to getting someone to do things in a particular order.

Update: Also known as Useful Landmines in the 43 Folders world – thanks Pantufla!

Mistake-proofing – poka-yoke – can be as simple as encouraging/forcing yourself to do things in a sequence, to avoid forgetting or avoiding intermediate steps. If you’re the sort of person who hangs a jacket or bag on the door handle, so it can’t be forgotten on the way out, puts things in front of the door so you can’t forget them when you’re going out, or at the top or bottom of the stairs so you’ll remember to carry them to their intended destination next time you’re using the stairs, you’re engaged in mistake-proofing. You’re introducing a behaviour-shaping constraint to assist your own effectiveness.

In the above photo, putting the mobile phone (on-charge) inside a shoe makes it more likely that it will be remembered when going out: the act of putting the shoes on requires the user to pick up the phone, which could otherwise be easily forgotten. Similarly, Mark Hurst (of Good Experience and ‘Broken’ fame) regularly features two very simple poka-yoke procedures in his Uncle Mark’s Gift Guide & Almanac:

How to remember if the batteries aren’t in your camera

Summary: If the batteries are dead, or aren’t in the camera, keep the battery compartment open.

Description: When you’re charging your camera batteries (in a wall charger, say), keep the camera’s battery compartment open. That way, if you pick up your camera to put it in your pocket or purse, you’ll see that the battery compartment is open and will remember that the batteries aren’t in it.

There’s also this:

How to make sure they see the papers you dropped off

Summary: Put the papers on their chair.

Description: Here’s a tip I learned years ago and have used ever since. If you want to make sure that someone sees the papers you dropped off at their desk, put the papers on their chair. The natural inclination is to drop the files on the keyboard, or beside the mousepad. What’s the first thing the person does when they get back to their desk? They shove the papers aside, onto a nearby pile. They want to check their e-mail immediately, and those papers are in the way!

But put the papers on their chair, and watch what happens: the person refuses to sit on them! They take a second to pick them up, and while they’re in-hand, the person takes a look at the files while they get comfortable in the chair. Bingo: you guarantee attention to your drop-off.

Of course the papers-on-chair method can also be used to remind (or discipline) yourself about dealing with important papers.

This kind of very simple sequencing poka-yoke comes almost naturally in our everyday lives, at least with certain tasks. Sometimes it’s simply reminding ourselves to do something (e.g. putting a Post-It note somewhere we can see it); other times it’s trying to prevent us proceeding until some action has been taken (e.g. putting a Post-It note right in the middle of the computer screen so we can’t ignore it). Donald Norman’s Things That Make Us Smart has some interesting discussion of the power of Post-It notes and their importance as “information in the world”, disburdening some of our mental load – also part of the whole Getting Things Done phenomenon.

Sometimes we even (consciously or otherwise) try to ‘trick’ ourselves into behaving how we want to (or know we should) – the random offset alarm clock (patent; Halfbakery discussion) and Gauri Nanda’s “runaway success” Clocky being examples that spring to mind. (I once had a bedside clock radio where the button to set the minutes no longer worked, which meant that I could only set it either on-the-hour, or, because I forgot to do it at the right moment, set it maybe between 5 and 30 minutes fast. That meant that there was an uncertainty built into every time I glanced at the display, and indeed every time the alarm went off. I was rarely late, as a result.)

I have a hunch that almost trivially simple sequencing poka-yokes (in particular) could be important in designing for sustainable behaviour, such as reducing energy use and waste generation. For example, if your rubbish bin had a recycling box built into the top, so that you had to lift it out of the way (hinged, perhaps, to make it hassle to remove entirely) before putting anything into the main bin, it would be difficult to ignore the recycling box. Hence, learning as much as possible about different methods people use to mistake-proof themselves, or shape their own everyday behaviour, is likely to be useful in exapnding this line of research.

So, what are the everyday home-spun (or otherwise) tricks you use to help mistake-proof yourself?

Now, this is not in the same league, but interesting nonetheless: a ‘game’ to teach children (4 years onwards) spelling using a poka-yoke technique. The Spellmaster, from J W Spear & Sons – the example here is from 1980 (the Enfield factory was closed after a Mattel takeover in 1994) featured eighty plastic letter tiles, Scrabble-like but larger, with raised pegs underneath, a different pattern for each letter.

The letter tiles are used to spell the names of objects and concepts (colours, numbers) illustrated on punched cards which fit onto a backing board, the tiles only fitting in their spaces correctly if the pegs pattern aligns perfectly with the punched holes. If the wrong letter is used, the tile doesn’t fit properly and sits at an angle rather than snapping neatly into place. The ’snap’ of a correctly positioned letter is actually pretty satisfying – surprisingly so, given the combination of plastic (urea formaldehyde, I think) and 30-year old cardboard.

Left: The wrong tile – the pegs do not align with the punched holes. Right: The correct tile – everything lines up. Below: The wrong tile here – note the extra peg on the left-hand edge of the tile, which doesn’t match up with the punched hole, and leads to the tile not sitting down properly.

Letters which could work either way up, such as ‘o’ and ’s’ have – as would be hoped – symmetrical peg patterns. It’s a simple system, but it’s clever and while not offering any ‘remedial’ function to the child, I would think it’s not too likely that many children would try all 25 other letters assuming the first one didn’t fit. Hence, there is some bias against pure trial-and-error. It’s interesting to think how immediately we might consider a computer-based solution to this kind of design brief today, where a purely physical one would work very well and give a different kind of tactile satisfaction.

The tilt is tiny, all of 3 degrees, and the net effect is very satisfying — you gradually add snacks to the “light” side until it makes a soft and very definite *click* as it falls.

This kind of ‘very mild persuasion’ example is a great demonstration of how a simple physical property can be used to inform the user – the conventional modern solution in this area might be to monitor users’ behaviour, e.g. by weighing the amount of food put into the bowl, and then display it electronically, with an indication of whether a pre-set portion size has been exceeded. But these bowls simply tilt, with no electronics or moving parts (other than the bowl itself) necessary. It’s an elegant poka-yoke style solution.

Portion perception (and unit bias) is a fascinating area – we’ve looked briefly at it a few times – but I hope to explore it in more detail in due course, along with a review of Dr Brian Wansink’s Mindless Eating – in a post about how cognitive biases could be used in designing behavioural change.

UPDATE: This 2-page PDF (produced summer 2008) introduces the research

I’ve taken the plunge, and will be starting a PhD in September at Brunel University, Uxbridge, in the School of Engineering & Design.

The chosen subject incorporates both a formal investigation and review of certain architectures of control in design, and practical application of them for what I see as a worthwhile purpose: reducing the environmental impact of consumer products. This is an area which has come up quite a few times on the blog and in my previous research, and which I feel is both timely and worthy of a detailed treatment. The initial official title of the research is Reducing the environmental impact of products by using design to change user behaviour, and I’ve quoted a slightly shortened version of my brief tentative proposal below:

Introduction

Much research has concentrated on reducing the environmental impact of consumer products through improving manufacturing methods, efficiency of operation, and end-of-life processes. Attention is also being turned to changing consumers’ behaviour to the same end, through public education, policy and taxation emphasis — and product design methods, on which this study will focus.

Various techniques allow the characteristics of a product’s use phase to be influenced in favour of increased sustainability or reduced environmental impact. In purely technological terms, increased efficiency of operation is clearly a major goal, yet it may also be equally — and independently — important to reduce or otherwise to alter the period or manner of the product’s use, and that means changing users’ behaviour. Methods of achieving this, by using design techniques, range from ‘hard’ coercive constraints (technology which ‘refuses’ to be operated in a certain manner) to ‘softer’ psychological constraints which encourage or guide the consumer to use the product in a different way. The field lies at the intersection of technology and human factors, with the limits of any approach’s impact being determined by both technological and interaction design issues.

The study

This study will, in the first phase, review and characterise existing and novel design- and technology-led approaches to changing users’ behaviour to reduce the environmental impact of products. Donald Norman’s concepts of forcing functions and behaviour-shaping constraints, Shigeo Shingo’s poka-yoke methods, and B.J. Fogg’s ‘captology’ research at Stanford are pertinent here as starting points, since while these have been developed in the contexts of interaction design, manufacturing engineering and computer science respectively, there is significant potential to apply similar thinking with environmental considerations in mind; as far as the author is aware, this has not previously been done systematically.

A few specific technological approaches include: use of interlocks to ensure users make decisions or perform actions in the ‘right’ order when the ‘wrong’ order can be detrimental environmentally; sensors to shut down functionality when a product is not being used (e.g. motion-detection for lighting); sensors which prevent unnecessary energy use (e.g. a vehicle throttle which prevents over-revving when stationary); and the use of designed-in obsolescence to produce ‘optimum environmental lifetime’ products which expire at predetermined lifetimes, perhaps even using active disassembly techniques.

The second phase will involve testing-out of selected approaches through user trials and simulated trials of a number of functional product prototypes incorporating the behaviour constraints to determine levels of actual environmental benefit, and establish the technological and human factors affecting the ‘real-world’ applicability of these. Comparing life-cycle analyses of existing products’ use phases with those of the prototypes will allow a quantitative assessment of the benefits of different techniques in these contexts.

For example (illustrative only): A lot of electricity is wasted due to over-filling of electric kettles — a trial might compare prototypes ranging from the ‘soft’ constraint of a kettle with clearer visual/audio indications of fill level (prominent ‘x cups of water’ display) or financial implications of the energy use (‘Boiling this amount of water will cost you x pence’), through a kettle with a requirement to pre-select the water fill-level before filling (hence forcing the user to think about what he or she is doing), to a more extreme constraint of a kettle which will only boil one cup of water at a time — rapidly, but ensuring there can be no over-filling. Analysing the results of user trials of a range of prototypes such as these, and comparing with the energy usage of a conventional kettle, would allow actual energy savings to be quantified, and the limits of efficacy due to human factors (e.g. user frustration or misunderstanding) to be established. (The kettle examples described here are simplistic but this is the sort of approach intended.)

Another aim is to develop a ‘toolkit’ of tested design approaches, with relative efficacies and pertinent issues specified, to be of use to designers and engineers looking to create more environmentally friendly products. The outcome here would be an accessible publication (a short book, eBook and/or presentation, separate from the thesis) illustrating and detailing the techniques, made available to companies and students. It is hoped that government eco-design initiatives may also be interested in the practical implications of the work.

Background

The author studied Industrial Design Engineering at Brunel from 2000-4, and did a (taught) Cambridge-MIT Institute Master’s in Technology Policy from 2004-5. He has since worked in freelance design engineering and product design for a number of clients including, currently, Sir Clive Sinclair. His Master’s dissertation (and ongoing independent research in this area) investigated ‘architectures of control’: intentionally controlling user behaviour, mainly for political and commercial reasons, in a variety of fields, especially the built environment and digital rights. This forms a useful background to the proposed study.

Contribution to knowledge

The aim of the study will be to address these questions, reformulated as appropriate: How can users’ behaviour be changed, through redesign of products, to reduce environmental impact? Which methods are most suitable for specific situations? How significant are the impact reductions, and what technology and human factors issues affect the implementations? It is hoped that the process of investigating and answering these questions, together with an outcome synthesising the practical applications (the ‘toolkit’ described above), in addition to the thesis, will constitute an original, distinct and useful contribution to knowledge.

I’m excited: this gives me a fantastic opportunity to develop and extend the architectures of control research into what I consider to be a positive area (rather than the generally distasteful social engineering/’security’/designed-in-compliance/economic lock-in), which was otherwise going to be very difficult. I’m very lucky, thanks to the efforts of my supervisor, to have a studentship, which effectively means that this PhD is a job in environmentally sensitive design research, at one of the best technological design institutions in the UK.

I’ll continue to chart and examine all architectures of control via this blog, of course, but will now have the backing of some academic credibility – and resources – which should allow a more rigorous level of analysis, and exposure to expertise, precedents and inspirations.

The decision to go for a PhD wasn’t taken lightly; deciding how to progress professionally is something which has been taxing me for some time, alongside the challenges of freelance work (one reason why this blog has suffered over the last few months). I’m aware that it is not going to be easy, by any means (Tom Coates’ article – and the appended comments – and Rich Watts’ blog, for example, were very helpful in this regard), but it’s a long time since a project has excited me as much as this one, and I take that as a very positive sign.

Why Brunel? It’s where I did my undergraduate degree (although at the Runnymede campus, very different to Uxbridge), and many of the same staff, research strengths and commercial partnerships remain or have further developed. The university has greatly expanded the promotion of engineering and design and, as a future part of the University of London, seems a lot more confident about itself. While I very much enjoyed my time at Cambridge doing my Master’s, and it sparked my academic interest in architectures of control (specifically, in Frank Field’s lectures, both in person and via MIT videolink), I want (using my background) to develop the subject in a design context, which Cambridge does not offer in the same way.

The success of this blog in attracting some amazing, insightful comments (from what I can assume are amazing, insightful readers) has also given me a lot more confidence that taking this research further is not just worthwhile, but something I really must do, and I’m very grateful to all who’ve helped along the way so far.

The next post will review some of the ‘environmental architectures of control’ examples (both real and suggested) which I already have on my list, from this blog and elsewhere. Other than that, my girlfriend and I are off to Dublin for a few days’ break, and I’ve pledged not to take any work with me, physically or mentally, so let’s hope the spam filter can take care of the blog until next week!

These smokers’ bins from New Pig employ a very simple architecture of control – simply, sloping tops which prevent litter (including cigarette butts) accumulating. Compared with more conventional flat-topped cigarette receptacles this is presumably effective, although it does mean that anything placed on top will end up on the floor.

As with cone cups and wire-mesh bins, the success of the design in reducing the ‘undesirable’ behaviour must be down to people’s (conscious or otherwise) antipathy to an immediate ‘messy’ consequence of their actions. If you throw a cigarette butt on the ground straight-off, you can immediately forget about it. If you put it on top of a flat-topped bin, you can also immediately forget about it. But putting it on a sloping bin top and seeing it (or imagining it) falling off onto the ground somehow draws attention to your actions, just as leaving a paper cone cup with some liquid spilling out onto the table is rarely done, but leaving a conventional flat-bottomed paper cup is very common.

Incidentally, New Pig seems quite an interesting company with a playful approach to building its brand.

Mr Person at Text Savvy looks at an example of ‘Guided Practice’ in a maths textbook – the ‘guidance’ actually requiring attention from the teacher before the students can move on to working independently – and asks whether some type of architecture of control (a forcing function perhaps) would improve the situation, by making sure (to some extent) that each student understood what’s going on before being able to continue:

Image from Text Savvy
Is there room here for an architecture of control, which can make Guided Practice live up to its name?

This is a very interesting problem. Of course, learning software could prevent the student moving to the next screen until the correct answer is entered in a box. This must have been done hundreds of times in educational software, perhaps combined with tooltips (or the equivalent) that explain what the error is, or how to think differently to solve it – something like the following (I’ve just mocked this up, apologies for the hideous design):

The ‘Next’ button is greyed out to prevent the student advancing to the next problem until this one is correctly solved, and the deformed speech bubble thing gives a hint on how to think about correcting the error.

But just as a teacher doesn’t know absolutely if a student has really worked out the answer for him/herself, or copied it from another student, or guessed it, so the software doesn’t ‘know’ that the student has really solved the problem in the ‘correct’ way. (Certainly in my mock-up above, it wouldn’t be too difficult to guess the answer without having any understanding of the principle involved. We might say, “Well, implement a ‘3 wrong answers and you’re out’ policy to stop guessing,” but how does that actually help the student learn? I’ll return to this point later.)

Blind spots in understanding

I think that brings us to something which, frankly, worried me a lot when I was a kid, and still intrigues (and scares) me today: no-one can ever really know how (or how well) someone else ‘understands’ something.

What do I mean by that?

I think we all, if we’re honest, will admit to having areas of knowledge / expertise / understanding on which we’re woolly, ignorant, or with which we are not fully at ease. Sometimes the lack of knowledge actually scares us; other times it’s merely embarrassing.

For many people, maths (anything beyond simple arithmetic) is something to be feared. For others, it’s practical stuff such as car maintenance, household wiring, and so on. Medicine and medical stuff worries me, because I have never made the effort to learn enough about it, and it’s something that could affect me in a major way; equally, I’m pretty ignorant of a lot of literature, poetry and fine art, but that’s embarrassing rather than worrying.

Think for yourself: which areas of knowledge are outside your domain, and does your lack of understanding scare/intimidate you, or just embarrass you? Or don’t you mind either way?

Bringing this back to education, think back to exams, tests and other assessments you’ve taken in your life. How much did you “get away with”? Be honest. How many aspects did you fail to understand, yet still get away without confronting? In some universities in the UK, for instance, the pass mark for exams and courses is 40%. That may be an extreme, and it doesn’t necessarily follow that some students actually fail to understand 60% of what they’re taught and still pass, but it does mean that a lot of people are ‘qualified’ without fully understanding aspects of their own subject.

What’s also important is that even if everyone in the class got, say, 75% right, that 75% understanding would be different for each person: if we had four questions, A, B, C and D, some people would get A, B, and C right and D wrong; others A, B, D right and C wrong, and so on. Overall, the ‘understanding in common’ among a sample of students would be nowhere near 75%. It might, in fact, be small. And even if two students have both got the same answer right, they may ‘understand’ the issue differently, and may not be able to understand how the other one understands it. How does a teacher cope with this? How can a textbook handle it? How should assessors handle it?

I’ll admit something here. I never ‘liked’ algebraic factorisation when I was doing GCSE (age 14-15) A-level (16-17) or engineering degree level maths – I could work out that, say, (2x² + 2)(3x + 5)(x – 1) = 6x^4 + 4x³ – 4x² + 4x – 10 (I think! I don’t think there’s an HTML character code for a superscript 4, sorry), but there’s no way I could have done that in reverse, extracting the factors (2x² + 2)(3x + 5)(x – 1) from the expanded expression, other than by laborious trial and error. Something in my mathematical understanding made me ‘unable’ to do this, but I still got away with it, and other than meaning I wasted a bit more time in exams, I don’t think this blind spot affected me too much.

OK, that’s an excessively boring example, but there must be many much, much worse examples where an understanding blind spot has actually adversely affected a situation, or the competence of a whole company or project. Just reading sites such as Ben Goldacre’s Bad Science (where some shocking scientific misunderstandings and nonsense are highlighted) or even SharkTank (where some dreadful IT misunderstandings, often by management, are chronicled) or any number of other collections of failures, shows very clearly that there are a lot of people in influential positions, with great power and resources at their fingertips, who have significant knowledge and understanding blind spots even within domains with which they are supposedly professionally involved.

Forcing functions in textbooks

Back to education again, then: assuming that we agree that incompetence is bad, then gaps in understanding are important to resolve, or at least to investigate. How well can a teaching system or textbook be designed to make sure students really understand what they’re doing?

Putting mistake-proofing (poka-yoke) or forcing functions into conventional paper textbooks is much harder than doing it in software, but there are ways of doing it. A few years ago, I remember coming across a couple of late-1960s SI Metric training manuals which claimed to be able to “convert” the way the reader thought (i.e. Imperial to SI) through a “unique” method, which was quoted on the cover (in rather direct language) as something like “You make a mistake: you are CORRECTED. You fail to grasp a fundamental concept: you CANNOT proceed.” The way this was accomplished was simply by, similarly to (but not the same as) the classic Choose Your Own Adventure method, having multiple routes through the book, with the ‘page numbers’ being a three digit code generated by the student based on the answers to the questions on the current page. I’ve tried to mock up (from distant memory) the top and bottom sections of a typical page:

In effect, the instructions routed the student back and forth through the book based on the level of understanding demonstrated by answering the questions: a kind of flow chart or algorithm implemented in a paperback book, and with little incentive to ‘cheat’ since it was not obvious how far through the book one was. (Of course, the ‘length’ of the book would differ for different students depending on how well they did in the exercises they did.) There were no answers to look up: proceeding to whatever next stage was appropriate would show the student whether he/she had understood the concept correctly.

When I can find the books again (along with a lot of my old books, I don’t have them with me where I’m living at present), I will certainly post up some real images on the blog, and explain the system further. (It’s frustrating me now as I type this early on a Sunday morning that I can’t remember the name of the publisher: there may well already be an enthusiasts’ website devoted to them. Of course, I can remember the cover design pretty well, with wide sans-serif capital letters on striped blue/white and murky green/white backgrounds; I guess that’s why I’m a designer!)

A weaker way of achieving a ‘mistake-proofing’ effect is to use the output of one page (the result of the calculation) as the input of the next page’s calculation, wherever possible, and confirm it at that point so that the student’s understanding at each stage is either confirmed or shown to be erroneous. So long as the student has to display his/her working, there is little opportunity to ‘cheat’ by turning the page to get the answer. No marks would be awarded for the actual answer; only for the working to reach it, and a student who just cannot understand what’s going wrong with one part of the exercise can go on to the next part with the starting value already known. This would also make marking the exercise much quicker for the teacher, since he or she does not have to follow through the entire working with incorrect values as often happens where a student has got a wrong value very early on in a major series of calculations (I’ve been that student; I had a very patient lecturer once who worked through an 18-side set of my calculations about a belt-driven lawnmower which all had wrong values, based on something I got wrong on the first page.)

Overall, the field of ‘control’ as a way of checking (or assisting) understanding is clearly worth much further consideration. Perhaps there are better ways of recognising users’ blind spots and helping resolve them before problems occur which depend on that knowledge. I’m sure I’ll have more to say too, at a later point, on the issue of widespread ignorance of certain subjects, and gaps in understanding and their effects; it would be interesting to hear readers’ thoughts, though.

Footnote: Security comparison

We saw earlier that there seems to be little point in educational software limiting the number of guesses a student can have at the answer, at least when the student isn’t allowed to proceed until the correct answer is entered. I’m not saying any credit should be awarded for simply guessing (it probably shouldn’t), just that deliberately restricting progress isn’t usually desirable in education. But it is in security: indeed that’s what most password and PIN implementations use. Regular readers of the blog will know that the work of security researchers such as Bruce Schneier, Ross Anderson, Ed Felten and Alex Halderman is frequently mentioned, often in relation to digital rights management, but looking at forcing functions in an educational context also shows how relevant security research is to other areas of design. Security techniques say “don’t let that happen until this has happened”; so do many architectures of control.

Stick

If we define ’stick’ as ‘punishing the user for attempted deviation from prescribed behaviour’, then many of the architectures of control we’ve examined on this site demonstrate the stick approach. They’re not explicitly ‘technologies of punishment’ in Foucault’s phrase, but rather a form of structural punishment. The thinking seems to be (for example):

…and so on. There are numerous other examples from software and urban planning, especially.

The thing is, though, for each of those ’sticks’, a large percentage of people will not be obedient in the face of the ‘punishment’. They’ll try to find a way round it: a way of achieving their original objective but avoiding the punishment. They’ll search for what others in similar situations have done (e.g. DeCSS in the DVD example) or ask among friends until they find someone with the required expertise or who knows about an alternative. They may even actively destroy the ’stick’ that punishes them. In some cases they might not even understand that they’re being punished, simply seeing ‘the system’ as beyond their comprehension or stacked against them.

Equally, there isn’t always a rational strategy behind the ’stick’ in the first place. The anti-homeless bench doesn’t ’solve’ the ‘problem of homelessness’. It just punishes those who try to lie down on it without offering an alternative. It’s punishment with no attempt at resolving the problem.

If a stick does get people to change their behaviour in the intended way, it will be accompanied by resentment, anger and dissatisfaction. It may only be fear of the consequences which prevent actual rebellion. In short: using sticks to change people’s behaviour is not a good idea.

Image from image.frame

Carrot

A ‘carrot’ means offering users an incentive to change their behaviour. This moves away from actual control to something closer to some aspects of captology – making a persuasive case for behaviour change through demonstrating its benefits rather than punishing those who disobey.

To some extent, control and incentives may be incompatible. Taking away functionality from users then showing them how they can get it back (usually by paying something) might be a classic combined “carrot and stick” technique, but it’s also bordering on a protection racket, and it doesn’t fool many people.

However, can control be used in conjunction with genuine incentives to serve the agendas of both sides? Electric lights that turn off automatically if no-one’s in the room take some control away from the user, but also offer benefits to both the user (lower electricity bills) and society as a whole (less energy used). But if they turn off automatically, is there actually any incentive for the user to change his or her behaviour? If we’re always spoon-fed, will we ever learn?

Perhaps mistake-proofing measures or forcing functions which allow a user to increase his or her productivity or safety, in return for giving up some ‘control’ – which may not be highly valued anyway – fit the definition best. If I’m working in a factory painting coachlines on hand-built bicycles, a steady guide arm that damps my arm vibrations – but only if I also take care as well – takes some control away from me, but also prevents me making mistakes, allowing me to paint more coachlines per hour, more accurately. It also helps my employer.

But that’s a very weak degree of control. Unless anyone can come up with any counter-examples, I would suggest that providing real incentives for users to change their behaviour is fundamentally a very different approach to the ‘control mindset’ (unless you are trying to trick people by offering false incentives, or by understating what they could lose by changing their behaviour).

I’ll get round to speedometers in a future post, since this approach is worthy of a deeper treatment.

*The phrase “carrot and stick” seems now universally to imply “offering incentives with one hand and punishment with the other” (though not necessarily at the same time), rather than the “carrot dangling from a stick, just out of reach” meaning (i.e. “motivating people to perform with incentives which will never be fulfilled”) which I first assumed it to have when I heard the phrase as a kid (I’m not the only one with this issue). In this post, I’ll use “stick” to mean “punishment”.

These ‘pinch point’ barriers at London’s Heathrow Airport prevent the baggage trolleys from the Bus Station being taken down the escalators which lead to Terminals 1, 2 and 3. Mistake-proofing (for safety reasons: a trolley down the escalator would be dangerous) but also unnecessary if the airport had been designed differently from the start. Is forcing users to load baggage on and off multiple trolleys whenever their path descends or ascends really desirable? A lift (elevator) may be available, but how many people – and their trolleys – can fit in it at once?

An inclined travelator (as used elsewhere at Heathrow) would be a better solution.

From McGazz (who also has some great music to listen to on his website):

“As I was getting myself a cup of tea in work this morning, I overheard a colleague talking about a problem at the tanning salon his wife runs. Each cubicle has a bin in it, and a regular customer has apparently taken to vomiting and urinating in it (the guy reckoned the tannee in question might be bulimic).

I suggested he get round the problem by using wire mesh bins. While he was chuffed with this idea, I’m slightly worried that I managed to devise an ‘architecture of control’ after only a few seconds thought. I must have authoritarian tendencies…”

This is a clever, non-invasive, psychological deterrent to the undesirable behaviour. I wouldn’t call it authoritarian: it’s guiding behaviour without outright control. This is good design.

The closest parallel example I can think of is the use of cone-shaped paper cups for water-coolers (see image below left): besides being simpler & cheaper to make than flat-bottomed cups, people (generally) have a much lower tendency to leave them lying around once they’re empty. The psychological resistance to leaving the cup on its side (since it can’t stand up on its own) on the table, in case that last drip of water leaks out, is – oddly perhaps – fairly high. Especially when in company, people just don’t do it, whereas they’ll happily leave empty coffee cups and screwed-up cake wrappers on the tables. (I spent a lot of time in the Judge Business School, in Cambridge, where the Common Room – below right – had a water cooler using cone cups. It was rare to find them left on the tables, but common to find other litter.)

How could this type of design thinking be used in more situations to guide people into better behaviour? Littering seems an obvious theme to target, but also perhaps energy waste? Can devices which show us our energy usage in real time, such as the Wattson, really change people’s behaviour, or is it better to embarrass them into change? Roadside CO2-readers which flash up to you (and other drivers) just how much damage you’re doing to the environment?

This is a water damage sticker, which changes colour if moisture gets into this bit of the phone, and will be used to void your warranty if your phone stops working for any reason.

A single droplet of water placed on the sticker turns it bright red (in the case of my phone, anyway):

WikiHow’s ‘How to save a wet cell phone’ (found via Consumerist) recommends that you:

“Place a piece of satin finish scotch tape over your water damage sticker before you drop your cell phone in the water to prevent the water damage sticker from voiding your warranty… Remove the tape if you ever have to return your phone for repairs or warranty.”

Now, it’s a clever idea on the part of the phone companies, and presumably water-damaged phones being returned under warranty were enough of a problem to make such stickers ‘necessary’.

However, we all know that in practice, any non-working phone where the sticker has changed colour will be immediately classified as ‘water-damaged’ and the customer’s rights voided, even if the actual phone was independently defective.

As a designer, I would much prefer to look at the problem as “How can we improve the sealing of phones so that water ingress is no longer a major problem?” than “How can we design something to cover our backs and shift all the blame onto the user for our design fault?”

But maybe I’m naïve.

P.S. My Motorola, shown above, began to work intermittently just a month after the warranty expired, completely unrelated to any water issues, hence I don’t mind getting the sticker wet.

P.S. Hi, visitors from Nokia. Please note, my intention wasn’t to have a go at phone designers (or the engineering teams); and your phones seem superior on the water-protection front anyway. It’s just a commentary on the mindset which says “it’s easier/cheaper to catch users out than it is to solve the problem.”

Ryan G Coleman kindly sent me a link to this very interesting New Urban News story, ‘Research: trees make streets safer, not deadlier’. The gist is that roads planted with trees cause drivers to put themselves in state of greater alertness, which makes them generally more cautious about driving and generally slow down:

“Proposals for planting rows of trees along the roads — a traditional technique for shaping pleasing public spaces — are often opposed by transportation engineers, who contend that a wide travel corridor, free of obstacles, is needed to protect the lives of errant motorists…

[However], Eric Dumbaugh, an assistant professor of transportation at Texas A&M… looked at accident records and found that, on the contrary, wide-open corridors encourage motorists to speed, bringing on more crashes. By contrast, tree-lined roadways cause motorists to slow down and drive more carefully, Dumbaugh says.

Dumbaugh examined crash statistics and found that tree-lined streets experience fewer accidents than do “forgiving roadsides” — those that have been kept free of large, inflexible objects. He points to “a growing body of evidence suggesting that the inclusion of trees and other streetscape features in the roadside environment may actually reduce crashes and injuries on urban roadways”…

Dan Burden, senior urban designer for Glatting Jackson and Walkable Communities Inc. in Orlando, notes that there is research showing that “motorists need and benefit from tall vertical roadside features such as trees or buildings in order to properly gauge their speed.”

The article goes on to mention the ‘Shared Space‘ work of Hans Monderman, Ben Hamilton-Baillie and others, which includes removing road markings as part of a wider scheme to change the perceived emphasis of an environment and, again, put drivers into a state of greater awareness. From the BBC article on the ‘naked road’ experiment in Seend, Wiltshire:

“Motoring psychologists and urban planners seem to agree that, overall, “naked roads” appear to have a positive effect on motorists…

“This approach draws on behavioural psychology involving the way drivers respond to their surroundings,” [Ben Hamilton-Baillie] says. “It removes the sense of security provided by barriers – such as kerbs, and traffic lights. Instead of relying on the street system for security, drivers are forced to use their reactions.”

According to Mr Hamilton-Baillie, the removal of a psychological safety net encourages drivers to exercise caution and restraint. He believes that the lack of clear markings encourages drivers to slow down and mingle with pedestrians, forcing them to make eye contact with one another.”

Why are these techniques so much better than this kind of thing?

As so often, I feel it’s better to put users of a system into a state of mind where they are actively, intelligently thinking about what’s going on, and how they can respond to dangers or risks in the environment, than to remove that option for awareness or action planning, and deliberately force them into a state of ignorance of the risks ahead just to compel them to slow down. The driver in the tree-lined or Shared Space road situation can read the road ahead, and adjust his or her behaviour based on the risks that are perceived, whereas just blocking drivers’ vision so they can’t read road hazards ahead and must therefore actually come to a stop, does much less to help safety, and instead merely causes frustration.

Something with an interesting ‘forcing function’ story has been right in front of me all this time: the QWERTY keyboard, developed by Christopher Sholes and then Remington, with the intention of controlling the user’s behaviour. Until typists became proficient with the QWERTY system, the non-alphabetical layout with deliberate, if arbitrary, separation of common letters allowed the maximum typing speed to be slowed to something approaching writing speed, which reduced the amount of keys sticking and thus benefited both the manufacturer (less product failure, fewer complaints) and the customer (less product failure, less irritation). It also locked users who learned on a Remington QWERTY typewriter into staying with that system (and manufacturer, at least until the patents expired).

Whether or not QWERTY is a real example of market failure (in the sense that it’s an ‘inefficient’ system which nevertheless came to dominate, through self-reinforcing path-dependence, network effects, lock-in, etc), it’s an interesting design example of a commonplace architecture of control where the control function has long become obsolete as the configuration becomes the default way of designing the product.

Would designers today dare to create anything so deliberately idiosyncratic (even if clever) for mass consumption? (Systems that have evolved collaboratively to create complex, powerful results, such as UNIX, probably don’t count here.) The individualistic interfaces of some 1990s modelling software (e.g. Alias StudioTools, Form Z, Lightwave) which required a significant learning investment, were presumably designed with making the user experience easier “once you got used to it” (hence not really architectures of control) but have increasingly fallen by the wayside as the ’standard’ GUI model has become so commonplace.

Today’s architecture of control is more likely to be something more robust against the user’s adaptation: if for some reason it was desirable to limit the speed at which users typed today, it’s more likely we’d have a keyboard which limited the rate of text input electronically, with a buffer and deliberate delay and no way for the user to learn to get round the system. Indeed, it would probably report the user if he or she tried to do so. Judging by the evidence of the approaches to control through DRM, such a wilfully obstructive design seems more likely.

Returning to the idea of slowing down users for their own benefit, as commenter ‘Apertome’ points out on Squublog:

“One way in which some such designs [i.e. architectures of control] can be GOOD is when mountain biking – a lot of times, they’ll put a tight curve before an obstacle to force you to slow down.”

Note how this is a somewhat different practice to deliberately reducing visibility at junctions: using a bend to slow down a rider before an obstacle does not impede riders who are already travelling at a lower speed, while it makes the higher-speed riders slow down and hence keeps them safe, whereas wilfully removing sightlines at roundabouts would seem in many cases to work to the detriment of drivers who like to assess the road ahead well before the junction, and force all to stop instead.

Scott Adams recounts an anecdote illustrating the ‘illusion of control’ and how important it is to many people – even to the extent that it is the single defining characteristic of mankind which one might use to explain human behaviour to aliens:

“The maintenance man is moving the thermostat in our office today. I started talking with him about the “Thermostat Wars” [from Dilbert comics]. He told me about one office with 30 women where they could never get the temperature to an agreeable level. At his suggestion they installed 20 dummy thermostats around the office. Everyone was told that each thermostat controlled the zone around itself.

Problem solved. Now that everyone has “control” of their own thermostat there is no problem.”

To what extent is the illusion of control, rather than real control, what most people really want in their products?

Do they care that their personal data may be encrypted and held to ransom by a software company, so long as they feel ‘in control’ in everyday use (e.g. the ability to change the colour scheme)?

And how should designers respond to this issue? Are there any examples of products (other than, say, children’s toys) deliberately designed with fake controls to make the user feel in charge even though he/she isn’t? (Fake solar cell calculators are interesting, but not quite the same issue)

P.S. On the other hand, it’s worth considering the opinion expressed by the Audi A2 owner, that she didn’t find it a disadvantage having to take her Audi to a ’specialist’ in order to open the bonnet (hood). Is even that basic level of control (being able to see the engine) too much for some people? Is it because, say, a thermostat affects people personally (temperature) whereas a car engine is something dirty, difficult, complex, for someone else to worry about?