Placebo buttons, false affordances and habit-forming

Elevator graph

This is a great graph
from GraphJam, by ‘Bloobeard’. It raises the question, of course, whether the ‘door close’ buttons on lifts/elevators really do actually do anything, or are simply there to ‘manage expectations‘ or act as a placebo.

The Straight Dope has quite a detailed answer from 1986:

The grim truth is that a significant percentage of the close-door buttons [CDB] in this world, for reasons that we will discuss anon, don’t do anything at all.

In the meantime, having consulted with various elevator repairmen, I would say that apparent CDB nonfunctionality may be explained by one of the following:

(1) The button really does work, it’s just set on time delay.
Suppose the elevator is set so that the doors close automatically after five seconds. The close-door button can be set to close the doors after two or three seconds. The button may be operating properly when you push it, but because there’s still a delay, you don’t realize it.

(2) The button is broken. Since a broken close-door button will not render the elevator inoperable and thus does not necessitate an emergency service call, it may remain unrepaired for weeks.

(3) The button has been disconnected, usually because the building owner received too many complaints from passengers who had somebody slam the doors on them.

(4) The button was never wired up in the first place. One repair type alleges that this accounts for the majority of cases.

Gizmodo, more recently, contends that:

…the Door Close button is there mostly to give passengers the illusion of control. In elevators built since the early ’90s. The button is only enabled in emergency situations with a key held by an authority.

Door close button

This is clearly not always true; I’ve just tested the button in the lift down the corridor here at Brunel (installed around a year ago) and it works fine. So it would seem that enabling the functionality (or not) or modifying it (e.g. time delays) is a decision that can be made for each installation, along the lines of the Straight Dope information.

If there’s a likelihood (e.g. in a busy location) that people running towards a lift will become antagonised by those already inside pressing the button (deliberately or otherwise) and closing the door on them, maybe it’s sensible to disable it, or introduce a delay. If the installation’s in a sparsely populated corner of a building where there’s only likely to be one lift user at a time, it makes sense for the button to be functional. Or maybe for the doors to close more quickly, automatically.

But thinking about this more generally: how often are deceptive buttons/controls/options – deliberate false affordances – used strategically in interaction design? What other examples are there? Can it work when a majority of users ‘know’ that the affordance is false, or don’t believe it any more? Do people just give up believing after a while – the product has “cried Wolf” too many times?

Matt Webb (Mind Hacks, Schulze & Webb) has an extremely interesting discussion of the extinction burst in conditioning, which seems relevant here:

There’s a nice example I read, I don’t recall where, about elevators. Imagine you live on the 10th floor and you take the elevator up there. One day it stops working, but for a couple of weeks you enter the elevator, hit the button, wait a minute, and only then take the stairs. After a while, you’ll stop bothering to check whether the elevator’s working again–you’ll go straight for the stairs. That’s called extinction.

Here’s the thing. Just before you give up entirely, you’ll go through an extinction burst. You’ll walk into the elevator and mash all the buttons, hold them down, press them harder or repeatedly, just anything to see whether it works. If it doesn’t work, hey, you’re not going to try the elevator again.

But if it does work! If it does work then bang, you’re conditioned for life. That behaviour is burnt in.

I think this effect has a lot more importance in everyday interaction with products/systems/environments than we might realise at first – a kind of mild Cargo Cult effect – and designers ought to be aware of it. (There’s a lot more I’d like to investigate about this effect, and how it might be applied intentionally…)

We’ve looked before at the thermostat wars and the illusion of control in this kind of context. It’s related to the illusion of control psychological effect studied by Ellen Langer and others, where people are shown to believe they have some control over things they clearly don’t: in most cases, a button does afford us control, and we would rationally expect it to: an expectation does, presumably, build up that similar buttons will do similar things in all lifts we step into, and if we’re used to it not doing anything, we either no longer bother pressing it, or we still press it every time “on the off-chance that one of these days it’ll work”.

How those habits form can have a large effect on how the products are, ultimately, used, since they often shake out into something binary (you either do something or you don’t): if you got a bad result the first time you used the 30 degree ‘eco’ mode on your washing machine, you may not bother ever trying it again, on that machine or on any others. If pressing the door close button seems to work, that behaviour gets transferred to all lifts you use (and it takes some conscious ‘extinction’ to change it).

There’s no real conclusion to this post, other than that it’s worth investigating this subject further.


  1. From what I can tell, most door close buttons are wired up, it’s just that the way they’re programmed means that there aren’t many cases where it actually speeds up the doors closing. In most lifts, there’s a sensor by the door which keeps the doors open when it detects someone walking through them, and then closes the doors after a few seconds, say 3s. In these circumstances, the door close button doesn’t achieve anything, as the lift still waits 3s after the sensor goes clear.

    However, if someone presses the ‘door open’ button, the doors will remain open for longer, say at least 10s, even if the sensors don’t detect anyone in the way. In these circumstances, the door close button can reduce the wait time back to the original 3s or so.

    Well – that what I’ve observed in a few lifts anyway!

  2. Who the hell presses the “close door” button 10 times?!?!

    Nice text, by the way.
    Elevators seem to be an ever guaranteed theme to explore, don’t they? I just love them.

    Loved the extinction burst thing…

  3. Interestingly, we’ve just had a new lift installed in our building which has no door close button, only door open. I’m positive I’ve encountered the lack of a close button in other lifts as well…

    Is it an acknowledgement that the door close button is a source of anti-social behaviour (avoiding a colleague you don’t want to join you) and frustration (as the door does not instantly close anyway)? Or just tight-fisted landlords?

    • Mr. Briggs

      In some elevators without a door-close button, pressing a button twice will serve as a door close.

      It works on the elevator I use to get to my dentist’s office.

  4. Wouter

    First of all, before posting this, why not contact some elevator constructors for some ‘real’ information?

    A disabled ‘Close door’ button because it would upset the person trying to catch the elevator; so now someone’s frantically pushing the ‘Close’ button after which the person DOES get in – will that result in a friendlier conversation?

    Maybe the ‘Close door’ button sometimes is used to cancel the ‘Open Door’ button? Which means – ‘Open door’ opens the door and keeps it opened for 10 seconds. When you push ‘Close door’ in those 10 seconds they are set back to 3.

  5. Pingback: ..:: Meta Matters » Placebo buttons, false affordances and habit forming

  6. Pingback: Placebo Buttons | Interaction Design Umeå

  7. James Smith

    I’ve heard the same thing about the buttons for the Walk/Don’t Walk signs. These have to be keyed to the traffic lights and it’s unlikely they would also control those. More likely is that they are there to allow pedestrians to feel they are doing something while waiting for the light to change.

    Personally, I have never noticed any difference in the sign changing when pushing the button.

    • Daniel

      Those buttons actually do something in many intersections. In such intersections, the walk/don’t walk signs are *always* on Don’t Walk until somebody pushes the button. This allows traffic to proceed more smoothly in large intersections with protected rights and lefts. When somebody pushes the button it waits until the next time when the pedestrian will be disrupting the least cars, and then allows them to go, blocking the cars that would otherwise be allowed to travel across their crosswalk. Intersections like this are “no turn on red”, for obvious reasons.

  8. Pingback: INFLUX library user experience > placebo buttons

  9. Pingback: Placebo Buttons « You Are Not So Smart

  10. Pingback: Кнопки с эффектом плацебо « Hanygan's Blog

  11. GMCourtney

    Watched NOVA the other night. Closed door button does absolutely nothing. It’s there to reduce tensions. It may light up, but that’s it.

  12. Pingback: Placebo buttons and the illusion of control « Another angry woman

  13. Pingback: My new enemy. « C'est La Vie

  14. Pingback: “The Placebo Button”: Evil Tricks Your Community Plays to Make You Feel Like You’re In Control–when you’really not. | The Red Panda Press

  15. Pingback: Placebo | Alfonso Morcuende

Comments are closed.