Today I was standing in the elevator and caught myself pressing the "Close doors" button repeatedly, although the button lit and I knew that it had received my intent. I know, this is common behaviour and that it won't make the doors close faster.

But then I wondered: Is this learned behaviour? If so, what systems in real life DO work faster when you repeat an input quickly?

  • 3
    The elevator button phenomenon is not about speed. It's about feedback: the user presses the button to request the elevator but is then given poor or no feedback about the status of their request - this leads them to think that the system might be faulty or that it has some other priority. The only recourse they have is to push the button again... and again... etc. – Andrew Martin Jul 3 '17 at 10:50
  • But the button lit up. I knew that the elevator received my signal. I edited the question to include this. – problemofficer Jul 3 '17 at 10:51
  • 1
    No, you only know that the button was pushed - you still don't know what the elevator is doing with your request – Andrew Martin Jul 3 '17 at 11:04
  • 1
    Door bells tend to provide feedback faster if you push them repeatedly. Also, other humans may give feedback if you touch them repeatedly. The latter may be where this behavior actually originates from. – Michael Schumacher Jul 3 '17 at 14:29
  • 1
    @Altainia if poking is the only way of touching someone repeatedly you can think of, then the feedback and the lesson you learn from it is well-deserved. – Michael Schumacher Jul 3 '17 at 19:06

I don't know of any system that is more responsive the more rapidly you give it identical input outside of, say, games that are specifically designed to that end (the faster you mash the 'A' button, the faster your racecar will go!).

But I think it is learned behavior. Suppose for a moment that your name is Clodor and you're in an elevator. You press the 'close' button. The button lights up, but nothing happens. Huh. So you press it again. And again, nothing happens. Then you press it a third time and well what do you know! The doors closed. Third time must've been the charm, right? So, the next day you do it again. 'Close the door'. 'Close the door'. 'Close the door'. Door closes. Yup. Must be your repeated input. Then the next day, used to this odd design choice these elevator designers went with, you press the button more rapidly than before. Except now it takes four times. Huh. It must not have a set number of inputs before it does what I want, it must just be a case of "The more I press it, the more likely it will be that it will respond".

This is called Confirmation Bias and it plagues us humans. We assume the elevators work this way and we do tests to confirm it, never trying to prove ourselves wrong (being patient, in this example). Apply this to our interactions with inputs in general, and, well, you get the idea.

Note: The confirmation bias is not the culprit here. The real way to solve this problem would be to either make the elevator doors close immediately or to provide some clear-cut message that says in no uncertain terms "We hear you, brah. We be closing the doors in short order", possibly even a recording that says something like that (preferably, in my opinion, exactly that). I know that those buttons on elevators tend to light up to indicate they got the message, but apparently humans don't get the message that the elevator got the message, so we may need something a bit more blatant.

Also, though elevators are used in this answer, this really applies to any interaction where the action requested does not happen immediately after the request. How many times have we clicked "OK" before that dialog box finally went away, even when the button made the pop-in pop-out animation? How much better would it be if we got a spinner at that moment to at least let us know it was working on it?

But as far as elevators go, I don't really see this "problem" being addressed by the industry, as it's not really a "problem" whether you press the button once or twenty times.

  • was going to answer more or less the same, but this answer is great, +1 – Devin Jul 3 '17 at 18:01
  • Sounds a bit far fetched to me. While the confirmation bias certainly could be an influence, I find it hard to believe that it is the primary reason. – problemofficer Jul 4 '17 at 3:28
  • 1
    @problemofficer No, I suppose the primary reason is that the doors do not close immediately after pressing the button or offer an alternative no-doubt-about-it confirmation that the closing action will be performed, which then leads to people trying multiple times, which then leads to the appearance that third time's the charm, at which point confirmation bias kicks in to cement this as a "this must be how it works". I'll edit my answer to mention that. – Altainia Jul 4 '17 at 15:24
  • BF Skinner saw this kind of behaviour in pigeons - the master of confirmation bias! io9.gizmodo.com/5746904/how-pigeons-get-to-be-superstitious – Andrew Martin Jul 5 '17 at 7:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.