Saturday 29 October 2011

The Affordances of Everyday Things

I ramble on about affordances quite a bit; a big chunk of those posts was dedicated to figuring out whether they are relations or dispositions (answer: they're dispositions), and then there was all that arguing with Ken Aizawa about whether they are even anything at all (answer: yes, they are).

The thing I've been a little worried about is that people might see all this and come away thinking, "Why should I care one way or another?". I wanted to spend a post reminding everyone why all this talk about affordances matters, and the reason is simply this: affordances are important because they influence our behaviour, for good and ill, all day, every day. I'd like to illustrate this with a story, and a few photos I've been accumulating recently.

The McDonald's on O'Connell St in Dublin, with doors open for a change
Many years ago, I had come to Dublin to visit a good friend of mine, and we'd arranged to meet in town at this McDonald's; the bus from the airport stops there, so it was handy. I arrived a little early, and went inside to sit down. The doors that day were both closed, and presented me with vertical bars for handles. I then went through a routine to get in, and spent the next 20 minutes or so watching everyone else who entered the store do exactly the same routine, which went like this:
  1. Reach out with right hand to pull on the right door's handle. The door doesn't move.
  2. Think, 'you must have to push', and so push the door. The door doesn't move. This door is locked.
  3. Check that the store is open. It is.
  4. Reach out with left hand to pull on the left door's handle. The door doesn't move.
  5. Swear.
  6. Try pushing the left door, not expecting anything to happen. Door moves, letting you stumble into the store in an extremely bad mood.

I find this sequence very revealing. People tried to most obvious thing first: pulling the door closest their favoured right hands (I blame me doing this on living in a right handed world). Pulling was an attempt to effect an affordance; going right reflects a bias in design. They then pushed; people are used to dealing with stupid doors and know they often need pushing. They then went to the other door and didn't try the push, the last thing they had done; instead, they immediately tried to effect the affordance and pull again. Finally they push and get it right, but only because there was no other combination available. 

Donald Norman and "The Psychology of Everyday Things"
If you've ever read Donald Norman's classic book, The Psychology of Everyday Things, you will be very aware of the existence of doors with handles that you have to push. Norman talks explicitly about the affordances of objects, and how the things we interact with all present a way of interacting with them to us. (Norman has read his Gibson and taken the concept from there, although he is very up front about considering them to be cognitive, mental entities. He knows this isn't what Gibson meant, though, he simply disagrees; I can respect this, because he's clearly paid attention.) The book is just a wonderful catalogue of objects that have been poorly designed and solutions based in affordances for fixing them; if you've never read it, you should because it's a classic and extremely readable, and you'll never look at doors and taps the same way again.

These failures amaze me. Take that McDonald's. They had gone to some expense to present me with two doors, each with handles just begging to be grasped and pulled by a human hand. They had then gone to the trouble of making the doors work by pushing, and, for good measure, made one door into a wall by locking it. None of these actual facts about the doors were perceptually available. Quite the contrary: the affordances of the doors were the direct opposite, for pulling, and I acted accordingly (as did at least 20 people after me).

I've always thought that McDonald's (and everyone else responsible for the terrible doors in the world) should care more about this. Handles cost more than no handles; two doors cost more than one door; and making people feel like idiots because they can't operate one of the simplest devices in the world on their way to a burger they might feel a bit guilty about eating anyway is just terrible PR. And, as Norman ably describes, it just doesn't have to be this way. You can make doors whose function is clearly specified; you can make taps you only need one attempt to get water from, and oven element controls that you get right first time. These 'gulfs of expectation and execution' are not inevitable, and good design is about making these gulfs as small as possible by bringing the function of an object to where it can be perceived by the user.

I've uploaded and captioned some photos I've taken recently to a Picasa album, which you can view in full here (if you have pictures of your own, send them to me, I collect them to use in lecture!). Once you start seeing these things, you'll never go back; and the thing that you will notice the most is just how common poor design is. The vast majority of doors you will interact will suck at being doors, and everywhere you go, taps will require you to do something different to get water from them.

I highlight all this to make my point: affordances are important because they influence our behaviour, for good and ill, all day, every day, and nothing makes that more obvious than the bizarre failures of design we interact with all the time.

This label means they did it wrong
The 'undesigned' world tells fewer lies, fortunately, because their affordances are typically grounded in anchoring properties that lead to reliable, even lawful consequences for what the object looks like. Of course, it's not perfect: the problem of perceiving the affordances for locomotion on ice reminded me quite forcefully that not all anchoring properties have consequences are specified in information, and we only perceive what there is information for. In the case of ice, the critical property (friction) doesn't exist until my foot contacts the icy surface, so there is no information ahead of time, only once I'm on the ice. And this is essentially Donald's lesson too: the key properties of the object always exist, but when poor design or no information hides these properties from perception, disaster ensues.

Donald has a simple rule for designers: "If you have to add verbal instructions, you designed it wrong"; next time you're out, count how many simple things like doors have verbal instructions, and you'll wonder how we ever let it get this way.

Norman, D. (1988). The Psychology of Everyday Things. Basic Books.


  1. I think fire safety codes typically require exterior doors of certain establishments to open outward, and it appears from the photo that the once offending doors did so at the time of the photo. This together with the fact that only one door was openable at the time of your earlier (?) visit suggests the possibility of a temporary malfunction rather than a design flaw.

    In any event, the unopenable door seems to be yet another example of Ken's unactualizable (non-?)affordances.

  2. The concept of affordances can get ugly in the grey areas, but it is beautiful in its natural habitat. I really like the examples of bad design, though they speak to the ugly grey area.

    The problem is that the doors of the McDonald's DID afford pulling-on (as evidenced by the fact that you realized the affordance). However, much to your frustration, they did not afford opening-towards-you-from-the-street.* So, does your firm tug reveal correct perception or incorrect perception?

    This is an interesting levels of analysis problem, and I'm not sure what to do about it. Often times in these discussions, when the notions of affordances gets shaky, one can seemingly find solid ground by shifting the scale of the behavior and outcome. However, shifting scales is typically a cheating move, because it typically changes the subject of the conversation rather than answering the question. (Typically, but not always.)

    I've tried to wrestle with this before and never found a satisfactory solution.

    *To be even more generous, the doors might well have afforded opening-outward-by-pulling-the-handle... just not to someone of your strength level. I'm envisioning a robotic Arnold Schwarzenegger taking the door against its hinges and casually walking in as if nothing unusual had happened. Even were this true, the question still remains: If you were mistaken about your ability to enact a specific type of change on the environment, does that prove a PERCEPTUAL error has occurred, or could the perceptual part of the story still have been error free?

  3. In the case of the door, I correctly perceived that the door handle afforded grasping and pulling, and did not perceive (because there was no information) that this would not work. Like friction, there was no information about the underlying properties available ahead of time; only when I pulled on the door did information become available,and then I acted accordingly.

    Ken's problem is he can't work time into his thinking. You always need to ask, who knew what, and when, and on what basis? Affordances are real and persist, but the information for those affordances can swing in and out of 'view' as you move through the world.

    Charles, the fire code thing works except when I was there I had to push the door in.

  4. "Ken's problem is he can't work time into his thinking."

    I think he works time in but on the wrong scale. The counterexamples I recall all involve viewing a distant scene and judging that some element of that scene offers an affordance. But that seems to confuse a long-term goal and an affordance, the latter - but not the former - offering immediate actualizability. If one thinks of the problem as being to assess the behavior of a continuous time adaptive system with feedback, the time interval of interest is the small delta t of calculus, not the large delta T of advanced planning. So, I think eric is right to identify the affordance as "pull-on-able", an immediately and continuously actualizable affordance over some finite time interval.

    Viewing it like that also solves the problem I've had with persistence. If one interprets a goal (here, opening the door) as an affordance, the problem of actor-dependence arises: as eric notes, that "affordance" is offered to the terminator (or whatever character he has in mind - I don't know Arnold's movies) but not to us wimps. But pull-on-able is offered to any critter that can effect graspable. And Andrew's continuous flow of information, manifest in the resistance experienced while effecting that affordance, will be different for Arnold from what it is for us, resulting in our ultimate continuing behavior diverging from Arnold's since he will be offered new affordances different from those offered us.

  5. That all seems about right to me.

  6. Actually, this made me think of something. Your description, which is basically right, is what Tony was trying to achieve with his Affordances 2.0. I still think that information is the solution to the problems Tony was raising; information ebbs and flows, affordances don't. So I'm on board with the need for their to be dynamic, time-varying access to affordances, I just don't think affordances themselves have to be relations to achieve this.

    Just a connection I wanted to make explicit.

  7. Needless to say, I'm happy to think that I'm finally nore-or-less on board the eco-psyche train! And it took only a year+. Once again, your patience is appreciated.

    Assuming that in your post on "Chemero Chap 7: Affordances" the initial three point summary of Chemero's objections to affordances as dispositions is reasonably accurate, I see these problems:

    Point 1 - If affordances must be immediately effected, there clearly can be only one at any specific time. Thus, if Chemero really says "I am not currently trying to effect all the affordances in my vicinity", either he is wrong (there is only one and it must be effected) or he is not assuming that affordances must be immediately effected. As for dispositions, I agree with Chemero that they are context dependent, but also agree with you that the context can (generally will) be complex and time-varying.

    What I take to be the mistake - which many of us make, but as a determinist I have no excuse for making - is to assume that at any time there is a choice to be made among affordances/dispositions. Again, I think it's a matter of time scale. Over any finite time interval T, there may be many actions that conceivably could have been taken during T if one considers only those aspects of context that were constant throughout T: the size of the door handle relative to the subject's hand, the subject's strength, et al. But at a specific instant, the context includes a multitude of elements that are time-varying over T. I submit that considering the total context at any instant, there is only one possible move, viz, to effect the one affordance/disposition that everything that has happened up to that instant has resulted in being offered to the subject - or in your preferred vocabulary, the move that all information available at that instant dictates. (Using the analogy of a continuous-time system, at the beginning of T the system could conceivably follow any of multiple paths. At any instant, however, it's immediate future (on the scale of the delta-t of calculus) presumably is determined by the input up to that instant (assuming "typical systems", eg, no "quantum-weird" stuff going on).

    Point 2 - From the perspective of the above discussion of Point 1, the complement of a disposition/affordance is present by definition since an affordance/disposition must be immediately effected. Ie, no effectivity, no affordance/disposition. But perhaps I'm missing some subtlety.

    Point 3 - I really don't get the "overlapping minds" bit. I skimmed through some of the various cited papers but gave up when I encountered what seemed to be an argument based on Berkley-style idealism - seemingly quite irrelevant to the issue at hand.

    I gather that the envisioned problem is, in short, that if two minds perceive (in some sense) the same object, they have (with respect to that object) the same mental content, which would violate some aspect of the assumed privacy/authority/whatever of first person perspective. I certainly can't make a coherent argument refuting this, but Donald Davidson could, and did. In fact, contrary to minds overlapping due to similarly perceiving a common entity (object or event) being a problem, he argued that it is a requirement for communication. Here too perhaps I'm missing a subtlety, but to me this objection seems a non-starter.

  8. Needless to say, I'm happy to think that I'm finally nore-or-less on board the eco-psyche train! And it took only a year+. Once again, your patience is appreciated.
    Woohoo! I always thought we should get sent chocolates when we talk someone into being ecological :)

    Thus, if Chemero really says "I am not currently trying to effect all the affordances in my vicinity", either he is wrong (there is only one and it must be effected) or he is not assuming that affordances must be immediately effected.
    Mostly the latter. He presents the fact that there are multiple affordances available and that we aren't trying to effect them all as evidence that they aren't dispositions. I think your time scale issue gets to the heart of that, though.

    Point 2: the subtlety is that Tony is concerned that we don't know enough to characterise the complementary effectivity precisely. He has a point about body scale not cutting it, but I tend to think this is a state-of-the-art problem more than something we can never solve, because, as you say, the complement is in the system somewhere, it's just up to us to find it.

    Overlapping minds: you've described it correctly. Chemero's read Heft and noticed that you need relations to solve the problem, but he didn't read Heft close enough to realise that the relation is the act of perception, not the thing perceived.

    Also I tend to agree it's a feature, not a bug, for the reasons you suggest, although that's only ever been a hunch. Nice to know there's some support :)

  9. I should note that it isn't just "some support". Not, of course, terribly "scientific", but FWIW the key Davidson paper arguing this - "3 Varieties of Knowledge" yields 266K google hits. The much more famous Quine's "Two Dogmas" and Sellars' "Empiricism and Phil of Mind" get 330K and 376K respectively. So, at least by that measure, "3 Varieties" is in roughly the same league.

    So, I'd say your hunch was pretty good!

  10. Two quick thoughts:
    Regarding dispositions - So far as I can tell, the only way to make the dispositions argument work is to off load a lot of nuance into the notion of "effectivities". If, given many things a person of my general body type could potentially do, I reach out to grab the door nob and open the door, ipso facto, at that moment I was, at a much more detailed level of analysis, the type of organism that was afforded door opening... and nothing else. This line of thinking makes me feel worried for lots of reasons, but it works. I don't think this is what Gibson was saying, but it is a potential foundation for ecological psychology. I just wish that Turvey et al were willing to own the implications more strongly so that a better conversation could be had.

    Regarding the two minds problems: This is a very long standing problem in philosophy. For example, William James and his students... the ones who would influence the future behaviorists and ecological psychologists... were very concerned with it 100 years ago. If you are a realist, the problem is entirely "scientific" and not at all "philosophical" as the terms are usually understood. However, if you are stuck trying to make a case for realism, some philosophers will still think it is kind of a big deal.

  11. If ... I reach out to grab the door nob and open the door ... at that moment I was ... the type of organism that was afforded door opening

    Perhaps the problem I see with this way of looking at it would be clearer if put into the vocabulary of intentionality.

    I see the affordances offered (actually continuous but sequenced for convenience) as something like: approachable, reach-for-able, graspable, pull-on-able, but not openable. A primitive critter (adaptive robot) might be "disposed" (designed) to execute that specific context-dependent sequence notwithstanding having no intentionality, ie, propositional attitude: wishing or desiring that the door be open, hoping or believing thatsomething yummy is behind the door, etc. And that's where I think Ken's examples all go wrong. They mix a description of a subject's behavior in the psychological vocabulary of intentionality with a description in the ecological vocabulary of affordances. And opening a door is similar. If the closed door is understood to offer the openable affordance - initially distal, later proximal - the paradox of the subject's being unable to immediately effect a proximal "affordance" arises. But not if opening the door is viewed instead as the object of an intentional posture. By definition, an affordance can't be thwarted; a hope, wish, or desire can.

    I'm clearly shooting from the hip here, trying to integrate a bunch of ideas grasped fuzzily at best. But perhaps there's a kernel that someone better positioned than I can do something with. Eg, I can kind of see this fitting into Andrew's robot post (representational robots pursuing intentions, dynamical systems robots pursuing affordances??) but not clearly enough to suggest how.

    Obvious question: did Gibson make some analous distinction between the automaton-like behavior of a creature qua "dynamical system" and the (supposedly) "intentional" behavior of a creature qua "mindful system"?

  12. After reading through the posts on affordances, it seems to me that Turvey must be confused about effectivities. Surely the actualisation of the pull-ability of a door handle is its being pulled. The effectivity, therefore, is pulling and not, say, the shape and size of the organisms hands. The shape and size of the organisms hands are not dispositions at all and therefore cannot be effectivities either. They are instead the vehicle that enables the pulling, just as the microphysical properties of salt enable it to be a solvent. They are complementary with those properties of the handle that enable its pull-ability: its size, shape, rigidity, etc. The compulsory effecting problem goes away on this account since pulling is the actualisation of our ability to pull, which we can choose whether or not to exercise.

  13. In the Turvey framework, pulling the door is effecting the affordance, which you can do by virtue of your effectivities; these are the properties of the organism which complement the affordance and thus allow the disposition to be played out.

    Body scale was one initial swing at finding effectivities. It's had mixed success.

    The shape and size of the organisms hands are not dispositions at all and therefore cannot be effectivities either. They are instead the vehicle that enables the pulling, just as the microphysical properties of salt enable it to be a solvent.
    This makes them dispositions.

  14. @scientism
    The Turvey formula is a trivial idea wrapped up in an unnecessarily sophisticated jargon.

    Turvey says: "Realization of the affordance" because "object affords", and "we have effectivity".

    Which can be read as: We behave when there is a thing that can be changed, and we have the ability to change it.

    I suppose one of the reasons I don't like Turvey's formula, which I don't think I've brought up too much on this blog, is that if Ecological Psychology = that formula, then there there would be nothing interesting or controversial about ecological psychology. That I find many people interesting, and that many more people find it highly controversial, suggests there is something more going on. Alternatively, many, many otherwise very smart people could be quite confused... which happens.

    The typical alternative proposal for Eco Psych, that affordances are the relation, and that the relation itself can be perceived, is at least a little more interesting, and more understandably controversial.

  15. Turvey does get carried away with his words; but I think the dispositional framework has a nice discipline to it. I also don't agree that it makes anything trivial; me grasping something does mean a) the object is graspable and b) I can grasp that thing, but there are numerous empirical questions that remain (how is the affordance composed? What is the information for it? etc).

  16. @Andrew Wilson

    Sorry, I got confused there. I should have said the effectivity would be the organism's pull. The actualisation (the effecting) would be pulling. The shape and size of the hand (among other things) would be the vehicle of the organism's pull. (This probably makes more sense with graspability. The effectivity would the organism's grasp. The vehicle of its grasp is the shape and size of the hand.)

    "This makes them dispositions."

    If I can say the effectivity is just the shape and size of the hand, then surely I can say the disposition is just the shape and size of the object, and then why do I need to speak of affordances at all? Is it just a convenience? But surely the whole point is that graspability, and not merely the shape and size of the object, is perceptible. If affordances are to be real things, then we shouldn't reduce them to their vehicles, and we shouldn't reduce effectivities to their vehicles either.

  17. Not all dispositions are affordances, but all affordances are dispositions (under this scheme). So we're highlighting a set of action relevant dispositions and calling those ones affordances; standard science move.

    The effectivity is not necessarily the size and shape of the hand. As I said, body scale is just the first thing people looked at (eg Warren expressing stair riser heights as proportions of leg length). I'm inclined to agree that body scale is unlikely to contain the full effectivity; but as yet we don't have a rigorous alternative.

    Graspability is indeed the property we want to get to. But understanding how that property is composed is important, because that then determines how that affordance can be specified in, say, light, and the affordance is of no use if it produces no information. Information is the critical thing for Gibson, and it gets left out of the analysis a lot because it's hard. This is a problem.

  18. It's funny how we still tend to want to frame affordances as things we're supposed to grasp based on information we get just from standing still & looking, when Gibson goes to great lengths to say we get information only through *action* within the environment.
    As you say, you had insufficient information by just walking up to the door and seeing it visually - you had to act on the object in order to experience that the event your perceptual system expected was not going to happen. Grasping the door and pulling *is* part of your perception. That's how you get to the other layers of the nested structures in your environment, no?
    (I realize this is an old post and this was probably settled, but I'm in the thick of it, so couldn't help but toss my thoughts in ...)

    1. I agree entirely. Now, ecological psychology has not been great about tackling this fact; we are a little guilty of talking carelessly about affordances. That's a fixable problem, though,

    2. The "actions of perception" are tremendously understudied, given their crucial importance in Gibson's system.

  19. The pull-handle affords the shop-owner a convenient manner of closing the door firmly when he goes home at night.