Which Affordance? Three questions for navigating different ways of thinking about affordances

Eric Cheung
12 hours ago
10 min read

I first came across the concept of affordance while reading about it in The Urban Design Reader (2007), where it was presented as a concept with an interesting potential for architecture and urban design. In a chapter originally published by Lang (1994), affordance is described as a concept “borrowed” from Gibson and used to clarify the relationship between the built environment, human behaviour, and the fulfilment of values and needs. Lang argues that patterns of the built environment afford certain activities or aesthetic interpretations, and that their overall configurations and properties can enlarge or constrain the possibilities for behaviour.

This framing seemed particularly useful for spatial design because it connects the configuration of the built environment to the possibilities available to people as they try to fulfil their needs. It suggests a design-oriented question:

What range of possibilities should an environment provide so that cities can accommodate what people need now, as well as what they might need in the future?

From this perspective, changing the configuration of the built environment can change the possibilities for behaviour available within it, considering different aspects of what is valued and needed (see article).

Then I thought, what do we know about affordance?

If the concept originated with Gibson (1977) and has subsequently been taken up across different fields, there should be a body of empirical research that could provide evidence relevant to architecture and urban design. But when I began looking at the research related to this concept, I became confused. The difficulty is not that the term is defined differently across fields. The more I looked at the literature, the more it seemed that different uses of affordance involved different objects of study, different research questions and different kinds of evidence. A staircase experiment, an observation of activity in a public space and a theory of situated social practice may all be described in terms of affordances, but they do not appear to establish the same kind of knowledge.

So I started trying to map the territory.

'Which Affordance? Version 1' is my current attempt. It groups some of the research traditions I have been reading into broad families. It is deliberately incomplete. There are other important lines of research that I am leaving aside for now because I want to understand these first before expanding the map. The grouping is not intended as a history of affordance theory, nor as a definitive taxonomy. It is a way of making visible some of the different questions that emerge when the concept moves between traditions.

For now, I am using three primitive questions to organise my exploration.

Can an affordance be measured and directly perceived?
What possibilities does an environment support, and which are taken up?
How do we make an intended action possibility apparent?

These are not necessarily stages in the development of affordance theory. They are different kinds of research questions that become visible when affordance is used across different traditions.

Uptake of affordance differs depending on the user's abilities. — Photo by Jukan Tateisi on Unsplash

1. Can an affordance be measured and directly perceived?

Perception-Action boundary

One way into the exploration is through Warren's well-known work on stair climbing. Gibson (1979) described affordances relationally: possibilities for action exist in relation to the capabilities of an organism. A staircase is therefore not climbable or unclimbable as an intrinsic property of the staircase. Whether it affords climbing depends on the relationship between its properties and those of the person encountering it. Warren (1984) turned this theoretical proposition into something that could be investigated empirically. The question was:

Can we identify an action boundary relation between environmental properties and organismic capabilities?

Stair riser height provided a measurable property of the environment. Leg length provided a corresponding property of the person. By relating the two, Warren could investigate the point at which stair climbing moved from possible to no longer possible and establish an action boundary relation.

But establishing an action boundary raises a second question:

Can an organism perceive whether an action is possible?

This is not necessarily the same question as whether the action is possible. Warren's experiments also examined participants' judgements about whether stairs appeared climbable, allowing perceived climbability to be compared with the action boundary established from the organism-environment relation.

Warren and Whang's (1987) study of walking through apertures extended this approach to passability. The width of an aperture could be related to the dimensions of the person, while participants' perceptual judgements and their walking behaviour provided ways of investigating whether this body-scaled relation was available to perception. The importance of this research programme is therefore not that particular stairs or apertures can be assigned a numerical threshold. It shows how a relational theory of affordance can be turned into an empirical problem in which environmental properties, organismic capabilities, action boundaries and perceptual judgements can be related to one another.

From an architectural perspective, there may nevertheless be relatively little to translate directly from the particular findings into routine design guidance. The practical dimensions of ordinary stairs or apertures are already addressed through established design knowledge and building regulations. That is not a weakness of the research, nor was producing architectural guidance its purpose. Its significance lies in showing what must be specified and measured when a relational theory of affordance is turned into an empirical account of perception and action.

Neuroscience conceptualisation

More recent neurophysiological work provides another way of investigating the perceptual side of this question. Djebbara, Fich and Gramann (2021) used mobile EEG to examine brain dynamics as participants encountered architectural transitions with different affordance characteristics. Their findings suggest that affordance-relevant spatial configurations are reflected in sensorimotor neural dynamics during perception, including before participants interacted with the environment. This provides evidence consistent with the idea that action-relevant properties of spatial structure can be registered without first requiring an explicit judgement about what action to take.

What it adds is a different level of evidence: alongside asking whether people can explicitly judge an action possibility, researchers can ask whether the environment is already modulating action-relevant neural processing as it is encountered. For architecture, this distinction shifts the question away from whether a space permits an action. It raises the possibility of asking whether spatial configuration itself makes an action possibility perceptually available and behaviourally anticipatable.

2. What possibilities does an environment support, and which are taken up?

Functional classification

When affordance enters architectural planning relevent research, one way of conceptualising Gibson's idea is to move from experimentally specified action boundaries towards a broader characterisation of the possibilities an environment makes available.

What possibilities can we recognise?

This is particularly visible in Heft's (1988) formulation of environmental properties in terms of their functional significance for action specifically for children activity. Rather than treating an environment as a collection of physical properties, it can be described in terms of the activities that its properties support for a particular organism. Examples in Heft's preliminary functional taxonomy (1988) includes a flat, relatively smooth surface affords walking, running; moldable material e.g. dirt or sand affords construction of objects. This provides a way of describing environments not only in terms of possible action like climbing but also functional actions. It is particularly useful for analysing everyday settings, where many possibilities coexist and where the same physical arrangement may participate in several different forms of activity.

Importantly, however, these possibilities are not necessarily fixed categories attached to an environment. They arise from the relation between environmental properties and the capabilities, practices and circumstances of an organism. A space that supports one set of activities for one population may support different activities for another, while changes in the physical environment can also alter the possibilities available. Lerstrup and Konijnendijk van den Bosch's (2017) study of outdoor settings for preschool children provides a useful example. By revisiting Heft's functional taxonomy, they use the categories not simply as a list of properties that environments possess, but as a way of examining the action possibilities that different outdoor settings provide for children. The same environment therefore cannot be understood independently of the population and activity context through which its possibilities are considered.

The taxonomy is therefore better understood as a way of characterising organism-environment relations than as a definitive list of environmental properties. What counts as an affordance can shift with the scale and purpose of the analysis.

Use and uptake of available possibilities

Once these possibilities have been identified, a further empirical question follows:

Which of the possibilities an environment supports are actually taken up in behaviour?

This shifts the research problem from identifying what an environment affords to examining the relationship between available possibilities and their actualisation. An environment may support sitting without people actually sitting. Conversely, people may appropriate an environment in ways that were not anticipated by its designer or researcher. The existence of an affordance and its actualisation are therefore not necessarily the same empirical phenomenon. This provides a useful connection between Heft's functional taxonomy and Kyttä's approach. Heft's taxonomy provides a way of describing the different functional possibilities that an environment makes available to children, while Kyttä (2004) shifts attention towards how many of these possibilities are actually taken up. She uses the number of actualised affordances, together with children's independent mobility, as criteria for evaluating child-friendly environments. The focus therefore moves from the range of possibilities available in an environment to the range of possibilities that children actually realise.

The empirical object consequently becomes the relationship between an identified environmental possibility and its actualisation. Observing an action can provide evidence that a possibility has been taken up, but it does not by itself establish the full set of possibilities supported by the environment. Conversely, identifying a theoretically plausible affordance does not establish that people will actualise it. The distinction is therefore between saying that an environment affords an action and demonstrating that people take up that possibility. The first concerns the organisation of action possibilities within an organism-environment relation; the second concerns their empirical actualisation in behaviour. This makes the Heft-Kyttä approach an important transition in the use of affordance in environmental research: affordance becomes not only a way of describing what an environment makes possible, but also a basis for investigating how those possibilities are realised in everyday activity.

3. How do we make an intended action possibility apparent?

Design & Communication

Norman (1988) use of affordance became influential because designers face a practical problem:

How does someone know what they can do with something?
Should this door be pushed or pulled?
Where should someone enter?
Which control operates which function?

Norman's design-oriented use of affordance shifts attention towards a design practical problem from a designer point of view:

If I want someone to perform a particular action, how can the environment/object communicate or cue that action?

Norman (1999) later distinguished between real affordances and perceived affordances. An object may physically permit an action without making that possibility apparent to the user. Design conventions can therefore shape how available actions are perceived and interpreted. For example, indicating whether a door should be pushed or pulled. For architecture, this distinction is useful precisely because it demonstrates the danger of allowing affordance to absorb every question concerning action and environment.

So which affordance?

Perhaps this is the wrong question. What I am finding is that the different uses of affordance are not competing definitions of the same thing. They often make different aspects of the organism-environment relationship available for investigation. We can ask:

Can an action possibility be specified as a measurable organism-environment relation, and can that relation be directly perceived?
What range of possibilities does an environment support, and which of those possibilities are taken up in activity?
How can a designed environment make a particular action possibility apparent or discoverable?

These questions overlap, but they are not interchangeable. A stair-climbing experiment, an observation of activity in a public space, and an investigation of how possibilities are communicated are not different ways of measuring the same affordance. They involve different objects of study, different forms of evidence, and produce different kinds of knowledge.

This is perhaps the most useful thing I have taken from making this map.

Rather than asking “What is the correct definition of affordance?”, I think it may be more productive to ask:

What question is this particular account of affordance trying to answer?
What can the evidence actually tell us about the environment-organism relationship?

A measurable action boundary might tell us something about whether a particular configuration is compatible with a bodily capability. Research on environmental activity might tell us how possibilities are taken up in everyday use. Research on perceived affordance might tell us how a designed environment communicates intended action.

There are still parts of the map I have deliberately left unresolved. In particular, the questions above still leave open a more fundamental issue:

What kind of thing is an affordance?

Is it best understood as a relational property between organism and environment? A possibility existing in the environment-organism system? Something emerging through situated activity? How do they still relate to physical space?

Those are not further empirical questions about whether an affordance can be perceived or taken up. They begin to challenge the ontology of the affordance itself. (see Ontology 1 and Ontology 2 in the map)

That is where I want to go next. For now, the useful conclusion is not that we have found the affordance. It is that before asking what affordances a building or city provides, we may need to be much more precise about what kind of affordance question we are asking in the first place.

Eric P.Y. Cheung is a system developer and researcher with a background in architecture. He is an ARB-registered architect and holds a PhD in Architecture, with research interests in analytical computation of possibility spaces arising from the relationship between people and the built environment.

Hello. Thank you for stopping by, I hope you have enjoyed your reading! If you have any questions or feedback on this article, please don't hesitate to drop me a line on LinkedIn or via email.

References

Lang, J. (1994). Functionalism. in Carmona, M., & Tiesdell, S. (Eds.). (2007). Urban design reader. Routledge. page 213
Warren, W. H. (1984). Perceiving affordances: visual guidance of stair climbing. Journal of experimental psychology: Human perception and performance, 10(5), 683.
Warren Jr, W. H., & Whang, S. (1987). Visual guidance of walking through apertures: Body-scaled information for affordances. Journal of experimental psychology: human perception and performance, 13(3), 371.
Djebbara, Z., Fich, L. B., & Gramann, K. (2021). The brain dynamics of architectural affordances during transition. Scientific reports, 11(1), 2796.
Heft, H. (1988). Affordances of children's environments: A functional approach to environmental description. Children's environments quarterly, 29-37.
Lerstrup, I., & Konijnendijk van den Bosch, C. (2017). Affordances of outdoor settings for children in preschool: Revisiting heft’s functional taxonomy. Landscape research, 42(1), 47-62.
Kyttä, M. (2004). The extent of children's independent mobility and the number of actualized affordances as criteria for child-friendly environments. Journal of environmental psychology, 24(2), 179-198.
Norman, D. (1988). The design of everyday things (originally published: The psychology of everyday things). The Design of Everyday Things (Originally published: The psychology of everyday things), 20.
Norman, D. (1999). Affordance, conventions, and design. interactions, 6(3), 38-43.