I have a Bengal cat who is turning five years old. He’s extremely cute. He’s so cute, clever, and cool that I think he may be the best cat in the world.

Despite my cat’s cuteness, I have the same problem that cat owners around the world have: my very cute cat likes to sharpen his claws on the sofa. I’ve heard that the claws won’t catch on a leather sofa, but what if that didn’t work for my cat? My concern led me to deal with the problem by putting a cover over the sofa that can be scratched at any time.

The strange thing is that my office chair has a similar fabric but is entirely unharmed. I think it may be the difference in shapes, but the dining room chair with a similar shape has also fallen prey to scratching. What is the difference? Is a small difference in fabrics leading to my cat’s decision on whether or not he can sharpen his claws there?

Whenever I look at my tattered sofa, one word comes to mind: “affordance”.

What are Affordances?

Even if you’re not a designer, you may have heard this word before. It was originally a concept proposed by American psychologist J.J. Gibson, and later it was written about by cognitive scientist Donald Norman in his book The Design of Everyday Things and became widely known among UX designers. Affordance is defined in the book as:

The term “affordance” refers to the relationship between a physical object and a person (“relationship” meaning the agent interacting with the object can be anything, including an animal, a person, or even a machine or robot). An affordance is the relationship between the properties of an object and the capabilities of the agent that determine just how the object could possibly be used.

Source: Donald A. Norman, “The Design of Everyday Things: Revised and Expanded Edition – Design Principles of a Cognitive Scientist” (Shin-yo-sha Publishing, 2015)

The important part here is that affordance refers to the relationship between the object and the agent who undertakes the action. Regardless of whether or not it is recognized, the relationship itself exists, and in some cases multiple types can exist. Different affordances can exist for the same object depending on the person.

For example, between me and a chair the affordances of sitting, placing an item, or standing on the chair exist. There is also the affordance of picking up the chair. However, for a small child the affordance of picking up that same chair may not exist.

As it happens, affordance is talked about slightly differently in the initial publication (The Psychology of Everyday Things) and the revised edition (The Design of Everyday Things). In the initial publication, it is explained in a somewhat confused way: objects have a special property which decides how they can be used and informs the relationship between people and objects and their cognition thereof; additionally, design can be used intentionally to convey a message to the user. Consequently, one often sees examples of the latter even to this day, such as: “Adding a shadow tells the user that there is an affordance of being able to push the button.”

Later, the author Norman cleared up any misunderstandings about the term “affordance” in the revised edition and this article. To more clearly explain the concept in the revised edition, he added a new term: “signifier”.

What are Signifiers?

Whereas “affordance” is a word which indicates a relationship which simply exists whether or not it is recognized, “signifier” is a signal which tells us that an affordance is there. In the revised edition, he expresses this as follows.

Signifiers give suggestions about things. In particular, they indicate what actions are possible and how they should be performed. Signifiers must be perceived. When they are not perceived, they do not function well.

Affordances decide which actions are possible. Signifiers convey where those actions should be performed. We need both of them.

Source: Donald A. Norman, “The Design of Everyday Things: Revised and Expanded Edition – Design Principles of a Cognitive Scientist” (Basic Books; Revised edition November 5, 2013)

For example, between a user and a touchscreen there is an affordance of “you can push with your finger.” A designer can then make a visual graphic such as a button and add a shadow to apply a signifier of “you can press here”.

Just as “affordance” is a new application of a psychology term, “signifier” is a term originally used in semantics which was re-purposed to explain a design concept. Norman writes that he used this word slightly differently in a design context, but if signifiers function as signals telling someone about an affordance, then the word is being used in a similar fashion to semantics.

Cats, Affordances, and Signifiers

Let’s return to talking about my cat. Between a cat and a sofa, there are several affordances.

1. Sitting
2. Sharpening claws
3. Jumping up onto and walking along the back of the sofa
Etc.

However, if we consider the relationship between me, a human, and the sofa, (1) exists but (2) does not. I can scratch the sofa if I try, but I can’t sharpen my “claws” (I don’t know if cats are actually sharpening their claws this way). (3) is also just beyond the realm of possibility. If I jumped on the back of the sofa, it would probably tip over from my weight.

Conversely, I have the affordance of:

4. Using the armrest as a pillow (and lying down face-up).

What, then, are the signifiers which signal these affordances? For (1) and (3), it may be the sofa’s size, height, and shape. For (4), the soft material and rounded shape of the armrest seem to be calling me to use it as a pillow. But what about (2)? Is it the material, the height, the shape? Considering that the chair is unharmed despite having the same height and shape, it may be the material. I’d like to conduct a user interview with my cat, but since he will only respond with “meow” we may never know the truth.

Whether intended for communication or not, signifiers are an important communication mechanism for the recipient. Whether a useful signal was placed intentionally or accidentally is not at issue. Distinguishing between the two is unnecessary.

Source: Donald A. Norman, “The Design of Everyday Things: Revised and Expanded Edition – Design Principles of a Cognitive Scientist” (Basic Books; Revised edition November 5, 2013)

As Norman writes, a signifier does not need to be arranged intentionally by a designer to be effective. Although my sofa’s designer did not intend to say to my cat “sharpen your claws here”, by using a specific material in a specific location they did in fact signal to my cat that he can sharpen his claws on the sofa.

Intuitive Signifiers and Learned Signifiers

Most interactions between a cat and an object are done intuitively by the animal. In other words, interactions from the natural world can be found within the house as well, and the cat is simply jumping on things and scratching with his claws without needing to learn anything in particular. No matter how many signifiers are prepared for a cat, it cannot understand a complex interface, such as inserting a key in a keyhole and turning it to open a door (which is how we are able to prevent them from getting into all sorts of trouble).

However, there are some cats which can turn a door handle or open a drawer, and it’s interesting that even animals can pick up on some signifiers. For example, my cat can turn the fan on and off. At first, I thought he had simply sat on it by accident, but looking more closely I saw that he was extending his forepaw and aiming for the button. He would do this when he wanted to attract the attention of a human (me), so perhaps he had learned that pushing the button would elicit a reaction from humans.

Some dogs have been trained far more than my cat, and I wonder just how much they recognize the signifiers prepared by humans.

Once I started considering that question, many more began to surface.

• If the “meaning” is not understood at all (e.g., push here to turn on the fan), then can the signifier be said to have worked?
• Are there any signifiers which humans can recognize without learning, i.e., they are truly intuitive?
• Most examples of signifiers are visual, but what kind of examples could there be of signifiers which use the other four senses?
• Is it possible for someone to have a residual sense of a signifier due to their past experience, even though the affordance no longer exists for some reason (e.g., the acting agent has become older)?
• Are there signifiers which exist only in the digital domain and are not at all metaphorical?
• Could you create a signifier which would only make sense to AI or robots?

What do you think?

Well then, I think I’ll peruse a catalog and look for a sofa which can’t be used for sharpening claws. Anti-affordance…

Special thanks to Yukio Ando