In the last post I introduced the concept of affordances. I will admit that my take on the concept is somewhat idiosyncratic. In the design of user interfaces (my favorite book on this topic is still The Design of Everyday Things by Donald Norman) the concept is mainly restricted to perceptions of function: door handles are for pulling (not pushing), door panels are for pushing (not pulling), and raising an electric window makes more “sense” when the switch is pushed forward rather than pulled backward. Having started using computers prior to touchscreens in “makes sense” to me that pulling down with my mouse will move me forward in a document. For my students pushing up is the more natural way of moving forward in a document. In other words, those who design objects for various functions do well to be aware of the responses that are typically occasioned by the perception of the user. We shouldn’t need a user’s manual for every on of the tools in our lives!
The active design of objects, though, also changes the affordances of a situation. Consider the design of the “bendy straw.” The illustration above (Patent #: US2094268 A) is from 1937, and was granted to Joseph B. Friedman. Mr. Friedman’s journey began with taking his young daughter to soda fountains. At the time milkshakes and other drinks were generally served out of tall bottles and / or glasses. Since straws were straight they elevated where drinking occurred. For a young drinker, tipping a glass filled with milkshake in order to get the straw to the mouth sometimes caused milkshake to spill out of the glass. However, bending the straw to the mouth was also not a solution. As Friedman writes in his patent application:
In order to drink in any satisfactory manner through the straw, it is either necessary to take the glass or bottle in the hand and tilt it, or to lean forward and bend the face down to contact the straw with the mouth. Either of these procedures is awkward and it is extremely common for the straws to become bent and broken over the rim of the bottle or glass during the consumption of the liquid. As the straws are substantially cylindrical such bending greatly obstructs the lumen of the straw and reduces the amount of liquid which can be drawn there through. (Friedman 1936, US patent application 2094268).
Friedman’s “aha” moment involved taking one of the paper straws of the day, inserting a screw, and tightly winding a string around the screw’s ridges. After removing the screw, one was left with a bendable straw. Mr. Friedman had intended that his bendy straw be used at soda fountains. One of the earliest adaptors, though, were hospitals because it became evident that bendable straws enabled bed-ridden patients to drink liquids while lying in bed.
However, rather than focusing on technology design, the psychologist who first proposed affordances as a term, James Gibson, mainly used the concept to capture a sense of what the external environment (i.e., its perception by the organism) “offers” the organism (see an excerpt here and Wikipedia here). Actions connected with threat and play, drowning and swimming, bathing, splashing, drinking, cooling off… all might be “afforded” to an individual by the sight of a pond. Gibson’s point was to stress that an animal is rarely, if ever, passively situated in its environment, and that action is a fundamental component of perception. As the environment changes so too do the potential behaviors.
My own use of the term in the last post leaned more in the design direction, while emphasizing that our evolutionary history, learning and maturation layer our perceptions with actionable meaning. But importantly, I tried to make the point that what is obviously the “correct” course of action – at least in terms of the most likely behavior to appear in a situation — might in fact be akin to an illusion. There are instances in which our perceptions conspire like a classical artist to restrict our attention to what resides within the painting’s ornate frame.
Given that brief aside, what I would like to do in this post is to consider the affordances of emotions and feelings. Just as we should be aware of and question the perceptual frame that weights our actions with regard to the external environment, so too should we question the perceptual frame that weights our actions with regards to our interoceptive environment. Notice that I am NOT referring to the affective affordances of external objects (see Carava & Benenti, 2024). I might feel the need to wallow in bout of melancholy romanticism and turn to the movie Before Sunset to help me experience this emotion. Or I might doom scroll through social media as a way of mitigating anxiety. Such affective affordances are not what I am referring to here. Rather, I would like to consider interoceptive feelings as their own perceptions that are capable of affording particular actions in their own right.
My stance will be the same as that taken in the last post – to consider interoceptive perceptions as differentially weighting different types of action. But also, I will want to hold on to a very broad definition of afforded actions. Just as an emotional response (fear of failure, fear of exposure, etc.) is an affordance that might be “suggested” to you by the prospect of having to speak up in class, our recall of particular memories and interpretative stances are also actions. Consider Duchamp’s urinal above. Part of the fame attached to this work of art concerns the interpretive expectations triggered by our perceptions. A Michelangelo sculpture affords a certain esthetic reverence. A urinal affords thoughts of contamination and excrement. Duchamp’s insight was to purposefully conflate the two. In considering the potential affordances of our visceral beliefs we will likewise attempt to hold on to an unrestricted notion of the potential action space.
For example, we all know that social behaviors such as communication are actions afforded to us by our interoceptive cues. A doctor asks us to report how much pain we are feeling on a scale of 1 – 10, or a friend asks us how we are feeling. “About a 7” and “Tired” might be the responses that our interoceptive perception affords. But of course, feeling possess more than just verbal affordances. Happiness affords smiles and anger affords scowls, and an infant cries in order to be held or because of gas pains or because of hunger (and it is up to the exasperated parent to determine which is which). In fact, we spend considerable time trying to train other behaviors to our internal perceptions in order to hide and control how and when we express our internal states. Since communicative actions are so obviously suggested by our interoceptive perceptions, I am not going to address them here. Instead, I would like consider three other, not so obvious actions afforded to us by our interoceptive perceptions: behavior systems, state-dependent memory, and interpretive projection.
Let’s begin just as we did with the last post with a puzzle. A rat in a cage is given a running wheel. Anyone who has had a pet rodent knows what happens. The rat will spend some time each day scurrying in the running wheel. What we are going to do, though, is run an experiment in which we force the animal to go hungry for some amount of time each day. Maybe we provide food for the animal for 4 hours per day, or maybe 2 hours per day, or maybe just 1 hour per day. Being good scientists, we have two groups of rats. For one group we do not provide a running wheel, and what we will observe is that all of the animals manage just fine. Those provided with food for only an hour per day, spend the majority of that time eating, while those given access to the food for 4 hours, will eat intermittently during those 4 hours. The important thing to note, though, is that all of these rats will maintain a stable body weight.
For our other group, we make one change. We provide these rats with a running wheel. What we will see in this group is a phenomenon known as activity-based anorexia (see Carera et al, 2014). First, greater food restriction will lead to greater time spent in the running wheel. In the graph below, the rats who are receiving 1 hour of food per day are running 10 km in their running wheels! What has also been shown, though, is that the rats in the 1 hour of food per day, will cease to maintain a stable body weight. In fact, if the experiment is not stopped, many of these rats will continuously lose weight, and eventually run themselves to death. How do we explain this?
One potential approach is via what is known as behavior systems theory (see Carbrera et al, 2019) Behavior systems can be thought of as the loose set of reflexes that are attached to functional states such as foraging, reproduction, status seeking, the rearing of young, and so forth. The figure below shows a set of behavior systems for foraging. Depending on the animal, the individual might start a bout of foraging by socializing before setting out with the group to look for food. Or the animal might search out a spot to lie in wait, hidden. Once a food item is located, though, a different set of behaviors come to the fore, and once a food item is being consumed, yet another set of behaviors become more likely. An astute observer will notice that behavior systems are simply affordances by another name.
So, let’s consider what perceptions might underpin the functional state that affords a particular behavior system. The sight of prey. The feel and smell of the food item, for sure. But what of the sensation of hunger, itself? To be placed into a functional state is to exist within a particular interoceptive perceptual space, and it is this perceptual space that is affording the behaviors in a behavioral system. Bucks during rutting season compete with other males and attempt to copulate with females. Horny teenagers congregate in packs and compete for status and attention. A child in a dark basement hears a sound and flees for the safety of the lighted steps. In each case, a functional state presumably correlates with a feeling state, and this feeling state affords actions just as much as the sight of a tea pot.
For example, from the figure above, an animal experiencing hunger is most likely to show general search behaviors. And (if you are rat) that means movement. Normally, these behaviors would cause prey or food to appear, and then normally the exteroceptive perception of this prey would lead to consumption behaviors. But in the laboratory, our poor animal is stuck. We have removed the normal contingencies of the behavioral system. The hungry animal increases its locomotion, but this does not produce a food item, and only serves to increase the interoceptive hunger cues, which in turn “suggest” the behavior of locomotion. It’s a vicious cycle, but one that is at least partially made sense of by appreciating that interoceptive perceptions (in this case, hunger) afford particular sorts of behavior just as much as exteroceptive perceptions do. It’s just that for whatever reason we tend to change our terminology. We don’t say that the hammer hitting our knee compelled us to kick out. But we do say that we felt compelled to eat the cupcake that was in front of us. The only difference, at least from within our discussion of affordances, is that in one case the perception is external while in the other the perception is internal. In both cases a behavior or set of behaviors is being strongly weighted (or primed) by the perceived situation.
So, when we speak of interoceptive “feeling states” (e.g., emotions, motivations) we should be aware that systems of behaviors are likely to be afforded to these internal “situations.” We feel guilt and behaviors that involve making amends will be suggested. We feel anger and behaviors that remove obstacles or exact payment will be suggested. We feel disgust and behaviors that lessen the likelihood of contamination will be suggested. We feel love and behaviors that care for the other will be suggested.
[Aside. I was educated in the behaviorist tradition, and therefore feel somewhat “dirty” using terminology such as “affordances” and “feeling states” and so forth. However, there are many flavors of behaviorism and I am much more aligned with the behaviorism of Clark Hull than B.F. Skinner. Hull made much use of “drive stimuli” in many of his models of learning, and what is a “drive stimulus” but an interoceptive perception / feeling state? Also, contra Skinner, I don’t think that operant conditioning is the be-all-and-end-all for understanding a behaving organism. For a nice summary of some of the current approaches to decision making and behavior see Redish, 2015). Finally, a crude behaviorism has a way of absenting the particular organism from the explanatory situation. Affordances to my thinking are more species specific, and do a better job of conceptually situating the individual organism in their immediate, perceived environment (internal and external).]