Minds on Monday: Sense and Un-sense in the Animal Kingdom

“The senses, or sensory modalities, constitute the different ways we have of perceiving the world.” So writes Fiona Macpherson in the Introduction to her new anthology, The Senses, a visionary work, which will spark a whole new literature on individuating the senses. At first glance, you might think “perceive” is unhelpful: the connection between “sensing” and “perceiving” seems too close and too porous for one to illuminate the other. Yet, there is a crucially important point here. Our bodies monitor blood oxygen levels and control breathing accordingly; this is a response to the (internal) environment. Yet, we do not perceive blood-oxygen levels. And this, presumably, is at least part of what Macpherson is getting at. But what does it mean? Colour perception is a response to the (external) environment, after all. Why is it right to say we perceive colour, but not blood-oxygen levels? (The internal/external distinction is unimportant, by the way: hunger and thirst are perceptions, aren’t they?)

Many functional responses to this problem are, at first sight, unhelpful. In a beautiful and deservedly influential paper, Brian Keeley offers us the following: a sense modality is an organ historically dedicated to “facilitating behaviour with respect to an identifiable physical class of energy.” Keeley explains “historically dedicated” by contrasting electric fish and humans with regard to electric current. “Humans are easily capable of discriminating fully charged nine-volt batteries from “dead” ones, simply by sticking them to the tongue,” he writes—the charged batteries produce a characteristic sensation. But humans don’t do with electricity anything like sharks do: produce a map of the surrounding electric field rich enough in detail to detect and track prey. Sharks are capable of electroreception, Keeley says, because they have a sensory data-processing system dedicated to electric fields. We just detect electric currents in an undifferentiated and behaviourally irrelevant way. Sharks have an electric sense; we don’t. This seems right, but it doesn’t help us with the blood-oxygen example. The data-processing is dedicated and the response specific.

Some ways of defining the senses focus just on humans (the “we” of Macpherson’s obiter dictum). One might observe, for instance, that colour perception gives the subject reason to believe that things are coloured, whereas blood-oxygen monitoring controls breathing autonomously without giving reason for belief. This is important. Take Keeley’s example of the vomeronasal “sense”: most humans can apparently determine the sex of other humans by smelling their breath. But since they don’t even know that they have this ability, they certainly cannot form a reasoned belief on this basis. This makes a good distinction—standard seeing comes out on the perceiving side, as Macpherson wants; the vomeronasal system, blindsight, and blood oxygen levels come out on the other. Intuitively, a pretty good result—though certainly contestable. But it doesn’t help us with animals. As Aristotle said, all animals have senses. But clearly not all animals form reasoned beliefs. Sharks don’t; yet they have sense modalities. The shark electric system ought to be a candidate sense modality. What would let it in?

In cases like this, an evolutionary perspective has merit. We have a human-based distinction, but it would be objectionably anthropocentric either to apply it to animals generally or to deny that animals sense because they don’t conform to humans. In such a case, an appeal to comparative animal studies is helpful. They help us identify a functional continuum that generalizes the human-based distinction. This can be applied to animals, though not in a binary way.

Here is one suggestion. The plasticity of information-governed behaviour is a key evolutionary innovation. In higher animals, plasticity is evident because context determines how information results in behaviour. A well-trained dog may see food, but not eat it until permitted to do so by its trainer. And even lower animals move toward sensed sources of food in ways modulated both by their own state of nourishment and by obstacles in their paths.  Human response to perception is not just context-relative but also reasoned. Visually, it may seem that some object in my vicinity is blue (say), but knowing that the illumination is unusual, I may “withhold assent” from this determination of the visual sense. In all of these cases, the information-capture facility doesn’t directly account for action. Rather, it produces an output state that is available to interact with other bodily states to produce action. It is convenient to regard such an output state as the bearer of “descriptive content” concerning the organism’s (external and internal) environment. 

Here is a second suggestion. The senses modify response patterns through conditioning. But information-capture for regulatory purposes does not feed into conditioning. A person might go running at 7 most mornings, and thus she might lower her blood oxygen levels and begin to breathe hard. However, she would not, on a morning when she does not run, breathe hard at 7 am, however conditioned she might be to running at 7. By contrast, if she eats every morning just when the coffee is brewing, her stomach will growl at the smell of coffee, even if she has decided to fast. Her food-anticipating reaction gets attached to the smell of coffee by conditioning. We may take it as a mark of a sense that it supports learning in this way. The information that regulates breathing is not sensory, on this way of looking at things, but the smell of the coffee and the sense of time are.

Two suggestions, one bottom line: the senses are information-capture facilities that can be used flexibly. Evolution throws light on how flexibility develops.