Perceiving Actively: A Reflection on a Sensorimotor Construction

Justin Broackes is THE expert on the history of colour science—alright, John Mollon is up there too!—and something of a bibliophile. Cruising the Internet one day, he was delighted to chance upon this book by a famous 19^th century Swedish physiologist.  (The author, Frithiof Holmgren, advocated testing for colour vision, and campaigned to exclude the colour blind from signalling work on the railways.)

 

Some time after it had arrived from the Netherlands, Justin proudly displayed the book to dinner guests, who naturally enough remarked on the striking cover design. Justin is somewhat colour-blind (an anomalous trichromat) and he is apt to confuse certain reds and greens. He hadn’t noticed the nifty colour alternation above: the letters just looked brown. But in light of his guests’ compliments, he examined it carefully. Then—but only slowly, after a couple of minutes of looking closely at the book from different angles under a desk-lamp—he saw it. The alternating letters finally looked red and green.

But wait! Justin is colour-blind. How did he manage to see the colours?

There is a useful truism in the psychology of vision: Information that is unavailable to a subject at a moment may become available with movement and activity. Three-dimensional shapes are a perfect example. An immobilized person can’t tell whether something is a disc or a sphere. But if she walks around it (or for close-by objects, if she just moves her head) it becomes obvious. Externalization is another example. When you bump against a nail sticking out of a wall, you feel a stinging pain, but you don’t feel the nail as such. But if you probe and poke and feel the nail with your finger, you perceive it, not your body—you sense that it is pointy and sharp, objectively so. (Note that pointy and sharp are not qualities you can attribute to the part of your body that is in contact with the nail.) As Broackes writes, “there are things one can identify by touch in the dark but only by moving and rolling them around in one’s hand.” (J. J. Gibson’s 1962 paper “Observations on Active Touch” systematizes this.)

Broackes experiences colour by actively perceiving it. Green things have a different brightness profile than red things. Reddish light (such as that from an incandescent bulb or the setting sun) brightens the facing side of a red thing more than that of a green thing. Consequently, the brightness difference between the facing side and the away side is greater for the red object. This is the kind of cue that colour-blind people exploit when they perceive colour. They look at shaded as well as illuminated parts of things; they look at things from different angles. Colour-blind people can perceive colour, but they have to do so actively. Normally sighted people view 3-d shapes in the same way; they are shape-blind, one might say, when immobilized. Touch and the other senses probe the externality of certain stimuli similarly: it is by stroking, probing, palpating, and feeling that we discern the external causes of touch-sensations.

Broackes makes a bold move here: “Might the ‘sensation’ of red not occur, not merely in response to a momentary stimulus, but in response to a suitable pattern of stimuli?” This is an extremely interesting speculation about sensations of colour, very much in line with the above observations on active perception. But Broackes, and Alva Noë following him, converts it into a sensorimotor speculation about colour. Noë writes: “The colour of a surface is its disposition to change its appearance as relevant viewing conditions change.” This recalls Maurice Merleau-Ponty’s infinitely elegant and seductive remark: “Smoothness is how a surface utilizes the time occupied by our tactile exploration or modulates the movement of the hand.”

The leap that Broackes and Noë make is perhaps more apparent when it comes to shape. (Noë doesn’t distinguish in this respect between primary and secondary qualities.) I said earlier that we perceive three-dimensional shapes by moving around and looking at things from different angles. Following Broackes, one could conclude that the sensation of shape is a response to a suitable pattern of stimuli. Question: Does this make it plausible that shape is itself a disposition to change appearance as viewing angles change? Surely not. We understand a cube as it really is—the 90° angles and equal edges that frame it. We use its appearance profile—the way it changes its appearance as we view it from different angles—to verify that the thing we see is in fact this way.

There is a difference between saying that we recognize something as F by doing A, and saying that our conception of F is simply that of doing A. Active perceiving is sensorimotor. What we actively perceive is not. This complements my message from last week. The visual guidance of motion is sensorimotor. But the senses are not just for the visual guidance of motion.