The Relationship Between Stimulation and Stimulus Information (Gibson, 1979, Chapter 4)

The first three chapters describe the world to be perceived, at the ecological scale. This chapter opens the section on how we visually perceive that world - via information. Remember, a key part of the ecological analysis is doing things in this order (see the Introduction). 

The work of the next few chapters is to lay out a theory of ecological optics. Lots of science studies light, but as with the physical world/environment distinction, Gibson will insist on distinguishing between the physics of light and the ecology of it; only the latter will be relevant to a theory of perception. 

Sabrina also blogged this chapter here. 

Gibson does the usual thing of making some key distinctions. 

The first is between a luminous body and an illuminated body. The former emits light, the latter reflects it - but we can see both. Ecologically they are very distinct, because illuminated bodies are typically textured surfaces that structures the light they reflect. 

This leads to the second distinction, between radiation and illumination. The former is how physics treats light. Illumination is what happens when radiation strikes a surface and undergoes scatter reflection. This structured light fills spaces and illuminates surfaces not directly in the light from the source - it reverberates through the medium, reflects off multiple surfaces, and completely surrounds every possible point of observation. This makes it ambient light, vs radiant light. 

Ambient light quickly reaches a steady state, but this state shifts with any changes to the amount of light in the environment. There is a lot of variation in the specific intensities coming from all directions to a potential point of observation, although the higher-order structure remains the same. Specifically, that structure is a nested set of solid angles - more on this invariance-over-transformation later. The fact that these angles are different depending on where they came from means that we need to consider the light around a point of observation as an ambient optic array. 

Gibson now distinguishes between stimulation and stimulus information. The former is a physiological concept, and is the energy striking a receptor. If the amount of stimulation crosses a threshold, the receptor reacts (transducing the energy from light to neural impulses). But this is not sufficient for perception; you can have light without structure (e.g. whiteout conditions) but there is no information in this set up and so vision fails. Receptors can be stimulated by light, but perceptual systems must be activated by information. Gibson then discusses the idea that we do not see light, per se; we see the environment (see Turvey, Lecture 11). Situations where we experience light (e.g. being dazzled by the sun) are actually us experiencing things about our bodies (that the visual system is overstimulated). 

Stimulation cannot be the basis for perception, because it has features that perception doesn't. It's not specific to it's source (contrary to Helmholtz) - retinas typically react to light but can react to mechanical stimulation. Stimulation is temporary, and receptors adapt and stop reacting to continued stimulation - but the environment is perceived as persisting. Stimulation involves thresholds, but perception does not have thresholds; we do not perceive or not as a function of amount of information, but as a function of our learning and development and motivation, amongst other things. 

So, perception does not begin with stimulation, nor the formation of a retinal image out of that stimulation. Gibson discusses the notion of the retinal image but notes the common problems; an infinite regress of homunculi to look at them, the fact you cannot build an image that contains what perception requires, and the fact that a whole class of animals (insects) have eyes that cannot form images but that still allow the animals to visually perceive their environments. 

What is the beginning of perception then? Optical information. The details will come in later chapters, but for now Gibson notes what optical information is not. It is not information in the Shannon, communication sense. It is not transmitted by a sender to a receiver, and it is not encoded nor decoded. It is not like human-made acts of communication (writing, images etc) which are a special case involving mediation by the first observer. Optical information will be very different from this.