Lecture 20: Ecology: The Science that Reasons Why (Turvey, 2019, Lectures on Perception)

Turvey has finally arrived at the ecological approach, but all the preceding chapters have made it such that there are preliminaries to deal with first. This first Lecture sets up a very broad, all-encompassing motivation for an ecological approach, summarised in the first sentence: "A theory of perception ought to be a theory for all organisms.". Turvey will spend some time identifying that previous work has focused on humans, we are one of a countless number of organisms and not especially representative, and that the why of perception-action should be rooted in the laws of thermodynamics. 

Turvey frames the challenge like this: ecology is "the science that reasons why", and there are many why questions. Why are there so many forms of life? Why do they come in discrete sizes? His answer is 'thermodynamics'. 

Natural processes increase entropy (they have to, thanks to the 2nd Law of Thermodynamics). This means the processes take 'free energy' (energy available to be used, not the FEP concept) and convert it inti less useful forms (heat). Processes vary in how well they do this, and processes that do this well are favoured (again, to conform with the 2nd Law). The world has many sources of energy that need to be processed (habitats) and within those, many ways to do the processing (niches). Life looks like it does because it is in the business of entropy at all the available scales and with all the available sources of energy. InformationL (lawful, specifying ecological information) is going to play a key role here because it opens up access to opportunities to dissipate energy that would otherwise be unavailable. Perception lets life dissipate energy with more than just the world it is in physical contact with, and once opened up these opportunities will inexorably be exploited leading to all the variety of life that we see.

All living things share a key property - agency. Turvey (p. 305) defines it as follows:

Agency encompasses (a) variation of means to bring about an end (flexibility), (b) coordinating current control with upcoming states of affairs (prospectivity), and (c) coordinating current control with preceding states of affairs (retrospectivity)

He gives them names that describe what they do, but that do not yet entail a theory about how these are achieved. He notes that agency can't come from evolution and natural selection, which only work because they presume agency exists. Agency will also need a thermodynamic origin story. 

(ADW note: In traditional schemes, prospectivity is accomplished by prediction, retrospectivity is accomplished by memory. Traditional researchers tend to claim that prediction and memory are necessary features of cognition, but actually they are just one way to accomplish what is actually necessary, namely prospectivity and retrospectivity). 

Turvey ends by defining ecological realism as "realism defined at nature's ecological scale in a niche-specific way". This kind of follows from the preceding discussion: perception and action must be understood at the scale of the organism in question, and must be consistent with the physical demands of the laws of thermodynamics. There can be no dualism at all between mind and body. Of course, Cartesian approaches all have a dualism at their heart, so we need a radically new approach; the ecological approach. 

I'll note that this analysis is very specific to Turvey and his colleagues. A key part of his later work was trying to make perception-action emerge as inevitable from the laws of thermodynamics, so that things like agency and intentionality could be accounted for without any dualism of mind separate from physical world. It is quite a detailed analysis, and a very powerful approach, but the work is far from complete and not everyone in the ecological community feels it's the only way to do ecological psychology (see Bruineberg et al, 2024).