Two Ecological Approaches 30 years later

Abstract:  In 1982, Cutting pointed out that two distinct theoretical programs were developing within the tradition rooted in James J. Gibson’s work on perception: Ecological Psychology. Thirty years later, the two traditions are alive and well. While the distinctions between them are still rarely discussed in print, they have become even more obvious, as the traditions have continued to develop in the directions predicted in Cutting’s paper. Updating the status of the “two ecological perspectives” requires both an assessment of the research generated by the two different perspectives, the theoretical arguments in which the differences are most salient, and the reasons why no rift has formed in the field. Part of the reason ecological psychology has stayed unified seems to be the focus on, and respect for empirical progress. Another important factor seems to be ecological psychologists’ diversification into research areas more concerned with methodological sophistication than with theoretical baggage.

---------------Below are excerpts from a paper I am preparing. The paper is mostly finished. This is a summary, and I would greatly appreciate any suggestions. In particular, I am looking for papers or books that continue to evidence these issues as important to the field - Chemero's book and Andrew and Ken's online debates are clear examples, as well as the explicit debates about afforadnces in Ecological Psychology. However the issues are deeper than that. Any suggestions and comments welcome. ---------------

Two Ecological Approaches 30 years later

(Excerpts from a draft)

Most controversy between ecological psychologists and those outside the ecological tradition revolve around descriptors of the above phenomenon, particularly the notion of “direct perception.” These controversies have often been highly public (e.g. the in-press debates variously including Gibson, Mace, Reed, Shaw, & Turvey on the one hand, Fordor, Gyr, and Pylyshyn on the other). Less public have been the controversies within the ecological psychology community, with the notable exception of James Cutting’s 1982 article “Two Ecological Perspectives: Gibson vs. Shaw and Turvey.”

Robert E. Shaw and Michael T. Turvey were knights triumphant in the first wave of Gibson supporters, crucial in the advancement of the discipline, and extremely influential on the current generation of ecological psychologists. Commenting on the eight-year-old collaboration between Shaw and Turvey, Cutting claims to have “hard evidence that Shaw and Turvey are not systematic followers of Gibson. They are revisionists who present a new ecological perspective.” (p. 200) The differences between Gibson’s perspective and Shaw and Turvey’s perspective were newly emerging at the time that Cutting wrote his article, but his insights were keen. Indeed, the differences are even more extreme and obvious now. Nevertheless, ecological psychology remains a stable, and fairly unified approach. The main implication of Cutting’s analysis, that a paradigmatic split of some sort was coming, seems not to have materialized. After thirty years, it is worth reassessing the two ecological perspectives.

Cutting’s Two Perspectives

Cutting labeled the first approach as that of Gibson, and as the ecological “formulation.” I keep the initial label. The second approach was labeled as the Shaw and Turvey perspective, and as the ecological “reformulation.” Due to their role in founding and leading the University of Connecticut’s Center for the Ecological Study of Perception and Action, it would now be easier to label their perspective the Connecticut approach. This term has the added benefit of being the term used in casual conversation amongst ecological psychologists.

Cutting lists several similarities and distinctions between the Gibson’s perspective and the Connecticut approach. Not all are relevant in comparing the two approaches today, and new distinctions have emerged. The similarities, discussed briefly, include use of metaphor, domain specificity of terminology, and penchant for formalisms. The criticisms, discussed at length, include:

Distinction 1: Necessity of Direct Perception
While Gibson’s perspective argues for direct perception being the normal, or at least natural, state of perception, it allows for indirect perception. In contrast, the Connecticut approach declares all perception to be direct (Cutting, 1982, p. 204). This is not just a definitional problem, but a methodological assertion – the Connecticut approach seems to hold that all situations that seem to involve indirect perception will be revealed to involve direct perception following further analysis.

Distinction 2: Possibility of Perceptual Error
Similarly, for Gibson perceptual error is rare. In contrast, the Connecticut approach declares perceptual errors non-existant (Cutting, 1982, p. 210). This might be an issue of definition, but if so it is unclear whether the definition in question is that of “perception” or that of “error.” It is also unclear what place this leaves in the Connecticut approach for perceptual learning. Perceptual learning is normally presented as a process of becoming better attuned to invariants, which seems to necessitate some degree of initial.

Distinction 3: Multiplication of Terms
Gibson was extremely careful with language, and though he did add terms unique to his system, they were rare. In contrast, the Connecticut approach introduced several new terms, quite rapidly (Cutting, 1982, p. 207-208). While this is not obviously problematic, in and of itself, Cutting’s additional assertion that the novel terms jeopardize the internal consistency of the system is worth further discussion.

Distinction 4: Effectivities
In example of the last point: For Gibson, affordances pointed both ways, but the Connecticut approach adds the term ‘effectivity’, which creates a duality of the type that Gibson seemed to eschew (Cutting, 1982, p.211-213). While Cutting brings up several additional terms of the reformulation that seem to have this flaw, they are no longer common in ecological psychology’s discourses.

Distinction 5: Term Meanings
The above points coalesce in different meaning that the two approaches give to the basic terms of the ecological approach, including “invariants”, “affordances”, and “constraints” (Cutting, 1982, p. 214-216). In each case Cutting asserts that the different meanings are incompatible, and he further claims that the meanings used by the Connecticut approach can lead to uninformative, circular definitions.

Conclusion
In conclusion Cutting notes that Turvey and Shaw proffer a position quite close to Gibson, and that the two approaches “fully deserve to be classified as kindred perspectives on psychology.” (p. 217) None the less, he believes he has done a service if has made more psychologists aware that a difference exists.

It is my belief that what the reformulation says is not always consistent with what Gibson said, and that few people currently recognize this…. It is my belief that both views of Gibson and of Shaw and Turvey should be known, and be known to be different in some important areas. (Cutting, 1982, p. 210, footnote 1)

-------------

Thirty Years Later

[The field has diversified tremendously.] As this diversification and maturation has occurred, the presence of two distinct approaches has remained. If anything, the two approaches have continued to develop in disparate ways, leading to growing awareness of the distinction. Despite this, and even with growing discussion – in print and at conferences – about the alternative approaches, the field has not fissured, and shows no signs of doing so in the immediate future. An update on the distinctions between the perspectives, and an evaluation of the solid base of the discipline is in order.

Reassessment of the Two Perspectives

Distinction 1: Necessity of Direct Perception
Opinions about the necessity of direct perception continue to distinguish the two approaches. It is a typical first step of an ecological research program to pick a behavioral task and try to identify the perceptual support for that task, with the hope that the researcher will find a single mathematically-describable visual (auditory, haptic, etc.) variable that can guide accurate behavior. This step is used both by those following the Connecticut approach and Gibson’s approach. However, the Connecticut approach treats the identification of such variable an absolute necessity for research to continue in a sensible manner, and this is best represented in what might be called Turvey’s dictum: If you have not found the invariant, you have not looked hard enough.

... the Connecticut approach seems to intend the dictum as an ontological and metaphysical axiom: “Given any task, properly conceived, there exists an invariant that is used to support the behavior in question.” ... In contrast, Gibson’s original system both allowed the possibility that some tasks lacked specifying invariants, that is, that the tasks had imperfect perceptual support; and allowed the possibility that some tasks have specifying variables that go unused.   There are two ways to view Turvey’s dictum in relation to the more egalitarian nature Gibson’s original system.... The second possibility, which seems to be favored by the Connecticut school, is view it as axiomatic that sufficient investigation will result in all behaviors being supported by direct perceptual, and thus to view any alternative suggestion as a failure to be sufficiently orthodox.

Distinction 2: Possibility of Perceptual Error
In the converse of the above distinction, the Connecticut school’s requirement that all perception be direct seems to imply that perceptual error never occurs. This is not to say that perceptual error might not seem to occur, but to assert that all such instances can be re-conceptualized to reveal the invariant the organism is sensitive to, and thereby show us how the behavior was guided by direct perception... [But] Gibson, in contrast, regularly wrote about the possibility of perceptual error and the conditions under which it might be expected. This is especially obvious in Gibson’s discussion of restricted perception... [and] is also apparent in the discussion of perceptual learning that characterized Eleanor Gibson’s work.....

Distinction 3: Multiplication of Terms
While it is true that Gibson was very careful in crafting terms, and did so rarely, and that the advent of the Connecticut approach lead to a slew of new terms being introduced, I do not believe this distinction remains. Most of the terms concocted during the great proliferation have gone by the way side, and those that remain are used by members of both approaches. Further, there are now members of both approaches that are very conservative with their terms, and members of both approaches that are very adventurous. Thus, what began as a clear stylistic difference between the leaders of the two approaches, did not become an enduring distinction between the approaches. The only Connecticut-introduced term that remains notably problematic is “effectivities”, and that is primarily problematic because of the way it shapes the Connecticut approach’s understanding of affordances.

Distinction 4: Effectivities
A key term in both approaches is “affordance,” a noun made out of the verb “to afford.” .... Gibson’s approach to psychology defined affordances as a relational property, a higher-order property of the organism-environment system, one that required specification of the situation, the action, and the organism.... In contrast, the Connecticut approach treated affordances as a higher-order property of the object itself – as a combination of object features.... This minor difference led to the creation of a new term, a term for the organism-properties that complement the object properties - effectivities. Effectivities are higher-order properties of organisms...

Disagreements between the two approaches thus often become debates over how to use the term “affordance” and whether or not we should even have the term “effectivities.” ...

Distinction 5: Term Meanings
As discussed above, the two approaches still seem to use some key terms in distinct ways. Indeed, the debates about term usage that continue to be published every few years in Ecological Psychology might lead one to believe that there is a deep internal schism in the field, of the type Cutting predicted might develop. However, the view from within the field is quite different. These issues rarely garner explicit discussion at meetings, and while there are strongly held views, I think most researchers understand that these issues are secondary, and that the agreed upon issues are far more important. Routinely, at conferences, I have witnessed a talk that used terminology in the manner preferred by the Connecticut approaches alternating with talks that used terminology in the manner preferred by Gibson’s approach, and no one listening flinched.

----------------

Looking Forward

The field of Ecological Psychology will continue to go forward, driven by its empirical success in tackling problems of perception and action. The two approaches will be reconciled, now that Turvey and Shaw have retired. As a new generation of energetic people develop further extensions of Gibson’s system, while also trying to return to the original tradition.