Ecological psychology is radical enough: A reply to radical enactivists

ABSTRACT

Ecological psychology is one of the most influential theories of perception in the embodied, anti-representational, and situated cognitive sciences. However, radical enactivists claim that Gibsonians tend to describe ecological information and its ‘pick up’ in ways that make ecological psychology close to representational theories of perception and cognition. Motivated by worries about the tenability of classical views of informational content and its processing, these authors claim that ecological psychology needs to be “RECtified” so as to explicitly resist representational readings. In this paper, we argue against this call for RECtification. To do so, we offer a detailed analysis of the notion of perceptual information and other related notions such as specificity and meaning, as they are presented in the specialized ecological literature. We defend that these notions, if properly understood, remain free of any representational commitment. Ecological psychology, we conclude, does not need to be RECtified.

Abbreviations

EP = Ecological Psychology REC = Radical Enactivism

KEYWORDS:

• Ecological psychology
• radical enactivism
• information
• affordances
• meaning
• specificity

Acknowledgements

We thank Nick Brancazio, Frank Faries, Paco Calvo, Tony Chemero, Manuel de Pinedo, Daniel D. Hutto, Michael Kirchhoff, and three anonymous reviewers for fruitful suggestions on earlier versions of this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Please note that our claim is not that EP is logically incompatible with a representational account (see Golonka & Wilson, 2019 for a discussion of this possibility). Rather, our claim is that such an interpretation is not implied by the theory itself. Thanks to an anonymous reviewer for the pointer.

2. Hutto and Myin use “basic cognition” to refer to the cognitive capacities that are shared across species and that are both ontogenetically and phylogenetically prior to sociocultural scaffolding. These basic cognitive abilities include “central forms of human cognition, such as perceiving, imagining, and remembering both in children and adults” (2017, p. 90).

3. In what follows, we will stick to this notion of content. We argue that the kind of information that EP invokes does not include such features.

4. The phenomenon of all reflections and refractions of light on the surfaces of the environment generates what Gibson dubs the “ambient optic array” (see Gibson, 1979/2015, Chapter, p. 5).

5. Hutto and Myin (2013, p. 66) borrow this example from Jacob (1997, p. 45), and present it as a genuine instance of covariant information (for subsequent uses of this example see Hutto & Myin, 2017, p. 30). An anonymous reviewer, by contrast, points out that covariant information is more technically referred to as Shannon information (Shannon, 1948), whereas this example refers to what Grice dubs “natural meaning” (1957). We thank the reviewer for the pointer. However, we prefer to remain neutral with regards to this discussion and to describe the notion as Hutto and Myin present it.

6. Although Hutto and Myin present the Hard Problem of Content as a general problem for representationalism in cognitive science, they examine different possible options to solve it. According to them, the most promising strategy is Millikan’s teleosemantics (Millikan, 1984, 2005), which appeals to biological teleofunctions as derived from natural selection to explain content. The advantage of this approach is that it does not commit to the existence of representational contents in nature that need to be gathered via the senses. Instead, for teleosemanticists, content-related properties are (partly) given by the interpretive activity of the system. The guiding idea of teleosemantics is that a device or an internal state S has the teleofunction of representing X if it is used (interpreted, consumed) by the system because S has the proper function of indicating the presence of X. This, however, is not enough to naturalize content. As Hutto and Myin argue, even though appealing to biological teleofunctions can serve to describe biological norms, they do not suffice to account for the kind of normativity implied by content – such as the property of misrepresenting X or being wrong regarding X. For instance, determining that a given state S in the visual system of frogs has the function of indicating the presence of flies does not suffice to determine under what description this state represents flies – that is, whether it represents them as “flies,” “moving dots,” “food,” and so on. It follows that “even if we can specify what is meant to be targeted [by the state S] that would give us no reason to think that the targeted item [the fly] is represented in a truth-conditional, referential, or otherwise semantic way” (2013, p. 80).

7. As van Dijk et al. (2015) explain, “they worried that the ecological notion of information that grounded information in the correspondence between the structure of ambient light and the environment, required too little participation of the animal. Environmental correspondence alone could not guarantee that ecological information allowed for perception without mediating content, and the process of information pick up did not alleviate trafficking content” (p. 211).

8. According to Baggs and Chemero (2019), to the distinction between physical world and environment we must add the distinction between environment and “umwelt.” As they point out, “we need to make a further distinction, between: the environment as it exists for a particular member of a species, a habitat; and the environment as it exists for a particular living animal, an umwelt” (p. 7, emphasis original).

9. “A sharp distinction will be made between the ambient array at an unoccupied point of observation and the array at a point that is occupied by an observer, human or other. When the position becomes occupied, something very interesting happens to the ambient array: it contains information about the body of the observer” (Gibson, 1979/2015, p. 59, emphasis added).

10. To explain the difference between the notion of information invoked by cognitivism and that of EP, some Gibsonians appeal to Runeson’s metaphor of a “smart perceptual device” (Bingham, Schmidt, & Rosenblum, 1989; Michaels & Palatinus, 2014; Runeson, 1977; Zhu & Bingham, 2008). Runeson offered the polar planimeter as a metaphor to make the idea of smart perceptual devices more intuitive. The main difference with the cognitivist notion of information is that, in contrast with cognitivist explanations of perception, a smart device detects higher-order information; it does not detect stimuli to be processed and enriched. In the same way that the design of the planimeter allows it to measure the area of a plane through moving over the perimeter, living beings are expected to detect information through acting upon the environment instead of doing mental calculus.

11. It is worth mentioning that the notion of law as used by Gibsonians must also be understood at the scale of the O–E systems. As Warren (2005) puts it, “ecological information is lawful not in the Newtonian sense of being universal in space and time, but in an ecological sense of being regular within an ecological context or constraint” (pp. 242–243).

12. We are indebted to an anonymous reviewer for calling our attention to this quote.

13. “Ecological information cannot be transmitted: it is ambient and available, not something put over a channel; it is something to be detected or used (or not) in regulating action…Information pick up is not a process of “internalizing” information” (Reed, 1996, p. 155).

14. For more examples of sensory substitution devices applied to research in ecological psychology, see Lobo, Travieso, Barrientos, and Jacobs (2014), Lobo et al. (2018).

15. It is worth mentioning that some Gibsonians have proposed that the concept of information needs to be expanded so as to include variables that do not relate to the environment in a lawful (1:1) manner (Chemero, 2009; Golonka, 2015). According to this idea, non-specifying variables, that is, variables that are contingent on conventions or reliable enough regularities, can support direct perception as well. Whether or not this hypothesis is tenable is not an issue we can address in this paper. Our claim, instead, is that the risk of conflating specifying information with information-as-content is not a good reason to abandon specificity.

16. Costall (2012) has coined the notion of “use-meaning” to distinguish the ecological notion of ‘meaning’ from semantic meaning and the meaning associated with representations.

Additional information

Funding

Manuel Heras-Escribano has been funded thanks to a Juan de la Cierva Postdoctoral Fellowship granted by the Spanish Ministry of Science, a 2018 Leonardo Grant for Researchers and Cultural Creators, BBVA Foundation (The Foundation accepts no responsibility for the opinions, statements, and contents included in the project and/or the results thereof, which are entirely the responsibility of the authors), the Project  [FFI2016-80088-P] granted by the Spanish Ministry of Science, and the FiloLab Group of Excellence granted by the Universidad de Granada, Spain.

Notes on contributors

Miguel Segundo-Ortin

Miguel Segundo-Ortin is a Ph.D. student at the University of Wollongong, Australia.

Manuel Heras-Escribano

Manuel Heras-Escribano is Juan de la Cierva Formación Research Fellow at the University of the Basque Country, Spain.

Vicente Raja

Vicente Raja is a Postdoctoral Fellow at the Rotman Institute of Philosophy at Western University, Canada. He is also a member of the MINT Lab at University of Murcia, Spain.