Report on the Extended Cognition workshop

Here is a short report on the Extended Cognition Workshop, which just took place over the last days in Amsterdam. The general goal of the workshop was to bring together people interested in the concept of extended cognition in the spirit of ‘second wave EM’ (EM: extended mind) (Sutton) and ‘cognitive integration’ (Menary). The main characteristic of this general approach is emphasis on what is often described as the complementarity principle:

In extended cognitive systems, external states and processes need not mimic or replicate the formats, dynamics, or functions of inner states and processes. Rather, different components of the overall (enduring or temporary) system can play quite different roles and have different properties while coupling in collective and complementary contributions to flexible thinking and acting. (Sutton, ‘Exograms and Interdisciplinarity’)

The contrast is with versions of EM which emphasize the parity principle:

Cognitive states and processes extend beyond the brain and into the (external) world when the relevant parts of the world function in the same way as do unquestionably cognitive processes in the head.(Sutton, ‘Exograms and Interdisciplinarity’)

Most of the debates on the topic of the extended mind and extended cognition have revolved around the Adams & Aizawa objections to the very idea of minds going beyond the intracranial realm, and the replies offered by defenders of the idea. While this is undoubtedly an important debate, it seemed to me that it could be productive not to focus specifically on this debate for once; the goal was thus to gather people who all share a belief in the basic premises of the notion of extended cognition, and in particular those emphasizing complementarity rather than parity. Indeed, parity is particularly important within the bounds of the mind debate, but less so if what one is interested (as I am) in is the general impact of using external cognitive artefacts on cognitive processes.

Moreover, rather than as a thesis, ‘second wave EM’ can also be seen as a general framework allowing for the analysis of phenomena which might otherwise remain difficult to explain (such as e.g. the impact of specific properties of mathematical notations for the development of mathematical theories). It seems to me that putting the framework to use in specific cases may be one of the best ways to argue for its cogency and fruitfulness, and at least as far as I am concerned, the workshop of the last two days provided very convincing evidence in this direction.

We started with Richard Menary on ‘cognitive transformations’. Richard spelled out the general framework of his concept of cognitive integration, emphasizing some of its key elements such as cognitive niches and cognitive norms, to then address the transformative power that operating with external devices can have on the mind itself. So rather than just having the mind leaking into the world, in Clark’s famous formulation of the thesis, we also have the world ‘invading the mind’. We then had Erik Myin, arguing that, as long as EM’ers, second wave or not, still rely on the external vs. internal dichotomy, there is still too much reliance on parity; a radical formulation of second wave EM would also need to break free from this dichotomy. He prefers to speak of ‘extensive cognition’ rather than ‘extended cognition’. After Erik we had Mirko Farina, arguing that attention to the diachronic processes involved in how we deal with external cognitive artefacts is crucial; he thus makes a plea for more focus on developmental aspects within the extended cognition framework. The last talk of the day was by me; I offered an analysis of the cognitive impact of using formal languages and formalisms in general from an extended cognition point of view. I argued that manipulating formalisms as a form of bodily engagement is an integral part of the reasoning processes in question, and that it truly modifies these processes when compared to reasoning without the aid of formalisms or with different formalisms.

The Tuesday began with John Protevi talking about killing from a philosophical point of view, and pointing out how, given that the vast majority of human beings are unable to perform cold-blood killing ‘in natura’, a wide range of devices are deployed (e.g. in army training) in order to make some people capable of such deeds. He further discussed the important ethical implications of this extended perspective on killing. We then continued with Bryce Huebner pursuing the line initiated by Protevi, and arguing that the very concepts of moral judgment and moral responsibility needs to be revised in light of the experimentally detected influence of the environment in our moral judgments (e.g. when holding a cup of warm coffee, we tend to form a more favorable impression of people than when holding a cup of cold coffee). Joel Krueger then focused on the infant-caregiver relationship to argue that the coupling of infant and caregiver truly constitutes one integrated cognitive and affective system, in particular in that infants are not able to control their attention to external stimuli; the caregiver’s guidance is what determines the focus of their attention in their first weeks of life.

After lunch, the first talk was by Helen de Cruz, offering an analysis of the role of mathematical notation in mathematics from the point of view of extended cognition. She presented a case study on early Chinese algebra, arguing that Chinese mathematicians discovered a precursor of the method of Gaussian elimination, specifically prompted by the particular perceptual properties of the counting rods they used to calculate. We then made a huge leap from Chinese ancient mathematics to the 21^st century, with a talk on Twitter (yes, Twitter!) by Jurgis Skilters. He argued that the extended cognition framework can be applied to social networks such as Twitter to explain phenomena such as what he described as ‘extended selves’. The next talk was by Pierre Steiner, who argued that the idea of extended cognition is simply incompatible with a representationalist account of mental processes; basically, if thinking is done ‘in the world’, the idea of internal representations becomes at best superfluous but possibly even incoherent. The last talk of the workshop was by Julian Kiverstein, arguing that Clark’s conception of extended cognition is indeed overly individualistic, as some recent critics have suggested; Julian contented (against Clark) that cognition is not only not organism-bound, but it should not be viewed as organism-centered either. He suggested that a Heideggerian Dasein perspective could represent the missing element towards the overcoming of individualistic conceptions of extended cognition.

All in all, the workshop has clearly shown that, even among those who agree on some basic premises (the very idea of extended cognition, and focus on complementarity), there is still plenty of room for different interpretations and applications of the framework, generating lively and healthy disagreement. My prediction is that the framework is here to stay, at least for a while.

(A side remark: it was also the first time (to my knowledge, at least) that four NewAPPS bloggers found themselves under the same roof: John Protevi, Helen de Cruz and myself as speakers, and Eric Schliesser who attended some of the talks. Maybe one day we’ll have a big NewAPPS reunion?)