Embodied Cognition at the Crossroads of
Philosophy, Linguistics, Psychology and Artificial Intelligence

May 13-15, 2021, Cluj-Napoca, ROMANIA

Section 9B


9B: Embodiment of abstract thinking



Saturday 15/5, 13:40-14:40



Anco Peeters

Affiliation: Ruhr University Bochum
Title: Enactive remembering through the lens of sensorimotor contingencies
Abstract: Abstract Enactivists have recently joined a growing number of philosophers of mind who agree, for various reasons, that memory traces are best understood as contentless. But enactivists who understand remembering as fundamentally embodied and reconstructive, have so far remained vague on the details of how episodic memory works if it is not grounded in the storage of semantic information – as is commonly believed. In this paper, I tap into current discussions in enactivism on sensorimotor action to further develop enactive accounts of remembering. This move depends on the assumption that mind arises out of the embodied and active coupling with an organism's environment. A number of enactivists have recently developed a detailed account of how this coupling can be understood in terms of different types of sensorimotor contingencies, in an approach to cognition that does justice both to phenomenological experience and scientific rigor. I apply this sensorimotor account to enactive remembering to argue that even episodic memory need not be understood as depending on the processing of contentful information, but can instead be seen as an active and environment involving process. I do this by establishing that episodic remembering can be understood as re-enacting scenarios from the past. The construction of such scenarios is fed by practised, familiar sensorimotor interactions with our environment, including other people and artefacts. This enactive approach provides helps solve lingering problems that have plagued the two main approaches in current philosophical literature on remembering: the causal theory and the simulation theory. On the one hand, while the causal theory of memory is able to ground the epistemological value of memory by linking the memory trace to the remembered event causally, it has problems incorporating an increasing amount of empirical findings that suggest remembering is reconstructive. On the other hand, while the simulation theory is able to take the constructive nature of memory on board, it looses grip on the causal connection between the remembered event and the remembering event. By positing that familiar sensorimotor patterns feed scenarios in the memorizing process, enactivists can account for both the causal links and the reconstructive nature of memory, without relying on mental representations. An enactive approach to memory thus provides a novel solution to current fundamental issues in the philosophy of memory. Bibliography Di Paolo, E., Buhrmann, T., & Barandiaran, X. E. (2017). Sensorimotor Life: An enactive proposal. Oxford University Press. Gallagher, S. (2017). Enactivist interventions: Rethinking the mind. OUP. Hutto, D. D., & Peeters, A. (2018). The roots of remembering: Radically enactive recollecting. In K. Michaelian, D. Debus, & D. Perrin (Eds.), New Directions in Philosophy of Memory (pp. 97–118). Routledge. Michaelian, K. (2016). Mental time travel: Episodic memory and our knowledge of the personal past. MIT Press. Michaelian, K., & Sant’Anna, A. (2019). Memory without content? Radical enactivism and (post)causal theories of memory. Synthese. Peeters, A., & Segundo-Ortin, M. (2019). Misplacing memories? An enactive approach to the virtual memory palace. Consciousness and Cognition, 76, 102834. Werning, M. (2020). Predicting the past from minimal traces: Episodic memory and its distinction from imagination and preservation. Review of Philosophy and Psychology.


Prakash Mondal

Affiliation: Indian Institute of Technology Hyderabad
Title: The Constraints of Embodiment and Language-Thought Relations
Abstract: Language and thought are intimately related to one another, but the level or degree of connectedness between language and thought is not clear due to the fact that the influence of language over thought can be more context-specific or general enough (see Zlatev and Blomberg, 2015). This reflects general assumptions from the Sapir-Whorf Hypothesis as well (Whorf, 1956). If the influence of language over thought, thinking and reasoning is very context-specific in being applicable to specific modes/modalities of cognition such as color, space, visual motion etc., this may suggest that the constraints of embodiment determine how modal linguistic symbols come to be grounded in neurally instantiated modality-specific systems (Barsalou, 2008). Prominent here is the idea that we can gain entry into the territory of human thoughts and reasoning by examining the structures of specific languages. It seems reasonable to assume that this supposition risks taking language to be the entry point rather than an entry point for the exploration of human thoughts and reasoning. Many cognitive consequences as claimed to ensue from the language-specific conceptualizations of number, color categories, motion, space and other categories (see Gentner and Goldin-Meadow, 2003; Levinson, 2003; Casasanto et al., 2004; Majid et al., 2004; Casasanto and Boroditsky, 2008; Wolff and Holmes, 2011; Lupyan, 2012) can be traced to the properties of our cognitive organization itself. For instance, an important notion on how knowing (or learning) a language different from the first language we acquire in childhood is linked to the rewiring of the brain (see Bylund and Athanasopoulos 2017). This particular view emphasizes that learning a new way of talking about time encoded in a language makes the user of the language adopt a new way of thinking which was not available to that language user who was ingrained in a distinct way of thinking encoded in her first language. The case for cognitive flexibility in bilinguals can be accounted for in a way that reflects the cognitive reality rather than any linguistic version of reality. Thus, for instance, when bilinguals switch from one way of thinking about time to another while shifting from the context of one language to another, it is not the language that induces the bilinguals to do so. Rather, it is the raw cognitive imprint or the mental signature the word evokes/triggers that induces bilinguals to switch ways of thinking. The observed linguistic effects on cognitive strategies in thinking when using language are stabilized regularities of a fluctuating cognitive system. Evidence for such a stance comes from the fact that the activation of modal semantic features in both brain-damaged patients and normal people is not deterministic but rather dynamically governed by many factors some of which are contextual and some of which are purely cognitive in themselves (Kemmerer, 2019). This also means that the constraints of embodiment are not selectively and exclusively oriented and restricted to language. Rather, the aspects of the cognitive system minus language can project certain modes of thinking.


Fintan Nagle, Hugo Stevensen, Brian Ball

Affiliation: Imperial; NCH
Title: Food caching in scrub jays: symbolic memory or connectionist learning?
Abstract: Food caching in scrub jays: symbolic memory or connectionist learning? The food caching behaviour of scrub jays has been called the first conclusive evidence of episodic memory in non-human species. Clayton et al. (1998) have conducted a large body of experimental work which characterises, in the lab, scrub jays' ability to hide caches of food and retrieve them appropriately. Gallistel and King interpret these results in their influential work Memory and the Computational Brain (2009). Their central claim is that a symbolic read-write memory is necessary to support this kind of intelligent cache searching. Our paper challenges this interpretation, and suggests an alternative which does not require symbolic information or read-write memory retrieval. After closely reviewing the original scrub jay work, we look more closely at Gallistel and King's claims. We look first at the distinction they make between symbolic information and procedural information. We deal next with Gallistel and King's views on the encoding of time, suggesting that the timestamps on which their suggested mechanism relies are not plausible in neural systems and that more useful encodings for time are already known to neuroscience. We also suggest an alternative explanation that could allow jays to cache and retrieve food without requiring a symbolic memory. Finally, we address a more general contextual question. Gallistel and King seem to maintain that a computational system can either be a connectionist network or employ a symbolic read-write memory. However, as Marr pointed out long ago, systems allow descriptions at different levels of abstraction. Is it wise to assert that a system "is" procedural or symbolic without specifying at which levels this assertion is valid? We conclude by exploring the possibility that connectionist networks can implement key aspects of Gallistel and King’s account of the cognition underpinning the jays’ behaviour. Clayton, N. S., & Dickinson, A. (1998). Episodic-like memory during cache recovery by scrub jays. Nature, 395(6699), 272-274. Gallistel, C. R., & King, A. P. (2011). Memory and the computational brain: Why cognitive science will transform neuroscience (Vol. 6). John Wiley & Sons.

Print Email