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

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

Section 10A


10A: Epistemology and methodology of research in embodied cognition



Saturday 15/5, 14:50-15:50



Cecilea Mun

Affiliation: Independent Scholar
Title: The Problem of Recalcitrant Emotions: Reconceiving the Structure of Rationality
Abstract: Recalcitrant emotions, which Deigh refers to as “ineducable” emotions (Deigh 1994, 851), are typically thought of as presenting a problem for cognitive theories of emotion, including embodied cognitive theories of emotion. Such criticisms have typically focused on examples of infant or non-human animal emotions, which can be readily denied some sort of cognitive content. Yet such criticisms are not about whether or not an emotional experience does or does not have cognitive content. For example, one may have a recalcitrant kind of resentment towards a certain group of people or an individual which causes one to interpret whatever relevant situation that occurs in a way that simply reinforces one’s own resentment, and such resentments may continue to persist for the same reason even when ample evidence is given to suggest otherwise. One might judge such emotions to be irrational to the extent that they are unwarranted, but the concern about recalcitrant emotions is not about the lack of appropriate justification for an emotional experience. It is instead about the fact that some emotional experiences are simply not susceptible to any kind of rational updating. Thus such experiences may be regarded as noncognitive rather cognitive, despite having some kind of cognitive content; and in doings so, they may be taken as evidence against cognitive theories of emotion, including embodied cognitive theories of emotion. In this paper, I first provide a brief overview of why the most common strategy which cognitive theorists of emotion have appealed to—the strategy of rationalizing recalcitrant emotions—is an inadequate response. More specifically, I argue that it either 1) does not address the problem of recalcitrant emotions, as it is understood by noncognitive theorists (e.g., Deigh), or 2) it leads to the consequence of undermining the notion of rationality. Second, I provide an alternative response to the problem of recalcitrant emotions in support of cognitive theories of emotion, including embodied cognitive theories of emotion. I argue that the most reasonable response for a cognitivist theorists is to simply accept that such emotional experiences are irrational experiences to the extent that they are impervious to countervailing evidence, yet to also deny that doing so entails that such experiences are noncognitive experiences. I conclude with the introduction of an alternative framework for understanding the structure of rationality, which also includes a particular way in which we ought to conceive the notion of intentionality, in order to make sense of the response I offer. Please see the following two diagrams for more details: Scale of Emotional Rationality (https://www.dropbox.com/s/8k9q9gd5e9r22l6/Figure9.1_Scale%20of%20Emotional%20Rationality.tif?dl=0), and Good OL Intentionality (https://www.dropbox.com/s/himvy5dr7xntxeu/Figure6.2_Good%20OL%20Intentionality.tif?dl=0).


Samuele Sartori

Affiliation: phd student in Scienze Cognitive - Università degli Studi di Messina
Title: The concept of incorporation through the history of artificial limbs
Abstract: The purpose of this lecture is to think over the plasticity of the body schema through an historical and scientific analysis of artificial limbs and amputation techniques. At first this speech would like to highlight how in cognitivism there is a growing debate among those who argue that it is possible to incorporate prostheses, and those who argue against this possibility. Using and updating a vast number of neuroscientific experiments, cognitivists and neuroscientists have understood that the two main but still enigmatic conditions for incorporation are the transparency and the proprioception of the prosthesis. An instrument or a prosthesis is the more transparent the less it is felt by the subject who uses this, while proprioception is defined as the ability to perceive and situate pre-reflexively the space that one's body occupies without necessarily relying on visual or reflexive controls. Often referring to Merleau-Ponty’s concept of body schema, some cognitivists believe that endosomatization is possible and they recognize the ability to restructure the body and become hybrid (Murray, 2004; Parisi, 2014). Other theorists give a normative status to the form and performative abilities of the body-schema (De Preester, 2010; Besmer, 2015), because they affirm that transparency and proprioception are possible only when prostheses are modeled on the structure and physiognomy of the ontogenetic human body. It follows that these researches are always focused on a perspective that investigates which abilities belong to the human organism. They pursue an anthropocentric and ahistorical prospective which is unlikely to solve the question as it does not take into account part of the issue itself: the evolution of technologies. In order to answer to these epistemological problems, the second part of this speech finds in Lambros Malafouris’ Material Engagement Theory a suitable method. Thanks to his awareness of the history in which technologies transact and contribute to the ontological evolution of mankind, this theory can be considered non anthropocentric. As the author writes: «What is needed is to fully embrace and accept the consequences of the fact that the science of mind and the science of material culture are two sides of the same coin.» (Malafouris, 2013) To show the epistemological resources discovered thanks to this method, the latter part of the speech describes the prosthetic limbs and the amputation techniques that today provide solutions to the issue of proprioception and transparency. Bibliography Besmer, K. M. (2015), “What Robotic Re-embodiment Reveals about Virtual Re-embodiment: A Note on the Extension Thesis”, in Postphenomenological Investigations, ed. by R. Rosenberg and P. Verbeek, Lexington Books, London, pp. 55-72. De Preester, H. (2010), Technology and the Body: the (Im)Possibilities of Re-embodiment, in “Foundation of Science”, 16, pp. 119-137. Malafouris, L. (2013), How Things Shape the Mind, MIT Press, Cambridge. Murray, C. D. (2004), An interpretative phenomenological analysis of the embodiment of artificial limbs, in “Disability and Rehabilitation”, 26, pp. 963-973. Parisi, F. (2014), Corpi e dispositivi: una prospettiva cognitivista, in “Fata Morgana”, 24, pp. 45-56.


Bulcsú Sándor, Tim Koglin, Michael Nowak, Frederike Kubandt, Laura Martin, Claudius Gros

Affiliation: Department of Physics, Babes-Bolyai University, Cluj-Napoca; Institute for Theoretical Physics, Goethe University Frankfurt, Frankfurt am Main;
Title: Embodied locomotion by self-organized attractor trailing
Abstract: Locomotion is generated via the on-line interaction between the nervous system, body, and environment. For fast and predictable movements top-down controlled behavior may be effective enough. However, in an ever-changing and complex environment, the required locomotion-related computational power can increase tremendously. As an alternative approach one may consider embodied locomotion, which implies that part of the computation needed to generate movements is carried out by the compliant body. Here we present a dynamical systems framework, called "attractoring", which allows for the design and analysis of distributed control systems involved in locomotion [1]. Motion patterns are often generated via abstract dynamical systems for which the resulting limit cycles are tuned to become suitable sequences of open-loop motor commands. For attractoring, on the other hand, trajectories are generated in the overarching phase space of the agents' brain and body, which is augmented by the physical space of the surrounding environment. By using the actuator states as propriosensory feedback, both the body and the environment become part of the agent's cognitive architecture. Locomotion is generated hence via the continuous trailing of the resulting attractors, which are in turn embedded in the physical dimensions as well. As an example we consider a simple adaptive controller that implements the donkey and carrot dynamics, using simulated springs and rate encoding neurons. As proof of concept, we present simulated and real-world wheeled and legged robotic architectures for which self-sustained locomotion emerges in terms of coexisting self-organized attractors. Multistability opens up new ways of interacting with and controlling robots [2]. [1] Kubandt, F., Nowak, M., Koglin, T., Gros, C., & Sándor, B. (2019). Embodied robots driven by self-organized environmental feedback. Adaptive Behavior, 105971231985562. [2] Sánsor, B., Nowak, M., Koglin, T., Martin, L., & Gros, C. (2018). Kick Control: Using the Attracting States Arising Within the Sensorimotor Loop of Self-Organized Robots as Motor Primitives. Frontiers in Neurorobotics, 12, 40.


Print Email