Introductory Post

So this is my first post, and I will try to ‘set the stage’ for what I’m gonna write about on this blog. The blog’s subtitle refers (albeit somewhat jokingly) to a phrase coined by Pierre Poirier, Benoit Hardy-Vallée (who has a pretty cool blog, by the way) and Jean-Frédéric Depasquale in their article about “Embodied Categorization.” (Poirier et al. 2005, availabe here).
The authors themselves note that

“obviously, no one will ever use this cumbersome descriptor to denote the discipline, but we feel that this is where the field is moving and that, one day, this is what people will have in mind when they think of cognitive science” (Poirier et al. 2005: 762).

Though I myself do not really have a very clear idea about what embodied evolutionary-developmental computational cognitive neuroscientists do, I think the term covers many of the topics I’m interested in (and don't seem to really have any idea about as I must confess).

To give an impression of what EEDCCN (it’s a shame this doesn’t create a really cool acronym) is about, I will write a bit about the paper in which this phrase was coined.
In their paper, Poirier and his colleagues claim that the fact that

"some, and perhaps all, cognitive capacities essentially depend on the body and its environment” (Poirier et al. 2005: 741)

has important implications for the way the world is categorized and conceptualized.
This idea of ‘embodiment’ – the fact that our “minds have bodies that are situated in environments” (Poirier et al. 2005: 741) and that this fact is a preconditional a priori of all experience – can be found in even the most simple systems.

Poirier et al. first study simple reactive systems and come to the conclusion that

“[t]here are categorization tasks which can be solved by simple cognitive systems, when the use of sensorimotor coordination is allowed, but which cannot be solved when conceived of as problems of passive, disembodied perception in which the system is neither spatially situated nor strongly coupled to its environment“ (Poirier et al 2005: 741).

Embodiment also provides crucial components for categorization abilities in ‘reactive categorizers that learn’ and ‘representing categorizers.’
As an illustration for the latter category, they cite an experiment by Floreano and Mondada (1996), in which robots have to explore an arena and from time to time have to recharge their batteries. The neural architecture implemented in the robot enabled him to have a number of capacities of embodied categorization. The most interesting one was the way the area was mapped by self-specializing, hidden neurons in the robot’s neural net. It was not mapped in a disembodied, abstract bird’s-eye-view manner,

“but in relation to the state of the battery: […] The way the arena is categorized is completely dependent on a relevant “bodily marker”: the state of its battery. Hence, it is impossible to explain how the robot categorizes its environment without reference to this important bodily property. There is no categorization of the environment that is independent of it” (Poirier et al. 2005: 751).

The next set of categorizers, emulating and simulating categorizers, are especially interesting for (human) cognition. Emulation is the process ‘when one system performs in exactly the same way as another […]’ whereas Simulation is the attempt ‘to predict aspects of the behaviour of some system by creating an approximate […] model of it.’ If we apply such a view to cognition,

“the brain emulates the body (motor emulators reproduce in parallel the body’s behaviour and generate feedback, like real perception and action) and simulates the external world (it reproduces possible things, agents, and events)” (Poirier & Hardy-Vallée 2005: 45).

Human cognition (or better parts of it) is thus an ‘anticipatory system’, which ‘preselects’ and prepares possible actions (Poirier et al. 2005: 752).
This is in accordance with the conviction of many neuroscientists that the brain’s main task is comprehensive prediction in order to survive in dangerous, unstable environments (Ryder and Favorov 2001).
One of the most important aspects of anticipation and action selection is that of inference, “a cognitive process in which new information is derived from given information” (Hegarty 2004: 280), which draws heavily on the knowledge present in our conceptual system, a “system distributed throughout the brain that represents knowledge about the world“(Barsalou 2005: 621)” and the automatic completion of internalized patterns therein (Barsalou 2005: 645).

Whereas emulators plan and predict bodily states, simulators plan and predict states of the external world “by modeling in order to predict what will be the case.” (Poirier et al. 2005: 756). One of way of framing this is by postulating ‘perceptual symbols’, "records of the neural
activation that arises during perception” (Barsalou 1999: 583), which are integrated in multimodal frames and simulate the outside world, when needed, by reactivating and integrating it into a conceptual representation of a state in the external world (Barsalou 1999).

Thus, “Simulation is a way to anticipate what the world could be if a given action were undertaken“ (Poirier et al. 2005: 757). Interestingly, simulations and performed or experienced actions employ overlapping neural networks, and the main difference between them is that simulations bring about motor output-suppressing inhibitory processes (Poirier et al. 2005: 757, Lotze et al. 1999).

In my next post, I’ll write a little more about the idea of embodied cognition, before turning to Poirier et al.’s cool account of three different forms of categorization, which is inspired by Dan Dennett’s (1987) notion of the physical stance, the design stance and the intentional stance.

References:

Barsalou, Lawrence W. 1999. “Perceptual Symbol Systems.” Behavioral and Brain Scienes 22.4: 577–609.

Barsalou, Lawrence W. 2005. “Situated Conceptualization.” Handbook of Categorization in Cognitive Science. Eds. Henri Cohen and Claire Lefebvre. Amsterdam: Elsevier.619-650.

Dennett, Daniel C. 1987. The Intentional Stance. Cambridge, M.A.: Bradford Books.

Floreano, Dario, and Francesco Mondada. 1996. “Evolution of homing navigation in a real mobile robot”, IEEE Transactions on Systems, Man, and Cybernetics – Part B: Cybernetics 26:396–407.

Hegarty, Mary. 2004- “Mechanical Reasoning by Mental Simulation.” Trends in Cognitive Sciences 8: 280-285.

Lotze, Martin, Pedro Montoya, Michael Erb, Ernst Hülsmann, Herta Flor, Uwe Klose, Niels Birbaumer and Wolfgang Grod. 1999. “Activation of Cortical and Cerebellar Motor Areas During Executed and Imagined Hand Movements: An fMRI study.” Journal of Cognitive Neuroscience 11: 491-501.

Poirier, Pierre, and Benoit Hardy-Vallée. 2005. “Structured Thoughts: The Spatial-Motor View.“ The Compositionality of Meaning and Content: Volume II: Applications to Linguistics, Psychology and Neuroscience. Eds. Edouard Machery, Markus Werning & Gerhard Schurz. Frankfurt: Ontos.

Poirier, Pierre, Benoit Hardy-Vallée and Jean-Frédéric Depasquale. 2005. “Embodied Categorization.” Handbook of Categorization in Cognitive Science. Eds. Henri Cohen and Claire Lefebvre. Amsterdam: Elsevier, 2005.

Ryder, Dan and Oleg V. Favorov. 2001. “The New Associationism: A Neural Explanation for the Predictive Powers of Cerebral Cortex.” Brain and Mind 2.2. : 161-194.