Durante el primer semestre de 2020, los días viernes 14.30 a 17 (a partir del viernes 3 de Abril), como parte de la programación del doctorado en Filosofía de la Universidad Alberto Hurtado, Abel Wajnerman Paz va estar dictando un seminario de Introducción a la Filosofía de la Neurociencia. Mientras que la contingencia actual lo haga necesario, el seminario será dictado de manera virtual por medio de la plataforma Zoom. Para participar del mismo o para más información, escribir a email@example.com
Cognitive science was one of the dominant approaches to psychological processes during (part of) the second half of the twentieth century. One of its main insights is the hypothesis that mental capacities could be explained through the theoretical tools provided by computer science. Despite its success in accounting for numerous cognitive phenomena, a crucial shortcoming of this research program was its inability to connect the computational and information-theoretic characterization of mental processes to their underlying neurological basis, thus casting doubt on the biological plausibility of its fundamental concepts.
During the last three decades we have witnessed an unprecedented development of the technological and theoretical tools for manipulating and analysing the neural mechanisms underlying cognitive function and dysfunction, which was considered to constitute a ‘neurocognitive revolution’. Neurocognitive models seem to build a progressively stronger connection between cognition and its biological basis by providing increasingly complex and detailed descriptions of neural mechanisms in computational and/or information-theoretic terms. However, the nature, goals and implications of cognitive neuroscience are still a matter of intense philosophical debate.
In the first place, there is a discussion regarding the norms of neurocognitive explanation. The mechanistic approach, which has been the dominant view during the last 15 years, identifies the goal of cognitive neuroscience with the modelling of mechanisms underlying cognitive functions. A mechanism is a structure performing a function in virtue of its component parts, component operations, and their causal organization. Despite its ability to characterize (and evaluate) a relevant part of the explanatory practices in the field, a number of philosophers have recently argued that mechanism constitutes a very narrow view of neurocognitive explanations, failing to account for the explanatory power of, for instance, structural models (which do not refer to causal relations) or dynamical models (which do not describe fixed components or operations).
In the second place, there is a widespread metaphysical discussion concerning the cognitive ontology of neural mechanisms, that is, which cognitive properties can be attributed to neural components and activities and organization.
A first ontological question is related to the identification of the basic building blocks of neurocognitive processes. The classic view, which is the functional version of the ‘neuron doctrine’, posits that individual neurons are not only anatomical but also cognitive basic components of our mental capacities. On the contrary, a ‘population doctrine’ claims that the contribution of individual cells to cognitive processes can only be understood when these are viewed as part of neural assemblies, populations or circuits.
A second ontological question is how we should understand these building blocks and the mechanisms they constitute. In this regard, a crucial concern is whether neural processes are better characterized in information-theoretic and computational terms or, on the contrary, cognitive processing can be implemented by network properties that do not involve information or computation. For instance, network neuroscience is a thriving field in which cognitive capacities are often explained by purely graph-theoretic or dynamical properties of neural structures.
Finally, there is a debate regarding the relationship between cognitive neuroscience and classic cognitivism. A central issue is whether cognitive neuroscience supports the classic cognitivist view that cognition is representation manipulation. If the kinds of (dynamical, information-theoretic, graph-theoretic or computational) properties implemented by neural structures are inconsistent with a minimal notion of representation, then it is possible that cognitive neuroscience will not contribute to bridging the gap between classic cognitive and neural levels but rather characterize them in a completely new manner.