In the past decades, various approaches to study the cognitive system have emerged in response to the orthodox computationalist hypothesis. One of the heated debates takes place between the computationalist and the dynamicist. Proponents of the dynamicist hypothesis claimed that cognitive system should be studied as a situated agent in continuous, simultaneous and mutually determining interaction with the changing world. A particularly strong case was made by van Gelder (1995) with his Watt governor example to show that cognition is a state-space evolution which requires no computations and representations.
In this paper, Chemero (2000) is trying to offer an alternative perspective on the dynamical approach, focusing specifically …show more content…
The problem is demonstrated by comparing the representationalist vs. anti-representationalist debate with the debate between atomists and phenomenologists in physics. Atomist Boltzmann (1900)’s guide to discovery argument asserts that phenomenology is inferior to atomism in physics because the presupposition of atoms provides a guide to discovering new equations that could capture physical phenomena more accurately (as cited in Chemero, 2000). In comparison, phenomenological physic excludes presuppositions of the underlying structure of beings, making it a fact-dependent and ad hoc approach that provides no guide to discovery (Chemero 2000). Likewise, representationalists can argue that representations provide critical insights in understanding the more complicated cognitive functions, such as linguistic processes and decision making. In contrast, dynamical systems theory seems rather ad hoc and fact-dependent. Chemero reasons that the solution to this challenge is to prove that dynamic accounts can provide a guide to discovery as well, such as by postulating a generally applicable dynamical model that accounts for a wide range of cognitive phenomena. A preliminary example would be Haken, Kelso and Bunz (1985)’s model, which uses macroscopic patterns of finger wagging to predict behaviors of the complex systems …show more content…
His argument serves as a persuasive objection to van Gelder’s perspective, which urges to eliminate representation on the grounds of both hypotheses. However, besides Chemero and van Gelder’s theoretical divergence on the nature hypothesis, they seem to achieve a consensus on requirements for the knowledge hypothesis. That is, dynamical accounts need to offer a convincing explanation for agent-environment systems without involving representations (i.e. the Watt governor), and they need to be extendable to other cognitive phenomena. Although both Chemero and van Gelder agree on the presence of preliminary models that satisfy the criteria, much work needs to be done to capture the more complex cognitive processes such as language. The viability of the knowledge hypothesis is still an open-ended question until we have more fruitful evidence at