Neural Machinery and Realization

Polger, Thomas W.

doi:10.1086/425948

Cited by 29 publications

(8 citation statements)

References 12 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As has already been stressed repeatedly, on Marr's account all three levels of analysis are needed to "completely understand" a cognitive system and explain its behavior. Accordingly, investigations at the computational and algorithmic levels must be supplemented by investigations at the implementational level, which are driven by questions about the way in which the constructs of a cognitive model are "realized physically" (Marr 1982: 25; see also Polger 2004). Unfortunately, it remains unclear what it actually takes to show that the production rules, symbols, equations and/or variables specified by a cognitive model are realized by complex and potentially unruly physical systems such as the brain.…”

Section: The Implementational Level: "Where?"mentioning

confidence: 99%

“…Since a mechanism's abstract mathematical properties are instantiated by its concrete physical properties, why should the former need to be cited in an explanation, in addition to the latter? Polger (2004) argues that mechanisms can compute information-processing functions or state-space trajectories just in virtue of their physical properties, regardless of whether these properties are also said to instantiate any particular abstract mathematical properties. In the same vein, Bickle (2015) advances a conception of mechanistic explanation in which a characterization of a mechanism's physical properties suffices to explain a wide variety of behavioral and cognitive phenomena; whether or not a cognitive mechanism can also be characterized in abstract mathematical terms at the algorithmic level is explanatorily irrelevant.…”

Section: The Implementational Level: "Where?"mentioning

confidence: 99%

See 1 more Smart Citation

Mechanisms in cognitive science 1

Zednik¹

2017

The Routledge Handbook of Mechanisms and Mechanical Philosophy

View full text Add to dashboard Cite

Section: The Implementational Level: "Where?"mentioning

confidence: 99%

Section: The Implementational Level: "Where?"mentioning

confidence: 99%

Mechanisms in cognitive science 1

Zednik¹

2017

The Routledge Handbook of Mechanisms and Mechanical Philosophy

View full text Add to dashboard Cite

“…Others have lumped together the computational and algorithmic levels, describing them as sketches that are “elliptical or incomplete mechanistic explanations” (Piccinini & Craver, , p. 284) to be later filled in by full‐blown mechanistic explanations. Yet others have associated the computational level with an idealized competence and the algorithmic and implementation levels with actual performance (Craver, ; Frixione, ; Horgan & Tienson, ; Polger, ; for a teleological variant, see Anderson, this volume). Finally, Egan () associates the computational level with an explanatory formal theory, which mainly specifies the computed mathematical function (see also van Rooij, ).…”

Section: The Computational Levelmentioning

confidence: 99%

The Non‐Redundant Contributions of Marr's Three Levels of Analysis for Explaining Information‐Processing Mechanisms

Bechtel

Shagrir

2015

Topics in Cognitive Science

View full text Add to dashboard Cite

Are all three of Marr's levels needed? Should they be kept distinct? We argue for the distinct contributions and methodologies of each level of analysis. It is important to maintain them because they provide three different perspectives required to understand mechanisms, especially information processing mechanisms. The computational perspective provides understanding of how a mechanism functions in broader environments that determine the computations it needs to perform (and may fail to perform). The representation and algorithmic perspective offer an understanding of how information about the environment is encoded within the mechanism and what are the patterns of organization that enable the parts of the mechanism to produce the phenomenon. The implementation perspective yields an understanding of the neural details of the mechanism and how they constrain function and algorithms. Once we adequately characterize the distinct role of each level of analysis, it is fairly straightforward to see how they relate. 1. Introduction The term level is used in a wide variety of ways. One use refers to levels of organization-a whole system is at a higher level than the parts that constitutes it and the activities the whole performs are at a higher level than those of its components (Craver & Bechtel, 2007). This is not the use Marr (1982) had in mind in distinguishing the computational, representational and algorithmic, and implementational levels. These are levels of analysis, not levels of organization and so, contrary to French et al. (this issue), Marr is not addressing topics such as the emergence of higher levels. The main question raised by the various papers in this issue is whether all three types of analysis are required and, if so, how they relate. The question of how they relate, we will argue, is fairly straightforward once we (1) characterize the distinctive role of each type of analysis and (2) show how each type of analysis makes an important, non--redundant contribution to understanding information processing mechanisms. As the explanatory relevance of the implementational level is generally not contested, we will not dwell on it. Within the context of a mechanistic framework of explanation, it specifies the parts (e.g., neurons) and operations (excitation or inhibition of other neurons) that are employed or recruited in the generation of the phenomenon of interest. Challenges to Marr's conception of levels focus on the representational and algorithmic and the computational levels of analysis. The idea of algorithms governing the transformations of representations is familiar from computer science but, as we will see, is not limited to the sorts of representations and algorithms used in digital computers. Rather, it allows for the

show abstract

“…As Polger () notes, there need not be a tight fit between the algorithm and the implementation. So long as the implementation produces the outputs the computational description requires, given the inputs it specifies, then the algorithmic level need only approximate how the brain actually transforms its representations.…”

Section: The Role Of Neuroscience In Information Processing Theoriesmentioning

confidence: 99%

“…In 1982, David Marr advocated for three independent levels of understanding for any information‐processing device, like the human brain: the level of computational theory, the level of representation and algorithm, and the level of hardware implementation (p. 25). Cognitive scientists and philosophers have discussed the viability of these distinctions since then (see, e.g., Butler, ; Cummins, ; Dennett, ; Egan, , , ; Gilman, , ; Harnish, ; Horgan & Tienson, , ; Kitcher, ; McClamrock, ; Newell, , ; Poggio, ; Polger, ; Pylyshyn, ; Shagrir, ; Sterelny, ; Stevens, ; Verdego & Quesada, ). Now, more than 30 years out, we can ask: What can we learn from contemporary research in computational neuroscience that might shed light on the feasibility of Marr's original proposal?…”

Section: Introductionmentioning

confidence: 99%

Marr's Levels Revisited: Understanding How Brains Break

Hardcastle

2015

Topics in Cognitive Science

View full text Add to dashboard Cite

While the research programs in early cognitive science and artificial intelligence aimed to articulate what cognition was in ideal terms, much research in contemporary computational neuroscience looks at how and why brains fail to function as they should ideally. This focus on impairment affects how we understand David Marr's hypothesized three levels of understanding. In this essay, we suggest some refinements to Marr's distinctions using a population activity model of cortico-striatal circuitry exploring impulsivity and behavioral inhibition as a case study. In particular, we urge that Marr's computational level should be redefined to include a description of how systems break down. We also underscore that feed-forward processing, cognition disconnected from behavioral context, and representations do not always drive cognition in the way that Marr originally assumed.

show abstract

Neural Machinery and Realization

Cited by 29 publications

References 12 publications

Mechanisms in cognitive science 1

Mechanisms in cognitive science 1

The Non‐Redundant Contributions of Marr's Three Levels of Analysis for Explaining Information‐Processing Mechanisms

Marr's Levels Revisited: Understanding How Brains Break

Contact Info

Product

Resources

About