This chapter subsumes David Marr’s levels of analysis account of explanation in cognitive science under the framework of mechanistic explanation: Answering the questions that define each one of Marr’s three levels is tantamount to describing the component parts and operations of mechanisms, as well as their organization, behavior, and environmental context. By explicating these questions and showing how they are answered in several different cognitive science research programs, this chapter resolves some of the ambiguities that remain in Marr’s account, and (...) shows that many different areas and traditions of cognitive scientific research can be unified under the mechanistic framework. (shrink)

Psychoneural reduction has been debated extensively in the philosophy of neuroscience. In this article I will evaluate metascientific approaches that claim direct molecular and cellular explanations of cognitive functions. I will initially consider the issues involved in linking cellular properties to behaviour from the general perspective of neural circuits. These circuits that integrate the molecular and cellular components underlying cognition and behaviour, making consideration of circuit properties relevant to reductionist debates. I will then apply this general perspective to specific systems (...) where psychoneural reduction has been claimed, namely hippocampal long-term potentiation and the Aplysia gill-withdrawal reflex. (shrink)

This paper discusses the relevance of models for cognitive science that integrate mechanistic and computational aspects. Its main hypothesis is that a model of a cognitive system is satisfactory and explanatory to the extent that it bridges phenomena at multiple mechanistic levels, such that at least several of these mechanistic levels are shown to implement computational processes. The relevant parts of the computation must be mapped onto distinguishable entities and activities of the mechanism. The ideal is contrasted with two other (...) accounts of modeling in cognitive science. The first has been presented by David Marr in combination with a distinction of “levels of computation”. The second builds on a hierarchy of “mechanistic levels” in the sense of Carl Craver. It is argued that neither of the two accounts secures satisfactory explanations of cognitive systems. The mechanistic-computational ideal can be thought of as resulting from a fusion of Marr’s and Craver’s ideals. It is defended as adequate and plausible in light of scientific practice, and certain metaphysical background assumptions are discussed. (shrink)

A single physical process may often be described equally well as computing several different mathematical functions—none of which is explanatorily privileged. How, then, should the computational identity of a physical system be determined? Some computational mechanists hold that computation is individuated only by either narrow physical or functional properties. Even if some individuative role is attributed to environmental factors, it is rather limited. The computational semanticist holds that computation is individuated, at least in part, by semantic properties. She claims that (...) the mechanistic account lacks the resources to individuate the computations performed by some systems, thereby leaving interesting cases of computational indeterminacy unaddressed. This article examines some of these views, and claims that more cases of computational indeterminacy can be addressed, if the system-environment interaction plays a greater role in individuating computations. A new, long-arm functional strategy for individuating computation is advanced. (shrink)

Optogenetics and DREADDs (Designer Receptors Exclusively Activated by Designer Drugs) are important research tools in recent neurobiology. These tools allow unprecedented control over activity in specifically targeted neurons in behaving animals. Two approaches in philosophy of neuroscience, mechanism and ruthless reductionism, provide explicit accounts of experiments and results using tools like these, but each offers a different picture about how levels of mechanisms relate. I argue here that the ruthless reductionist’s direct mind‐to‐cellular/molecular activities linkages “in a single bound” better fits (...) with both the experimental designs using these tools and some of the scientists’ own judgments about their results than does the mechanist’s “nested hierarchies of mechanisms‐within‐mechanisms.” So at least some important work in current neuroscience appears to be ruthlessly reductive. Mechanism may not correctly characterize all current work in neuroscience, despite its recent popularity. (shrink)

Neurobiologists talk of linking mind to molecular dynamics in and between neurons. Such talk is dismissed by cognitive scientists, including many cognitive neuroscientists, due to the number of “levels” that separate behaviors from these molecular events. In this paper I explain what neurobiologists mean by such claims by describing the kinds of experiment tools that have forged these linkages, directly on lab benches. I here focus on one of these tools, gene targeting techniques, brought into behavioral neuroscience from developmental biology (...) more than a quarter-century ago. Discussion of this tool does more than illuminate these claims by neurobiologists, however. An account of its development shows the doubly dependent role that theory plays in neurobiology. Our best current theories about “how the brain works” depend entirely on the experiment tools neuroscientists have available. And these tools get developed via the solution of engineering problems, not the application of theory. Theory is thus of tertiary importance in neuroscience, not of the primary importance that many cognitive scientists assume it to occupy. (shrink)

Since the emergence of our species at least, natural selection based on genetic variation has been replaced by culture as the major driving force in human evolution. It has made us what we are today, by ratcheting up cultural innovations, promoting new cognitive skills, rewiring brain networks, and even shifting gene distributions. Adopting an evolutionary perspective can therefore be highly informative for cognitive science in several ways: It encourages us to ask grand questions about the origins and ramifications of our (...) cognitive abilities; it equips us with the means to investigate, explain, and understand key dimensions of cognition; and it allows us to recognize the continued and ubiquitous workings of culture and evolution in everyday instances of cognitive behavior. Taking advantage of this reorientation presupposes a shift in focus, though, from human cognition as a general, homogenous phenomenon to the appreciation of cultural diversity in cognition as an invaluable source of data. (shrink)

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 an understanding of how a mechanism functions in broader environments that determines the computations it needs to perform. The representation and algorithmic perspective offers 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. (shrink)

One of the most lasting contributions of Marr's posthumous book is his articulation of the different “levels of analysis” that are needed to understand vision. Although a variety of work has examined how these different levels are related, there is comparatively little examination of the assumptions on which his proposed levels rest, or the plausibility of the approach Marr articulated given those assumptions. Marr placed particular significance on computational level theory, which specifies the “goal” of a computation, its appropriateness for (...) solving a particular problem, and the logic by which it can be carried out. The structure of computational level theory is inherently teleological: What the brain does is described in terms of its purpose. I argue that computational level theory, and the reverse-engineering approach it inspires, requires understanding the historical trajectory that gave rise to functional capacities that can be meaningfully attributed with some sense of purpose or goal, that is, a reconstruction of the fitness function on which natural selection acted in shaping our visual abilities. I argue that this reconstruction is required to distinguish abilities shaped by natural selection—“natural tasks” —from evolutionary “by-products”, rather than merely demonstrating that computational goals can be embedded in a Bayesian model that renders a particular behavior or process rational. (shrink)

The Frame Problem is the problem of how one can design a machine to use information so as to behave competently, with respect to the kinds of tasks a genuinely intelligent agent can reliably, effectively perform. I will argue that the way the Frame Problem is standardly interpreted, and so the strategies considered for attempting to solve it, must be updated. We must replace overly simplistic and reductionist assumptions with more sophisticated and plausible ones. In particular, the standard interpretation assumes (...) that mental processes are identical to certain kinds of computational processes, and so solving the Frame Problem is a matter of finding a computational architecture that can effectively represent relations of semantic relevance. Instead, we must take seriously the possibility that the way in which intelligent agents use information is inherently different. Whereas intelligent agents are plausibly genuinely causally sensitive to semantic properties as such (to what they perceive, desire, believe intend, etc.), computational systems can only be causally sensitive to the formal features that represent these properties. Indeed, it is this very substitution of formal generalizations for genuinely semantic ones that is responsible for the way current AI systems are brittle, inflexible, and highly specialized. What we need is a more sophisticated way of investigating the relationship between computational information processing and genuinely semantic information use, so that these two senses of using information are not conflated, but instead the question of how they are related to one another can be studied directly. I apply the generative methodology I have developed elsewhere for cognitive science and AI research (Miracchi, 2017, 2019a) to show how the Frame Problem can be appropriately updated. (shrink)