A defense of the computational explanation of cognition that relies on mechanistic philosophy of science and advocates for explanatory pluralism. In this book, Marcin Milkowski argues that the mind can be explained computationally because it is itself computational—whether it engages in mental arithmetic, parses natural language, or processes the auditory signals that allow us to experience music. Defending the computational explanation against objections to it—from John Searle and Hilary Putnam in particular—Milkowski writes that computationalism is here to stay but is not what many have taken it to be. It does not, for example, rely on a Cartesian gulf between software and hardware, or mind and brain. Milkowski's mechanistic construal of computation allows him to show that no purely computational explanation of a physical process will ever be complete. Computationalism is only plausible, he argues, if you also accept explanatory pluralism. Milkowski sketches a mechanistic theory of implementation of computation against a background of extant conceptions, describing four dissimilar computational models of cognition. He reviews other philosophical accounts of implementation and computational explanation and defends a notion of representation that is compatible with his mechanistic account and adequate vis à vis the four models discussed earlier. Instead of arguing that there is no computation without representation, he inverts the slogan and shows that there is no representation without computation—but explains that representation goes beyond purely computational considerations. Milkowski's arguments succeed in vindicating computational explanation in a novel way by relying on mechanistic theory of science and interventionist theory of causation.

This paper centers around the notion that internal, mental representations are grounded in structural similarity, i.e., that they are so-called S-representations. We show how S-representations may be causally relevant and argue that they are distinct from mere detectors. First, using the neomechanist theory of explanation and the interventionist account of causal relevance, we provide a precise interpretation of the claim that in S-representations, structural similarity serves as a “fuel of success”, i.e., a relation that is exploitable for the representation using system. Then, we discuss crucial differences between S-representations and indicators or detectors, showing that—contrary to claims made in the literature—there is an important theoretical distinction to be drawn between the two.

Replicability and reproducibility of computational models has been somewhat understudied by “the replication movement.” In this paper, we draw on methodological studies into the replicability of psychological experiments and on the mechanistic account of explanation to analyze the functions of model replications and model reproductions in computational neuroscience. We contend that model replicability, or independent researchers' ability to obtain the same output using original code and data, and model reproducibility, or independent researchers' ability to recreate a model without original code, serve different functions and fail for different reasons. This means that measures designed to improve model replicability may not enhance (and, in some cases, may actually damage) model reproducibility. We claim that although both are undesirable, low model reproducibility poses more of a threat to long-term scientific progress than low model replicability. In our opinion, low model reproducibility stems mostly from authors' omitting to provide crucial information in scientific papers and we stress that sharing all computer code and data is not a solution. Reports of computational studies should remain selective and include all and only relevant bits of code.