文档介绍：Bayesian models and simulations
in cognitive science
Giuseppe ignone1, Roberto Cordeschi2
putation Lab
Dipartimento di ’Informazione e Ing. Elettrica
Universit`adi Salerno
via Ponte Melillo, 1 -84084 Fisciano (SA), Italy
******@
2Dipartimento di Studi Filosofici ed Epistemologici
Universit`adi Roma ”La Sapienza”
via Carlo Fea 2, I-00161 Roma, Italy
******@
Abstract
Bayesian models can be related to cognitive processes in a variety of
ways that can be usefully understood in terms of Marr’s distinction among
three levels of explanation: computational, algorithmic and implementa-
tion. In this note, we discuss how an integrated probabilistic account of
the different levels of explanation in cognitive science is resulting, at least
for the current research practice, in a sort of unpredicted epistemological
shift with respect to Marr’s original proposal.
1 Introduction
Sophisticated probabilistic models are finding increasingly wide application across
the cognitive and brain sciences.
It has been argued [Knill et al., 1996, Chater et al., 2006] that probabilistic
models can be related to cognitive processes in a variety of ways. This variety
can be usefully understood in terms of Marr’s [1982] widely known distinction
between three levels at which any agent carrying out a task must be under-
stood, the what/why level (computational theory), the how level (algorithm),
the physical realization (implementation):
• Computational theory. What is the goal of putation, why is it
appropriate, and what is the logic of the strategy by which it can be
carried out?
1
• Representation and algorithm. How can putational theory be im-
plemented? What is the representation for the input and output, and
what is the algorithm for the transformation?
• Implementation. How can the representation and algorithm be realized
physically?
Figure 1: The three levels of explanation suggested by Marr.
Indeed, in