| Title & Author | Abstract | |
|---|---|---|
| 7(01) | AN OPTIMAL YARDSTICK FOR COGNITION
Target Article by Worden on Optimal-Cognition Robert P. Worden Logica UK Ltd Betjeman House 104 Hills Road Cambridge CB2 1LQ, UK wordenr@logica.com |
Abstract:
Firstly, by analysing its biological function, I show
that there is a best possible form of cognition, which gives
greatest fitness. The results of this optimal cognition can be
calculated in any domain -- such as foraging, navigation, vision,
or conditioning. Secondly, internal representations and symbols
emerge as a feature of near-optimal cognition. This gives a
biological basis for cognitivism. Finally, we do not expect brains
to evolve to the precise optimum. However, there are indications
that animal cognition usually comes close to optimum efficiency,
while today's cognitive models (such as neural nets or
reinforcement learning models) have a wide range of efficiencies.
Keywords: Bayes, conditioning, evolution, foraging, navigation, neural nets, optimality, representation, situated action. |
| 7(03) | RECONCILING THE OPTIMAL BRAIN WITH VARIABILITY
IN NEURAL ORGANIZATION AND COGNITION Commentary on Worden on Optimal-Cognition Roslyn Holly Fitch Center for Molecular and Behavioral Neuroscience Rutgers University 197 University Ave. Newark, NJ 07102 holly@axon.rutgers.edu |
Abstract:
The concept of an optimal brain which underlies an
optimal cognitive strategy for a given species in a given situation
is apparently at odds with the enormous within-species variation in
cortical organization and cognitive function, and specifically with
replicable group differences in brain organization and/or cognitive
performance. One such example concerns sex differences in neural
lateralization and spatial navigation strategy; differences which
cannot be reconciled with the notion of an optimal evolutionary
path. Moreover, on a more general level the term "optimal" seems
impossible to define with respect to human cognitive behavior,
since a specific outcome can rarely be defined as the "ideal" one.
Keywords: Bayes, conditioning, evolution, foraging, navigation, neural nets, optimality, representation, situated action. |
| 7(10) | OPTIMALITY: EFFICACY, EFFICIENCY AND EFFECTIVENESS
Commentary on Worden on Optimal-Cognition David M. W. Powers AI and Cog. Sci. Group Department of Computer Science Flinders University of South Australia powers@acm.org |
Abstract:
Worden's definition of optimality hides two components:
efficacy in satisfying the needs of the organism, and effectiveness
or robustness across a range of environmental contexts. There is
however, in between, another criterion which relates to the
efficiency of the solution in terms of the organisms resources,
which is closely related to the parsimony of the model employed.
In addition, there is a question as to how significant and useful
the bounds are which Worden proposes, but he assumes that they are
fairly tight and his proof of monotonicity assumes a rather
simplistic model. Part of the problem here is the failure to
recognize efficiency directly or to allow direct comparison of
orthogonal mechanisms.
Keywords: Bayes, conditioning, evolution, foraging, navigation, neural nets, optimality, representation, situated action. |
| 7(26) | COMPARING YARDSTICKS FOR COGNITION
Commentary on Worden on Optimal-Cognition Csaba Szepesvari Bolyai Institute of Mathematics Jozsef Attila University Szeged, 6720 Aradi vrt tere 1. and Department of Adaptive Systems, Institute of Isotopes Hungarian Academy of Sciences and Jozsef Attila University Budapest 1525, Pf. 77. HUNGARY szepes@math.u-szeged.hu |
Abstract:
Worden (1996) has suggested that cognitive science could
be built around a model for optimal cognition. He has also proposed
an equation, called the Requirement Equation (RE), which should
describe the biological requirement brains must meet. Here I
analyse the limitations of his equation in detail. The analysis is
based on a more general class of models, namely, models of optimal
sequential decisions, of which Worden's equation is a special case.
It turns out that the RE can describe optimal behaviour if there is
no perceptual aliasing. If the RE is able to capture all aspects of
cognition then animal cognition is probably modular and thus it is
more likely to be suboptimal than optimal. I also show that --
contrary to Worden's suggestion -- the optimisation problem faced
by evolution may have many local minima, depending on the genetic
encoding. Nevertheless, one can show, purely on the basis of the
theory of optimal sequential decisions, that brains of animals
probably use internal representations and that cognition has a
universal limit when one considers its biological function.
Keywords: evolution of cognition, immediate rewards, internal representation, modularity of cognition, optimal sequential decisions. |