samer@74: ** Information dynamics and temporal structure in music **
samer@74: 
samer@74: 
samer@74: It has often been observed that one of the more salient effects
samer@74: of listening to music to create expectations within the listener,
samer@74: and that part of the art of making music to create a dynamic interplay
samer@74: of uncertainty, expectation, fulfilment and surprise. It was not until
samer@74: the publication of Shannon's work on information theory, however, that 
samer@74: the tools became available to quantify some of these concepts. 
samer@74: 
samer@74: In this talk, we will examine how a small number of
samer@74: \emph{time-varying} information measures, such as entropies and mutual
samer@74: informations, computed in the context
samer@74: of a dynamically evolving probabilistic model, can be used to characterise
samer@74: the temporal structue of a stimulus sequence, considered as a random process 
samer@74: from the point of view of a Bayesian observer.
samer@74: 
samer@74: One such measure is a novel predictive information rate, which we conjecture
samer@74: may provide an explanation for the `inverted-U' relationship often found between
samer@74: simple measures of randomness (\eg entropy rate) and
samer@74: judgements of aesthetic value [Berlyne 1971]. We explore these ideas in the context
samer@74: of Markov chains using both artificially generated sequences and
samer@74: two pieces of minimalist music by Philip Glass, showing that even an overly simple
samer@74: model (the Markov chain), when interpreted according to information dynamic
samer@74: principles, produces a structural analysis which largely agrees with that of an 
samer@74: human expert listener and improves on those generated by rule-based methods. 
samer@74: