|   |   | |||||
![]() |
|
Submitted by pscully on Fri, 10/30/2009 - 14:16.
11/05/2009 - 16:10 11/05/2009 - 17:30 Short Title: STA/BST 290: Petr Janata (UC Davis) Short Desc: Music, Probability, and the Brain
THURSDAY, November 5th, 2009 at 4.10pm, MSB 1147 (Colloquium Room) Refreshments: 3.30pm, MSB 4110 (Statistics Lounge)
Speaker: Petr Janata (UC Davis, Dept Psychology, Center for Mind and Brain) Title: Music, Probability, and the Brain
Abstract: The system of major and minor keys on which Western tonal music is based defines a music-theoretical, perceptual, and statistical space that can be represented on the surface of a ring (torus). Keys are defined by probability distributions across the 12 possible pitch classes (the notes C, C-sharp, D, etc.). Local regions on the torus represent key regions (tonal centers) that are identified with similar probability distributions across the pitch classes. Keys that are close to each other in a music theoretical and perceptual sense, e.g. C major and G major, are situated close to each other on the toroidal surface because their pitch probability distributions are similar. A piece of music creates a pattern of movement on the toroidal surface. The pattern of movement is governed by the time window over which the occurrence probabilities of the 12 pitch classes are integrated. Using a leaky integrator with a 200 ms time constant, the transitions between notes in a melody or between chords are emphasized, whereas longer time constants of 2 or 10 s emphasize short chord sequences and the established sense of key. As we engage with music, different psychological processes transpire at various timescales. Sensorimotor processes, such as moving or singing along with a piece of music, are well-characterized at the shorter timescales, whereas mnemonic and emotional effects are more characteristic of the longer timescales. I will describe results from a functional magnetic resonance imaging (fMRI) experiment seeking to understand how music-evoked autobiographical remembering experiences are manifested in the brain. Using the model described above, I identify networks of brain areas that track the tonality changes at each of the timescales. The networks partially overlap and consist of brain areas whose known functions correspond to different facets of hearing a familiar song that evokes memories and positive emotions. Given the brain's penchant for extracting probability structures in its sensory inputs and responding to violations of expectations based on those probabilities, the sensitivity of brain regions to the time-varying probability structures that define music's movement in tonal space across multiple timescales is auspicious for our understanding the way in which the human brain couples complex musical stimuli with the many psychological processes they engender. » |
![]() |
|||