On the mind's time : production, perception, and pianists performance of rhythms : neural correlates and neuroanatomical changes

Human movements are impressively flexible and precise. They are also coloured by the characteristics with which each person performs an act. This is evident in everyday activities, in sports, in art, and in music performance. How does the human brain control a movement pattern and give it its character, given all possible degrees of freedom?

In this thesis we have investigated the cortical organization of two fundamental aspects of sequential movements; the explicit temporal control of when to move, and the spatial control of where to move, with a particular focus on the temporal control. As model behaviour we have used sequence production with the fingers, and with functional resonance imaging we localized the associated neural activation. We measured the behaviour of the participants' rhythmic motor performance to enable interpretation of the brain activation. Finally, we studied the plasticity of the white matter as a function of practicing in pianists using diffusion tensor imaging.

Firstly, we investigated if the relative timing and the ordinal sequence of spatial locations of a movement can be handled independently by the brain. We found evidence for a parallel processing of the two aspects. In study I, we found that once a rhythmic sequence is learned it can be executed with different spatial movement patterns with little extra training. In study II we showed that the two aspects mainly activate different brain regions. Processing the temporal information activated motor areas of the medial wall, the inferior frontal cortex, the temporal lobe, and the lateral cerebellum, whereas the ordinal sequencing activated the parietal cortex, the dorsal premotor cortex, the cerebellum and the putamen. In study III, we found that pianists use two parallel visual pathways to guide their movements when playing from written music. The dorsal visual stream appears to be involved in the spatial melody processing, whereas the ventral visual stream processes the rhythmical information of the music. We argue that the parallel processing of the relative timing and the ordinal sequence of spatial locations of a movement gives flexibility in performance since each aspect could be manipulated individually.

Secondly, we investigated if rhythm sequence control relies on a central mechanism, or if it processed within the structures required for a particular task. In studies I and IV we observed that individuals differed from each other in their rhythm production profile, when performing a rhythm with different body parts and different spatial movement patterns, but showed remarkably high intra-individual consistency. This suggests that the same representation of the explicit rhythmic structure is used to time movements. From the brain imaging results we propose a set of brain areas involved in rhythm control: the supplementary motor area (SMA) and preSMA, the dorsal premotor cortex, the superior temporal lobe, the inferior frontal cortex, the insula, and the cerebellum. These areas were active during rhythm production, irrespective of body part used (study IV) or spatial movement pattern (study II). In addition, listening to rhythms without motor performance engaged these areas (study V).

Finally, we investigated if the structure of the white matter, containing myelinated axons that transmit neural signals between brain regions, changes as a consequence of longterm training of rhythmic finger movements. In study VI, we found an increased degree of organization of the white matter in professional pianists, which correlated with the number of hours practiced in different ageperiods. The data suggest that childhood practising is of particular importance, since the most widespread changes were seen after intense childhood practising.

List of scientific papers

I. Ullen F, Bengtsson SL (2003). Independent processing of the temporal and ordinal structure of movement sequences. J Neurophysiol. 90(6): 3725-35.
https://doi.org/10.1152/jn.00458.2003

II. Bengtsson SL, Ehrsson HH, Forssberg H, Ullen F (2004). Dissociating brain regions controlling the temporal and ordinal structure of learned movement sequences. Eur J Neurosci. 19(9): 2591-602.
https://doi.org/10.1111/j.1460-9568.2004.03269.x

III. Bengtsson SL, Ullen F (2005). Dissociation between melodic and rhythmic processing during piano performance from musical scores. Neuroimage. Oct 20: Epub ahead of print.
https://doi.org/10.1016/j.neuroimage.2005.09.019

IV. Bengtsson SL, Ehrsson HH, Forssberg H, Ullen F (2005). Effector-independent voluntary timing: Behavioural neuroimaging evidence. European Journal of Neuroscience. [Accepted]
https://doi.org/10.1111/j.1460-9568.2005.04517.x

V. Bengtsson SL, Ehrsson HH, Toshihiro H, Tomonori K, Ullen F, Naito E, Forssberg H, Sadato N (2005). Listening to rythms - Roles of the prefrontal cortex, motor areas and the cerebellum. [Manuscript]

VI. Bengtsson SL, Nagy Z, Skare S, Forsman L, Forssberg H, Ullen F (2005). Extensive piano practicing has regionally specific effects on white matter development. Nat Neurosci. 8(9): 1148-50.
https://doi.org/10.1038/nn1516


https://doi.org/000231483800013