A neural model of neuromodulatory (dopamine) control of arm movements in Parkinson's disease (PD) bradykinesia was recently introduced [1, 2]. The model is multi-modular consisting of a basal ganglia module capable of selecting the most appropriate motor command in a given context, a cortical module for coordinating and executing the final motor commands, and a spino-musculo-skeletal module for guiding the arm to its final target and providing proprioceptive (feedback) input of the current state of the muscle and arm to higher cortical and lower spinal centers.

The neuromodulatory model is successful at offering an alternative explanation to what other models suggest about the causes of Parkinson's disease bradykinesia. More specifically, it focuses more on the effects of dopamine (DA) depletion in cortex and spinal cord and less on its effects in basal ganglia (as other models have done).

The neuromodulatory model provides a unified theoretical framework for PD bradykinesia and it is capable of producing a wealth of neuronal, electromyographic and behavioral movement empirical findings such as:

Recently the model of PD bradykinesia [1, 2] was extended in two ways: (1) Incorporated the spindle feedback not only in the spinal cord as in [1, 2], but also in cortex and examined its effects on the activities of specific types of cells found in primary motor cortex both in normal and in DA depleted cases, and (2) Examined the effects of DA depletion not only in alpha motoneuronal (MN) and Renshaw activities as in [1, 2], but also in the activities of type Ia and Ib inhibitory interneurons (IN) and primary spindles, in order to inverstigate whether abnormal reciprocal inhibition of spinal IaINs plays a significant role in PD rigidity.

The new model [3] predicted that the reduced reciprocal disynaptic Ia inhibition in the DA depleted case doesn't lead to the co-contraction of antagonist motor units. Furthermore, the model predicted that although the co-contraction of antagonist muscles might be a mechanism for PD rigidity, the co-contraction isn't due to abnormal reciprocal inhibition at the spinal level. The causes of MN co-contraction ought to be searched more centrally, potentially in the microcircuit of the motor cortex and/or the basal ganglia.

References:
[1] V. Cutsuridis, S. Perantonis (2006) A Neural Model of Parkinson's Disease Bradykinesia. Neural Networks 19(4): 354-374

[2] V. Cutsuridis (2006) Neural Model of Dopaminergic Control of Arm Movements in Parkinson's Disease Bradykinesia. In: Artificial Neural Networks - ICANN 2006, Lecture Notes in Computer Science, LNCS 4131 (Springer-Verlag, Berlin) 583-591

[3] V. Cutsuridis (2007) Does Abnormal Spinal Reciprocal Inhibition Lead to Co-contraction of Antagonist Motor Units? A Modeling Study. International Journal of Neural Systems, in press

Model usage:
Extract the folders in this archive, start matlab, add the Cutsuridis_PDmodel folder to the path,and run main.m
This will generate a few figures associated with the publications:

fig1.jpg Figure 1 depicts the basal ganglia-thalamus output (GO signal) that drives the motor cortical cells in the model and alpha-MN activity in both normal and PD cases.

fig2.jpg Figure 2 depicts the position, velocity and force curves in both normal and PD cases.

fig3.jpg Figure 3 depicts the DVV and P activities of M1 cells in normal case, whereas

fig4.jpg" Figure 4 depicts the same activities but in PD case.
This version of the files is June 28, 2007 supplied by Vassilis Cutsuridis.