| Riccardo Dalla Volta |
 |
Istituto di Fisiologia Umana, Universitą degli Studi di Parma, Via Volturno 39, 43100 Parma, Italy
e-mail: ridallav@tin.it |
Poster Presentation: |
|
| CHRONICAL MULTIELECTRODE RECORDING APPLIED TO MONKEY VENTRAL PREMOTOR CORTEX. |
| R. Dalla Volta (1)(2); F. Grammont (1); A. Bruzzo (1); B. Engel (3); J. Krueger (3) |
|
|
Istituto di Fisiologia Umana, Universitą degli Studi di Parma, Via Volturno 39, 43100 Parma, Italy |
|
| Chronical multielectrode recording of neural activity allows to follow for several months the firing patterns of a specific population of neurons. We are recording from the ventral premotor area of monkey using 64 chronically implanted microelectrodes. The target area (F5) has been identified on a 3D reconstruction of the brain from MRI slices. The microelectrodes were implanted by reaching F5 from below via a remote trepanation. The electrode tips were located in the gray matter approximately within a sphere with a volume of 2 cube mm. The microplug to which the electrodes were connected was cemented on the top of the head so that the 64 channel headstage can be easily plugged. During the recording sessions the monkey can freely move the head and the limbs through small windows of the monkey chair.
Functional properties of the neurons are tested for motor, visual, acoustic and tactile properties. The data are analysed by performing either single neuron and multineuron analysis. In particular for the latter kind of analysis we determined relevant time sections for each action and represented the multineuronal excitation patterns as multidimensional “vectors”. Differences between vectors can be determined in a straightforward way. The method of multidimensional scaling (MDS) allows to visualize the ensemble of differences. As expected, vectors belonging to repetitions of actions differ little from each other. In contrast there are larger differences for different actions. It was surprising that the overall distribution of differences given by MDS was similar for different subsets of neurons, although the sensory and motor properties of the neurons in these subsets were different.
The stability of the recordings is sufficient for directly pursuing the multineuronal correlates of learning processes over several days. Our technique will be applied to study the effect of learning a voluntary motor behaviour in the premotor areas. |
|
