چكيده به لاتين
Understanding neural coding is a pivotal concern across various brain regions. Over the years, extensive research has delved into spiking neural activity and local field potentials (LFP). Recent discoveries have shed light on the informative relationship between these signals. One intriguing facet of this relationship is the temporal coordination between spike occurrences and the phase of LFP signals. While this coordination has been scrutinized in diverse regions of the human brain and in animal models like rats and monkeys, it remains relatively uncharted in motor areas. To address this gap, our project leveraged a freely available database that encompassed force-motor tasks. We meticulously examined the temporal alignment between spike timing and LFP phase in the primary motor cortex across different time points and frequency bands. Notably, our analysis revealed that, within the beta frequency band, there is a notable lack of coordination during movement phases. Conversely, during periods of inactivity and force application, a discernible synchronization between spike occurrences and LFP phases was evident. This observation suggests that, during active movement, there is a deficiency in spike-phase coupling within the local field potential. Furthermore, our exploration of the delta band unveiled an increased temporal coordination between spikes and LFP phases during force application. These findings indicate that certain neurons in this region are responsive to different phases of the task. Subsequently, these neurons were harnessed for classifying grip types, force levels, and a four-class mode that combines grip type and force level. The utilization of spike coupling values in relation to LFP phase yielded remarkable results, demonstrating the potential for accurate classification across a spectrum of frequencies, including low (1-4 Hz) and high frequency ranges (approximately above 60 Hz). The accuracy achieved was 90% for grip type, 76% for force type, and 62% for the four-class mode, albeit slightly lower compared to classification based on firing rates and superior to LFP signal amplitudes. Moreover, this feature exhibited promise in the continuous decoding of forces, achieving an r2 value of 0.61, and hand movements with an r2 value of 0.81. Notably, the monkeys' reaction time was significantly correlated with grip type, which could be classified with 94% accuracy using the spike-LFP phase coupling feature. In conclusion, our study provides insights into the intricate relationship between spiking neural signals and LFP phases within the primary motor cortex, shedding light on their roles in motor control and offering potential applications for classifying and decoding motor-related information.
