Orientation selectivity in cat and agouti primary visual cortex as reflected in firing rates, LFP gamma oscillations and phase relationships

Abstract

Even though several mammalian orders have orientation-selective neurons, these neurons can be organized in different functional cortical layouts. A prominent example is the columnar orientation preference map found in carnivores (cats, ferrets, tree shrews) and all primates (e.g. human, macaque, mouse lemur) studied so far, as opposed to the seemingly random arrangements commonly referred to as salt-and-pepper configurations, observed in rodents. Orientation selectivity is generally deduced from stimulus-driven spiking activity. However, it has also been observed in oscillations of the local field potential (LFP), especially in the gamma band. In several of the species expressing a columnar map, preference orientation of spiking activity and gamma power are similar. This is compatible with the interpretation that the LFP samples activity from contiguous cortical patches responding to the same orientation. Spikes can be strongly locked to certain phases of LFP oscillations, and the latter are thought to not only establish temporal windows coordinating spiking activity of local neuronal populations but to also enable communication between distant neuronal populations. We thus hypothesized that a columnar/non-columnar functional layout might also be reflected in spike-field interactions. In the present thesis, we examined spikes and LFP oscillations in the low gamma (low-γ) band of +/-30-50 Hz obtained from multiple parallel recordings in visual cortex during grating stimulation in two representatives of different mammalian order and similar V1 size but different functional layouts – the domestic cat for carnivores, and the Amazonian agouti for rodents. Previously, we observed that although agoutis possess orientation selective neurons they do not express columns as cats do. We correlated the orientation tuning of spikes with that of the low-γ power of the same electrode. This analysis revealed that units exhibiting both orientation selectivity in spikes and LG power (OSI > 0.1) exhibit similar preference in cats (n = 95, circular correlation = 0.56, p = 1.44e-12), but much less in agouti (n = 39, circular correlation = 0.40, p = 0.0067). Further, we evaluated if single unit spiking activity is coordinated with the LFP in the Gamma range recorded from the electrode as reflected in phase-locked events. In cats, phase-locked events frequently occurred in an orientation selective manner and with similar orientation preference as the spiking activity (n = 33, circular correlation = 0.57, p = 0.0035). In contrast, agouti showed only very few phase-locked events and those were not correlated with spike orientation selectivity (n = 6, circular correlation coefficient = 0.1093, p = 0.78756). Our results support the notion that the coordination of spiking and LFP gamma power in cats could be related to their orderly columnar layout. Along the same line of argumentation, we propose that the lower similarity of spike and gamma tuning and the lack of selective phase-locked events in agouti reflects a lesser or absent organization of orientation preference in a contiguous columnar layout in this species.