More than expected: extracellular waveforms and functional responses in monkey LGN

Abstract

AbstractUnlike the exhaustive determination of the retina, key populations in the lateral geniculate nucleus of the thalamus (LGN) may have been missed. Here, we have begun to characterize the full range of extracellular neuronal responses in the LGN of awake monkeys using multi-electrodes during the presentation of colored noise visual stimuli to identify any previously overlooked signals. The extracellular spike waveforms of single units were classified into seven distinct classes: four negative-dominant classes that were narrow or broad, commonly associated with inhibitory or excitatory neurons, respectively; one triphasic class; and two positive-dominant classes. These units had their receptive fields (RF) mapped using spike-triggered averaging and were classified into magnocellular (M), parvocellular (P), koniocellular (K), and non-RF (N) neurons. Spike classes were correlated with their mapped RF and response characteristics to identify any relationships between spike shape and cell type: negative and narrow spiking waveform units had RFs that were mostlyP; negative and broad waveform units, mostlyN; and positive waveform units, mostlyM, as well as having shorter response latencies, larger RF sizes, and larger eccentricities than the other waveform classes.Ncells, those without an estimated RF, have not been regularly reported before. We observed that mostNcells consistently responded to the visually presented noise stimulus at a lower and more sustained rate than units with an RF. Our observations indicate that the population of LGN cells may be broader than traditionally held.Significance StatementThis study aims to extensively survey the extracellular space in the LGN of rhesus macaques. We found significant functional differences across extracellular waveform classes, such as their receptive field (RF) and spiking characteristics (e.g., firing rate, burstiness, latency). We also found a set of single units that did not yield an RF but reliably responded to the visually presented stimuli. Our findings reveal that the neural population of LGN may extend beyond the classical divisions of magnocellular, parvocellular, and koniocellular responses. This new finding will have interesting implications for our understanding of LGN function and interpretation of extracellular signals, especially significant with the recent advancement of dense multi-electrode arrays.