INFLUENCE OF LOW AND HIGH-FREQUENCY INPUTS ON SPIKE TIMING IN VISUAL CORTICAL-NEURONS

Cortical neurons in vivo respond to sensory stimuli with the generatio n of action potentials that can show a high degree of variability in b oth their number and timing with repeated presentations as well as, on occasion, a high degree of synchronization with other cortical neuron s, including in the gamma frequency range of 30-70 Hz. Here we examine d whether or not this variability may arise from the intrinsic mechani sms of action potential generation in cortical regular spiking, fast s piking and intrinsic burst-generating neurons maintained in vitro. For this purpose, we performed intracellular recordings in slices of ferr et visual cortex and activated these cells with the intracellular inje ction of various current waveforms. Some of these waveforms were deriv ed from barrages of postsynaptic potentials evoked by visual stimulati on recorded in vivo; others were artificially created and contained va rious amounts of gamma range fluctuations; finally, others consisted o f swept-sinewave current (ZAP current) functions. Using such stimuli, we found that, as expected given the resistive and capacitive properti es of cortical neurons, low frequencies have a larger effect on the me mbrane potential of cortical neurons than do higher frequencies. Howev er, increasing the amount of gamma range fluctuations in a stimulus le ads to more precise timing of action potentials. This suggests that di fferent frequencies play different roles, low frequencies being effici ent for depolarizing cells with high frequencies increasing the precis ion of action potential timing. In parallel to increases in temporal p recision, the addition of higher frequency components increases the ra nge of interspike intervals present in the action potential discharge, These results suggest that higher frequency components such as gamma range fluctuations may facilitate the generation of action potentials with a high temporal precision while at the same time exhibiting a hig h degree of variability in interspike intervals on single trials. This temporal precision may facilitate the use of temporal codes or the ge neration of precise synchronization for the transmission and analysis of information within cortical networks.