This paper explores the information content of neural spike trains in the fly's visual system, focusing on a motion-sensitive neuron, H1. The nervous system encodes time-dependent signals as sequences of action potentials or spikes, with information carried in spike timing. The authors quantify the information in spike trains in bits, without assuming which features are most important. They apply this method to analyze experiments on H1, showing that it transmits information at up to 90 bits/s, nearly reaching the physical limit set by the entropy of the spike train. Spike trains encode sensory information through temporal sequences, and the information content is measured by the difference between the total entropy and the conditional entropy. The efficiency of information transmission is defined as the ratio of information rate to entropy rate. The authors use a model-independent approach to estimate entropy and information in spike trains, discretizing them into time bins and analyzing segments of the spike train. They find that the entropy rate for H1 is 157 ± 3 bits/s, with an information rate of 78 ± 5 bits/s, or 1.8 ± 0.1 bits/spike. This indicates that the neuron uses about 50% of its available information capacity. The analysis shows that spike timing is crucial for information transmission, and the results are consistent across different time resolutions. The study demonstrates that spike timing plays a significant role in encoding dynamic signals in the nervous system, without assuming any specific structure of the neural code.This paper explores the information content of neural spike trains in the fly's visual system, focusing on a motion-sensitive neuron, H1. The nervous system encodes time-dependent signals as sequences of action potentials or spikes, with information carried in spike timing. The authors quantify the information in spike trains in bits, without assuming which features are most important. They apply this method to analyze experiments on H1, showing that it transmits information at up to 90 bits/s, nearly reaching the physical limit set by the entropy of the spike train. Spike trains encode sensory information through temporal sequences, and the information content is measured by the difference between the total entropy and the conditional entropy. The efficiency of information transmission is defined as the ratio of information rate to entropy rate. The authors use a model-independent approach to estimate entropy and information in spike trains, discretizing them into time bins and analyzing segments of the spike train. They find that the entropy rate for H1 is 157 ± 3 bits/s, with an information rate of 78 ± 5 bits/s, or 1.8 ± 0.1 bits/spike. This indicates that the neuron uses about 50% of its available information capacity. The analysis shows that spike timing is crucial for information transmission, and the results are consistent across different time resolutions. The study demonstrates that spike timing plays a significant role in encoding dynamic signals in the nervous system, without assuming any specific structure of the neural code.