The paper by Nikos K. Logothetis reviews the neural basis of the blood-oxygen-level-dependent (BOLD) functional magnetic resonance imaging (fMRI) signal. It discusses the relationship between BOLD fMRI signals and neural activity during sensory stimulation, using simultaneous intracortical recordings of neural signals and BOLD responses in monkeys. The study found a moderate to strong correlation between neural activity measured with microelectrodes and the BOLD signal averaged over a small area around the microelectrode tips, but the BOLD signal had significantly higher variability than neural activity, indicating that traditional statistical methods underestimate the reliability of neuronal activity. The paper also examines the contribution of different types of neuronal signals to the hemodynamic response, comparing local field potentials (LFPs), single- and multi-unit activity (MUA) with high spatio-temporal fMRI responses in monkey visual cortex. Only LFPs were significantly correlated with the hemodynamic response at transient response sites, and LFPs had the largest magnitude signal and better linear systems analysis performance compared to MUA in predicting fMRI responses. These findings suggest that the BOLD signal primarily measures input and processing of neuronal information within a region rather than output signals to other brain regions.The paper by Nikos K. Logothetis reviews the neural basis of the blood-oxygen-level-dependent (BOLD) functional magnetic resonance imaging (fMRI) signal. It discusses the relationship between BOLD fMRI signals and neural activity during sensory stimulation, using simultaneous intracortical recordings of neural signals and BOLD responses in monkeys. The study found a moderate to strong correlation between neural activity measured with microelectrodes and the BOLD signal averaged over a small area around the microelectrode tips, but the BOLD signal had significantly higher variability than neural activity, indicating that traditional statistical methods underestimate the reliability of neuronal activity. The paper also examines the contribution of different types of neuronal signals to the hemodynamic response, comparing local field potentials (LFPs), single- and multi-unit activity (MUA) with high spatio-temporal fMRI responses in monkey visual cortex. Only LFPs were significantly correlated with the hemodynamic response at transient response sites, and LFPs had the largest magnitude signal and better linear systems analysis performance compared to MUA in predicting fMRI responses. These findings suggest that the BOLD signal primarily measures input and processing of neuronal information within a region rather than output signals to other brain regions.