Reza Azadi¹ (), Timothy Ma¹, Emily Lopez¹, Josh Ebel¹, Arash Afraz¹; ¹Laboratory of Neuropsychology, National Institute of Mental Health, NIH, Bethesda, MD 20892, USA

We have recently shown that detectability of cortical stimulation in high-level visual areas, such as inferior temporal (IT) cortex in monkeys, varies by the content of the visual input presented during stimulation. However, the neurophysiological mechanisms underlying this interaction remain unexplored. Since IT neurons respond differentially to various visual stimuli, it is possible to hypothesize a simple explanation: visual stimuli that maximally activate the neurons, result in a narrower dynamic range for the impact of artificial stimulation, making the behavioral detection of stimulation more difficult. This hypothesis predicts a negative correlation between image-induced neural activity and behavioral detectability of neural stimulation. To study this interaction, we implanted a multi-electrode Utah Array in central IT cortex of a macaque monkey, allowing recording and stimulation in the same sites across multiple sessions. The animal learned to detect and report IT cortical stimulation while fixating on different images. We also collected the neural responses while the animal passively fixated on a series of visual stimuli. The results revealed a strong positive correlation (r = 0.81, p = 0.001) between the neural responsiveness to visual stimuli and detectability of artificial cortical stimulation. Contrary to the simple hypothesis mentioned above, the behavioral performance for detection of cortical stimulation is better when the visual input induces higher neural activity in the simulated site. Additionally, we included trials in which multiple electrodes simultaneously delivered electrical stimulation. The results revealed a positive correlation between the average visual response across stimulated sites and the detectability of multi-electrode cortical stimulations. These findings suggest that the interaction between the neural activity evoked by visual input and cortical stimulation is complex, and that perturbability of a neural state might be dampened in the absence of its corresponding visual stimuli.