Neuromatch 3 / / Track 1 / Interactive talk
Watch now on YouTube

Representation of Spatial Frequencies in Neural Populations of Human Temporal Cortex

Hojjat Azimi Asrari

Mohammad Rabiei Ghahfarokhi, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran; Niloufar Shahdoust, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran; Farideh Shakerian, Department of Brain and Cognitive Sciences, Cell Science Research Center, Royan Institute for Stem Cell Biology and Technology, ACECR, Tehran, Iran, Department of Physiology, Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran; Mohammad-Reza A. Dehaqani, School of Electrical and Computer Engineering, University of Tehran, Tehran, Iran, School of Cognitive Sciences, Institute for Research in Fundamental Sciences, Tehran, Iran

Abstract

Multiple brain areas in temporal lobe, associated with human vision, represent the visual stimuli and regulate object recognition. Moreover, EEG and fMRI studies suggest that brain visual representation of low versus high spatial frequency is significantly noticeable. Recently, SEEGs (stereo-electro-encephalography electrodes), a technique used for invasive recording of brain activity via three dimensional recordings, have been standing out for its clinical applications in seizure surgeries. The high quality of recoding in SEEGs data gives the opportunity to combine Long Time Monitoring (LTM) with controlled stimulus presentation and investigate the neural representation of objects in human brain. Although, massive body of non-invasive imaging studies have been addressed the neural representation of objects in human brain, the neural representation of spatial filtered objects in human brain using intra-cortical recording is poorly understood.
During LTM, a pharmacologically intractable epileptic patient with implanted SEEGs participated in a face and non-face detection visual task. Electrodes were implanted in the right hippocampus, right and left supplementary motor area, right pre-supplementary motor area, and right anterior cingulate gyrus to evaluate an anatomo-electro-clinical correlate in the generation and propagation of seizures. The stimulus set is comprised of 155 grayscale images including 73 animates (face and non-face) and 82 inanimates at different level of frequencies to probe the selectivity of neural population across the five areas of the brain implanted by electrodes.
Our results support the existence of distinct neural representation of different spatial frequencies for face and non-face stimuli within temporal lobe of human brain. Furthermore, the interaction of category representation and spatial frequencies suggests a major role of spatial dynamics in face and object processing across ventral visual stream at human brain.
Our findings provide insight into the fine-grained functional architecture within the human brain pertaining to spatial dynamics in face and object processing.