Object EEG response: Recording Human Steady State Visually Evoked Potential for Object Images

Abstract

The brain mechanisms behind object recognition are complex. In recent years, interest has grown in discovering the process behind complex and natural stimuli e.g., real object images. Steady State Visual Evoked Potential (SSVEP) is a measure of a natural reaction to visual stimuli at a specific frequency. Since SSVEPs can easily be quantified to frequency and higher signal to noise ratio (SNR) than the usual VEP, it could be the answer to many questions within the vision scientist community. This study was conducted using stimuli from the THINGS, a database which is a global collection of object images and data collected on them. The goal is to bring together researchers and expand understanding of object recognition. A great amount of EEG data will be collected on three participants (approximately 12 hours per participant), on 650 real object images. The images will be randomly assigned to pairs. Each trial, participants performed a fixation task while image pairs were shown. One image (“standard”) was presented at a frequency of 6 Hz, while the other (“oddball”) was shown at 1.2 Hz, every fifth image. Power was estimated at 6 Hz for both image types and at 1.2 Hz for the oddball. Oddball and standard power were analysed to determine, 1) if correlation was between them, on different electrode clusters and for first/second time viewing the pairs, 2) if there was a correlation with other similarity measures on the object pairs. The goal of this study was to answer if SSVEP can be used to better understand how the human brain processes real object images. In conclusion the data gathered in this study did not show evidence of SSVEPs correlating with similarity measures. However, there is a lot more that can be discovered with further analysis of the data, and the hope is that it will be of benefit to other studies and the development of THINGS database.