Brain Computing Lab
Project 3
Read More

Spatial and Temporal Dynamics of Cortical Object Processing


The information carried by different spatial frequency (SF) has long been recognized to represent in a coarse-to-fine sequence. The visual system uses the combination of global and local features, which are carried by low SF (LSF) and high SF (HSF) respectively, to facilitate the processing of the upcoming stimulus. To understand the mechanistic role of SF processing in temporal dynamics of category representation, we examined neural responses of the inferior temporal (IT) cortex of macaque monkeys viewing SF filtered face and non-face stimuli with two different exposure times (fast trials: 20ms and slow trials: 200ms). The stimuli were six faces and three inanimate objects and each of which had been filtered in five frequency ranges (band-pass SF filter). In addition, for all the stimuli, corresponding phase-scrambled images were generated. All stimuli were equalized to have same luminance and contrast.


We directly addressed spatiotemporal dynamics of object and face representation by investigating how the network of IT neurons processes distinct SF over time. We applied a combination of encoding and decoding methods to the single and population of recoded neural data to explore the role of SF coding properties in IT category representation. Consistent with the coarse-to-fine nature of perception, we found an earlier representation of LSF in IT population compared with HSF. Furthermore, our results showed a systematic category representation difference over SFs in fast and slow trials, while the category information was similar for unfiltered stimuli in both fast and slow trials. These observations suggest a direct link between SF processing and evolution of category information over time. Together, these results provide evidence for the existence of an effective interaction between SF coding and category representation in IT cortex.



Mohammad-Reza Abolghasemi Dehaqani

Ehsan Rezayat

rezayat [at] ipm [dot] ir

Farideh Shakerian


Ramin Toosi


Behnam Karami


Roxana Koushki




Abdol-Hossein Vahabie

vahabi [at] ipm [dot] ir

Mohammad Ali Akhaee