“School of Cognitive Sciences”

Back to Papers Home
Back to Papers of School of Cognitive Sciences

Paper   IPM / Cognitive Sciences / 14471
School of Cognitive Sciences
  Title:   Enhansive and suppressive face neurons form two distinct physiological networks for course and fine face representations in inferotemopral cortex
  Author(s): 
1.  S. Salehi
2.  MR. A.Dehaqani
3.  H. Esteky
  Status:   In Proceedings
  Proceeding: Program No. 530.09 / WW22, Society for Neuroscience Meeting. San Diego, 2016.
  Year:  2016
  Supported by:  IPM
  Abstract:
A network of clusters of face selective neurons in the inferior temporal (IT) cortex of monkeys has been found mostly by studying the excitatory responses to face stimuli. But the spiking activities of a relatively large proportion of neurons in IT are suppressed following presentation of faces. Most of these neurons are located outside the face clusters. Contribution of these suppressive neurons in face representation is not clear. We recorded the spiking activities of single neurons in the inferior temporal cortex of two macaque monkeys while they passively viewed face and non-face images and compared the contribution of enhanced (ENH) and suppressed (SUP) face selective neurons in face representation. Fifty-six of 176 neurons selectively responded to faces by either increasing or decreasing their spiking activities compared to their baseline. The neural code for fine and coarse face information was largely different between the ENH and SUP face selective neurons at single cell and population levels. Compared to ENH neurons, SUP neurons showed higher (lower) sparseness values for highly similar individual faces (face category). Pattern of activity in the SUP neural population discriminated highly similar face images better than ENH face neural populations. But ENH neural population discriminated faces from non-face objects better than SUP neural population. Our data show that ENH and SUP neurons form two distinct physiological neural networks in IT cortex with widely different coding properties: a course-scale (ENH neurons) and a fine-scale (SUP neurons) system. These face processing functional subunits may index fine-to-course shape similarity in parallel depending on task at hand.

Download TeX format
back to top
scroll left or right