The first face and body-selective neurons in the brain were discovered nearly 50 years ago. Nonetheless, it is only recently that studying social signals in the primate brain has become popular, in part because early researchers were skeptical that the brain would explicitly respond to, or represent, individual faces. This initial skepticism is mostly dissipated, and the recent burgeoning of the field of social neuroscience is of great importance not only for understanding the construction of the brain, but also for understanding the basis of social deficits that accompany many psychiatric illnesses. Our recent research in the macaque has asked how neurons in the high-level visual cortex respond to social stimuli presented in different contexts. Over the past year, we have extensively used a microwire bundle array developed and modified in our laboratory, in collaboration with Prof. Igor Bondar, that is able to longitudinally track the activity of cells in the brain that reside in so-called face patches. Face patches are small, circumscribed regions of the temporal and prefrontal cortex showing greater fMRI responses to faces than to other categories of stimuli. The longitudinal nature of the recordings allows us to investigate individual neurons for weeks or months at a time. Since our early work established that neurons in these areas are remarkably stable in their responses, testing over this period means that we are able to examine how single cells respond to a very large array of artificial and natural stimuli and conditions. We are currently completing three studies related to the selectivity of neurons within macaque face patches. In one of these studies, we are investigating the curious role of an internalized average face in the encoding of facial identity. The associated theory, which is often described as norm-based encoding, posits that the brain responds to individual faces as deviations from an internally stored average prototype. This average prototype is presumably learned through experience during childhood experience with a wide array of different faces. An intuitive view of how the average face comes into play comes when one thinks of the job of a caricaturist: their job is to accentuate the brain features that distinguish an individuals face from average. For example, the caricature of a person with particularly bushy eyebrows and a pointed chin will accentuate these features. In an early study, our laboratory showed evidence that this norm-based encoding was present among face-selective neurons in the macaque brain. This year we have finished a major follow-up study to that, systematically investigating these responses among local populations in three face-selective areas of the temporal cortex. We found strong evidence for norm-based tuning for morphed macaque faces, as well as morphed human faces. Moreover, this selective signal only arose after >200ms, long after the neurons initial response latency. Based on several observations, including measures of cell-to-cell activity coherence, we posit that this tuning is the product of late inhibition, whereby a subset of specialized neurons transmit broad inhibition to the population when the average norm face is shown. In a related study in macaques, we asked which facial features are the critical elements for driving neurons in the different face patches. One of the findings in that study, which is currently underway, is that the face patches differ markedly in this regard. This is notable since they appear to respond very similarly to the norm-based tuning paradigm described above even when the same neurons are considered. In the anterior fundus face patch, for example, face selective neurons are most strongly influenced by the outer contour of the face, whereas in the anterior medial face patch face selective neurons are most strongly influenced by internal facial structure. Moreover, these face patches also differ in regard to the plasticity that emerges with growing familiarity of stimuli over time. Whereas the anterior fundus face patches show little if any response change across days and weeks, those in the anterior medial face patch show the selective elimination of late-phase activity that arises after several days. These two aspects of single-neuron activity that differentiate the two face patches are currently being studied intensively in the lab, and we hope to have definitive results and a manuscript submitted within the coming months. Our work in marmosets is less well established, but has grown substantially in the last year, during which time we have made great headway in areas such as surgical procedures, training regimen, behavioral tracking, optical imaging, and viral expression and histology. With this progress, our marmoset laboratory is now beginning to generate its first results in the domain of optical imaging. The marmoset laboratory has two main branches. In the first, we are investigating acting within face selective regions of the brain using Ca++ imaging and implanted lenses, technically similar work widely carried out in mice. We are now at the stage of collecting data from marmosets expressing the associated fluorescent signals in their brain, and we anticipate that within the next year we will have results related to the diversity and longitudinal stability of face selective responses in the marmoset brain. In the second branch, we are investigating the natural social interplay between marmosets, using a highly precise tracking system to measure the position and gaze direction of multiple individuals. Within the next year, we plan to measure neural activity within socially related brain areas in this freely moving environment. The marmoset work is part of a strong collaboration with several extramural laboratories. In the coming year, we plan to additionally begin research into the neurodevelopment of these social circuits. This past year, several intramural investigators were awarded a DDIR Innovation Award to study neurodevelopment in marmosets, and much of our current work is building toward that goal. We will begin by attempting to identify the critical windows and learning principles through which the brain specializes itself through experience dependent plasticity to identify individual faces.
| Mundinano, Inaki-Carril; Fox, Dylan M; Kwan, William C et al. (2018) Transient visual pathway critical for normal development of primate grasping behavior. Proc Natl Acad Sci U S A 115:1364-1369 |
| Leopold, David A; Russ, Brian E (2017) Human Neurophysiology: Sampling the Perceptual World. Curr Biol 27:R71-R73 |
| Dougherty, Kacie; Cox, Michele A; Ninomiya, Taihei et al. (2017) Ongoing Alpha Activity in V1 Regulates Visually Driven Spiking Responses. Cereb Cortex 27:1113-1124 |
| Toarmino, Camille R; Yen, Cecil C C; Papoti, Daniel et al. (2017) Functional magnetic resonance imaging of auditory cortical fields in awake marmosets. Neuroimage 162:86-92 |
| Taubert, Jessica; Wardle, Susan G; Flessert, Molly et al. (2017) Face Pareidolia in the Rhesus Monkey. Curr Biol 27:2505-2509.e2 |
| Park, Soo Hyun; Russ, Brian E; McMahon, David B T et al. (2017) Functional Subpopulations of Neurons in a Macaque Face Patch Revealed by Single-Unit fMRI Mapping. Neuron 95:971-981.e5 |
| Russ, Brian E; Kaneko, Takaaki; Saleem, Kadharbatcha S et al. (2016) Distinct fMRI Responses to Self-Induced versus Stimulus Motion during Free Viewing in the Macaque. J Neurosci 36:9580-9 |
| Murphy, Aidan P; Leopold, David A; Humphreys, Glyn W et al. (2016) Lesions to right posterior parietal cortex impair visual depth perception from disparity but not motion cues. Philos Trans R Soc Lond B Biol Sci 371: |
| Kaskan, P M; Costa, V D; Eaton, H P et al. (2016) Learned Value Shapes Responses to Objects in Frontal and Ventral Stream Networks in Macaque Monkeys. Cereb Cortex : |
| Miller, Cory T; Freiwald, Winrich A; Leopold, David A et al. (2016) Marmosets: A Neuroscientific Model of Human Social Behavior. Neuron 90:219-33 |
Showing the most recent 10 out of 27 publications
