Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Cross-species representation of semantic information in prefrontal cortex at single-cell resolution
Poster Session D, Saturday, October 26, 10:30 am - 12:00 pm, Great Hall 3 and 4
Mohsen Jamali1, Irene Caprara1, Benjamin Mash1, Raymundo Báez-Mendoza2, Ziv Williams1; 1MGH, Harvard Medical School, 2German Primate Center, Göttingen
The capacity to anticipate events in their environment through the categorization of objects and experiences based on semantic information is a shared trait among both humans and many animals. It is believed that conserved neural mechanisms exist across species for processing semantic information, serving as a common foundation to navigate their environments. While previous research has revealed neuroanatomical similarities between species, understanding whether and to what extent neurons similarly represent semantic information has posed a major challenge. To address this, single-unit recordings were conducted in the homologous prefrontal cortex areas involved in semantic processing and categorization in macaque monkeys (n = 2) and human participants (n = 2) undergoing awake neurosurgery. Various images spanning different semantic categories were presented as stimuli while recording neuronal activities. To enable cross-species comparison, similar sets of images with diverse semantic content, valence, and themes were used, and normative ratings were obtained. For further comparisons, human participants were also presented with audio stimuli consisting of sentences covering similar themes and semantic categories. Vectorial representations of the stimuli were obtained and used for clustering the images and modeling and decoding neuronal responses. Our preliminary results showed that neurons in both monkeys and humans responded selectively to specific semantic domains, and their collective activities decoded categorical information about the presented images – with striking similarities in the relationships between certain semantic domains and their hierarchical organization. However, we also find that there was variability in how semantic information was represented within neuronal ensembles across species and how stimuli mapped across population response patterns. These findings demonstrate a striking cross-species representation of semantic information at the neuronal level, suggesting conserved mechanisms for processing semantic knowledge. They also highlight species-specific differences in the organization and mapping of semantic information, offering together new insights into the evolutionary divergence and convergence of semantic representations at a cellular level.
Topic Areas: Multisensory or Sensorimotor Integration,