Abstract

Understanding how neural populations encode sensory information thereby leading to perception and behavior (i.e., the neural code) remains an important problem in neuroscience. When investigating the neural code, one must take into account the fact that neural activities are not independent but are actually correlated with one another. Such correlations are seen ubiquitously and have a strong impact on neural coding. Here we investigated how differences in the antagonistic center-surround receptive field (RF) organization across three parallel sensory maps influence correlations between the activities of electrosensory pyramidal neurons. Using a model based on known anatomical differences in receptive field center size and overlap, we initially predicted large differences in correlated activity across the maps. However, in vivo electrophysiological recordings showed that, contrary to modeling predictions, electrosensory pyramidal neurons across all three segments displayed nearly identical correlations. To explain this surprising result, we incorporated the effects of RF surround in our model. By systematically varying both the RF surround gain and size relative to that of the RF center, we found that multiple RF structures gave rise to similar levels of correlation. In particular, incorporating known physiological differences in RF structure between the three maps in our model gave rise to similar levels of correlation. Our results show that RF center overlap alone does not determine correlations which has important implications for understanding how RF structure influences correlated neural activity.

Highlights

  • Encoding sensory stimuli and integrating this information into suitable behavior is a highly complex task that is continuously solved by our nervous system

  • Growing evidence across nervous systems and species shows that the activities of neighboring neurons are not independent but are correlated with one another, which has important implications for neural coding. Such correlations are generally thought to be due to shared input. How this shared input is integrated by neurons in order to give rise to correlated activity is not well understood in general

  • We focused on understanding how differences in receptive field (RF) center size and overlap influence correlated electrosensory lateral line lobe (ELL) pyramidal cell activity

Read more

Summary

Introduction

Encoding sensory stimuli and integrating this information into suitable behavior is a highly complex task that is continuously solved by our nervous system. There is a growing body of evidence that perception and behavior are determined by the integrated activity of large neural populations [1]. Our understanding of such population codes is complicated by the fact that neural responses are often correlated and not independent of one another. Noise correlations have been observed ubiquitously across systems and species [1,3] and there is general agreement that they will strongly influence sensory processing [2,3]. The fact that these strongly depend on factors such as the organism’s behavioral state [1,4] as well as stimulus statistics [5,6,7,8] greatly complicates understanding their impact on coding in general

Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call