I am broadly interested in the mechanisms of sensory perception. For every animal the extraction of information from the environment allows for the generation of purposeful behavior and thus is key to survival.
Using weakly electric fish as a model system, my investigations focus on:
(1) the neural code , i.e. the neural mechanisms that allow the brain to encode behavioral relevant information
(2) sensory-motor interactions, i.e. behavior that is used to optimize sensory acquisition & analysis.

1. The neural code

Correlated variability is found almost ubiquitously in the CNS, and we know that their presence substantially impacts the efficacy with which information can be encoded in neural populations.
I am investigating the different aspects of population coding in the medulla of apteronotoid weakly electric fish. My current questions are:

  • what are the mechanism by which correlated variability arises?
  • how does correlation structure in a heterogeneous population of cells with different “polarity” (ON vs OFF) influence information transmission?
  • how does correlated variability directly contribute to information encoding?

suggested reading:
Averbeck et al 2006; Chacron & Bastian 2008; Hofmann & Chacron 2017

CorrelationStructureFigure: Examples of correlation structure (signal vs. noise correlations) in ELL of Apteronotus albifrons. A pair of pyramidal neurons with positive signal and negative noise correlations (left) and positive signal and positive noise correlations (right). Probability distributions (solid circles) separate better for opposing correlation structure (left) than if signal and noise correlations have the same sign (right).

2. Sensory-motor interactions

Active sensing is a term that depicts the fact that sensory acquisition of information is strongly influenced by the contextual behavior issued during the acquisition phase: every motion generates re-afferent sensory input that, after being processed is used to decide on the next purposeful motor-step suitable to disambiguate the sensory input in context of a specific task. We know of numerous examples from various different sensory systems where animals use active behavior to enhance their sensory “performance”.
Weakly electric fish are a true champion-animal model system to investigate those aspects of sensory-motor interaction and integration. My current questions are:

  • which aspects of electro-sensory behavior (temporal, spatial) can aid the extraction of sensory information and how? 
  • how does the interplay between sensory and motor components change during learning of a sensory task?

suggested reading:
Schroeder et al. 2010; Hofmann et al. 2013; Loeb & Fishel 2014; Grant et al. 2014;

Pic_ActiveSensingApproach copy
Figure: Modeled electro-sensory input (current density modulation) perceived by Gnathonemus petersii during the final phase of the approaching an metal cube for spontaneous inspection. (picture from Hofmann et al 2017