Experience, circuit dynamics, and forebrain recruitment in larval zebrafish prey capture

Claire S. Oldfield, Irene Grossrubatscher, Mario Chávez, Adam Hoagland, Alex Huth, Elizabeth C. Carroll, Andrew Prendergast, Tiange Qu, Jack L. Gallant, Claire Wyart, Ehud Y. IsacoffeLife 2020 · DOI

TL;DR

Larval zebrafish learn to hunt more effectively after experiencing live prey, not by improving their visual processing of prey location, but by strengthening connections to forebrain areas like the habenula. This experience primes the brain to more readily initiate capture behaviors in response to visual cues, effectively lowering the threshold for action.

What Problem Does This Paper Address?

How does experience change the brain to improve performance in complex behaviors? While we know experience matters, the specific neural circuits and dynamic changes that lead to enhanced skills are often unclear. This paper investigates this question in the context of prey capture in larval zebrafish, a fundamental survival behavior. Zebrafish are known to hunt from a young age, but the study asks if and how prior exposure to live prey refines this innate ability at the neural level. Specifically, they want to understand if experience alters how the brain processes visual information about prey or if it changes how that information is translated into motor commands.

Methods

The researchers employed a multi-pronged approach:

  • Behavioral Assays: They compared "prey-experienced" zebrafish (fed live paramecia) with "prey-naïve" zebrafish (fed inert flakes) in both free-swimming and semi-immobilized virtual prey capture scenarios. This allowed them to quantify capture success, initiation probability, and response to virtual stimuli.
  • Wide-field Calcium Imaging: Using genetically encoded calcium indicators (GCaMP6f), they imaged neural activity across large portions of the zebrafish brain in awake, behaving larvae during prey encounters. This allowed them to observe which brain regions were activated and when.
  • Granger-Causality Analysis: This statistical method was used to infer directional information flow between different brain regions based on their calcium activity patterns. This helped them understand how neural signals propagate through the prey capture circuit.
  • Forebrain Disruption: To establish a causal link, they used a genetic tool (nitroreductase system activated by metronidazole) to selectively ablate neurons in the habenula, a forebrain structure, and observed the impact on prey capture behavior in experienced fish.

Datasets

The study used larval zebrafish (Danio rerio) as the model organism. No specific external datasets were used for training or evaluation; the data was generated from experiments conducted on these fish.

Key Insights and Ideas

  • Experience Enhances Initiation, Not Just Execution: The key finding is that experience doesn't make the zebrafish better at the physical act of capturing prey once initiated, but rather makes them more likely to initiate the capture sequence in the first place.
  • Forebrain as a "Gain Control" Mechanism: The study points to the telencephalon and habenula (forebrain structures) as crucial for this experience-dependent improvement. These areas show increased activity in experienced fish and their disruption impairs capture. This suggests the forebrain acts as a "gain control" mechanism, amplifying relevant signals and lowering the threshold for action.
  • Visual Processing Remains Similar: Interestingly, the basic visual processing of prey location in areas like the pretectum and optic tectum appears unchanged by experience. This implies the learning doesn't happen at the initial sensory input stage.
  • Strengthening of Specific Pathways: Granger-causality analysis revealed that while basic visual-to-motor pathways are present in both experienced and naïve fish, the connection from tectal PVNs to the telencephalon is significantly stronger in experienced fish. This highlights a specific circuit modification.

Key Results

  • Prey-experienced fish showed significantly higher rates of prey pursuits and successful captures compared to prey-naïve fish.
  • In a virtual prey capture assay, experienced fish exhibited a higher discriminability index (d') and eye convergence rate, especially at higher stimulus contrasts.
  • Calcium imaging revealed similar activation patterns in visual (pretectum, optic tectum) and motor (cerebellum, hindbrain) areas for both groups when responding to prey.
  • However, prey-experienced fish showed greater visually-evoked activity in the telencephalon and habenula during prey capture initiation.
  • Disrupting habenular neurons in prey-experienced fish significantly reduced their prey capture initiation and performance.
  • Granger-causality analysis indicated that while visual-to-motor area connectivity was similar, the link from tectal PVNs to the telencephalon was stronger in experienced fish.

Limitations and Future Work

  • Model Organism: While zebrafish are a powerful model, the findings may not directly translate to all species, especially mammals.
  • Causality of Forebrain: While habenula disruption showed a causal role, the precise mechanisms by which the telencephalon and habenula modulate the prey capture circuit require further investigation.
  • Generalizability of Experience: The study focused on live prey vs. inert food. Future work could explore other types of experience and their impact on different behaviors.
  • Specific Neuronal Populations: The study used broad imaging and disruption. Future research could delve into the specific neuronal populations within the telencephalon and habenula that are critical for this learned improvement.
  • Neuromodulation: The authors acknowledge that neuromodulatory systems (e.g., dopaminergic, noradrenergic) might also play a role in gain control and could be explored further.

Summary updated

Abstract

Experience influences behavior, but little is known about how experience is encoded in the brain, and how changes in neural activity are implemented at a network level to improve performance. Here we investigate how differences in experience impact brain circuitry and behavior in larval zebrafish prey capture. We find that experience of live prey compared to inert food increases capture success by boosting capture initiation. In response to live prey, animals with and without prior experience of live prey show activity in visual areas (pretectum and optic tectum) and motor areas (cerebellum and hindbrain), with similar visual area retinotopic maps of prey position. However, prey-experienced animals more readily initiate capture in response to visual area activity and have greater visually-evoked activity in two forebrain areas: the telencephalon and habenula. Consequently, disruption of habenular neurons reduces capture performance in prey-experienced fish. Together, our results suggest that experience of prey strengthens prey-associated visual drive to the forebrain, and that this lowers the threshold for prey-associated visual activity to trigger activity in motor areas, thereby improving capture performance.
BibTeX
@inproceedings{oldfield2020experience,
  author = {Claire S. Oldfield and Irene Grossrubatscher and Mario Chávez and Adam Hoagland and Alex Huth and Elizabeth C. Carroll and Andrew Prendergast and Tiange Qu and Jack L. Gallant and Claire Wyart and Ehud Y. Isacoff},
  title = {{Experience, circuit dynamics, and forebrain recruitment in larval zebrafish prey capture}},
  booktitle = {eLife},
  year = {2020},
  doi = {10.7554/elife.56619},
}

Loading related content...

Figure