ZFIN Image: Torigoe et al., 2021, Fig. 8

Image

Figure Caption

Fig. 8 a Schematic circuit diagram to select swimming forward. The neural network comprises ensembles encoding the perception of blue (BP, blue) and swimming forward (SF, red). Moreover, two additional units represent mutually different prediction errors. The reward prediction error (RPE, green) ensemble computes the difference between the (negative) reward prediction (RP) and actual punishment. Whereas, the scenery flow prediction error (SFPE, cyan) computes the difference between the scenery flow prediction (SFP) and actual backward flow. Here, the RPE ensemble took a positive value when the fish could avoid the punishment contrary to its expectation; otherwise, it took zero. The SFPE ensemble self-organized in the early stage of training to take a positive value when the fish sensed the SFPE at a given time point or take zero otherwise. The synaptic potentiation of $W$

Acknowledgments

This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ Nat. Commun.