A Behavioural Assay to Investigate Judgment Bias in Zebrafish.

In this protocol, we describe for the first time a judgment bias paradigm to phenotype the way zebrafish assess ambiguous stimuli. We have developed and validated a protocol for a judgment bias test based on a Go/No-go task, and performed using a half radial maze. After a habituation phase, fish are trained to discriminate between two reference arms [positive (P) and negative (N)]. For this purpose, they experience a positive event (food reward in P), when presented with a specific location/color cue, and a negative event (chasing with net in N), when presented with a different location/color cue. Acquisition of the discrimination learning between P and N is revealed by the latencies to enter the experimental arms of the behavioral maze being significantly lower for the P arm than for the N arm. Once zebrafish are able to discriminate between P and N arms, their latency to enter other maze arms spatially located between P and N [(Near Positive (NP), Ambiguous (A) = half-way between P and N, and Near Negative (NN)] is analyzed. Latencies (L) to enter NP, A and NN maze arms are interpreted as indicating the individual expectancy to experience a reward/punishment on each of them. A judgment bias score (JBS) is calculated from the latencies to enter the P, N, and A arms for each fish [JBS = (L