The three-chamber paradigm is a widely used task to evaluate spatial and non-spatial learning as well as memory in zebrafish. The watertight apparatus from MazeEngineers is outlined by a dark panel; covering one side of each compartment as a visual cue to provide an axis of orientation for right/left discrimination. The apparatus is primarily used for learning and memory but can also be used for various toxicity experiments.


Price & Dimensions


$ 1990

  • Total Width: 22cm
  • Total Length: 44cm
  • Total Height: 30cm
  • Middle chamber Width: 22cm
  • Middle chamber Length: 18cm
  • Door: Height: 12cm. Width: 10cm.
  • Choice Area: Width: 22cm. Length: 13cm



The three-chamber paradigm is a widely used task to evaluate spatial and non-spatial learning as well as memory in zebrafish. (D. Arthur, E.D. Levin, 2001)

The development of this test has helped to determine the molecular mechanisms of cognitive functions in zebrafish.

The subjects are trained to move in a particular direction when introduced into the central compartment of the chamber (spatial learning) or to approach a particular color regardless of the location (non-spatial learning).

This task can also assess reversal learning through aversive stimuli. The three-chamber maze is a simplified version of T-maze with an additional characteristic of aversive stimulus. (Levin ED, Cerutti DT, 2009)

  1. Arthur and E.D. Levin developed the original version of the three-chamber maze.

In a three-chamber maze, the subjects can either swim to the right or left compartment of the chamber. The choice of a correct compartment is rewarded by enhancing the swim area for the fish. On the contrary, incorrect choice leads to aversive stimuli in which the swim area is reduced. (Bailey JM 2015)

Spatial discrimination learning in zebrafish is helpful in differentiating response latency from choice accuracy. Acute and long-term drug effects can also be determined by the three-chamber task with choice accuracy and response latency serving as the reliable measures.

Apparatus and Equipment

The test apparatus consists of a Plexiglas maze fitted into a fish tank approx. 22 cm (w) 44 cm (l) 30 cm (h) in dimensions.

The tank is divided into three equal-sized compartments, the central compartment for the entrance of the subjects and the two adjacent side compartments for the subsequent movement of the subjects. (Levin ED 2003)

The Plexiglas maze is outlined by a dark panel; covering one side of each compartment as a visual cue to provide an axis of orientation for right/left discrimination.

On either side of the central compartment, vertical sliding windows approx. 12 cm (h) and 10 cm (w) are present that lead to right and left compartments.

The sliding partitions are mounted on rails to restrict the movement of the subjects as punishment. The sliding partitions can be moved as close as 2 cm from the end wall of the tank.

Training Protocol

The purpose of the three-chamber paradigm is to evaluate spatial and non-spatial learning behaviors in zebrafish in a vehicle-treated control vs. disease model/intervention group, by assessing their choice accuracy and response latency in the three-chamber maze.

Pretreatment for three-chamber task

Zebrafish are kept in tanks with recirculating distilled water (instant ocean salt and vitamins added).

The water temperature is maintained at approximately 28.5 °C, and the subjects are kept in 12-12 hour light/dark cycle.

The subjects are usually housed in tanks with proper aeration and containing recirculating deionized water. The subjects are fed with flakes or brine shrimp twice a day.

Every behavioral test is favored during the light phase between 8 a.m. to 5 p.m.

Evaluation of Spatial learning

Initially, the subject should be introduced into the central compartment of the three-chamber maze. After one minute, both sliding doors are opened simultaneously. Half of the subjects are trained to swim to the right, and half of them are trained to swim to the left compartment.

If the subject makes a right choice, the sliding window is closed, and the subject is allowed to swim freely (rewarded with more space) for 30 seconds. On the other hand, a wrong choice leads to punishment by limiting the swim space to 2cm from the end wall for 10 seconds.

If the subject fails to make a choice in 20 seconds, a fishnet is waved across the central chamber parallel to the doors as a threatening stimulus until the choice is made.

Choice accuracy and response latency are recorded as an average of 5 testing sessions.

Evaluation of Non-Spatial learning

Non-spatial learning includes color discrimination.

For color discrimination, colored inserts are added at the ends of each adjacent compartments. The positioning of the inserts is randomly rotated so that the subject chooses the color regardless of the location.

Evaluation of Reversal learning

Reversal learning can be evaluated for both spatial and non-spatial discrimination. Initially, the subjects are trained to swim to one side (spatial discrimination) or color (non-spatial discrimination). Afterward, the contingencies are switched to evaluate reversal learning behavior in zebrafish. (Levin ED, 2011)


The original version of the maze was rectangular in shape. (D. Arthur, E.D. Levin, 2001)

The most recent version of the three-chamber maze was developed by Eddins 2009 for investigating the effects of nicotine on learning in zebrafish.

The testing apparatus consists of a semi- cylindrical pipe unit divided into three chambers. The chambers are named as the start chamber, right choice chamber, and the left choice chamber. Plexiglas rods are extended through both sides of the apparatus.

The inner side of the rod is provided with rotatable plastic partitions that permit the entry of the subjects.

The partitions are circular in shape and vertical in position. The partitions are attached to approx. 13 cm long rails. The rails can move freely from the central chamber to the end wall of the apparatus.

In the original version, a fishnet was waved, but in this modified version, a plastic beaker is dropped just above the start chamber as a threatening stimulus in order to evoke a response.

Sample Data

A sample data is represented by plotting spatial, on-spatial, and reversal learning as a function of response latency in control versus drug administered group. It is evident from the graph that drug administered group [Chlorpyrifos (CPF)] depicts delayed response. Therefore the response latency is enhanced.

Strengths & Limitations


One of the most exceptional advantages of the three-chamber task is that the subjects can be tested repeatedly because of the learning ability of the subjects.

The three-chamber task is preferred over the two-chamber shuttle box because the activity of the fish is not confused with the choice accuracy as the fish needs to cover the same distance for both correct and wrong choices.

However, this task is unrelated to either active or passive avoidance in the two-chamber shuttle box task.

This task can be used to determine the effects and toxicity of different drugs on the learning behavior of zebrafish. For instance, the molecular mechanisms underlying nicotine effects on learning (Eddins et al. 2009) and CPF-induced developmental, behavioral toxicity can be assessed. (Levin ED 2003)


As response latency is recorded manually, therefore skilled personnel supervision is necessary for the accomplishment of this task, otherwise the interpreted results may vary.

Summary and Key Points

  • The three-chamber paradigm is an extensively used task to evaluate spatial and non-spatial learning as well as memory in zebrafish.
  • The apparatus consists of three equal-sized compartments for the evaluation of learning behaviors in zebrafish.
  • Behavioral testing is ideally conducted during the light phase between 8 a.m. to 5 p.m.
  • Spatial, non-spatial, and reversal learning can be evaluated with the help of the three-chamber test.
  • Besides learning behavior, the effects and toxicity of different drug compounds can also be evaluated using this paradigm.
  • The most recent version of the three-chamber maze is a three compartment semi-cylindrical pipe unit.


Arthur, E.D. Levin, Spatial and non-spatial discrimination learning in zebrafish, Anim. Cogn. 4 (2001) 125– 131

Levin ED, Cerutti DT. Behavioral Neuroscience of Zebrafish. Methods of Behavior Analysis in Neuroscience. 2nd edition (2009)

Bailey JM, Oliveri AN, Levin ED. Pharmacological analyses of learning and memory in zebrafish (Danio rerio) Pharmacol. Biochem. Behav. 2015

Levin ED, Chrysanthis E, Yacisin K, Linney E. Chlorpyrifos exposure of developing zebrafish: effects on survival and long-term effects on response latency and spatial discrimination. Neurotoxicol Teratol. 2003; 25:51–57

Levin ED. Zebrafish assessment of cognitive improvement and anxiolysis: filling the gap between in vitro and rodent models for drug development. Rev Neurosci. 2011; 22:75–84

Eddins D., Petro A., Williams P., Cerutti D. T. & Levin E. D. Nicotine effects on learning in zebrafish: the role of dopaminergic systems. Psychopharmacology (Berl). 202, 103–109 (2009)