Testing spatial memory in the Morris water maze

Spatial memory is the memory of where things are. In the Morris water maze it is tested by putting an animal in a round pool of opaque water from which it can escape onto a small platform hidden just below the surface (Morris, 1984). The platform cannot be seen, so on the first trials the animal finds it only by chance. Over days of training it learns where the platform is from large cues fixed around the room, and it swims there more and more directly. A video camera above the pool records every path.

Rats with lesions of the hippocampus, a brain region needed for memories of places and events, were impaired at this place navigation (Morris et al., 1982), and infusing AP5, a drug that blocks NMDA-type glutamate receptors, impaired place learning without affecting visual discrimination learning, while blocking long-term potentiation, a lasting strengthening of synapses (Morris et al., 1986). These studies used rats; the simulated study uses mice in a smaller pool.

Faster escape over training is not enough to show memory: animals also stop swimming round the wall and swim more efficiently. Standard procedures therefore add a probe trial, in which the platform is removed and the time spent searching where it had been is measured, and cued trials with a visible platform, which reveal differences in vision, swimming or motivation that have nothing to do with place learning (Vorhees and Williams, 2006).

The simulation records what happened to every mouse, so the effect of each decision, including each mistake, on what the study can show is known.

Learning objectives

• Explain how the water maze tests spatial (place) memory, and why the water is opaque, the platform hidden and the cues fixed on the walls
• Plan a study with a justified sample size, a pre-specified primary outcome and exclusion rules, random allocation and blinding
• Run a training trial correctly: release facing the wall, guide a mouse that fails, record a capped latency and keep the mouse warm
• Interpret the probe trial and the visible-platform control, and say what each shows that training latencies do not
• Apply exclusion rules blind, then analyse learning curves with a mixed ANOVA and the Greenhouse–Geisser correction and compare probe-trial scores with chance
• Judge how far a result can be trusted from the way the study was designed and run

Ethics and safety

Water-maze studies are regulated procedures. They need approval by an Institutional Animal Ethics Committee; in India such work is governed by the CCSEA under the Prevention of Cruelty to Animals Act 1960, with equivalent frameworks in the European Union (Directive 2010/63/EU), the United States (Animal Welfare Act and PHS Policy) and the United Kingdom (Animals (Scientific Procedures) Act 1986). Drug administration follows the approved protocol. The 3Rs apply: Replacement where possible, Reduction of the number of mice to that given by a sample-size calculation, and Refinement, for example handling the mice before testing, keeping the water at a constant, recorded temperature, limiting trial length, guiding mice that fail, and drying and warming them after every trial. Randomisation, blinding, the sample-size justification, exclusions with their reasons and all outcome measures are reported following the ARRIVE 2.0 guidelines, developed by the NC3Rs in the United Kingdom and endorsed by journals internationally (Percie du Sert et al., 2020).

Key terms

• Spatial (place) memory: Memory of where something is, here the position of the hidden platform relative to the room.
• Distal cues: Large, fixed landmarks around the room, far from the pool. None marks the platform; its place is learned from their arrangement.
• Hidden platform: A small platform about 1 cm below the surface of opaque water (for mice usually 0.5 to 1 cm; Vorhees and Williams, 2006); reaching it lets the mouse escape from the water.
• Escape latency: The time from release until the mouse reaches the platform. A mouse that fails is given the maximum (here 60 s, the usual limit for mice), a capped value.
• Thigmotaxis: Swimming close to the wall of the pool, typical of naive or anxious animals; it fades with training.
• Acquisition: The training phase: several trials a day over several days, from different start positions, with the platform in the same place.
• Probe trial: A trial with the platform removed. Time spent in the target quadrant, where the platform was, measures memory of its place.
• Quadrant: One of four imaginary quarters of the pool (NE, NW, SW, SE) used in the analysis. A mouse with no memory of the platform spends roughly a quarter of the time in each (chance, about 25%), although not exactly, because it lingers near its start point and the wall. The four times add up to 100%, so they are not independent of one another.
• Cued (visible-platform) trials: Trials with a visible, flagged platform moved between quadrants; they test vision, swimming and motivation, not memory.
• Vehicle: The solution in which the drug is dissolved, given without the drug to the control group.
• Allocation concealment and blinding: The groups are hidden behind coded mouse IDs, so that whoever tests, scores and excludes the mice does not know who had the drug.
• Exclusion criteria: Rules written before the study that say which animals will be left out of the analysis, for example persistent floating.
• Mixed ANOVA: An analysis of variance with a between-mouse factor (group) and a within-mouse factor (day), used for learning curves.
• Greenhouse–Geisser correction: An adjustment of a repeated-measures test for data that violate sphericity, that is, when the differences between pairs of days are not equally variable. It multiplies the degrees of freedom by a factor ε (at most 1), which makes the test more cautious.