Current Protocols in Neuroscience
Featured Protocol

Behavioral tasks must be evaluated in terms of the cognitive functions they require in order to be performed. Each task is a tool that allows the researcher to achieve a specific goal--i.e., to determine the consequences of manipulations of specific brain regions. These manipulations usually fall into one of four categories: stimulation of a single brain region by drugs or small electrical current, impairment of normal function by production of a lesion or administration of appropriate pharmacological agents, recording of brain activity during the performance of a specific behavioral task, or behavioral phenotyping of transgenic and knockout mice for genes expressed in specific brain regions. All of the tasks described in this chapter can be used with each of these four experimental manipulations.

Performance of the radial arm maze task (see Basic Protocol 1 and Alternate Protocol 1) requires intact spatial memory abilities. Normal performance is sensitive to the effects of hippocampal damage, normal aging, and a variety of pharmacological agents. Performance of the water maze task (see Basic Protocol 2 and Alternate Protocols 2 and 3) also requires intact spatial memory abilities and is particularly sensitive to the effects of aging. The major advantage of the water maze task over the radial arm maze task is that the rats do not need to be water or food deprived; they are quite motivated to escape from the water. The task is also free from errors of omission or abortive choices--i.e., the rat makes an attempt to find the platform on every trial.

Rats are generally used in research employing these three behavioral tasks because a considerable amount is known about their brain anatomy and chemistry. Previous experimental studies have clearly shown that rats can be used to investigate the structure/function relationships between selected brain regions and learning or memory. Eight to ten rats are included in each experimental group, a number that is sufficient for statistical analyses (e.g., an analysis of variance, ANOVA) of the behavioral data and any neurochemical assays or histopathological studies that might be performed for confirmation of the manipulations.
BASIC PROTOCOL 1
USE OF RADIAL ARM MAZE TASK TO TEST BASIC WORKING MEMORY
The radial arm maze task has been most extensively used to investigate specific aspects of spatial working and reference memory. This task is based upon the premise that animals have evolved an optimal strategy to explore their environment and obtain food with the minimum amount of effort.

The radial arm maze can easily be built by the investigator, so the cost for this equipment can be quite low compared to other behavioral testing equipment. Using the maze, however, is quite labor-intensive and requires that a tester be present throughout the task. The maze is typically used for rats, but it can be scaled down in size (by ~75%) for use with mice.

Materials

Rats
Radial arm maze (Fig. 8.5.1), handmade or fully automated (Coulbourn Instruments or Columbus Instruments)

Train rats

1. Weigh each rat daily throughout training and testing to monitor health and degree of food deprivation.
2. Restrict food available to rat so that its body weight attains 85% of that prior to training. During testing and training, allow rat to gain ~5 g body weight per week
3. Allow rat to become comfortable with the experimenter (see Critical Parameters, discussion of rat handling).
4. Give food reward in home cage for a few days prior to training in order to acclimate the rat to the reward in a familiar environment.

The food reward is typically a small piece (10 mg) of normal chow or a flavored (chocolate is a favorite) or sweetened breakfast cereal. Liquid rewards, such as chocolate milk or water, can also be used. Liquid rewards are preferred if the rat will be given a drug, e.g., scopolamine, that might make swallowing dry food uncomfortable.

5. Set up radial arm maze (see Fig. 8.5.1).

The wooden or Plexiglas maze is set up ~1 m above the floor for easy access to the rat and food cups. It is composed of a central octagonal platform with eight arms extending from it like the spokes of a wheel. Guillotine doors that can be opened and closed individually separate the central platform from the arms. The maze should be placed in a room that contains various external cues that are visible to the rat while on the maze--e.g., a doorway, overhead lights, a noisy radio, and large simple designs on the walls. The maze can be handmade or a fully automated setup purchased commercially. The latter, including data analysis software, can be obtained from either Coulbourn Instruments or Columbus Instruments.

6. Place a well-handled pair of rats (preferably cagemates) on the maze at the same time.

Using two rats reduces the time it takes for each to acclimate to the maze.

7. Spread food rewards around the entire maze to encourage exploration.

Acclimation period should require only 1 or 2 days.

8. On subsequent days, place food only on the arms, then only at the ends of the arms.
9. Finally, place rat alone on maze and food only in the food cup at end of arms. Testing can begin when rat is comfortable being picked up by the experimenter and, when placed alone on the maze, explores without hesitation and without excessive defecation or urination.

A typical rat will be ready for testing--i.e., food-restricted and acclimated to the maze--within ~7 days. Rats should be run in the maze once a day every day (including weekends, ideally)during training and testing).

Test rats

10. Place food reward at end of each arm before each test session.
11. Place rat on central platform with all guillotine doors closed.
12. Raise all doors simultaneously. Allow rat to enter an arm. Close doors to all other arms.
13. Allow rat time to eat food and to return to central platform.
14. Close door to that arm and confine rat to the central platform area for a set time (from 0 sec to many min; 5 sec is ideal to begin).

Longer waits make the task more difficult to solve--i.e., increase the length of time for which the rat must remember which arms it has entered.

15. Repeat steps 12 to 14 until all food pellets have been retrieved or until a predetermined length of time has elapsed.
16. Record the following data:

Which arm the rat entered each time and whether it received a food reward.

Time elapsed between the beginning of the test session and the rat's obtaining all eight food rewards.

Number of correct arm choices: i.e., those that are chosen the first time.

Number of incorrect arm choices: i.e., visitation to the same arm more than once during a single test session.

Time elapsed since the beginning of the session is considered in order to determine how fast the rat is making choices and finding the food rewards. This is also an indirect indication of motivation.

Visitations to a previously chosen arm is considered a working memory error. Normal healthy young rats will perform this task almost perfectly every time.

A healthy, happy, and motivated rat should learn this task, such that working memory errors are <15%, within 15 days.

17. Begin studies with pharmacological agents or lesions when performance is stable and choice accuracy is >85%.

Alternatively, one can study the effects of drugs or lesions upon the acquisition of performance of this task. Drugs can be administered (or lesions produced) either prior to training to assess their effects upon acquisition, or after acquisition in order to assess their effects upon performance.

ALTERNATE PROTOCOL 1
USE OF RADIAL ARM MAZE TASK TO TEST WORKING VERSUS REFERENCE MEMORY
This protocol allows a disassociation to be achieved between working and reference types of memory, whereas the previous protocol is primarily sensitive to impairments in working memory. Working memory is operationally defined as information that is only useful to a rat during the current experience with the task, whereas reference memory is information that is useful across all exposures to the task--i.e., on any day of testing. This protocol should be performed after initial training of rats (Basic Protocol 1) has been completed.

1. Train rats in radial arm maze (see Basic Protocol 1, steps 1 to 9).

2. Place food reward at the end of only four arms of the radial arm maze before each test session.

The arms chosen to be baited must be the same for a given rat but should vary between rats. For example, bait arms 1, 3, 6, and 8 every time that rat #1 is on the maze, bait arms 1, 4, 5, and 7 every time that rat #2 is on the maze, and so on. This task can be conducted on mazes with more or fewer arms in much the same way.

3. Place rat on maze with all doors raised and allow it to explore the maze completely and retrieve all food rewards

.
4. Remove the rat from the maze for a set period of time (1 hr to 1 day).

Longer delays make the task more difficult for the rat.

5. Bait appropriate four arms of maze for rat.

In the example discussed above, bait arms 1, 3, 6, and 8 for rat #1.

6. Repeat step 3.

7. Record the following:

Number of correct entries into baited arms.

Number of entries into unbaited arms.

Number of reentries into baited arms.

Time elapsed between the beginning of the test session and the rat's obtaining all available food rewards.

Entries into unbaited arms are reference memory errors; reentries into baited arms are working memory errors.

8. Repeat steps 2 to 7 until performance is stable and choice accuracy is high (>85%).

9. Begin studies with pharmacological challenges or lesions once the rat makes few, or no, reference or working memory errors.

Normal healthy young rats will perform this task almost perfectly every time. A healthy, happy, and motivated rat should learn this version of the task, such that working and reference memory errors are <15%, within 10 days.

BASIC PROTOCOL 2
USE OF MORRIS WATER MAZE TASK TO TEST SPATIAL MEMORY
The water maze task has been most extensively used to investigate specific aspects of spatial memory. This task is based upon the premise that animals have evolved an optimal strategy to explore their environment and escape from the water with a minimum amount of effort--i.e., swimming the shortest distance possible. The time it takes a rat to find a hidden platform in a water pool after previous exposure to the setup, using only available external cues, is determined as a measure of spatial memory. Studies with pharmacological agents or lesions are initiated when performance is stable; the water maze task is particularly sensitive to the effects of aging (Brandeis et al., 1991). Alternatively, one can study the effects of drugs or lesions upon the acquisition of performance of this task. Drugs can be administered (or lesions produced) prior to training to assess their effects upon acquisition, or after acquisition in order to assess their effects upon performance. The maze (Fig 8.5.2) can easily be built or purchased by the investigator, so the cost for this equipment can be quite low. Using the maze is quite labor-intensive and requires that a tester be present, or nearby, throughout the task. This maze is typically used for rats, but it can be scaled down in size (by ~50%) for use with mice.

Materials

Rats
Water maze apparatus
Tracking system and software (Columbus Instruments, HVS Image, San Diego Instruments, or CPL Systems)

Set up apparatus and begin acquisition testing

1. Set up water maze (see Fig. 8.5.2).

A water-tight pool, painted white, should be positioned in a room with various external cues that are visible to a rat swimming in the pool, e.g., a doorway, overhead lights and camera (if desired), and large simple designs on the walls. Make water opaque by adding powdered milk or nontoxic white paint to the water. The pool should be designed so that it can be easily drained on a regular basis.

2. Insert platform into one quadrant of the pool.

3. Place rat into water with its head pointed towards the side of pool.

The starting position should be at a different, and randomized, location each day of testing, e.g., north, south, etc.

4. Record time (in sec) it takes the rat to find the submerged platform. Guide rat to platform on first few trials if it requires >120 sec.

A tracking camera, positioned ~200 cm above the center of the pool, can be used to quantify the distance swam on each trial and thereby determine swimming speed when combined with latency measurements. The tracking system can also display swim path and distance and provide additional information on search efficiency and exploration patterns during acquisition and probe trials. This equipment and associated computer software can be obtained from several commercial manufacturers.

5. Allow rat to remain on platform for 10 to 15 sec.

This allows the experimenter time to return to an appropriate place at the side of the water pool in order to be ready for step 6.

6. Remove rat from pool. Wait 5 min.

7. Release rat into pool (from the same location) with platform in same location. Record time for rat to find platform.

8. Give each rat four trials on the first day.

Perform trials

9. On second day, insert platform in same location as on the first day.

10. Release rat with its head pointed towards the side of the water pool.

11. Record time it takes rat to find platform.

12. Give rat eight to ten trials per day with 5-min intertrial intervals for several days until performance is stable and latency to find the platform is low (<5 to 7 sec).

13. Begin studies with pharmacological agents or lesions.

ALTERNATE PROTOCOL 2
USE OF WATER MAZE TASK FOR SPATIAL PROBE TRIAL
This alternate protocol should be performed after the acquisition phase of testing in Basic Protocol 2 has been completed. It is important that the rats know the location of the hidden platform before beginning this protocol. This knowledge is demonstrated by the rat swimming quickly and directly to the hidden platform. The spatial probe trial is used to test the rat's knowledge of the precise location of the platform. An accurate direction of the swimming behavior provides evidence that the rat has learned the spatial location of the platform relative to the available external cues (Sutherland et al., 1982). Although the spatial probe trial is typically performed only after performance has stabilized, some researchers perform a spatial probe trial at the end of each day of training throughout the acquisition phase, and average performance across trials.

1. Train rats in water maze (see Basic Protocol 2, steps 1 to 12).

2. Set up water pool without platform.

3. Release rat with its head pointed towards the side of the water pool.

4. Remove rat after 90 sec.

5. Record time rat spent in the quadrant that previously contained the platform and calculate as a percentage of total time in pool. If possible (i.e., if using a computer tracking system), also record the percentage of time spent in the other quadrants.

ALTERNATE PROTOCOL 3
USE OF WATER MAZE TASK TO TEST WORKING MEMORY
This alternate protocol can be performed after the acquisition phase of testing in Basic Protocol 2 has been completed. It is important that the rats have demonstrated that they know the location of the hidden platform before beginning this next protocol. This protocol has also been referred to as a "reversal test."

1. Train rats in water maze (see Basic Protocol 2, steps 1 to 12).

2. Release rat with its head pointed towards the side of the water pool.

Start position should be at same location each day.

3. Record time it takes rat to find submerged platform. Allow rat to remain on platform for 10 sec. Remove rat from pool and place in holding cage for 15 sec.

4. Move submerged platform to new location.

5. Release rat from same location as in step 2.

6. Allow rat to swim for up to 120 sec. Record time it takes for rat to find platform.

Guide rat to platform if necessary.

7. Allow rat to remain on platform for 10 sec. Remove rat from pool and place in holding cage for 15 sec.

8. Repeat steps 5 to 7 until rat swims directly and quickly to the platform. Record time on each attempt.

9. Repeat steps 4 to 8 once per day for 4 days, with the platform in a different quadrant of the pool each day.

Latency to find the platform should decrease with each day of testing.

COMMENTARY
Background Information

Radial arm maze task

The radial arm maze task was introduced and popularized in its present form by Olton and co-workers (Olton and Samuelson, 1976). The task is a logical extension of the multiple, simultaneous choice tasks originally described by Hamilton (1911) and Tolman et al. (1946); it has been used to measure the effects of various brain manipulations upon specific aspects of memory, such as spatial working and reference memory, and can be adapted for use with rats, mice, and pigeons (Bond et al., 1981; Levy et al., 1983; Wenk et al., 1986). The large number of sequential locations that the rat can visit to obtain a reward makes the task ideal for investigating the effects of drugs or lesions upon serial order memory--i.e., whether locations visited first or last are remembered better (Kesner and Novak, 1982). The task is sensitive to the effects of brain lesions (Becker et al., 1980) and to numerous drugs (for review see Levin, 1988) that either impair or enhance performance, including inebriants such as ethanol (Devenport et al., 1983), endogenous neuropeptides such as vasopressin (Buresova and Skopkova, 1982), amnestic drugs such as scopolamine (Stevens, 1981; Watts et al., 1981; Okaichi and Jarrard, 1982), and neurotoxins such as trimethyltin (Walsh et al., 1982).

Water maze task

The water maze task was designed to address theoretical controversies that arose from using the radial arm maze task (Brandeis et al., 1989): i.e., the concept that memories about spatial information are handled by the brain quite differently than information on other forms of learning. The water maze task is not a better or more sensitive task than the radial arm maze task, it simply asks many of the same questions of the animal under different circumstances. Differences between the two tasks, walking versus swimming, include the nature of the locomotion; the nature of the motivation; food deprivation versus avoidance of drowning; the location of available cues for finding the reward, local and distant cues versus distant cues only; the visibility of the location of the reward, a visible food cup on the radial arm maze versus a submerged platform in the water maze task. Sometimes asking the same question in a different way has allowed researchers to discover subtle differences in the contributions of different brain regions or the effects of specific lesions or drugs. The water maze task was introduced by Morris (1981) and colleagues as a spatial localization or navigation task. The task has been extensively used to study the neurobiological mechanisms that underlie spatial learning and memory, age-associated changes in spatial navigation (Gage et al., 1984; Rapp et al., 1987; Pitsikas et al., 1990), and the ability of psychopharmacological agents (Sutherland et al., 1982; Hagan et al., 1983, 1986; McNaughton and Morris, 1987; Brandeis et al., 1991; McNamara and Skelton, 1991), lesions (Morris et al., 1982; Kolb et al., 1983), or gene mutations (Tsien et al., 1996; Crawley et al., 1997) to influence specific cognitive processes.

Critical Parameters

Essence of rat handling

Remain relaxed. The rat can sense your nervousness. Be consistent with your treatment and handling. Use the rat's native intelligence and your skills to provide him with the specific knowledge that he needs to perform the task. Take the rat's point of view. Handle the rat the way that you would like King Kong to handle you! Go slowly and be gentle. Do not approach the rat from behind and above; this is how a predator would attack the rat. Do not grab the rat tightly around the abdomen, as its internal organs have very little protection. Support the rat from underneath its body and hold it against yours. Run rats on the maze at the same time every day; rats are like people in that they grow accustomed to a particular schedule. Allow the rat to gain ~5 g each week during testing and training, even though it is food-restricted. Pay attention to how well groomed the animal is. A well-groomed rat is a healthy rat. If the rat is undernourished, it will feel cold to the touch. Sick rats do not provide useful data and may be in danger of dying. Find out why the rat is sick and correct the problem.

Radial arm maze task

Rats will use whatever sensory cues are available to solve the radial arm maze task and obtain a reward. Removing these cues will make performance difficult and impair choice accuracy. Reducing the salience of stimuli around the maze--e.g., completely enclosing the arms or placing the maze in a homogeneous environment--can greatly influence performance. This may force the rat to use other, more egocentric, information to solve the task, such as a sequence of left turns after every arm choice, or to depend upon intramaze cues, such as odor trails.

The guillotine doors are a critical feature of the maze in that they confine the rat to the central platform area between choices. Otherwise, the rat may develop a biased response pattern, which makes interpretation of the performance difficult: i.e., it becomes impossible to determine whether the rat remembered the correct choice or a response habit. For example, without temporary confinement between each arm choice, the rat could successfully solve this task by simply always turning right after each choice and entering the first arm away from the previously chosen one. This simple strategy does not require an accurate knowledge of the spatial environment or memory for a specific location. Unless the experimenter is primarily interested in studying response patterns, it is best to have the rat confined to the central platform prior to making each arm choice.

The number of arms the maze contains can vary depending upon the goals of the experimenter. For example, having fewer arms requires that the animal remember fewer visited places on each trial. Increasing the number of arms increases the mnemonic demands of the task by increasing the list of spatial locations in memory. In addition, an increased number of arms introduces considerably more proactive interference, i.e., interference of previous learning on current memory. Most researchers choose to use eight arms in order to minimize proactive interference (if that is desired) or to shorten the length of time it takes to test each rat. Numerous variations have been introduced in order to automate the task, reduce testing time, or provide alternative interpretations of the psychological processes that underlie normal performance (Bond et al., 1981; Okaichi and Jarrard, 1982).

Most rats are quite fearful of exploring the arms during the initial training periods. Sometimes it is better to have somewhat taller sides (6 to 8 cm) along the edges of the arms to offer more support to the rat. A taller barrier may be attached along the edges of the arms near the central platform. This prevents the rat from jumping from one arm to another and forces the rat into the central platform area between choices. The guillotine doors should be attached by strings to an overhead pulley system that allows all doors, or individual doors, to be raised and lowered by one experimenter from a single location. Alternatively, individual electronic mechanisms could also be installed at a greater cost.

The food reward is typically a small piece (10 mg) of normal chow or a flavored (chocolate is a favorite) or sweetened breakfast cereal. Liquid rewards, such as chocolate milk or water, can also be used. Liquid rewards are preferred if the rat will be given a drug, such as scopolamine, that might make swallowing dry food uncomfortable.

Water maze task

Rats will use whatever sensory cues are available to solve this task and escape from the water. Removing external cues around the pool will make finding the submerged platform more difficult and increase the latency to escape. Reducing the salience of stimuli around the pool may force the rat to use other, more egocentric, information to solve the task. Rats will use a search strategy to find the platform when it is moved. They will swim close to the wall for a few laps then move further away and continue swimming in concentric circles until they bump into the platform. Latency to find the platform and swimming distance will then decrease quickly thereafter. The greatest advantage of this task over food-motivated tasks is that most rats are more highly motivated to escape from the water. In addition, food restriction is unnecessary, which is a great advantage when testing aged animals, to which such restriction is more stressful.

Performance in the water maze task can be influenced by many factors that should be considered carefully when comparing the results of one study with another. For example, the sex and strain of the rats, the dimensions of the pool and temperature of the water, and the particular training schedule can all affect performance. One should also take into account factors that affect swim speeds, which include body weight, muscle development, and age. See Brandeis et al. (1989) for a thorough discussion of the role of these factors in performance of the Morris water maze task.

Finally, because older rats or mice are frequently tested in the water maze, it is important to be sure that they can swim adequately and have sufficient visual acuity to use distant cues. To test this, place rat into the pool and allow it to swim to a platform that is supported above the water level. To assist rat, suspend a large visible cue above the platform. If the rat can swim directly to the visible platform without difficulty, it is ready to begin testing using the protocols outlined in this unit.

Troubleshooting

Radial maze task

The major problems associated with testing rats in open mazes are usually related to two competing factors for the rat: its fear of the maze (or experimenter) versus its motivation to explore and find the food that it knows is on the maze. Excessive levels of the first factor will prevent the rat from performing. It will usually remain frozen in one place on the maze and not explore. In addition, if it is frightened it will usually defecate and urinate on the maze and squeal when being picked up. If the rat is not making choices on the maze it is impossible to know whether its mnemonic abilities are normal or impaired. Fear can be overcome by considering some of the issues presented in the discussion of rat handling above (see Critical Parameters). The most important thing the experimenter can do is simply handle the animal more often. Also, higher barriers can be installed along the arms of the maze to provide the rat a better sense of security from falling.

A lack of the second factor will produce similar results--i.e., the rat will make little or no attempt to explore the maze and find food. Motivation can be increased by a slightly greater restriction of food intake. The rat's weight and general health must be carefully monitored during food restriction. Usually the rat's weight should not drop below 80% of its free-feeding weight. For most rats, it is only necessary to reduce their weight by 15%. If the rat is aware that a safe food source exists on the maze, they will usually explore the maze to find it.

Water maze task

There are only two problems that are typically associated with this task. First, immersion of the animal into the water may cause significant stress and subsequent endocrinological changes that may interfere with the purpose of the study. These problems are usually resolved with continued exposure to the pool. However, for aged animals this stress can be sufficient to induce cardiovascular system collapse leading to death or stroke. Second, the method by which the water is made opaque can produce problems. If powdered milk is added to the water, the pool must be drained regularly (e.g., daily) to avoid bacterial contamination and odor. If powdered white paint is used, care must be taken to ensure that it is nontoxic to the animal, who may frequently ingest small quantities of water while performing the task.

Anticipated Results and Time Considerations

Radial arm maze task

A healthy, happy, and motivated rat should learn this task within 15 days: i.e., working and reference memory errors should be <15%. This assumes that the rat is actively making choices and eating the food rewards on the maze. The rat should run quickly down the arms to retrieve the food reward and return to the central platform immediately after eating the reward. During the confinement period the rat will usually wander around the central area and quickly choose an unvisited arm as soon as the guillotine doors are raised. After a control rat has learned the task, a single test session--i.e., eight correct arm choices--should not require more than 5 to 10 min to complete. Rats given drugs or lesions may require two to three times longer to complete the test session. A healthy group of rats, including experimental and control groups, should be able to be trained and tested in Basic Protocol 1 and Alternate Protocol 1 in ~4 weeks. This estimate assumes that all rats are tested once per day.

Water maze task

Control rats may take ~30 to 60 sec to find the platform on the first day of testing. Latency should decrease significantly with subsequent trials to ~5 sec. On the probe test, the rats should spend a greater percentage of time exploring the quadrant that previously contained the platform than in the other quadrants. On both the reversal and working memory tests, the rats should first explore the quadrant that previously contained the platform and then inadvertently find the new location of the platform. On subsequent trials (usually less than five), their latency to find the platform will decrease significantly. All rats should be trained in the Basic Protocol 2 portion of this task within 1 week. The spatial probe trial (Alternate Protocol 2) can be completed in 1 day, and the reversal test (Alternate Protocol 3) within 5 days. Lesioned rats may require a few additional days of training in Basic Protocol 2 in order to reach an asymptotic level of performance. Drugs can be given each day prior to testing in this protocol; however, this will also extend the amount of time required to reach asymptotic performance. Drug administration will also prolong the number of days spent testing in the spatial probe and reversal trials; the number of days depends upon the number of doses to be tested. It is important to have drug-free and saline injection days interspersed between the drug testing days. Obviously, this will prolong the time required to complete this task.
Literature Cited
Becker, J.T., Walker, J.A., and Olton, D.S. 1980. Neuroanatomical bases of spatial memory. Brain Res.200:307-320.

Bond, A.B., Cook, R.B., and Lamb, M.R. 1981. Spatial memory and the performance of rats and pigeons in the radial-arm maze. Anim. Learn. & Behav.9:575-580.

Brandeis, R., Brandys, Y., and Yehuda, S. 1989. The use of the Morris water maze in the study of memory and learning. Int. J. Neurosci.48:29-69.

Brandeis, R., Sapir, M., Kapon, Y., and Borelli, G. 1991. Improvement of cognitive function by MAO-B inhibitor L-deprenyl in aged rats. Pharmacol. Biochem. Behav.39:297-304.

Buresova, O. and Skopkova, J. 1982. Vasopressin analogues and spatial working memory in the 24-arm radial maze. Peptides3:725-727.

Crawley, J.N., Belknap, J.K., Collins, A., Crabbe, J.C., Frankel, W., Henderson, N., Hitzemann, R.J., Maxson, S.C., Miner, L.L., Silva, A.J., Wehner, J.M. Wynshaw-Boris, A., and Paylor, R. 1997. Behavioral phenotypes of inbred mouse strains: Implications and recommendations for molecular studies. Psychopharmacology 132:107-124.

Devenport, L.D., Merriman, V.J., and Devenport, J.A. 1983. Effects of ethanol on enforced spatial variability in the 8-arm radial maze. Pharmacol. Biochem. Behav. 18:55-59.

Gage, F.H., Dunnett, S.B., and Bjorklund, A. 1984. Spatial learning and motor deficits in aged rats. Neurobiol. Aging5:43-48.

Hagan, J.J., Alpert, J.E., Morris, R.G., and Iversen, S.D. 1983. The effects of central catecholamine depletions on spatial learning in rats. Behav. Brain Res.9:83-104.

Hagan, J.J., Tweedie, F., and Morris, R.G. 1986. Lack of task specificity and absence of posttraining effects of atropine on learning. Behav. Neurosci.100:483-493.

Hamilton, G.V. 1911. A study of trial and error reactions in mammals. J. Anim. Behav.1:33-66.

Kesner, R.P. and Novak, J.M. 1982. Serial position curve in rats: Role of the dorsal hippocampus. Science218:173-175.

Kolb, B., Sutherland, R.J., and Whishaw, I.Q. 1983. A comparison of the contributions of the frontal and parietal association cortex to spatial localization in rats. Behav. Neurosci.97:13-27.

Levin, E.D. 1988. Psychopharmacological effects in the radial-arm maze.Neurosci. Biobehav. Rev.12:169-175.

Levy, A., Kluge, P.B., and Elsmore, T.F. 1983. Radial arm maze performance of mice: Acquisition and atropine effects. Behav. Neural Biol. 39:229-240.

McNamara, R.K. and Skelton, R.W. 1991. Diazepam impairs acquisition but not performance in the Morris water maze. Pharmacol. Biochem. Behav. 38:651-658.

McNaughton, N. and Morris, R.G. 1987. Chlordiazepoxide, an anxiolytic benzodiazepine, impairs place navigation in rats. Behav. Brain Res. 24:39-46.

Morris, R.G.M. 1981. Spatial localisation does not depend on the presence of local cues. Learn. Motiv. 12:239-260.

Morris, R.G., Garrud, J., Rawlins, N.P., and O'Keefe, J. 1982. Place navigation impaired in rats with hippocampal lesions. Nature 297:681-683.

Okaichi, J. and Jarrard, L.E. 1982. Scopolamine impairs performance of a place and cue task in rats. Behav. Neural Biol.35:319-325.

Olton, D.S. and Samuelson, R.J. 1976. Remembrance of places passed: Spatial memory in rats. Exp. Psychol. [Anim. Behav.]2:97-116.

Pitsikas, N., Carli, M., Fidecka, S., and Algeri, S. 1990. Effect of life-long hypocaloric diet on age-related changes in motor and cognitive behavior in a rat population. Neurobiol. Aging11:417-423.

Rapp, P.R., Rosenberg, R.A., and Gallagher, M. 1987. An evaluation of spatial information processing in aged rats. Behav. Neurosci.10:3-12.

Stevens, R. 1981. Scopolamine impairs spatial maze performance in rats. Physiol. & Behav.27:385-386.

Sutherland, R.J., Whishaw, I.Q., and Regehr, J.C. 1982. Cholinergic receptor blockade impairs spatial localization by use of distal cues in the rat. J. Comp. Physiol. Psychol.96:563-573.

Tolman, E.C., Ritchie, F.B., and Kalish, D. 1946. Studies in spatial learning. I. Orientation and the short cut. J. Exp. Psychol. 36:13-24.

Tsien, J.Z., Huerta, P.T., and Tonegawa, S. 1996. The essential role of hippocampal CA1 NMDA receptor-dependent synaptic plasticity in spatial memory. Cell. 87:1327-1338.

Walsh, T.J., Miller, D.B., and Dyer, R.S. 1982. Trimethyltin, a selective limbic system neurotoxicant, impairs radial arm maze performance. Neurobehav. Toxicol. Teratol.4:177-183.

Watts, J., Stevens, R., and Clare, R. 1981. Effects of scopolamine on radial maze performance, male rats, implications for hippocampal role in spatial memory. Physiol. & Behav.26:845-851.

Wenk, G., Sweeney, J., Hughey, D., Carson, J., and Olton, D. 1986. Choline acetyltransferase inhibition does not impair radial maze performance in rats. Pharmacol. Biochem. Behav.25:521-526.
Key References
Brandeis et al., 1989. See above.

Provides a general review of the many ways this water maze task has been used to study brain function and the general theoretical principles that underlie its use.

Olton, D.S. 1985. The radial arm maze as a tool in behavioral pharmacology. Physiol. & Behav.40:793-797.

Reviews the many ways in which the radial arm maze task has been and can be used to investigate the effects of lesions or drugs upon the function of specific brain regions.

Current Protocols main page