4-1-1 T Maze

We observe that rodents on an end of the long arm (a bottom of T) of a T maze in which 3 arms are arranged with T-shape move until a branching point and select one of the 2 arms (right or left) to enter. In general method, a food pellet is placed on either end of right or left arms and entries into the arm with or without the pellet are counted as correct or error responses, respectively. Because correct responses are expected 50% by chance, differences from the chance level should be confirmed by multiple trials. Since rats and mice must be hungry due to food rewards, a food deprivation is needed.

A sample trial in which rodents can enter into only one arm with the pellet and a choice trial in which they can enter into the both arms a certain delayed time after it are conducted in the delayed matching-to-sample test and its entry into the same arm with the pellet in the sample trial is judged as correct. Mice and rats can into the both arms by turns to right or left and cannot see out of the T maze due to walls of the arms. Therefore, they do not memorize places in the space and it is not spatial learning.

4-1-2 Y Maze

We use a Y maze for a spontaneous alternation test in mice although the Y maze consists of 3 arms and can be used at the same manner as a T maze. The spontaneous alternation is an exploratory behavior that mice spontaneously enter into different arms, which can be conducted by their memories of the previous entry arm. In this test we use the Y maze that has 3 arms with the same size and is arranged with 120-degree intervals. Rewards and punishments are not necessary for the test, which is a simple and good method by using spontaneous behaviors. Actually, we record an order of entry into each arm for a short time such as 5 or 10 min in mice. The total arm entry (a summation of entries into every arm) and the alternation count (continuous entering into 3 different arms), which means the entries into all arms at minimal 3 trials in mice. The alternation count is divided by the value that the total arm entry minus 2 and it is evaluated as the alternation rate for a parameter of memories.

4-1-3 Elevated Plus Maze

We generally use an elevated plus maze consisting of enclosed arms and open arms which are crossed and placed high from a floor for evaluation of "anxiety". In measuring "anxiety", we placed a mouse in a central platform in the maze and count entries into and times spent in enclosed arms and open arms. In measuring learning and memories, a mouse is placed on an end of either open arm facing the end at the first day as an acquisition trial. Since the mouse will move to platform, know the enclosed arm with walls and enter into it, we measure a time until the first entry into either enclosed arm from the end of the open arm (an acquisition latency). We pick up the mouse and return into a rearing cage. At a next day (24 h after it), we measure the time to enter into the enclosed arm (a retention latency) at the same way as the acquisition trial. The retention latency shortened compared to the acquisition latency. We can think from it that mice learn existence of the enclosed arm in the acquisition trial and their memories hurry to go to it in the retention trial (a trial at the second day). We can evaluate learning and memories by the elevated plus maze as described above without food pellets and electric shocks same as the spontaneous alternation test.

4-1-4 Radial 8-arms Maze

A radial 8-arms maze consists of radial 8 arms projecting out from a central platform. We place rats or mice in the platform of the maze with a food pellet on an end of the arms and record an order of entry into the arms. Hungry rodents due to food deprivations eat a food pellet and enter into the other arm with the pellet. Re-entries into the arm previously entered, namely, the arm without the pellet are counted as an error (an error entry). Rodents can eat the all food pellets by 8 entries at minimum. One trial continues until rodents complete to eat the all pellets, and total entries (entries into all arms) and errors are counted in trials for multiple days (one trial a day or multiple trials a day). The total entry and the error gradually decrease until 8 and 0, respectively. We can detect gradually learning (a learning curve) by the multiple trials in the radial 8-arms maze.

In another procedure some of 8 arms are present of the food pellets and entries into the arm without the pellets are counted as the error. In this case the entries into the arms in which the food is originally absent and absent by the previous entry of rodents are separately counted. If the originally food-absent arms are same in all trials, a memory of these arms is a reference memory and the entry into these arms is counted as its parameter. On the other hand, a memory of the arms in which the food is absent by the previous entry of rodents is a working memory only valid in the trial and the entry into them is evaluated as a parameter of the working memory. If the originally food-absent arms are 4, rodents can enter into the arm with the pellet at a 50% chance level and a detecting rate of the reference memory is lowered. On the other hand, if we decrease the arms with the pellet to decrease the chance level, the detecting rate of the working memory which detected by the error entries into the arm previously entered also decrease. You should determine a number of the arms with the food pellet according to your purpose, the working or reference memories.

4-1-5 Morris Water Maze

In a Morris water maze a platform is placed just under water surface as a goal in a circular pool and a rodent is put into water from an edge of the pool and measured a time to reach the goal. Rodents learn a site of the goal by repeated trials and gradually decrease a swimming time until they go to the goal. Learning and memories are evaluated by the swimming time until the goal. Rats and mice spontaneously swim in water and rewards such as food pellets are not needed in this test. Stress in water in the Morris water maze becomes a punishment stimulus for rodents similar as an electric shock to motivate swimming behaviors to the platform in which they keep their almost whole bodies above water without swimming. If rodents can see the platform, they can easily reach the goal in spite of a memory on a site of the platform. Therefore, we must set the platform below the water surface for purpose that we hide it from rodents. Morris altered water to be opaque with milk for this purpose but rodents cannot see the goal under the surface of clear water in swimming with keeping their faces above water. Therefore, we can use tap water kept at a certain temperature with no alteration.

Although a time to reach the goal can be manually measured with a stopwatch, a video tracking by a camera above the Morris water maze with a PC analysis is also conducted. The latter can measure a swimming distance as well as the time and get the swimming trajectory. Although the video tracking is generally used in maze learning such as the radial 8-arms maze and the Y maze, it is especially valuable in the water maze because it can get the distance and trajectory but not manual observation. The video images are generally transformed to binary (monochrome) images and a rodent is detected by its differences from backgrounds in brightness. Because there is reflection of light on water surface which is changeable by surface movement, the rodent in the water maze is more difficult condition for correct detection by the video tracking than in the other maze. The light reflection may be suppressed by beads with different color from rodents float in water and by opaque water color. However, if the video tracking performance is ameliorated by the other means such as control of a room light using only the tap water, the test will be more easily conducted. Because a water temperature affects swimming performances same as a forced swimming test for antidepressants, we must keep the water temperature constant at the same temperature as the room in all trials.

The platform is fixed at the same site and rodents are put into water at different sites in multiple trials by one trial a day or multi-trials a day and recorded their swimming time until the goal. The rodents recognize landscapes out of the pool, memorize association between the site of the goal and the landscapes as spatial memories, and can reach the goal. We remove the platform and measure the swimming time in rodents in each quadrant of the circular water pool after the spatial learning was established by several trials. If the rodents have memories of the site of the platform, they swim for the longest time in the quadrant the platform placed in learning trials. The site of the goal constant between all trials becomes a reference memory valid between the trials. On the other hand, association between a start point and the site of the goal becomes a working memory valid in the trial if we put rodents into water at a different point by a trial. Because rodents learn spatial association from the landscape out of the pool, we must keep it constant. We should carefully conduct the test, e.g. the experimenter sits at the same site in every manual observation. Since the Morris water maze test is not needed food deprivation for motivation of learning, it is relatively easy but a big experimental room is needed due to a large size (more than 1 m diameter) of the pool even though in mice.

4-1-6 Barnes Maze

A Barnes maze has many circular holes with the same size and the same interval in a peripheral area of a circular disk (platform) and one of the holes has an escape box that is dark and set under the hole, and the maze is elevated. Since rodents do not prefer light and open spaces, they enter into the escape box with walls in which they can hide into it if they find. We can evaluate spatial learning like the Morris water maze by measuring time until rodents enter into the escape box and numbers of head-dipping into the other holes than it (error numbers).

Barnes Maze for Mice
A photo cited from Lafayette Instrument

Although a platform is large (ca. 120-cm diameter) like the pool for the water maze, the same size platform can be used in mice and rats different from the water maze. Because a size of the peripheral holes is different in the platforms for mice and rats, the number of hole is fewer for rats than for mice. We can change a starting point in an every trial in the Morris water maze test but cannot change it in the Barnes maze test because it is a center of the platform. A goal point is same in all learning trials in the both test and rodents develop spatial memories from landscapes surrounding the maze. In some Barnes maze tests association between the goal and the landscape is changed by daily rotation of the platform and multiple trials a day are conducted. In this case rodents cannot use their memories of yesterday trials (reference memories) as a goal cue in the today's first trial but memories in the today's previous trials (working memories) is valid from the second trial. Time to reach the goal (latency) and the number of error in all trials a day are averaged, respectively, in everyday and used to evaluate learning development. The results in the first trial and the other trials are separately averaged and differences between them can be evaluated as working memories (1). We remove the escape box after establishment of the spatial learning and can measure how long rodents stay in the site of the previous escape box as a probe test in the same way as the Morris water maze test (2).

Barnes Maze for Rats
A photo cited from Lafayette Instrument

Although both the Barnes maze and the Morris water maze tests evaluate spatial learning, the former is more usable than the later because it can conduct without water. It is ideal that rodents cannot see the escape box until they dip head into holes and the platform in which fake escape boxes set in the other holes than the goal is available. Motivation of behavior in the Barnes maze is property that mice and rats do not prefer open and light spaces same as the elevated plus maze test. Because rodents do not know the escape box in the first trial, they may stay in a central area around a starting point. Maze learning may be facilitated if you manually lead the rodents to the escape box in the first trial.

4-2-1 Operant Learning

Operant behaviors are spontaneous behaviors for specific purpose and spontaneous lever-pressing behaviors are measured in general experiments. A typical example is an experiment that hungry rats by food deprivation press lever to get foods as rewards in a special cage with a lever called as an operant chamber. Rats learn that lever pressing is rewarded by foods and become to press the lever. These behavior and learning are called as an operant behavior and operant learning, respectively. Because mice cannot physically press the lever for rats, measuring nose poke as operant behaviors in substitution of the lever pressing is developed in mice. Currently light levers that mice can press are developed and we can measure lever pressing in mice. Liquid rewards such as milk and sucrose solution as well as small globular pellets of foods are used as positive reinforcing stimuli. The former and the latter are supplied by liquid and pellet dispensers, respectively. Operant behaviors in rodents are also motivated by punishment stimuli (negative reinforcing stimuli) pressing the lever to avoid electric foot shocks. Other than foods self-administration of dependent drugs and electrical self-stimulation in rewarding neuronal systems in brains are also used in the positive reinforcement.

Modular Test Chamber
A photo cited from Lafayette Instrument

Experimental procedures in the operant learning in rodents are complicated and time-consuming such as food deprivation and learning lever-pressing and so on. However, we can get highly reliable data because lever pressing does not happen by chance and is the operant behavior with purpose. Rodents press the lever as results they learned associations between lever pressing and food rewards but it is difficult to use itself as a parameter of learning and memories after its establishment since the lever pressing is a simple behavior. It is used in a Geller-type conflict test evaluating anxiolytics, self-administration and self-stimulation tests measuring rewarding effects, and a drug discrimination test. For evaluation of learning and memories, we make rodents learn the lever press under the specific condition that lever pressing is rewarded only when conditioned stimuli combined a buzzer, a light and so on are presented (operant conditioning). Rewards (positive reinforcer) such as food pellets and punishment stimuli such as electric foot shocks gotten by responses (e.g. lever pressing) to the conditioned stimuli are called as unconditioned stimuli. If the unconditioned stimuli are the punishment stimuli, rodents learn the lever pressing under condition that the punishment can be avoided by it. Behaviors to avoid the punishment by the lever pressing are called as active avoidance responses. Behaviors that rodents avoid by a movement to the next compartment are called as shuttle avoidance responses. The both behaviors are used to evaluate learning and memories as active avoidance learning. Further, the delayed matching to sample test with a two-lever operant chamber is also used, which is described later.

We count lever pressing in absence and presence of the conditioned stimuli, which is an error response and a correct response, respectively, total lever pressing and total rewards (earned pellets) under conditions that only responses with conditioned stimuli (in presence of tones and lights) are rewarded in the operant conditioning, and evaluate the behaviors from these parameters. Various schedules of the operant conditioning are possible such as lever pressing without the conditioned stimuli is considered as a correct response and pressing an either lever is correct in the two-lever operant chamber, and further various combinations of conditioned stimuli and rewards are also available.

4-2-2 Passive Avoidance Response

A passive avoidance response is a behavior to avoid electric shocks without action and it is used a word of "passive" compared to the "active" avoidance learning. Actually we placed a rodent in a light compartment of a chamber consisting of light and dark compartments connected each other, and give the electric foot shock if the animal enters into a dark compartment. After the rodent return into the light compartment, we close a door dividing the both compartments and pick up the rodent (an acquisition trial). In a tomorrow trial (a retention trial) when the rodent is placed in the light compartment and the door is opened, the animal stays in it and does not enter into the dark compartment. Time until enter into the dark compartment (latency) is measured in the both trials. If latency at the second day (retention latency) is significantly longer than that at first day (acquisition latency), we consider as rodents acquired learning and memories. Because rodents does not often enter into the dark compartment in the retention trial for a long time, we set a cut-off time at 300 or 600 sec, stop the retention trial after the cut-off time, and latency is presumed as this time. Mice and rats naturally prefer the dark place to light place and spontaneously move into the dark compartment. Because the passive avoidance test is conducted based on this behavior, it is simple and widely used for screening of anti-dementia drugs. The passive avoidance test using the light/dark chamber described above is called as a step-through type and in a step-down type stepping down to the floor from a small platform is punished by the electric shocks. Because rodents are difficult to stay in the small platform and they spontaneously step down, we use this behavior in the step-down type and its retention latency is shorter than that in the step-through type.

4-2-3 Active Avoidance Response

An active avoidance response is learning avoiding electric shocks by active behaviors. The avoidance is generally conducted by lever pressing or by movement into the next compartment. The latter is called as the shuttle avoidance because rodents come and go between left and right boxes like a shuttle of a weaving machine. The shuttle cage is used in the shuttle avoidance test, which consists of the two same boxes with lights and connected each other and can be added electric shocks on the floor. If a rodent moves into the next box when the box in which the rodent currently places is present of lights as the conditioned stimulus and also tones are present, the animal can avoid the electric foot shocks from floor rods. Because in the beginning the rodents do not know that unconditioned stimuli (electric shocks) are given after the conditioned stimuli (lights and tones), the rodents show escape behaviors from the electric shocks to the next box after they received the shocks. Then, they gradually learn and become to move before the shocks to avoid them. Since the place movement is a behavior observed in mice and rats in spite of unconditioned stimuli, they may easily learn this behavior. The operant conditioning by lever pressing takes a while but escape or avoidance behaviors are soon observed in the shuttle avoidance. We measure a number of avoidance errors (a number that the avoidance behavior did not occur in presence of the conditioned stimuli) as a parameter of learning and memories and total movement (shuttle crossing) as that of locomotion.

The active avoidance system is available, which automatically presents the conditioned and unconditioned stimuli (electric shocks) and counts the avoidance errors. The automatic system makes this experiment relatively easy. Because avoidance responses are evaluated by movement behaviors, locomotive capability with no relation to learning and memories possibly affects the avoidance behavior. Therefore, we should carefully discuss results of the shuttle avoidance.

4-3-1 Taste Aversion

If LiCl is intraperitoneally administered in rodents after they lick sucrose solution, they become dislike the sucrose solution. Sweet taste of the sucrose solution is associated to aversive feeling by LiCl in rodents and they learn behaviors not to take sweet substances. Presentation of sweet taste is the conditioned stimulus and LiCl injection is the unconditioned stimulus and rodents can learn the taste aversion by one conditioning alone. Because food selection by taste is very important for animal survival, they are thought to easily learn behaviors that they do not eat foods making sick. Preferable taste (e.g. slight salty) in rodents other than sweet also can be used as the conditioned stimulus and LiCl are often used as the unconditioned stimulus but the other is possible.

Taste aversion is generally used for studies in central nervous systems associated with taste recognition and learning and not for evaluation of anti-dementia drugs.

4-3-2 Fear Conditioning

If inescapable electric shocks (unconditioned stimuli) are given in rodents under conditioned stimuli are presented, the rodents become to feel fear of the conditioned stimuli alone, which are thought to be results that they learned association between the conditioned stimuli and fear of the shocks. Tones and lights are sometime used as the conditioned stimuli but in contextual fear learning without them electric shocks are given in a specific chamber and then rodents learn association between circumstances of the chamber and the electric shocks. We confine rats or mice in chambers and give them the electric foot shocks for a certain time immediately after tones and lights are presented and then return to their home cages. We place the rodents in the same chamber more than 24 hrs after that and evaluate their fear as freezing behaviors. The freezing is immobility in animals except for minimum movements to maintain their lives such as breath, which is thought to be induced by fear. The freezing is evaluated as a freezing time by manual observation or analysis with video tracking and PC software. It is also evaluated by devices measuring locomotion. In the contextual fear learning without tones and lights, we place the rodents in the same chamber for the conditioning with the electric shocks and a retention trial more than 24 hrs after it without the shocks. Fear memories acquired by learning cause the freezing behavior but it is not usually used to evaluate ameliorating effects on learning and memories as well as the taste aversion. It is rather used to evaluate stress-induced disorders such as PTSD (Posttraumatic stress disorder).

4-4-1 Delayed Matching to Sample

In "delayed matching to sample", a selection of the same stimulus is judged as a correct response with a delayed time after a sample stimulus, which is a good rodent model at extrapolation to humans because the similar procedures can be conducted in rodents, monkeys and humans. Conversely the selection of the other than the sample becomes to be correct in "delayed nonmatching to sample". T mazes or 2-lever operant chambers are usually used in rodents due to one selection from 2 options. First, rodents can enter into only one arm with a food pellet and eat it as a sample trial in the T maze and in a retention trial after the delayed time, they can enter the both arms but an entry into the same arm is only rewarded with the food. Either of lights above the both levers is on in the sample trial in the 2-lever operant chamber and responses to the lever light on are only rewarded. The both lights are on in the retention trial and responses to the same lever in the sample trial are only rewarded as correct. Either procedure is selecting one from 2 options and a chance level of correct responses is 50%. An increase in the option is possible but makes difficult to learn it.

4-4-2 Drug Discrimination

Rodents learn drug stimuli in drug discrimination associated with each lever in the 2-lever operant chamber and memorize the drug stimuli in a similar way as the taste aversion. First, we generally train lever pressing to hungry rats with food rewards due to food deprivation. After rats acquired pressing the both levers, a training drug or a vehicle (physiological saline) is administered. A certain time in which central effects of the drugs are elicited (e.g. 15 min in ip or 30 min in po) after the administration, the rats are trained the lever pressing under conditions that responses on only one lever is rewarded. In another day saline or the drug, the other than the previous injection, is administered in the rats and the rats are similarly trained the lever pressing under conditions that another lever is only rewarded. The rats learn that only responses to the different lever are correct when the drug and the vehicle are injected. Administrations of the drug and the vehicle are generally changed according to a double alternation schedule, in which the injection is changed after the same injections for 2 days. If the administration changes every day, rats may press the different lever from the previous one by their memories of yesterday lever pressing. Therefore, we use the double alternation which thought to be the schedule that rats cannot memorize the alteration schedule. A number of pressing an invalid lever which is not rewarded is an error count and its rate compared to a count of valid responses or a total count is used to evaluate the drug discrimination learning.

If the drug discrimination learning is acquired in rats by the training, rats can feel subjective effects of the drugs and can be used to evaluate the subjective effects. Different drugs from the trained drug are injected in rats and they are observed which lever they press. If the rats press the trained drug-associated lever, it is told that stimuli of a test drug "generalize" to the trained drug. If rats select one from 2 options, we do not know that rats considered them same as the drug stimuli or they considered different from vehicle due to a presence of stimuli. Therefore, we can conduct more complicated methods in which rats are trained with 2 different drugs in substitution of vehicle or they are trained with 2 drugs and vehicle in the 3-lever operant chamber. However, it may need a long time to train it and finally may not be trained. Thus, the drug discrimination is used to study the subjective effects of drugs and does not generally used to evaluate learning and memories.

4-4-3 Object Recognition

Mice and rats show exploratory behaviors in novel circumstances such as a novel cage. According to time passed novelty decreases (acclimation) and it decreases the exploratory behaviors and also locomotion. Exploratory behaviors to objects are also decreased by recognition of objects. If one object is changed to a novel one and the rodents are explored to it after exploratory behaviors to 2 objects placed in a cage, exploration to the acclimated object decrease compared to that to the novel object by memories to the objects. Object recognition is a method to measure learning and memories by such exploratory behaviors to objects.

Actually we make rodents explore to the 2 same objects for a short time (e.g. 10 min) which are placed apart from each other in a cage after acclimation to the cage without any object. We measure time spent in their exploring behaviors (sniffing, touching objects with a fore paw(s), and so on) to each object, separately, and return to their cage. One object is changed to novel one having a different form a certain time, e.g. 24 hrs, after it and the time in the exploration is measured similarly. A ratio of the novel object to a summation of the both objects in exploration time is evaluated as a parameter of learning and memories. This ratio in the exploration time becomes close to a chance level, 50%, by dysfunction of learning and memories, and it becomes more than 50% by memories to the objects. We use 2 different objects in the first exploration in some methods. In this case exploration time to each object must not be largely different by its preference in rodents. Further, to avoid differences in the exploration time by sites of the objects, the site of the novel object is equally applied within the group. Smells attached to the object in exploration must be cleaned before a trial in the next animal not to be used as cues of the object memories in experiments using the same objects in multiple trials. An object recognition test is a simple test generally consisting of one acquisition trial (first exploration to objects) and a retention trial (exploration to objects of one novel and one same) a certain time after the acquisition trial at a similar way as passive avoidance test. However, we may take longer time for the acquisition trial or repeat the trials to surely establish object recognition. If we increase a number of objects to be explored, a chance level of the ratio for evaluation of learning and memories decrease less than 50% but we usually use 2 objects. Because we become to need a larger cage and a longer time to explore the objects well by increases in the objects, 2 objects are actually practical.

We can measure exploratory time by manual observation or a video tracking and PC analysis.

References