A preliminary study on memory attributes and methodology of Morris water maze detection

With the deepening of the study of the learning and memory brain mechanism and the need to find effective methods to prevent and treat cognitive disorders, objective and accurate evaluation of the learning and memory levels of experimental animals has become a very important research content in neurobiology and neuropharmacology. In the study of senile dementia, we believe that the Morris water maze can be used as an important tool to detect the learning and memory levels of experimental animals. To this end, we would like to make a preliminary discussion on the memory properties and related methodological problems detected by the Morris water maze.

1 Morris water maze detection program and memory properties

Learning memory is the advanced neurophysiological activity of the brain. Learning refers to the process of accepting external environmental information and affecting its own behavior. Memory refers to the process of storing, processing and extracting information or experience acquired in the brain. Learning and memory are both important aspects of cognitive activities and important elements of intellectual structure. In recent years, with the deepening of the study of the essential content of learning and memory, especially the establishment of the theory of "hierarchical multiple memory system" ^[1 , ^2] , the research process and research direction of learning and memory have been completely changed. It also makes the classification of memory in the past have a deeper connotation. For example, the distinction between transient memory, short-term memory and long-term memory, explicit memory and implicit memory (or procedural memory and declarative memory) is no longer just a difference in the form of learning and memory, but more important. The reason is that they may belong to different memory systems, and there are different ways of processing and extracting information, involving different neural mechanisms. It is only meaningful to study the mechanism of a particular learning and memory process in a particular memory system.

The Morris water maze was designed by the British psychologist Morris in the early 1980s and applied to the study of memory and brain mechanisms ^[3] . It consists of a circular pool filled with water, a platform hidden under the water, and an automatic image acquisition and processing system (camera, video recorder, display and analysis software, etc.). Since then, the labyrinth system has been widely used in basic and applied research in the field of neurobiology, and experimental animals are mainly rats. The more classic Morris water maze test program mainly includes two parts: positioning navigation test and space exploration test. The place navigation was carried out for several days, and the rats were placed into the water from the four water inlet points several times a day, and the time to find the platform hidden under the water surface (escape latency) was recorded. The spatial probe is to remove the platform after the positioning navigation test, and then select a water inlet point to put the rat into the pool, record its swimming trajectory within a certain period of time, and examine the rat's memory on the original platform. There are many parameters that can be analyzed in the experiment, including escape latency, swimming distance in each quadrant, ratio of swimming distance to total distance in the original platform quadrant, ratio of swimming time to total time in the original platform quadrant, platform deviation angle, 40cm piercing time, cross The number of platforms and the percentage of swimming distance between the middle and outer rings.

The classic Morris water maze detects that the rat learns to search for a fixed position of the hidden platform in multiple trainings to form a stable spatial positional cognition. This spatial cognition is formed by processing spatial information (external clues). The position of the platform is independent of the position and state of the rat itself. It is a reference cognitive (allocentric cognition) with a different reference. The memory formed is a reference memory. From the perspective of the processing and extraction of information, this spatial reference memory enters the consciousness system, and its storage mechanism mainly involves the limbic system (such as the hippocampus) and the brain regions related to the cerebral cortex, often accompanied by Hebb synaptic modification ^{[3, 4]} Should belong to declarative memory. In patients with clinical forgetfulness and dementia, declarative memory is firstly impaired and more prominent ^[5] . Therefore, from the detection of memory properties, the use of Morris water maze to conduct screening research on effective drugs for the prevention and treatment of such diseases is more appropriate.

The type of learning memory detected by any type of learning memory test tool can vary with the test procedure. Frick et al. perfected the classic Morris water maze test procedure ^[6] . After testing the spatial reference memory of the rat with a fixed position platform, they changed the position of the platform to be not fixed, and further tested the working memory ability of the rat. Obviously, the cognitive process of finding an unfixed platform is more complicated, requiring rats to integrate and judge information separated in time and space. A similar concept was proposed in the eight-arm maze testing program designed by Olton et al. ^[7] . It is worth noting that many times the changes in reference memory and working memory can be asynchronous. Frick's study suggests that rats have impaired reference memory after 17 months of age, and the working memory of rats is significantly different from the younger group at 24 months of age ^[6] . Our results also confirmed that there was no significant decrease in reference memory between the 15-month-old rats and the young rats (Fig. 1). There was no significant difference in the multivariate analysis of variance between the young group and the 15-month-old group. >0.05).

2 Morris water maze detection method and statistical analysis

The basic experimental procedures and methods of all Morris water maze are similar. The difference in methods may have a significant impact on the results mainly in the detection of time limits and statistical analysis of the data, and we will talk about our views based on our own experimental results.

2.1 Experimental environment requirements In the Morris water maze test, the experimental environment has a great influence on the results. The spatial clues provided by the objects around the maze and even the hanging objects must remain fixed, that is, the cognitive "reference system" should be consistent. The most easily overlooked here is the position of the experimenter during the experiment, because if the experimenter is within the field of view of the experimental animal, it will naturally become one of the reference objects of the experimental animal. If the experimenter often changes position or keeps moving during the experiment, it will naturally have a greater impact on the experimental results. Secondly, it is also observed in our experiments that the water temperature of the labyrinth is also an important factor affecting the "evacuation motivation" of experimental animals and affecting the experimental results. In general, the water temperature should be slightly lower and should be kept as constant as possible during the experiment. In addition, it is important to keep quiet and consistent light during the experiment. In addition, many researchers also stressed that milk should be added to the labyrinth water to make it turbid to help hide the platform below the water. However, from our experimental results, the absence of milk has no effect on the experimental results, which may be related to the hidden platform below the water surface by 1 to 2 cm, while the rat always raises the head during swimming escape (avoiding flooding). .

2.2 Detection time The cycle of the Morris water maze test is often determined according to the experimental requirements, generally 7 to 15 days. The main problem at present is that it is appropriate to allow the experimental animals to find the maximum time limit for the platform (ie, the maximum incubation period) in the positioning navigation test. Some now take 120s, some take 180s, and the most recent literature reports are mostly 60s ^[6] . But how much is appropriate, most of them are determined by the experimenter based on personal experience, and there is no statistical support. According to our preliminary statistics, the difference in the number of times a rat can find a hidden platform between 60 and 120 s is only about 5% of the total number of times it finds the platform. This can be regarded as a "small probability event" in statistics. . In other words, if the rat can't find a hidden platform within 60s, it will be difficult to find it if it is extended to 120s. From this point of view, it seems that the maximum latency of 60s can save experimental time. But the problem is not so simple, because in the statistics, many studies directly include the maximum latency into the statistics. Obviously, the difference in maximum latency will inevitably affect the statistical results. Therefore, it is statistically known how much different maximum latency has an impact on the results of the Morris water maze experiment. How to avoid the systematic errors caused by this problem remains to be further studied.

Figure 1 Average daily escape latency changes in different groups of rats

2.3 Statistical analysis The experimental data of Morris water maze is more complicated, and its statistical analysis is also more difficult. Especially, how to carry out statistical analysis of the escape latency measured by rats during positioning navigation has many problems. In the past studies, some directly calculated the average escape latency of each day and then performed t test or analysis of variance, or according to the escape latency of the rats, which stabilized on the 3rd day after the navigation test, only the average escape for the last 3 days. The incubation period was analyzed by t test or analysis of variance. Some studies conducted a one-way analysis of variance for the escape latency of 1 to 5 days. We believe that the above statistical methods are not correct. A more obvious error is that the average escape latency of the last 3 days of stabilization tends to represent the memory level. The first problem encountered is that the rats in the physiological state only show this, while the experimental memory impairment group tends to be 3 days later. The escape latency has continued to decline. As shown in Figure 1, the escape latency of normal young and early-aged rats stabilized at the last 3 days, but the escape latency of rats in the experimental injury group was higher due to learning and memory impairment, and escaped on days 3 to 5. The incubation period is still declining and does not tend to be as stable as the normal control group. Not only that, but also the above-mentioned statistical methods for the misunderstanding of the statistical characteristics of the escape latency data measured in the navigation test. The escape latency in the positioning navigation test has the following characteristics: 1 series repeated measurement design, that is, data obtained by repeated measurement from the same subject at different time points after giving a certain treatment (evacuation latency); The statistical analysis results of the incubation period (independent variable) have at least several effective factors such as animal group and detection time point, and there are interaction effects between the effect factors; 3 measured at different time points of the same test animal There is a high degree of autocorrelation in the escape latency; 4 there is truncated data (maximum latency). Based on this, we believe that the escape latency of the positioning navigation test in the Morris water maze should be more appropriate using the multivariate analysis of variance of repeated measurement data combined with the survival analysis method of censored data ^[8,9] .

3 Morris water maze application and evaluation

At present, the Morris water maze is mainly used in the study of the neurobiological mechanism of learning and memory and the study of neuropharmacology (including pharmacology of traditional Chinese medicine). Compared with the previous water maze and the detection tools such as diving platform and avoiding darkness, the main advantages include: providing more experimental parameters, systematically examining the process of spatial cognition processing of experimental animals, and objectively reflecting their cognitive level. Separating the learning and memory disorders of the experimental animals from the sensations and motor defects, reducing their interference with the detection of learning and memory processes; the operation is simple and the data error is small. The shortcoming is that the design of the experimental program needs to consider more factors, and the experimenter needs to have certain knowledge in neurobiology, cognitive psychology and mathematical statistics. Secondly, the statistical analysis is more complicated.