An apparatus for training and testing cooperative behavior in rodents

Abstract

The application provides a device for training and testing cooperation behavior of rodents, comprising an animal experiment box and an image-based control system; the animal experiment box comprises a coverless experiment box, a reward area is arranged at a corner of the bottom of the animal experiment box, an electric shock plate is arranged on the reward area, a water drinking device is arranged at a corner of the reward area, and the size of the reward area and the body length of the animal satisfy a preset condition, so that the animal can obtain a water drinking opportunity only by entering the reward area; the image-based control system comprises a monitoring camera for recording the movement of the animal in the experiment box, an image real-time analysis unit for analyzing the movement state of the animal in real time, and an electric shock control unit for controlling whether the electric shock plate gives an electric shock. The application can analyze the cooperation behavior of rodents in a free activity state without any physical separation, realize closed-loop and fully automatic training and testing, and realize cooperation behavior research of two or more animals.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of animal cooperation behavior research, and particularly relates to a device for training and testing cooperation behavior of rodents. BACKGROUND

[0002] The cooperation behavior of animals (such as nurturing, reproduction, cooperative hunting, etc.) is crucial for the growth and development of individual animals and the propagation of the group. From the cooperation of a lion group in hunting large prey to the cooperation of ants in carrying food that is much heavier than their own weight, cooperation among animals is ubiquitous in nature. Many mental illnesses are accompanied by abnormalities in social cooperation behavior, such as autism, schizophrenia, depression, drug abuse, and anxiety, etc. Rodents (such as mice, rats, etc.) are the most widely used model animals, and the study of cooperation behavior of rodents is crucial for revealing the neural mechanisms of cooperation behavior and the pathogenic mechanisms of abnormal cooperation behavior. In view of the behavioral characteristics of rodents, researchers have proposed various methods and systems for exploring the cooperation behavior of animals, including a food-pole-based system and a motion coordination-based system.

[0003] In the food-pole-based system, two animals are placed in two physically isolated cages, and the cages are generally made of transparent, grating or mesh materials, so that the two animals can see each other. Each cage has a food-pole, and the food-pole in each cage controls the food position in the other cage. After being deprived of food, the animals have a demand for food. One animal can deliver food to a position that the partner animal can touch by controlling the food-pole, i.e., each animal can only help the partner animal to obtain food, but cannot directly obtain food for itself. The two animals need to alternately pull the food-pole to provide the other with the opportunity to obtain food, thereby achieving cooperation to obtain food. The disadvantages of the above prior art are as follows: Disadvantage 1: The two animals are physically isolated and cannot interact freely without restraint, which does not conform to the cooperation behavior of animals in a natural scene. Disadvantage 2: The animals cannot directly obtain rewards themselves and need to obtain rewards in cooperation with partners. This rule is relatively obscure, and the animals need a long time of training to learn cooperation, resulting in low experimental efficiency. Disadvantage 3: It is not suitable for group cooperation behavior experiments of more than two rodents.

[0004] In a motion coordination-based system, two animals are placed in two physically isolated narrow cages, the two cages are placed side by side, and the cages are generally made of transparent, grid or mesh materials, and the two animals can see each other; the animals have a demand for sugar water after being deprived of water. One end of the narrow cage is the starting end, and the other end is the reward end; the two animals move synchronously from the starting end to the reward end and can obtain sugar water reward at the reward end; only one animal moves from the starting end to the reward end, or the movements of the two animals are not synchronized, and the sugar water reward cannot be triggered. After training, the two animals learn to move synchronously and simultaneously from the starting end to the reward end to obtain sugar water reward. The above prior art has the following disadvantages: Disadvantage 1: The two animals are physically isolated and cannot interact freely without restraint, which does not conform to the cooperative behavior of animals in a natural scene. Disadvantage 2: The animals cannot directly obtain the reward themselves and need to cooperate with their partners to obtain the reward. This rule is relatively obscure, and the animals need a long time of training to learn to cooperate, resulting in low experimental efficiency. Disadvantage 3: It is not suitable for group cooperation behavior experiments of two or more rodents. Disadvantage 4: The definition of animal cooperative movement in the technical solution is not specific enough, including the speed of cooperative movement and the degree of synchronous movement. Different animals (such as rats and mice) usually have different parameters, and there is a lack of universality for various rodents. SUMMARY

[0005] In view of the problems of the inability of animals to interact freely without restraint and the low experimental efficiency in the existing animal cooperation behavior technical solutions, the embodiments of the present application provide a device for rodent cooperation behavior training and testing, which can analyze the cooperation behavior of rodents in a free activity state without any physical separation, realize closed-loop automatic training and testing, and realize cooperation behavior research of two or more animals.

[0006] The embodiments of the present application provide a device for rodent cooperation behavior training and testing, which comprises an animal experiment box and an image-based control system; the animal experiment box comprises a coverless experiment box, the bottom surface of the experiment box is rectangular or square, and the reward area is arranged at the corner of the bottom of the animal experiment box; the reward area is covered with an electric shock plate, a water drinking device is arranged at the corner of the reward area, and the size of the reward area and the body length of the animal satisfy a preset condition, so that the animal can obtain a water drinking opportunity only by entering the reward area.

[0007] The image-based control system comprises a monitoring camera, an image real-time analysis unit and an electric shock control unit connected in sequence; the monitoring camera is used to record the movement of the animal in the experiment box; the image real-time analysis unit is used to analyze the movement state of the animal in real time; and the electric shock control unit is connected with the electric shock plate and is used to control whether the electric shock plate gives an electric shock.

[0008] The image real-time analysis unit is configured to:

[0009] S1: capturing images in the experimental box through a monitoring camera;

[0010] S2: detecting all animals in the experimental box and the positions of the animals according to the images captured by the monitoring camera;

[0011] S3: calculating the pixel area Area1 of the overlapping part of the body of each animal detected in S2 and the reward area and the body area Area2 of the animal, respectively; calculating the ratio Ratio = Area1 / Area2 of the two areas; setting a threshold value Thre for the area ratio, and if Ratio >= Thre, it is considered that the animal has entered the reward area; repeating the operation of this step for all animals detected in S2 to obtain the number Num-Reward of animals entering the reward area;

[0012] S4: determining whether two or more animals enter the reward area, if Num-Reward >= 2, then entering S5; if Num-Reward < 2, then entering S1;

[0013] S5: when it is detected that two or more animals enter the reward area at the same time, the image real-time analysis unit sends information to the electric shock control unit, the electric shock control unit generates an electric shock signal, and the electric shock plate causes electric shock to the animals entering the reward area, so that all animals that have entered the reward area leave the reward area to escape the electric shock, and at the same time, they can no longer obtain drinking water.

[0014] The threshold value Thre of the ratio of the pixel area Area1 of the overlapping part of the body of the animal and the body area Area2 of the animal is in the range of 0.75-1.

[0015] The image real-time analysis unit is further configured to: receiving the number of individual animals manually input during the detection of the animals.

[0016] The reward area is a rectangle, and the coverage area of the electric shock plate is consistent with the size of the reward area.

[0017] The size of the reward area is 2-3 times the length of the body of the animal, so that the animal can obtain the opportunity to drink water only when it enters the reward area

[0018] The imaging frame rate of the monitoring camera is selected according to the movement speed of the experimental animal, and the frame rate of the camera for monitoring the movement state of the rat and the mouse is selected to be 15-30 Hz.

[0019] The analysis speed of the image real-time analysis unit is described by the time required for analyzing one frame of image, which needs to be adjusted according to the average movement speed of the experimental animal. For mice or rats, the time required for analyzing one frame of image in the image real-time analysis unit is controlled in the range of 0.5-1.0 seconds.

[0020] The electric shock control unit adopts a grid electric shock plate.

[0021] The intensity of the electric shock is adjusted according to the type of the experimental animal. The electric shock intensity for mice is 0.1-0.5 mA, and the electric shock intensity for rats is 0.5-1.0 mA.

[0022] The device for training and testing the cooperation behavior of rodents according to the embodiments of the present application has the following beneficial effects:

[0023] The present application can analyze the cooperation behavior of rodents in a free activity state without any physical separation, realize closed-loop automatic training and testing, and realize the cooperation behavior research of two or more animals. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 Fig. 1 is a structural schematic diagram of the device for training and testing the cooperation behavior of rodents according to the embodiments of the present application;

[0025] Figure 2 Fig. 2 is a flowchart of the image real-time analysis unit according to the embodiments of the present application;

[0026] Figure 3 Fig. 3 is a statistical diagram of the number of times and total time of the mice entering the reward area after the individual training experiment;

[0027] Figure 4 Fig. 4 is a statistical diagram of the number of times and total time of the fine drinking water behavior of the mice after the individual training experiment;

[0028] Figure 5 Fig. 5 is a statistical diagram of the number of times and total time of the mice entering the reward area in the cooperation training stage;

[0029] Figure 6 Fig. 6 is a statistical diagram of the number of times and total time of the fine drinking water behavior of the mice in the cooperation training stage. DETAILED DESCRIPTION

[0030] The present application will be further described below in combination with the drawings and embodiments.

[0031] The following description provides multiple embodiments of the present invention. Different embodiments may be replaced or combined, and thus this application may be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, even though such embodiments may not be explicitly described in the following text.

[0032] Example 1

[0033] like Figure 1 As shown, the device for training and testing cooperative behavior of rodents in the present application includes an animal experimental box and an image-based control system. The animal experimental box includes an uncovered experimental box 101, the bottom surface of the experimental box 101 is rectangular or square, and a reward area 103 is provided in the corner of the bottom of the experimental box. The reward area 103 is covered with an electric shock plate 104, and a drinking water device 105 is provided in the corner of the reward area 103. The size of the reward area 103 and the body length of the animal meet the preset conditions, so that the animal can only get the opportunity to drink water when entering the reward area 103. The image-based control system includes a monitoring camera 201, a real-time image analysis unit 202 and an electric shock control unit 203 connected in sequence. The monitoring camera 201 is used to record the movement of the animal in the experimental box, the real-time image analysis unit 202 is used to analyze the movement state of the animal in real time, and the electric shock control unit 203 is connected to the electric shock plate 104 to control whether the electric shock plate 104 gives an electric shock.

[0034] This application provides a device for training and testing cooperative behavior in rodents. Water-deprived animals have a need for water. A single animal entering a reward area within the device can freely drink water. However, if multiple animals enter the reward area simultaneously, an electric shock is triggered within the reward area, causing all animals to flee the reward area and be unable to obtain water. Therefore, in this device, animals need to learn to cooperate and take turns entering the reward area to obtain water.

[0035] This application can analyze the cooperative behavior of rodents in a freely moving state without any physical separation, realize closed-loop fully automatic training and testing, and can realize the study of cooperative behavior of two or more animals.

[0036] Example 2

[0037] like Figure 1As shown, the device for training and testing the cooperation behavior of rodents in the present application includes two subsystems: an animal experiment box and an image-based control system. The animal experiment box includes a rectangular experiment box 101 without a cover, and the bottom of the experiment box can be rectangular or square. A rectangular reward area 103 is arranged at one corner of the bottom of the experiment box, and the non-reward area of the bottom of the experiment box is a free movement area 102. An electric shock plate 104 is covered on the reward area. A water drinking device 105 is arranged at one corner of the reward area. In some embodiments, the reward area 103 is a rectangle, the coverage area of the electric shock plate is consistent with the size of the reward area, and the size of the reward area is generally 2-3 times the length of the animal body. According to the size ratio of the reward area to the length of the animal body, the animal must enter the reward area to obtain the opportunity to drink water.

[0038] The image-based control system includes a monitoring camera 201 for recording the movement of the animal in the experiment box, an image real-time analysis unit 202 for real-time analysis of the movement state of the animal, and an electric shock control unit 203 for controlling whether the electric shock plate 104 gives electric shock. In some embodiments, the imaging frame rate of the monitoring camera 201 is selected according to the movement speed of the experimental animal, and the movement state of the rat and mouse can be monitored by selecting a camera with a frame rate of 15-30 Hz. The analysis speed of the image real-time analysis unit 202 is described by the time required to analyze one frame of image, which needs to be adjusted according to the average movement speed of the experimental animal. For mice or rats, the time required to analyze one frame of image in the image real-time analysis unit 202 can be controlled within the range of 0.5-1.0 seconds. The electric shock control unit 203 adopts a grid electric shock plate, and the intensity of the electric shock is adjusted according to the type of experimental animal. The electric shock intensity for mice is 0.1-0.5 mA, and the electric shock intensity for rats is 0.5-1.0 mA.

[0039] Design of animal cooperation water drinking paradigm rules: animals have the need to drink water after being deprived of water, so each animal has the motivation to enter the reward area 103 to obtain water reward. The animal cooperation water drinking experiment is divided into an initialization experiment stage, an individual training stage, and a cooperation training stage. In the initialization experiment stage, the experimental animal is not deprived of water. In the individual training stage and the cooperation training stage, the experimental animal needs to be deprived of water, that is, the water source in the feeding cage of the experimental animal is removed, and the experimental animal can only obtain water during the time period.

[0040] 1) Initialization experiment stage

[0041] The initial experiment is performed on day 0. In the initial experiment, one animal is placed in the experimental chamber at a time, and the animal's movement is recorded by the camera for a duration of T0. The purpose of the initial experiment is to familiarize the experimental animal with the experimental setup and to record the animal's movement in the absence of water deprivation for comparison with the individual training phase and the cooperative training phase. Preferably, the initial experiment is performed only once, denoted as P0; the duration of T0 is in the range of 5-10 minutes. After the initial experiment, each animal is returned to the home cage, and the water source in the home cage is removed.

[0042] 2) Individual training phase

[0043] After the initial experiment, the individual training phase is performed for a duration of N1 days. In the individual training experiment, one animal is placed in the experimental chamber at a time, and the animal's movement is recorded by the camera for a duration of T1. The purpose of the individual training experiment is to teach each water-deprived animal to go to the reward area 103 to drink water. The individual training phase is denoted as P1, and P1-k denotes the kth day of the individual training phase, k = 1, 2,... N1. Preferably, the individual training is performed for 2-3 days, i.e., N1 = 2-3; the duration of T1 is in the range of 5-20 minutes.

[0044] 3) Cooperative training phase

[0045] After the individual training phase, the cooperative training phase is performed for a duration of N2 days. A plurality of (n > 2) animals that have been trained in the individual training phase are placed in the experimental setup to perform a cooperative drinking experiment for a duration of T2. Each animal will spontaneously enter the reward area 103 to drink water. During the experiment, the image real-time analysis unit 202 monitors the movement of the animals in real time and analyzes the number of animals that enter the reward area; when the image real-time analysis unit 202 detects that no animal or one animal enters the reward area 103, the electric shock control unit 203 does not generate an electric shock signal; when the image real-time analysis unit 202 detects that two or more animals enter the reward area 103 at the same time, the electric shock control unit 203 generates an electric shock signal, and the electric shock plate 104 causes electric shock to the animals that enter the reward area; all animals that have entered the reward area 103 must leave the reward area in order to escape the electric shock, and at the same time, they can no longer obtain water. When the image real-time analysis unit 202 detects that all animals have left the reward area 103, the electric shock control unit 203 stops generating the electric shock signal, and the duration of the electric shock signal is in the range of 0.3-2.0 seconds. The cooperative training phase is denoted as P2, and P2-k denotes the kth day of the cooperative training phase, k = 1, 2,... N2. Preferably, the cooperative training is performed for 5-20 days, i.e., N2 = 5-20; the duration of each cooperative training experiment T2 is in the range of 20-40 minutes.

[0046] AsFigure 2 As shown in the figure, the flow and steps of the image real-time analysis unit 202 include:

[0047] S1: Capture an image frame by the camera 201.

[0048] S2: Animal detection, detect all animals in the experimental device and the corresponding positions of the animals; the animal detection method includes a traditional image processing and binarization method, a deep learning-based target detection method, a deep learning-based semantic segmentation method, etc. As a preferred embodiment, the number of animal individuals can be manually input during the experiment for optimizing the animal detection result. As a preferred embodiment, the traditional image processing and binarization method first acquires a background image without experimental animals, and when detecting animals, the acquired image is subtracted from the background image, and then the difference image is binarized and segmented to obtain the pixel area corresponding to each animal body. As a preferred embodiment, the deep learning-based target detection method includes a Faster R-CNN model, a YOLO model, an SSD model, etc., and calculates the bounding box corresponding to each animal individual. As a preferred embodiment, the deep learning-based semantic segmentation method includes a Mask R-CNN model, etc., and calculates the pixel area corresponding to each animal body.

[0049] S3: Calculate the number of animals entering the reward area, denoted as Num-Reward; for each animal detected in step S2, calculate the pixel area Area1 of the overlapping part of the animal body and the reward area, and the area Area2 of the animal body; calculate the ratio Ratio = Area1 / Area2 of the two areas; set a threshold Thre for the area ratio, and if Ratio≥Thre, it is considered that the animal has entered the reward area. As a preferred embodiment, the value range of Thre is 0.75-1. Repeat the operation for all detected animals in step S2 to obtain the number of animals entering the reward area Num-Reward.

[0050] S4: Determine whether two or more animals enter the reward area, i.e., determine whether Num-Reward≥2; if Num-Reward≥2, proceed to step S5; if Num-Reward<2, proceed to step S1.

[0051] S5: The image real-time analysis unit detects that two or more animals enter the reward area 103 at the same time, and the electric shock control unit 203 generates an electric shock signal, and the electric shock plate 104 causes electric shock to the animals entering the reward area, so that all animals that have entered the reward area must leave the reward area 103 to escape the electric shock, and at the same time, they can no longer obtain water.

[0052] Analysis process and statistical indicators: record the number of times and total duration of individual animals entering the reward area in the initialization experiment stage and individual training stage; and the number of times and total duration of individual animals entering the reward area and performing fine drinking behavior in the reward area. For multi-animal (n≥2) experiments in the cooperation training stage, analyze the number of times of triggering electric shock, the number of times and total duration of individual animals entering the reward area, and the number of times and total duration of individual animals entering the reward area and performing fine drinking behavior in the reward area.

[0053] In the above technical solution, the electric shock is used to trigger the drinking water conflict between mice, that is, when multiple mice are detected to enter the reward area, the system triggers electric shock, forcing all mice to leave the reward area and lose the opportunity to obtain drinking water. A feasible alternative method is to remove the electric shock control device and the electric shock plate, and replace it with a solenoid valve control system. When there is no animal or only one animal enters the reward area, the solenoid valve is turned on, the drinking water loop of the system is unblocked, and the animal can obtain drinking water in the reward area; when multiple mice are detected to enter the reward area, the solenoid valve is closed, the drinking water loop is closed, and all animals cannot obtain drinking water reward in the drinking water port.

[0054] The device for training and testing cooperation behavior of rodents provided by the present application can analyze the cooperation behavior of rodents in a free activity state without any physical separation based on deep learning technology, real-time image processing technology, and automatic control technology, and can realize closed-loop automatic training and testing, and can realize cooperation behavior research of two or more animals.

[0055] Example Three

[0056] 1) Example condition setting

[0057] In the described example, the model animal used is a male mouse, aged 2-3 months. All mice are raised in a 12-hour periodic light and dark environment; all experiments are carried out during the light period.

[0058] In the initialization experiment stage, the duration of each experiment T0=10 minutes; in the individual training stage, the duration of each experiment T1=10 minutes, and the individual training stage lasts N1=2 days; in the cooperation training stage, the duration of each experiment T2=30 minutes, and the cooperation training stage lasts N2=14 days. In this example, the data is expressed in the form of mean±sem.

[0059] 2) As Figure 3As shown, after individual training experiment, the number of times and total time of mice entering reward area significantly increased. The number of times of mice entering reward area in individual training stage was significantly higher than that in initial stage (P0, 17.77 ± 2.01; P1-1, 18.72 ± 1.69, p > 0.05; P1-2, 27.96 ± 3.00; p < 0.001; t test; n = 24); the total time of mice entering reward area in individual training stage was significantly higher than that in initial stage (unit: second; P0, 78.63 ± 10.97; P1-1, 206.51 ± 13.73, p < 0.001; P1-2, 238.45 ± 9.82, p < 0.001; t test; n = 24).

[0060] 3) As shown in FIG. 3, after individual training experiment, the number of times and total time of mice entering reward area significantly increased. The number of times of mice entering reward area in individual training stage was significantly higher than that in initial stage (P0, 17.77 ± 2.01; P1-1, 18.72 ± 1.69, p > 0.05; P1-2, 27.96 ± 3.00; p < 0.001; t test; n = 24); the total time of mice entering reward area in individual training stage was significantly higher than that in initial stage (unit: second; P0, 78.63 ± 10.97; P1-1, 206.51 ± 13.73, p < 0.001; P1-2, 238.45 ± 9.82, p < 0.001; t test; n = 24). Figure 4 4) As shown in FIG. 4, in cooperation training experiment, with the increase of cooperation training time, the number of times of mice entering reward area had a significant tendency to increase over time (Mann Kendall test, p < 0.05), and the total time of mice entering reward area had a significant tendency to increase over time (Mann Kendall test, p < 0.01).

[0061] Figure 5 5) As shown in FIG. 5, in cooperation training experiment, with the increase of cooperation training time, the number of times of mice entering reward area had a significant tendency to increase over time (Mann Kendall test, p < 0.01), and the total time of mice entering reward area had a significant tendency to increase over time (Mann Kendall test, p < 0.01).

[0062] 5) As shown in FIG. 5, in cooperation training experiment, with the increase of cooperation training time, the number of times of mice entering reward area had a significant tendency to increase over time (Mann Kendall test, p < 0.01), and the total time of mice entering reward area had a significant tendency to increase over time (Mann Kendall test, p < 0.01). Figure 6

[0063] ​​These embodiments fully verify the correctness and reliability of the device for training and testing the cooperation behavior of rodents. In the cooperation training experiment stage, with the increase of cooperation training time, the number of times and the total time length of the mice entering the reward area, and the number of times and the total time length of the fine drinking water behavior all have a significant upward trend, indicating that the mice learn to avoid electric shock through cooperation and obtain more drinking water opportunities. The device of the present application stably and reliably trains the cooperation drinking water behavior of mice.

[0064] The technical scheme of the present application brings the following beneficial effects:

[0065] 1) The present application provides a new method, system and device for training and testing the cooperation behavior of rodents, which provides a new method for studying the neural mechanism of animal cooperation behavior and the mechanism of abnormal animal cooperation behavior.

[0066] 2) The device for training and testing the cooperation behavior of animals provided by the present application can be applied to animals moving freely, and animals can interact freely without restraint in the experiment, which can more truly reflect the cooperation behavior of animals in nature.

[0067] 3) The method for training and testing the cooperation behavior of animals provided by the present application is stable and reliable, and the animals after water deprivation have a strong motivation to drink water, and the animal individuals can directly enter the reward area to obtain drinking water; the cooperation rule that animals need to learn is not to enter the reward area at the same time, which is more direct than the cooperation rule of the prior art (such as synchronous coordinated movement, pulling the pull rod to provide food for the partner), the efficiency of the cooperation training process is higher, and the cooperation behavior obtained by training is more stable.

[0068] The above introduction is only the preferred embodiment of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A device for training and testing cooperative behavior in rodents, characterized in that: The device comprises an animal experiment box and an image-based control system; the animal experiment box comprises an uncovered experiment box, the bottom of the experiment box being rectangular or square, a reward area being provided in a corner of the bottom of the animal experiment box, the reward area being covered with an electric shock pad, and a drinking device being provided in a corner of the reward area, the size of the reward area and the length of the animal's body meeting a preset condition, so that the animal can only have a drinking opportunity when entering the reward area; The image-based control system includes a monitoring camera, a real-time image analysis unit, and an electric shock control unit connected in sequence. The monitoring camera is used to record the movement of animals in the experimental box, the real-time image analysis unit is used to analyze the movement state of the animals in real time, and the electric shock control unit is connected to the electric shock plate to control whether the electric shock plate gives an electric shock.

2. The device for training and testing cooperative behavior of rodents according to claim 1, characterized in that: The real-time image analysis unit is used for: S1: Capture images in the experimental box through the surveillance camera; S2: Detect all animals in the experimental box and their corresponding positions based on the images captured by the surveillance camera; S3: For each animal detected in S2, calculate the pixel area Area1 and the animal body area Area2 of the overlapping part of the animal's body and the reward area respectively; calculate the ratio of the two areas Ratio = Area1 / Area2; set a threshold Thre for the area ratio; if Ratio ≥ Thre, it is considered that the animal has entered the reward area; repeat this step for all animals detected in S2 to obtain the number of animals that have entered the reward area Num-Reward; S4: Determine whether two or more animals enter the reward area. If Num-Reward ≥ 2, enter S5; if Num-Reward < 2, enter S1; S5: When two or more animals are detected entering the reward area at the same time, the real-time image analysis unit sends information to the electric shock control unit, and the electric shock control unit generates an electric shock signal. The electric shock plate causes electric shock to the animals entering the reward area, causing all animals that have entered the reward area to leave the reward area in order to avoid the electric shock, and they can no longer obtain drinking water.

3. The device for training and testing cooperative behavior of rodents according to claim 2, characterized in that: The threshold Thre of the ratio of the pixel area Area1 of the overlapping part of the animal body and the reward area to the animal body area Area2 is in the range of 0.75-1.

4. The device for training and testing cooperative behavior of rodents according to any one of claims 1 to 3, characterized in that: The real-time image analysis unit is further configured to receive manually inputted numbers of individual animals during the animal detection process.

5. The device for training and testing cooperative behavior of rodents according to any one of claims 1 to 3, characterized in that: The reward area is rectangular, and the coverage area of ​​the electric shock plate is consistent with the size of the reward area.

6. The device for training and testing cooperative behavior of rodents according to any one of claims 1 to 3, characterized in that: The size of the reward area is 2-3 times the length of the animal's body, so that the animal can get the opportunity to drink water only when it enters the reward area.

7. The device for training and testing cooperative behavior of rodents according to any one of claims 1 to 3, characterized in that: Select the imaging frame rate of the monitoring camera according to the movement speed of the experimental animals. For monitoring the movement status of rats and mice, choose a camera with a frame rate of 15-30Hz.

8. The device for training and testing cooperative behavior of rodents according to any one of claims 1 to 3, characterized in that: The analysis speed of the real-time image analysis unit is described by the time required to analyze one frame of image, and needs to be adjusted according to the average movement speed of the experimental animal. For mice or rats, the time required to analyze one frame of image in the real-time image analysis unit is controlled within the range of 0.5-1.0 seconds.

9. The device for training and testing cooperative behavior of rodents according to any one of claims 1 to 3, characterized in that: The electric shock control unit adopts a grid electric shock plate.

10. The device for training and testing cooperative behavior of rodents according to claim 9, characterized in that: The intensity of the electric shock is adjusted according to the type of experimental animal. The electric shock intensity for mice is 0.1-0.5 mA, and the electric shock intensity for rats is 0.5-1.0 mA.

Images