Final Paper: Cooperative Behavior in Rats

Introduction

The influence of teamwork has been studied by researchers over the years. The findings have shown interesting results on humans, monkeys, dolphins and rats. The studies included in this experiment focus on the ability of the white rat also known as rattus norvegicus to engage in cooperative behaviors.

Cooperation is defined as the event in which two or more laboratory rats engage in behavior that involves working together or using each other to coordinate behavior. Usually this behavior is beneficial to the other(s). One philosopher, William J. Daniel was interested in studying cooperative behaviors in rats. In 1942, he set out to test whether or not two paired rats would be able to learn to avoid shock and obtain food using one another for support. His apparatus consisted of a wooden cage with a grid floor layered with shocking circuits, a platform at one end of the cage, a food crock in the center of the cage near the entrance and a glass top. He taught each rat to step on the platform or to go over to the food crock to avoid being shocked. Then he would administer a shock to the rats in order to get them to switch from the platform to the food crock. Eventually the rats would switch back and forth with no shock. They learned that if the one on the platform stepped onto the grid before the other rat was on the platform, a shock would be given. Therefore, they were learning how to coordinate behavior in order to avoid pain (Daniel, 1942).

William Daniel continued studying cooperation in white rats the following year. In 1943, he came up with an experiment where three pairs of white rats were used and had to work together to get food and avoid an electric shock. He found that the rat pairs learned to cooperate with one another by taking turns stepping on the platform, which reduced their number of shocks. Shocks would be experienced by the rats if they did not step onto the platform, every time they failed to step on they would be shocked. Interestingly enough, the rats not only learned to step onto the platform, but they also learned to alternate stepping on the platform as well as taking turns at the food bowl, located at its center. Both animals worked together to avoid shock and obtain food (Daniel, 1943).

In conjunction with Daniel’s studies on problem solving in rats in order to avoid shock, a second study by Epley and Rosenbaum was conducted in 1971 and then again in 1975. They used a similar apparatus and used a shock method in order to teach the albino rats to interchange positions between the platform and the food. Obviously, the platform and the food were far enough apart so that each rat had to take turns stepping on the platform so that the other could

A third cooperative rat study by Schuster and Perelberg was set up differently than the studies mentioned above. In this experiment, two laboratory rats were taught to simultaneously press a lever. These researchers claim that cooperation results in unlimited social interaction which can lead to positive relationships, communication, and fitness. Social interactions such as these, they say, lead to reinforcement. Therefore, the rats were being reinforced by completing the cooperation tasks.

Not only do Schuster and Perelberg argue that cooperation is seen in the laboratory but they also say that the rats’ cooperative behavior in the lab is tied to behaviors in the natural world such as aggression and group hunting (2004).

A more recent study focused attention on the reciprocal altruism aspect of cooperation. Reciprocal altruism occurs when two or more individuals share behaviors that benefit their respective partner (Rutte & Taborsky, 2007). The first phase of this experiment revealed that rats are more helpful towards a partner from which they had received previous assistance from than towards a partner that had not helped. The second phase of the experiment went on to show that after receiving assistance from other rats, rats were more helpful towards them than they were toward a new partner. Rutte and Taborsky ran this experiment using all female rats. They were taught to pull a lever to release food for itself then they were placed in a box with a littermate but only received the food if they littermate pulled it for them. Therefore the rat that was pulling the lever did not have access to the food, only the littermate did.

In the present study, a test was created to observe the behavioral responses of two, female, albino rats. It was expected that this pair of rats would respond to a condition cooperatively; Rat1 and Rat 2 would chose the cooperation task of standing in a hoop together in response to a click.

Method

Participants

            For our experiment we used two female Sprague-Dawley laboratory rats. Neither rat was deprived of food or water throughout the experiment. They were housed in separate metal cages in a psychology laboratory containing other female Sprague-Dawley rats. The cage was about 36 cm by 24cm, containing about 15 grams of rat food (that were only given once a day) and an attached water bottle for drinking. The ages of the rats were approximately three months at the start of the experiment.

Design

            We used a 2×3 within subjects design. The independent variable for our experiment was the number of rats in the box (one or two) and the type of behavior that could potentially be performed in the box by each rat (individual, cooperation, or having the choice to decide between the individual or cooperation behavior). The dependent variable was the response to the click: either choosing to do what they have been trained to do individually, what they had been trained to do together, or making no behavioral response. This was measured by observing the rats’ behavior directly after the sound of a click.

Materials

            Materials included rat pellets (20 grams per rat per day) which were either left whole or broken up into small pieces. The small pieces (averaging about 0.16 grams) were used as the biologically salient unconditioned stimulus in classical conditioning. Using 5 grams of pellets per rat per experimental session, we performed about 40 trials where the rat would respond to the sound of a click. We then placed the remaining 15 grams in their cages. We did not deprive them of their normal 20 grams of food per day, and we performed the trails every 24 hours so that a day would pass before the rats were fed again during the trials. After we finished the day’s trails, we cleared out the box and let the rats spend 15 minutes together in the box before being placed back in their cages (and thus having access to the rest of their food). The pellets were also used as reinforcement for operant conditioning when shaping their behavior.

            Other materials included water in a water bottle for drinking (300mL), an animal-training clicker for classical conditioning, a large wood box (17.14”12.0612.06”cm), two small containers (c=17.94”cm), and one blue plastic hoop (c=40.38”cm). The apparatus used for the actual experiment was the large wood box containing the two small containers inside it and the blue hoop (see Figure 1, drawn to scale).

Procedure

            We employed four separate conditions to teach the rats two different tasks. In the first condition, Condition 1, the rats were taught to jump onto their container to receive a piece of food. As soon as the rats stood upright on the container (on their hind legs) they were reinforced (with food). The rats were not together in the box during Condition 1; we worked separately with each rat using the same box. Rat 1’s container was located in the center of the box and Rat 2’s container was located in the center- right of the box, about 2 cm from the right wall.

            In the second condition, Condition 2, we used the same conditioning procedure as in Condition 1 to teach each rat to jump up onto their specified container and stand on their hind legs while being physically together in the same box. The purpose of this condition was to get the rats used to being together in the same box at the same time without working together. They both heard the same click and were reinforced for making the correct response regardless of how the other rat responded. Either Rat 1 or Rat 2, both Rat 1 and Rat 2, or neither Rat 1 or Rat 2 were reinforced at any given trial in this condition. Again, the correct response was standing upright on their respective containers in order for reinforcement to occur. The containers were located in the same positions as they were in Condition 1 (center and center-right for Rat 1 and Rat 2, respectively).

            In Condition 3, we taught both rats to stand inside a blue hoop, located in the front left corner of the box. We shaped their behavior by giving them a piece of food and sounding the clicker every time they approached the blue hoop together. They then received a piece of food every time they made physical contact with the hoop. Once they were able to do this separately we reinforced their coordinating behavior of approaching the hoop at the same time and standing inside of it at the sound of a click. They were only reinforced when the two rats coordinated approaching and standing inside the hoop together. We defined “at the same time” or “together” in regards to coordinating behavior as occurring within no more than ten seconds: in other words, the rats were only allowed ten seconds to elapse from when the first rat stood inside the hoop until the second one did in order for reinforcement to occur.

In Condition 4 both rats were placed in the same box containing their specified containers and the blue hoop. We gave them the choice of task: either approaching the hoop together and standing inside of it or standing upright on their specified container. We did not condition their behavior as they had learned both in Conditions 1, 2, and 3. We then recorded which task each rat chose to do in response to the click: their own task of standing upright on their container, the cooperation task of standing inside the hoop together, or making no behavioral response. Together, they could choose to do any combination of the aforementioned three possible responses.

Results

Rat 1 and Rat 2 were both observed responding to the sound of a click in condition four for two consecutive days. On day one and day two of condition four there were forty trials on each day, so there were eighty total trials. Thirty-four of the eighty trials both Rat 1 and 2 were observed choosing the cooperation task, nine trials both Rat 1 and 2 chose the individual task, thirteen trials both rats failed to respond to the click, nine trials Rat 1 chose the individual task and Rat 2 failed to respond, two trials Rat 1 failed to respond and Rat 2 chose the individual task, six trials Rat 1 chose the individual task while Rat 2 chose the cooperation task, two trials Rat 1 chose the cooperation task while Rat 2 failed to respond, four trials Rat 1 chose the cooperation task and Rat 2 chose the individual, and one trial Rat 1 failed to respond while Rat 2 chose the cooperation task.

The frequencies of the combined data from day one and day two showed that 42.5 percent of the time both Rat 1 and Rat 2 chose the cooperation task of standing in a hoop that had a circumference of 40.38 centimeters. 11.2 percent of the time both rats chose the individual task and 16.2 percent of the time the rats did not respond. 11.2 percent of the time Rat 1 chose the individual task while Rat 2 did not show a response. 2.5 percent of the time Rat 1 showed no response while Rat 2 chose the individual task. 7.5 percent of the time Rat 1 chose the individual task while Rat 2 chose the cooperation task. 2.5 percent of the time Rat 1 chose cooperation while Rat 2 showed no response. 5.0 percent of the time Rat 1 chose the cooperation task while Rat 2 chose the individual task. 1.2 percent of the time Rat 1 showed no response while Rat 2 chose the cooperation task.  This is shown in Figure 2.

Our hypothesis states that both Rat 1 and Rat 2 would choose the cooperation task in response to a click. Results showed that the rats did choose to respond more often to the cooperation task because thirty-four of the eighty trials were observed showing cooperation. Thirty-four is more the double the number recorded for any other response.  This is shown on the graph in Figure 3.

Figures

Fig. 1apparatus

 

 

 

 

 

 

 

 

 

 

 

Fig. 2 fig-2-graph1

 

 

 

 

 

 

 

 

 

Fig. 3

figure-3

 

 

 

 

 

 

 

 

 

Discussion

Our results indicate that overall, rats respond by choosing the cooperation task more than their individual task, or even making no response at all. This supports our theory that, when given the choice, both rats will perform the cooperation task of approaching and standing inside a hoop at the same time. This was true for 34 trails, as compared to the 9 trails where they chose their individual task. Although we did not run a statistical analysis on our findings, we are confident that this difference is significant enough to address.

When both of the rats responded to their individual task during the same trial, we assumed cooperation was not involved, as each rat was taught that responding to this task was not contingent upon the other rat also responding to her individual task. That is why we added Conditions 1 and 2: in Condition 1, the rats were taught their individual task while alone in the apparatus, and in Condition 2 they continued practicing this task together in the box and were reinforced regardless of the other rat’s response to the click.

Some limitations of our study include small sample size (n=2) and short testing time (80 trials over two days).  Another limitation is the fact that the rats were reinforced in Condition 4 – the testing condition – regardless of which task they chose (individual or cooperation). This may have devalued the cooperation task, as the rats knew that they would be reinforced for either of the behavioral responses they were taught to make. If this study were to be replicated, a fifth condition should be created where the rats were only reinforced for choosing the cooperation task. This should strengthen the cooperation task and (hopefully) encourage the rats to cooperate, despite the presence of their individual containers in the apparatus. Also, it is important to note that our study focused on the behaviors used when the rats cooperate, but not why they do it. It is important to investigate why the rats chose the cooperation task, not just whether or not they do it. Our study also failed to address if cooperation resides in the contingency between the behaviors of two (or more) rats and outcomes, or if they would still continue to respond to the cooperation task regardless of this contingency?

This study also brings about certain implications that may encourage future research in this area. The fact that both subjects chose the cooperation task more frequently than their individual tasks suggests that even when time-consuming or uneconomic in the short-term, cooperation can still be economic in evolutionary terms when long-term consequences can be linked to such behaviors that influence the tendency to cooperate. Even though choosing the individual task results in obtaining reinforcement (economic) quicker and easier, the rats still performed the cooperation task more frequently. This relates to the aforementioned limitation of needing to understand why the rats chose to respond to cooperation task. One possible explanation is that coordinating behaviors and cooperating results in not only long-term economic benefits (reinforcement), but immediate benefits such as attending to their intrinsic social emotions and social relationships (assuming rats have an innate desire to be around other rats, that is).

            Our study brings both limitations and implications that can, on a small scale, better the field of behavioral psychology. Behavior in and of itself is interesting, especially when it comes to interacting with others. Our rats were given the opportunity to receive reinforcement quickly and immediately by performing their individual tasks, and yet they chose to coordinate their behavior and respond to the cooperation task. As experimenters, we taught the rats how to coordinate behavior, and they taught each other how to cooperate in order to be reinforced. This notion can only leaving one wondering – would the rats continue to cooperate even if they were not being reinforced?

References

Daniel, W. J. Cooperative problem solving in rats. 1942. Journal of Comparative Psychology,

34, 361-368.

Daniel, W. J. Higher order cooperative problem solving in rats. 1943. Journal of Comparative

Psychology, 35, 297-305.

Epley, S. W. & Rosenbaum, M. E. 1975. Cooperative behavior in rats: Effect of extinction of

shock onset condition drug acquisition. Journal of Personality and Social Psychology,

31, 453-458.

Perelberg, A. & Schuster, R. 2004. An animal model for studying the behavior of cooperating.

Behavioral Processes, 66, 261-277.

Rutte, C. & Taborsky, M. 2007. The influence of social experience on cooperative behavior of

rats (rattus norvegicus): Direct versus generalized reciprocity. Behavioral & Sociobiology, 62, 499-505.

Advertisements

Updated Method Section: 4/20/09

Our hypothesis in this experiment states that both Rat 1 and Rat 2 will perform the cooperation task of approaching and standing inside the hoop at the same time, in response to the sound of a click. 

 

Method

Participants

            For our experiment we used two female Sprague-Dawley laboratory rats. Neither rat was deprived of food or water throughout the experiment. They were housed in separate metal cages in a psychology laboratory containing other female Sprague-Dawley rats. The cage was about 36 cm by 24cm, containing about15 grams of rat food (that were only given once a day) and an attached water bottle for drinking. The ages of the rats were approximately three months at the start of the experiment.

Design

            We used a 2×3 within subjects design. The independent variable for our experiment was the number of rats in the box (one or two) and the type of behavior that could potentially be performed in the box by each rat (individual, cooperation, or having the choice to decide between the individual or cooperation behavior). The dependent variable was the response to the click: either choosing to do what they have been trained to do individually or what they had been trained to do together. This was measured by observing the rats’ behavior directly after the sound of a click.

Materials

            Materials included rat pellets (20 grams per rat per day) which were either left whole or broken up into small pieces. The small pieces (averaging about 0.16 grams) were used as the biologically salient unconditioned stimulus in classical conditioning. Using 5 grams of pellets per rat per experimental session, we performed about 40 trials where the rat would respond to the sound of a click. We then placed the remaining 15 grams in their cages. We did not deprive them of their normal 20 grams of food per day, and we performed the trails every 24 hours so that a day would pass before the rats were fed again during the trials. After we finished the day’s trails, we cleared out the box and let the rats spend 15 minutes together in the box before being placed back in their cages (and thus having access to the rest of their food). The pellets were also used as reinforcement for operant conditioning when shaping their behavior.

            Other materials included water in a water bottle for drinking (300mL), an animal-training clicker for classical conditioning, a large wood box (17.14”12.0612.06”cm), two small containers (c=17.94”cm), and one blue plastic hoop (c=??”cm). The apparatus used for the actual experiment was the large wood box containing the two small containers inside it and the blue hoop (see Fig.1, drawn to scale).

Procedure

            We employed four separate conditions to teach the rats two different tasks. In the first condition, Condition 1, the rats were taught to jump onto their container to receive a piece of food. As soon as the rats stood upright on the container (on their hind legs) they were reinforced (with food). The rats were not together in the box during Condition 1; we worked separately with each rat using the same box. Rat 1’s container was located in the center of the box and Rat 2’s container was located in the center- right of the box, about 2 cm from the right wall.

            In the second condition, Condition 2, we used the same conditioning procedure as in Condition 1 to teach each rat to jump up onto their specified container and stand on their hind legs while being physically together in the same box. The purpose of this condition was to get the rats used to being together in the same box at the same time without working together. They both heard the same click and were reinforced for making the correct response regardless of how the other rat responded. Either Rat 1 or Rat 2, both Rat 1 and Rat 2, or neither Rat 1 or Rat 2 were reinforced at any given trial in this condition. Again, the correct response was standing upright on their respective containers in order for reinforcement to occur. The containers were located in the same positions as they were in Condition 1 (center and center-right for Rat 1 and Rat 2, respectively).

            In Condition 3, we taught both rats to stand inside a blue hoop, located in the front left corner of the box. We shaped their behavior by giving them a piece of food and sounding the clicker every time they approached the blue hoop together. They then received a piece of food every time they made physical contact with the hoop. Once they were able to do this separately we reinforced their coordinating behavior of approaching the hoop at the same time and standing inside of it at the sound of a click. They were only reinforced when the two rats coordinated approaching and standing inside the hoop together. We defined “at the same time” or “together” in regards to coordinating behavior as occurring within no more than ten seconds: in other words, the rats were only allowed ten seconds to elapse from when the first rat stood inside the hoop until the second one did in order for reinforcement to occur.

            In Condition 4 both rats were placed in the same box containing their specified containers and the blue hoop. We gave them the choice of task: either approaching the hoop together and standing inside of it or standing upright on their specified container. We did not condition their behavior as they had learned both in Conditions 1, 2, and 3. We then recorded which task each rat chose to do in response to the click: their own task of standing upright on their container, the cooperation task of standing inside the hoop together, or making no behavioral response. Together, they could choose to do any combination of the aforementioned three possible responses.

 

 

 

*we are re-drawing our apparatus (Fig.1) since we changed the materials used in the cooperation task.

Condition 4 Data

Day One: Monday, April 20, 2009

 

Trail #

Louise’s Response

Thelma’s Response

1

Individual

No Response

2

Individual

Individual

3

Cooperation

Cooperation

4

Cooperation

Cooperation

5

No Response

No Response

6

No Response

No Response

7

Individual

Individual

8

Cooperation

Cooperation

9

Individual

Cooperation

10

Cooperation

Cooperation

11

Cooperation

Cooperation

12

Cooperation

Cooperation

13

No Response

No Response

14

Individual

Individual

15

No Response

No Response

16

Individual

Cooperation

17

Individual

No Response

18

No Response

Individual

19

Individual

Individual

20

Cooperation

No Response

21

Individual

Cooperation

22

Individual

No Response

23

No Response

No Response

24

No Response

No Response

25

Cooperation

Cooperation

26

Individual

No Response

27

Cooperation

Cooperation

28

Cooperation

Cooperation

29

Cooperation

Cooperation

30

Cooperation

Cooperation

31

Cooperation

Cooperation

32

Individual

Cooperation

33

No Response

No Response

34

No Response

No Response

35

No Response

No Response

36

Individual

No Response

37

Individual

No Response

38

Individual

Cooperation

39

Individual

Individual

40

No Response

No Response

 

 – 12 out of 40 times both rats responded to the cooperation task

– 10 out of 40 times both rats made no response

– 6 times out of 40 Louise responded to her individual task and Thelma made no response

– 5 out of 40 times both rats responded to their individual tasks

– 5 times out of 40 Louise responded to her individual task and Thelma responded to the cooperation task

– 1 time out of 40 Louise responded to the cooperation task and Thelma made no response

– 1 time out of 40 Louise made no response and Thelma responded to her individual task

Condition Four Videos: Day 1

Both Respond to Cooperation Task:

Both Respond to their Individual Tasks:

Both make No Response:

Louise Responds to Individual Task; Thelma to Cooperation Task:

Louise Responds to Individual Task; Thelma makes No Response:

Condition 3 Videos

Day1.

Day2.

Day3.

Day4.

*On Day 4 we decided to change the cooperation task: instead of teaching the rats to pick up wood chips and carry them from one corner to the other, we had them touch or stand inside of a small blue hoop. Again, they were only given 10 seconds to both get to the hoop (and stay there until the other rat arrived if they didn’t get there at the same time) if they were to receive food. We changed the cooperation task because towards the end of the third day of Condition3 the rats started chewing on the wood chips and breaking them apart with their teeth.

Day5.

Final Video for Condition2

We had to spend a lot of time working on Condition2 to ensure that both rats had a high behavioral response to the click while together in the experiment box. Above is another video of both animals making the correct response (standing upright on the ceramic container) to the sound of a “click”. During Condition2, we also separately worked on Condition1 to maximize exposure to this task (again, standing upright on the ceramic container after hearing a “click”). We are confident that both animals have “learned” this behavioral response, as they do so after hearing the click almost every time. We have now begun introducing Condition3, or the cooperation task. Again, both rats will be in the experiment box at the same time, except now without the two ceramic containers. Instead, there will be a small pile of 10 wood chips in the top left corner of the box. The rats will learn to pick up a wood chip and carry it to the back left corner as soon as they hear a click. We will use shaping to bring this behavior from a low operant level to a high operant level by using food as a reward. We will then use Classical Conditioning to get the animals to associate the sound of a click with the aforementioned behavior of bringing the wood chip from one corner to the other. The only way the animals will be reinforced (with food), however, is if they both make this response at the same time. They will be given 10 seconds from when the first rat drops the wood chip in the opposite corner until the second rat does so if both are going to be reinforced. If only one rat makes the appropriate response and the other does not (within a 10 second time frame) or neither rat makes the appropriate response, than reinforcement will not occur. Again, it is only if both rats make the correct behavioral response within at least 10 seconds of each other that they will be reinforced. We are beginning this Condition (3) this week and hope to have both rats exhibiting this behavior by the end of the week. We will then introduce Condition4 early next week and collect our data. We will collect our data over a 5 day period (Monday-Friday) and then will analyze our results that weekend. We are really excited to see how our experiment will turn out and hope that our rats will support our hypothesis and both choose to respond to the click by picking up the wood chips together!!

Experiment Update

Goal1. Conclude Condition 2 and begin Condition 3

Goal2. Revise and update Method section of paper

 

Since our last blog entry, we have revised our Method section for correction by our Professor and have received his remarks. We are currently working on updating our Method section, using both our Professor’s corrections and our own experiences actually conducting the experiment. We are constantly making minor changes as we develop more effective methods of performing our experiment. We hope that by the end of this week we will complete all revisions and be able to have it peer (or professor) reviewed by next week.

 

We are also still working on Condition 2. I even work on Condition 1 when my partner and I are not together and I am in the lab alone. I feel that it is necessary to continue the individual trick training over a period of time so that each rat has “learned” the trick well enough to make the same response after “learning” the cooperation task in Condition 3.

 

We are starting to implement Condition 3 and hope to have a video up by the end of this week or early next week.