Dear Sakthivel,

Any previous experience will change the behavior of an animal in the future. What effects you have to expect depend on the animal you use and the tasks you conduct, also on the order in which different tasks are run.

Whether you "only" train the animal, or also record or score the behavior does not make a difference for the experience the animal had made.

It is a bit difficult to answer your question specifically without knowing which animals you work with and what you would like to investigate.

Kind Regards, Laura

Hello Laura,

     Thank you so much for your suggestion.

  i am going to use Wistar albino rat for the study, and i am not going to record the any score.  and also i train the animal for learning and memory task, To check the maternal experience on offspring.

Do you already have a clear idea on which tests you would like to use? If you list possible test, I can think about the Problems you could have.

Cheers

Briefly: 

a) Yes, it's possible to do but you should consider that training process itself may stress out the individual. By putting it through several training periods may cause the subject to become so stressed that its subsequent behavior is merely a stress response to the training and makes it difficult to assess individual learning. 

b) It really depends on what your two experiments are and how the information from one experiment might affect the animal's behavior on another experiment. For example, if you have a learning test that present conflicting information in the second experiment your learning measure may not be the same between the two tests (e.g. the second one might be looking more at reversal learning). 

Just to add to the above - learning in one experiment may actually present a confound for a second experiment. Consider - 1) the animal is smart overall, learns tasks quickly vs. 2) learning in exp 1 had an effect on learning in exp 2. You may actually want to record data on learning in all exp & examine its impact in later exp. Still not sure how you would differentiate between the above 1/2 issues. Based on my undergrad experience with rats, this is often why psychologists I worked with did not use the same rats in multiple experiments/studies - as pointed out above, all experience affects later behavior & learning (and may confuse your study animal or study results - particularly if you are not systematically examining the prior learning).

Hi Laura!!!

i am going to use eight arm radial maze, T-maze, open filed test and novel object identity,,

thanks & regards

Sakthi 

Hi Paul,

Then which will be best, whether one experiment for longer period or training or two experiment training for  shorter time?

Hi Sakthivel,

It depends. I would recommend piloting a few tests and see if the individuals seem to do ok or if they act stressed the whole time. If they seem to exhibit normal behavior, then it's probably ok if you run both (in terms of stress masking the behavior of interest). However, if they are stressed, you will need to try tweaking it, following your ideas above. In my experience, staple tests in the field do not always work because of small but important details that are left out of the manuscripts or conditions particular to that study. Because of this, it's important to run pilots to figure out what works and doesn't work in your laboratory setting. 

That being said, you still have the issue mentioned above of prior experience affecting performance on subsequent tasks (I'm particularly thinking in terms of the learning tasks). For the open field test and novel object ones, you might take a look at Bell 2013 which discusses some of the issues of test order.

Bell, A. 2013. Randomized or fixed order for studies of behavioral syndromes? Behavioral Ecology 24:16–20.

Hi Sakthivel,

As others wrote, it all depends on what you do. If you are looking at the effect of some variable on behavior, often animals can be used in multiple experiments. See the large literature in operant conditioning, especially with pigeons as subjects, or chimpanzees as subjects. Because you are looking at acquisition and not stable behavior, indeed, there are likely t be sequence effects. 

Regarding multiple learning experiments and stress, one needs to remember that in nature, animals are always being exposed to new learning experiences. So, multiple learning tests will not necessarily produce a stress response. If each learning task produces a stress response, multiple learning tasks will make this more extreme.

I hope this helps,

Jim

Dear Sakthivel,

Sorry for my late reply. I would agree with Jim and not consider simply the performance of tasks per se as stressful. It can even be regarded as enrichment, as laboratory animals are pretty deprived in any sense of experiencing.

Of course, certain tests specifically induce stress, which is most likely what Paul was addressing. You would rather not test animals in several tests involving punishments.

To the specific tests you mentioned, I think Bell et al. (mentioned by Jim) discusses the issue of test batteries quite nicely.

The open field test addresses the response to a novel environment. If the rats were tested in novel environments before, they can react quite differently compared to when they are absolutely naive to novel environments.0

You can combine an open field test with a novel object recognition test, if you first obtain the response to the novel environment, and then add an object. In this way, you also rule out that the rats' behavior is due to a novel environment. For this to work, you best test the rats for a certain duration (let's say about 30 min). You should get a habituation response (decrease in overall locomotor activity; increase in time spent in the center) and a plateau phase (where the behavioral parameters you look at stay the same for a significant amount of time). Then retest the rats on the next day. You will probably see that they initially have an increased response again, but habituating faster. This also tells you something about how they remembered the new environment. You can then implement a novel object, either on the next day, or the same day, and see their interaction. By repeating this, you will be able to see the memory on the object.

If you want to investigate spontaneous alternation/exploration in the radial arm maze, the rats might behave differently after having experienced the open field test before. In general, this might not be bad. I can confound your results though, if your disease model reacts differently to being tested repeatedly compared to your wild type rats.

In case you want to run a reversal Trial (in order to assess flexibility of learning) in the radial arm or T-maze test, it would also be recommended to not repeat this in another test with the same animals.

In general, I would run tests with short protocols first (that would also speak for open field before maze tests).

To pick up Jims last comment, I would also highly recommend to run pilot studies. Use maybe 6 animals per group (wild type and disease model, I will call it transgenic in the following) and test them in open field first, and then a maze test. Take another 6 animals per group and test them in the maze test first. Like this, you can evaluate (with a two-way ANOVA): 1) if there is a difference in the behavior between the genotypes (genotype or between-group effect), 2) if there is a difference in the results when you compare the order of testing (order or within-group effect) and, 3) if there is a difference in how the two genotypes react to the different order (interaction effect).

That's all that Comes into my mind right now. Would be nice, if you shared your results with us, when you get them.

Best, Laura