I need advice; Correct statistical model in preclinical experiment

#1
Hi all,

I made a specific preclinical experiment to test an effect of new drug on memory. I used water maze test (behavioral test based on learning and orientation in space) with 4 groups of animals (each group n=7), 3 different treatment groups and 1 saline group. I measure three values; time spent finding a target (in sec), distance travelling (in cm) and procentual time spent in specific area divided to the total time (in %) and my main focus is on (i) 4 last days because one group was tested subchronically, and (ii) the last day, when I tested 2 different acute doses. Total duration was 12 days.

For the better validity of the design I used different measurements as I placed each animal every day into the labyrinth 8 times and always from the North, South, West and East and then again N, S, W, E. And this is now problem for me. If I used only one side, I think I would use 2-way ANOVA with repetitions to measure the difference but now I have a set of data which are different (4 sides twice each day).

Could you please suggest an idea how to approach to have a strong statistical method? I am a PhD student and I have limited knowledge in this field.

Thank you,
Marek