If I have used any confusing terms please forgive me. My father is an engineer/flight test conductor and I grew up around it.
I did not mean a leash, or a scientific control group, but a scientific control. A control is a part of the test designed to eliminate false positives or false negatives by exposing your test subject to the conditions of the test without exposing it the independent variable (in this case color) in the experiment. This allows you to see if the dependent variable (in this case a particular behavior) is exhibited or not. If it is then you know your test is invalid in this case. Of course you'll want to repeat the test about 20 times so you can get a feel for your error margin.
Control groups are a type of control that is often used in biomedical research where one group of subjects is given a treatment and the other is given a placebo (a fake drug). The control group is what you compare the group that is treated against.
To set up a control in your test I recommend using boxes all the same color, let's say gray, and see if your chinchilla can discern between them. If it can then you know something is wrong with the test. It could be detecting a difference in smell, or texture, or shape or something that has nothing to do with color but you wouldn't know without a good control. Chinchillas have a good sense of smell so you might want to rub a small drop of perfume on each box to overpower any subtle smells.