Step 1: The preparation
1. I have two main curly strand sizes (plus some straight hair too but I didn't use this for obvious reasons).
2. I have super curly hair that pretty much coils round as you can see in the picture below.
3. For a better view, I stretched out the length loosely and picked out shed hair that matched the smaller strand size A (mostly because there was more of it).
Step 2: The experiment
1. I took 4 individual stands of hair and washed them in shampoo to get rid of any hair product on them (oil or conditioner). I then let them air dry.
2. I took 2 strands and soaked them in plain water for 24 hours. The next two strands were soaked in a solution of baking soda for 24 hours (dissolved as much baking soda as the water could hold)
Step 3: The result
From my point of view (see the pic below), I could not see a difference between the water soaked strands and the baking soda treatment. The hair had the same curl and stretched out in the same way. The curl pattern looked the same before and after.
Step 4: The conclusion
From my experiment using my baking soda (Sainsbury's brand if you are interested), there was no 'relaxing' of the curl.
I think it matters that the baking soda I used had some corn starch in it (all the ones on the shelf did too). Its pH was 7-8 (as measured by me). Some people report that pure baking soda should have a pH of 9. Relaxers have a pH of 10-12. I do think people should know that the pH of the relaxer is important towards relaxing hair but more important is the ability to break the bonds of the hair. In short, I do not think that baking soda can break bonds.
I think much like the coconut lime relaxer, some people are confusing moisturised hair with relaxed hair.