A few weeks ago, somebody suggested that willing Chowhounds replicate the vanilla taste test from Cooks Illustrated. Cooks Illustrated investigated whether people could distinguish the difference between real vanilla and artificial vanilla (also called vanillin). Cooks found that when it comes to baked goods, tasters preferred vanillin. In things like custards, real vanilla was the winner.
I did my own version of this test over the weekend. I used Beranbaums pound cake recipe from Cake Bible. I made one pound cake with pure Tahitian vanilla extract. The brand I used was Nielsen-Massey, which cost about $15 for a 4-oz bottle. I made another pound cake with vanillin, which I bought from Smart & Final for about $4 per quart.
Aside from the vanilla-vanillin variable, I tried to keep all other variables constant. I used one mixer and two work bowls, cleaning and drying the beater before mixing next batch of batter. I dipped from the same batch of flour, sugar, beaten egg, etc. I weighed everything on a scale, timed the mixer so that each batch was beaten for the same amount of time at the same speed, and I baked the cakes side-by-side in identical pans. Okay, I know this sounds picky, but I wanted to be sure I was controlling the variables properly so that I didnt taint the results. I guess all that time in graduate school has really paid off! :)
I served the cakes to family and friends. I didnt tell them which was which until after the votes were in.
The difference between the two cakes was not huge, but it was noticeable. Three out of the four adults (including me) preferred the vanillin cake.
To my palate, the vanillin cake tasted and smelled more vanilla-y. Im convinced that for baked goods, vanillin is the way to go. It seems to hold up better after a stint in the oven, and its a heck of a lot cheaper than the pure stuff. But, if I were making, say, a pudding or a custard, I would use real vanilla extract or a vanilla pod because I think pure vanilla would shine under those circumstances.