How do soda taxes reduce consumption?

The WSJ writes about the controversy surrounding the proposed new soda tax under consideration in Richmond City, northeast of San Francisco.

In May, Richmond's City Council agreed to put a measure on the November ballot to charge businesses a penny for every ounce of those beverages they sell in the city. If it passes, it would be the first city tax of its kind in the nation and the first to be approved by voters.

Now, I see two possible effects of a soda tax packaged this way, both of which will work towards reducing consumption of soft drinks. One, the standard "price effect" arising from the higher prices payable by the consumer due to incidence of taxation. Second, the more subtle "signaling effect" due to the negative imagery on soft drinks consumption conveyed by the reason underlying the decision to impose a tax. This signaling may be more effective if the tax is described as a "sin" tax, in so far as it would signal to the religious and moral sentiments of its consumers.

Which is of these two effects is more powerful in reducing consumption? What are their respective contributions? This has important implications for public policy. If the signaling effect is reasonably powerful, then it opens up possibilities for limiting certain negative externalities. More specifically, it would mean that even a small tax (and therefore more politically acceptable) can be effective in mitigating a negative externality if packaged appropriately. 

It would be great to explore this through a randomized control trial experiment (RCT), preferably at Richmond itself, to segregate the relative consumption reductions achieved by the two effects.