Last I checked, pineapple juice isn't a highly-utilised behavioural manipulation channel.

The other mechanism is.

Even were the relationship not causal, it is at the very least likely exploitable.

Pineapple juice is exploitable as well! Sweetened beneverages are highly caloric and hyperpalatable and preferentially consumed by lower-SES customers. This is an actual criticism people have of our food supply! The point isn't that they're not exploitable, it's that you have to show that they are, otherwise we're not talking about a study anymore, and instead simply telling each other just-so stories.