The linked study is co-authored by the disgraced researcher/fraudster Brian Wansink. I guess this was one of the studies that didn’t get retracted, but I still wouldn’t put a lot of faith in anything with his name on it.
But is that a nit that changes nothing (e.g., there is significant other valid evidence that supports the same point) and so could be simply updated, or is it the linchpin of the entire argument, without which the entire premise and argument fails?
If it is the latter, it's good to draw our attention to it, but if the former, then this is kind of a meaningless distraction. I haven't got the time ATM to sort it out. Perhaps someone with knowledge could add the needed insight?
I do think the bad cite weakens this one point—if there were better studies in support of this point, why didn’t the author cite them instead? And since there’s no reliable evidentiary backing for this point, you have to go by whether it feels right to you. Which is fine! That’s how most advice works. But we shouldn’t delude ourselves into thinking that “studies say” this point is right without further evidence.
There is a reason they say you can judge a person by the company they keep.