I’m always wary of secondary sources for this kind of thing, especially when it’s in a non-technical field like marketing. Somewhere between Dr. Ambady, Malcolm Gladwell, Harry Beckwith, and Scott’s post here, the original findings of these studies have been garbled a bit.

Both papers are about the accuracy of thin slices (meaning first impressions, more or less). I can’t find much of anything in the papers about the durability of first impressions. I’ll let you know if I find anything, though.

Plenty of research has been done on the confirmation bias, which is more relevant to our conversation here, as well as the primacy effect (although, see also its opposite, the recency effect, at the same link).

But I’m confident none of this research suggests that your average “first impression” is immovable. (And the relevancy of Dr. Ambady’s research is dubious to begin with, as it’s done on people, not brands—a significant difference, in my opinion.) To the contrary, all of my experience studying the human mind suggests that it is remarkably malleable. The creation of new connections and associations is, at a basic level, how the brain works!

I’m sure Scott’s right that in many cases it is prohibitively costly to try to change customers’ minds about a brand. But determining whether or not it’s worthwhile to try is the name of the game. Assuming it will be impossible without analyzing the situation may prove the more costly mistake.