I am going through some of my old grad school notes, and in my microeconomics notes on transitive preferences, the teacher made a note of a behavioral economics result where when presented with two dates, A and B, a person might say they would strictly prefer A to B. Then when presented with A, B, and B', where B' was slightly uglier than B, then suddenly the preferences might turn into B preferred to A preferred to B'.
Does anyone know what paper this could be taken from? My notes here were written in September 2015, so the paper would have to have been earlier than this.