In addition to this generous discount the banner stated “4 jars per customer at max” and given the sudden rationing, I immediately felt inclined to get the maximum benefit out of the deal.

There were jars piled up in abundance, stocked to the ceiling, and I couldnt help but admire the insight that must have preceded this new deal, enhancing the discount-inherent time limit with an unexpected numerical one and by that suggesting a higher demand and harvesting a higher volume.

I’d be happy to read about experiments based on this setting as opposed to the same discount without suggesting the ideal number of purchases. After all, it might turn out that I’m the only one who would fall for these transparent schemings: I “only” bought 3 jars.

Don’t want to derail the current discussion, but your studies on how a less-attractive choice makes a similar candidate more attractive has me intrigued.

How do you think this (the idea of contrasting similar choices) would work in election systems?

For the last decade I’ve been interested in single- and multi-winner election systems that are robust according to several criteria. The primary motivation is to reveal the voters’ true preferences without excessive need for compromise or strategic voting. I realize that due to Arrow’s Paradox this is not possible, but some methods show promise.

In particular, Concorcet completion schemes such as Ranked Pairs (Tideman) or Beatpath (Schulze) satisfy the Immunity to Clones Criterion — a set of nearly identical candidates will not hurt each other by being on the ballot together.

What intrigues me is that according to your studies, a set of clones might actually increase the overall attractiveness of that faction.

But what I see in practice is just the opposite … e.g., radical extremist leftists cast a shadow on their more moderate liberal counterparts, and similarly on the other side of the spectrum.

Is your finding more broadly applicable in politics, or is there something else at work?