Sunk Cost inclinations are greater in problems with a social dimension

Interesting insight. Needs to be robustly replicated. From The Time and Social Context in Sunk-Cost Effect by Josip Hrgović and Ivana Hromatko. From the abstract.

In classic economic literature, decisions based on unrecoverable past investments instead of future gains are considered fallacies. They are known as the sunk-cost fallacy in humans and Concorde fallacy in other animals. Here, we argue that the aforementioned fallacies may in fact reflect adaptive heuristics, as well as that the evolutionary novelty of the dilemmas used in previous studies caused these seemingly suboptimal decisions. We hypothesized that (a) because humans are social animals, the tendency to keep investing should be stronger in social than non-social context (because consequences of such decisions might influence one’s reproductive success) and (b) if it is indeed an adaptive heuristic, the decision to keep investing should require less cognitive effort than the decision to terminate further investments. To test this, we developed several domain-specific tasks. Participants had to decide whether to stop or continue investing into an unprofitable project such as a business investment, romantic partnership, friendship, and sibling relationship. The time needed to reach a decision was used as an objective measure of cognitive effort. The pattern of responses was similar for social scenarios, where a majority of participants decided to prolong the investment. Immediate termination of any further investments was likeliest in the business scenario. Furthermore, participants were faster in reaching the decision to keep investing than the decision to terminate further investments, thus corroborating the notion that the decision to keep investing requires less cognitive effort.