Calendar
DR@W Forum - Bernadette Kamleitner (Queen Mary University of London)
Bernadette Kamleitner (Queen Mary University of London)
When Sunk Costs Matter Most. The Influence of Cognitive Associations on the Sunk Cost Effect.
People are more likely to engage in something costly, boring, or unpleasant, if (and because) they have paid for it. This tendency to consider past cost has been termed the “sunk cost effect”. The sunk cost effect drives decisions across domains and is hence of wide-ranging practical importance. Many explanations have been brought forward to explain why people are influenced by past costs that have no objective bearing on a decision. I add to these explanations and argue that sunk cost effects are a result of people’s tendency to mentally link costs and benefits (cost-benefit-associations); in particular the tendency to think of costs when first contemplating benefits (benefit2cost association). The implication is that natural variations in association patterns as well as priming of associations should influence the extent of the sunk cost effect. Two studies tested whether cost-benefit associations and in particular the direction of benefit2cost associations actually influence sunk cost effects. Study 1 primed (via matching tasks) cost-benefit associations on 2 decision problems, a managerial investment problem and a choice between 2 leisure activities. Decisions on both problems were influenced by benefit2cost associations but not by cost2benefit associations. Benefit2cost associations increased sunk cost sensitivity. Study 2 investigated whether trait cost-benefit-associations also influence actual behaviour. Participants answered items on trait cost-benefit-associations. After a filler task participants were asked to build a puzzle and the salience of the (sunk) cost of building the puzzle was manipulated. After the manipulation participants were offered to either keep the puzzle built or to get an alternative puzzle. If sunk costs were made salient, participants with a strong tendency to establish benefit2cost associations were significantly more likely to go for a new puzzle than participants with a weak tendency to establish benefit2cost associations. Implications for decreasing the vulnerability to sunk cost effects are derived.