I think that is the message behind a new study which contains this MTurk conclusion:
“First, we were able to show that individuals are heavily affected by search and evaluation costs, even when, by searching and evaluating properly, they would be able to find tasks with higher rewards. Most importantly, by gaming the market and moving our tasks to the first positions of the default task display, we were able to show that we can affect, using our model, the decision making process of individuals.”
I have to say, though, this whole article is over my head.
“The cost of search and evaluation in online problem-solving social networks with financial and non-financial incentives”