> making me immediately skeptical of this one From a fellow reader of Everything is Obvious, same here. It seems obvious (heh) to me that it's not the _apology

> making me immediately skeptical of this one From a fellow reader of Everything is Obvious, same here. It seems obvious (heh) to me that it's not the _apology_ that caused an increase in trust, but the addition of neutral or better contact. In all of the experiments, the control group did one action (e.g. asked for a cell phone or played their move) and the test group did something over and above that action which was a neutral or better action (a superfluous apology). I see no reason to think that the same results could not be reproduced if the control said "you have nice shoes" or "that was a nice move". That is, the conclusion should instead be that an additional amount of neutral or better contact with a person causes them to trust you more than they would without the statement--the contact does not have to necessarily be an apology. In fact, the article admits as much in second to last paragraph: >How trustworthy are these results? The accumulated findings from several experiments help build a convincing case, but unfortunately the field study - which had the potential to provide the most persuasive evidence - is seriously flawed. The actor apologised (sic) for the rain then asked to borrow a phone, or in the comparison condition he just asked to borrow the phone. There was no proper control condition. This means we don't know if the impact of the apology was specific to making an apology or merely an effect of uttering any kind of ice-breaker.