As it happens people underestimate control over situations. Take the classic example of a student who waits until the last minute to study for a final exam because “they have it all under control”. This example is a type of bias that surprisingly is more common than expected. Another example of bias is a person who walks into a café to only get a coffee and is temped to get a tasty pastry. The phenomenon referred to is the restraint bias, or the perceived ability to have control over an impulse. Apply this concept to any vice when someone feels or is biased into perceived control and a similar conclusion is likely to occur.
Take the new fad: the cookie diet. People are purportedly allowed to eat cookies in addition to one meal. And it is precisely because of the name that people underestimate their ability to control the impulse, according to the New York Times report. However because people see the feasibility they are likely to try the diet and nevertheless fall for the impulse of eating that extra cookie.
As an example, Nordgren, van Harreveld & van der Pligt (2009), asked satiated and hungry participants to select a snack which was to be returned a week later in exchange for money. The authors reported that the more restraint bias experienced by satiated participants the greater the likelihood of not returning the snack. More importantly is the fact that the satiated participants chose their first or second favorite snacks while the hungry participants reportedly accounted for the bias by selecting a second or third favorite snack. The cookie diet then is an important example because it sounds harmless and increases the likelihood of bias. However based on the experiment by Nordgren et al., (2009) it can be concluded that because a cookie seems harmless people are more likely to be biased for that extra snack.