It matters where money comes from. Logically, it shouldn t matter whether that hundred dollars in your wallet came from your grandmother for your birthday, from her will, from a tax rebate, from the lottery, or from selling an antique porcelain kitty on eBay. A hundred dollars is a hundred dollars. But your brain thinks it matters with implications for everything from how you shop to how the government distributes stimulus money.

Why is one hundred-dollar bill different from another (other than the narco-state-on-the-verge-of-a-coup trappings of the new bills Treasury unveiled this week)? Why do our brains push money into separate piles? And what do these piles tell us about our relationship with money more generally?

The short answer to all these questions is two words: mental accounting. A concept pioneered by behavioral economist Richard Thaler in the 1980s, mental accounting describes a process by which people tend to monitor their spending according to segregated accounts inside their head. While the accounts are of course nothing more than mental fictions, they have very real implications. Most noticeably, people are more likely to spend from certain accounts than they are from others.

A classic example of mental accounting is the guy who maintains a large credit-card balance racking up interest on the order of 18% while he has the money sitting in his checking account to pay it off. In purely rational terms, he should immediately do so. But because that money is labeled inside his head as savings or emergency fund he takes a major hit so as not to breach the walls of his mental accounts.

The same effect, in a different context, is seen in the investor who sells his winning stocks but holds onto his losers. Since it s more painful to close the losing account and take the loss, that investor will leave a sinking stock be while selling off a stock that s gone up in value.

Accounts for how people plan to spend money aren t the only type of mental accounting, though. People also assign money to different mental accounts based on how they obtained it.

A study of prostitutes in Oslo in the 1980s, for instance, found that the women were more likely to spend their money from sex work on drugs and alcohol and partying; government checks and other money from legitimate work was much more likely to be used to cover things like rent and groceries. Another well-known effect is the so-called house effect, where gamblers are more willing to take big risks with money they ve won at a casino. While all money should be treated as fungible, the gamblers clearly feel they re not really losing anything if they lose the house s money. People even spend money differently based on how they feel about how they obtained it an inheritance from Grandma, for instance, is less likely to be spent on something frivolous and fun than, say, a lotto winning.

One of the most important forms of mental accounting, though, for policy makers and businesses, is how people treat financial windfalls. The purpose of an economic stimulus bill, for instance, is to get people to spend money instead of saving it. So, if you give people the same amount of money, but frame it differently, can you get them to spend more?

Behavioral economics tells us framing does matter. In a 2006 review of the literature, Nicholas Epley and Ayelet Gneezy of the University of Chicago Graduate School of Business found that framing a sum of money as a bonus makes people much more willing to spend it than framing it as a rebate or a return to the status quo. What s more, larger windfalls make people more likely to save, and small windfalls make them more willing to spend an effect that s been observed, among other places, in Holocaust survivors paid large and small sums of reparations from the German government.

So, what does this mean for a president trying to stimulate the economy? Well, consider the case of the checks President Bush distributed in 2001, in sums of $300, $500, or $600. This money, meant to stimulate consumer spending, was sold as a rebate that was being returned to the taxpayers who earned it. While that may have been good politics, there s significant evidence that the $38 billion spent on this idea didn t have its desired effect. People seem to have overwhelmingly stuffed the money into savings a predictable outcome, perhaps, for a lump-sum rebate.

To try to get around this effect, President Obama has done his stimulus differently. Instead of one big rebate check, the Obama administration distributed their stimulus money by reducing the amount the government withheld from people s weekly paycheck, so that their weekly incomes would go up slightly. It s unclear so far how well this idea has worked one danger being that people might not have noticed the extra money in their paychecks at all.

Whether or not this particular tweak worked, it s clear policy makers need to pay attention not just to how much money they distribute, but how that money will be perceived by the people receiving it. One dollar is simply not the equivalent of any other at least not to our brains.

Ryan Sager writes the blog Neuroworld at TrueSlant.com.