How do you like them apples … now versus later?

Imagine yourself in the following scenario … a generous person offers you the following choice:

  • A small piece of chocolate (dark or milk chocolate … your choice!) in 365 days


  • A healthy apple (green or red … again, your choice) in 366 days?

Which would you choose?

Now imagine the scenario is a bit different. You can choose:

  • A small piece of chocolate (dark or milk chocolate) RIGHT NOW


  • A healthy apple (green or red) TOMORROW

If you’re like most people, your answers likely changed. People typically choose apple in 366 days in the first scenario, but then choose the chocolate when it’s immediately available. BUT, in both cases, there is the same one day delay between the options. This is known as dynamic inconsistency and is one of Richard Thaler’s ideas that inspired the birth of behavioral economics and resulted in his receipt of the Nobel Prize in economics. (And if you’ve not yet read the book, Nudge, stop what you’re doing and do so right away!)

So, what’s the fuss? Why care? The interesting thing about dynamic inconsistency is that it runs counter to traditional economic theory. If apples are worth more than chocolate, then they should ALWAYS be worth more. But they’re not. Context matters in how reinforcers are valued and consumed. This is the premise behind behavioral economics. When choices are immediate, organisms are tempted by immediate gratification and make “myopic” decisions (good for the short term, bad for the long term). But the temporal horizon is altered and the choices are both delayed, it is easy to make the healthier decision.

Delays are powerful and can be leveraged to motivate healthy decisions. Even minor delays can impact preference. A classic example is Van Houten, Nau, and Merrigan’s study showing that signage promoting stair usage near an elevator did not sufficiently change behavior; however, imposing a brief delay to the elevator door opening yielded substantial decreases in energy consumption associated with elevator operation. Thus, rather than riding the elevator, the researchers nudged stairway usage, thereby saving electrical energy AND promoting physical exercise.

Leveraging delays to nudge healthy decisions necessitates a scientific approach to quantifying effective delays. Behavior analysis has proliferated the science of delay discounting – the concept that outcome valuation is reduced as a function of delay. Delay discounting is widely recognized by the social and behavioral sciences and is a major success story of behavior analysis. Delay discounting uses parametric techniques to identify the exact ways delays impact value (the function is typically hyperbolic-like, meaning that very small increases in delay have profound effects on value, but these effects become relatively less profound at very high delays; for example, increasing the elevator door delay opening from 1 second to 30 seconds will have major effects, while changing it from 30 seconds to 60 seconds won’t be as robust because the outcome is already devalued substantially).

Recently, Bradley Appelhans and colleagues published a paper in the journal Appetite on the effects of small delays on snack food choices in vending machines. Using insights from delay discounting, the researchers identified a delay of 25 seconds as a potentially effective nudge to change snack food choices. The idea was to impose the 25 second time delay as a “tax” on unhealthy snack food choices. In one study, the researchers provided participants with a voucher for a free snack at the vending machine; thus, participants could get a free healthy snack, or a free healthy snack at varying delays. The unhealthy option was significantly discounted around 25 seconds, which informed delays used in later studies comparing delays with monetary taxes on unhealthy options, as well as discounts on healthy ones. Results across tens of thousands of vending transactions suggested a $0.25 tax on unhealthy snacks and a combination of the tax and delay on unhealthy snacks were the most effective interventions – but, keep in mind that the delay itself was effective (just not AS effective as when the tax was imposed along with it). Best of all, overall sales were not impacted by the interventions, suggesting that this intervention may be scalable without fear of economic harm (thus, one could argue that the persistence in sales indicates social validity of the intervention). Imagine the potential of translating this approach to school cafeteria choices!

Why my excitement over the use of minor delays in nudging healthy and sustainable behavior? There are many reasons. First, delay imposition is backed by decades of basic research, rendering it conceptually systematic with behavior analytic research. Second, it is a cheap (financially speaking) intervention. Reinforcers are not required and once the delay-imposing technology is designed and implemented, the intervention can persist without much human oversight. Third, delays – unlike costs and taxation – mitigate social justice concerns inherent in financial interventions where particular socioeconomic statuses are differentially sensitive to such manipulations. Fourth and finally, delays are consistent with Thaler and Sunstein’s notion of libertarian paternalism with respect to intervention (again … go read Nudge for more information). In this concept, delays promote the behavior people prefer without restricting personal rights/liberties. If people want to make wait out the delay to access the less-preferred alternative (e.g., unhealthy snacks, taking the elevator), they have the ability to do so. Such interventions do not prohibit responses or punish specific choices. The “unhealthy” choice is still present, and it’s available without penalty (with the exception of wait-time, of course).

Despite the awesome potential of leveraging delay discounting to promote healthy decisions, relatively little work in applied behavior analysis has done so, especially in the general population (under the dome). But imagine the possibilities … Imposing a delay on accessing social media in school or business settings where social media use is discouraged. Imposing a delay between spins on a slot machine. Pharmacologically imposing delayed effects of drugs with high abuse liability. Creating traffic control delays in non-HOV lanes on highways that feature HOV lane options. Imposing delays on the use of credit cards to purchase particular items for which patrons opt-into intervention. The possibilities are endless. Let’s not delay (pun intended) this research line any longer. Budgets are tight and there are plenty of significant social problems rooted in unhealthy decisions … delay-based interventions could be a fantastic solution.