Learn it → Apply it → Save as Skill
A reward due today can beat a much bigger reward due next week — and then lose to that same bigger reward once both are pushed a month away. We don't discount the future at a steady rate. We discount it steeply near the present, then almost stop, which is why plans made from a comfortable distance rarely survive contact with "now."
The Dr. Wallace's Favorite Mental Models library is designed to turn useful ways of thinking into reusable tools. Each model gives you a fast visual overview, deeper interactive learning, practical application exercises and an AI-ready Markdown skill you can save for future use.
Instead of merely reading about mental models, you can gradually build your own AI skills library: a collection of portable reasoning skills that can help ChatGPT, Claude, Grok and other AI assistants apply the mental models you value.
When a smaller reward available now is compared with a larger reward available later, the near-term option gets an outsized boost just for being close. Push both options equally further into the future, and that boost mostly disappears.
Notice: the gap is 7 days in both rows. Only the distance from today changed — and the preferred option flipped.
The model is most useful whenever a smaller, immediate option is being weighed against a larger, delayed one, and the comparison quietly changes once both options move further away in time.
The method does not tell you which option is objectively better. It tells you whether today's preference is really about value, or just about nearness.
State the smaller-sooner reward and the larger-later reward exactly as they are offered right now.
Imagine both options moved an equal distance further into the future. Would the choice flip?
If it flips, you likely won't stick with today's "reasonable" plan once the near-term reward actually arrives.
Build a structure now, while the reward still feels distant, that removes the choice at the moment temptation is close.
Open a use case to see what the model changes about how you'd think it through.
Ask whether the same-percentage discount would feel as urgent if the deadline were two months out. If not, part of the pressure is coming from the discounting curve, not from the deal's real merit.
An automatic transfer set up today, while payday still feels distant, tends to hold. The identical transfer decided on payday, with the cash already visible, is far easier to skip.
A plan made today about a distant Monday is judged calmly. The same choice, faced Monday morning, now competes with an immediate reward the earlier plan correctly discounted as unimportant.
Recognize that a subscriber's intention to cancel gets harder to act on the closer the renewal date gets, even though nothing about the product's value changed in the meantime.
An offer available immediately gets weighted more heavily than a similar offer a few weeks out. Separate the deal's real value from the psychological pull of its immediacy.
The fix's cost feels small today and its payoff feels distant, so each sprint repeats the same reasonable-sounding deferral. A standing policy decided in advance resists this better than a fresh decision every sprint.
This is an illustrative example built on the classic reversal experiments, using round numbers to teach the method.
Expand for the research behind the model and how it differs from the closely related idea of "present bias."
In research beginning in the 1970s with animal subjects, psychologist George Ainslie found that preferences between a smaller-sooner and larger-later reward reversed as both rewards were moved further from the present — a pattern a steady, exponential discount rate could not produce. He proposed a hyperbolic curve, one that drops steeply near the present and flattens with distance, as a better fit for the data.
In 1981, Richard Thaler asked people to name a future amount of money they would consider equivalent to a smaller amount today, and found the implied discount rate fell as the time horizon lengthened — human evidence echoing Ainslie's earlier findings. Thaler and Loewenstein later applied the pattern to real-world choices such as savings and self-control.
Economist David Laibson introduced a simpler two-parameter approximation, sometimes called beta-delta discounting, that captures the extra weight placed on the present without the full mathematical complexity of a true hyperbolic curve. It is the version most often used in formal economic models of savings and consumption.
They are close cousins. "Present bias" is the general, descriptive name for over-weighting the immediate option. Hyperbolic discounting, and its quasi-hyperbolic approximation, are specific mathematical descriptions of that bias. In everyday use the terms are often used interchangeably, but a formal paper will usually pick one specific curve.
The distinguishing feature is the reversal when both options shift equally, not merely that someone prefers "now" in general.
Steepness varies by person, by domain (money versus health versus the environment), and even by mood or scarcity at the moment of choosing.
A precommitment set up after temptation is already visible rarely holds. It works because it is built while the reward still feels distant.
This model concerns delay specifically, not uncertainty. Conflating the two leads to the wrong fix for the wrong problem.
Naming the bias describes what happened. It does not make the smaller-sooner option the right one to take.
People often discount small amounts more steeply than large ones, so a finding at one stake size may not transfer to another.
the decision is a straightforward financial comparison, such as loan APRs, where a single steady discount rate is genuinely the right tool.
a deadline is externally real — limited inventory, a contractual cutoff — rather than a psychological pull created purely by nearness.
the "immediate reward" involves substance use or compulsive behavior. This model explains part of the pattern; professional treatment frameworks go further than it can.
This browser tool structures your thinking. It does not send your information anywhere or perform AI analysis.
Save as Skill converts the model into portable Markdown instructions that teach an AI when and how to apply Hyperbolic Discounting, what questions to ask, what evidence to seek, what mistakes to avoid and how to structure its analysis.
Save the skills you find useful and organize them into your own reusable library. The Markdown is intentionally model-neutral, so it can be used wherever an AI system accepts uploaded files, project knowledge, custom instructions or reusable text context. Platform capabilities differ, so this is a portable skill file rather than a claim of one-click installation.
The classic form values a reward V received after delay D as V divided by (1 plus k times D), where k sets how steeply the curve drops. Economists more often use the simpler quasi-hyperbolic approximation for modeling, since it needs only two parameters and is easier to fit to data.
Knowing about the bias does not remove it in the moment; the near-term pull is still there when temptation arrives. A commitment device works by removing the choice itself, or raising its cost, while it is still being decided from a calm distance.
Often, yes. Experiments repeatedly find steeper discounting for small amounts than for large ones, sometimes called the "magnitude effect." A finding calibrated on a small reward should not be assumed to transfer directly to a much larger one.
No. The model describes a pattern in how people weigh time, not a rule that later is always better. A later reward can carry real risk, uncertainty, or opportunity cost that a rational decision should still weigh against waiting.