Behavioral Psychology

Intermittent Reinforcement and Habit Formation

How unpredictable rewards build persistent habits in apps, games and gambling, why such habits resist stopping, and ways to loosen their grip.

On this page
  1. A principle from learning research
  2. Why unpredictability sticks
  3. Schedules of reward
  4. It is not only about gambling
  5. How habits form
  6. What this does not mean
  7. Loosening the grip

A principle from learning research

Slot machines and similar games are games of chance with a built-in house edge. No strategy, system or timing changes the odds, most players lose money over time, and nobody should gamble money they cannot afford to lose. This article is about one reason that behaviors tied to chance can become habits that are hard to put down.

Psychologists who study learning have long distinguished between ways of delivering rewards. With continuous reinforcement, every action is rewarded. With intermittent (or partial) reinforcement, only some actions are rewarded, and the pattern can be predictable or unpredictable. A consistent finding in animal and human research is that behavior learned under unpredictable, intermittent rewards tends to be more persistent when the rewards stop than behavior learned under constant rewards.

Why unpredictability sticks

Consider two vending machines. One always gives a snack when you insert a coin. The other gives a snack only occasionally and at random. If both stop working, you will probably give up on the first one quickly, because the pattern has plainly changed. With the second, a dry spell looks like business as usual, since it was never clear when the next reward would come. You might try many more times.

Another factor is anticipation. Brain-imaging research is often summarized as showing that the expectation of a reward, especially an uncertain one, engages the brain’s reward system strongly, sometimes more than the reward itself. The “maybe this time” feeling is part of what keeps people engaged. This description is a simplification, but the broad idea that uncertainty can heighten engagement is well established.

Schedules of reward

Researchers describe several schedule types:

  • Fixed ratio: a reward after a set number of actions, such as a coffee card stamped ten times.
  • Fixed interval: a reward after a set amount of time, such as a paycheck.
  • Variable ratio: a reward after an unpredictable number of actions.
  • Variable interval: a reward after an unpredictable amount of time, such as occasionally receiving a message.

Variable-ratio schedules are generally associated with high, steady rates of behavior and strong resistance to stopping. Games of chance, in which each play has some probability of paying out, resemble this pattern. The specific design of slot payouts and their variable reward schedules is covered elsewhere on this site. Here the point is that the pattern is general.

It is not only about gambling

The same structure appears in many everyday technologies:

  • Social feeds. Pulling to refresh may or may not show something interesting. The unpredictability makes checking feel worth repeating.
  • Email and messaging. Most messages are routine, but occasionally one is important or pleasant, which keeps people looking.
  • Mobile games. Random rewards, daily bonuses and surprise drops encourage frequent return visits.
  • Loot boxes and gacha systems. Randomized item rewards in games follow a similar logic, which is part of why they draw regulatory attention.
  • Hobbies like fishing or hunting for bargains. Part of the pull is that results vary.

Most of these are harmless or mildly annoying. The same principle becomes more serious when each repetition costs real money.

How habits form

A habit typically has three parts: a cue (a time, place or feeling), a routine (the action) and a reward. With repetition, the link between cue and routine strengthens, and the action starts to happen with less conscious thought. A person might reach for their phone or head to a familiar spot when bored, tired or stressed, without actively deciding.

Add unpredictable rewards, and the loop becomes more resilient. The cue triggers the action, the action offers a chance at reward, and the occasional reward reinforces the whole sequence. Because losses occur between rewards, the experience of not winning does not signal that the habit should end. It becomes part of the rhythm.

What this does not mean

Intermittent reinforcement does not mean everyone who plays will develop a problem, and it does not describe a single cause of problem gambling. Habit formation interacts with personality, circumstances, stress and access. The topic of gambling addiction involves many factors beyond reward schedules. The principle explains why some behaviors are sticky, not who will be affected.

It also does not suggest that persistence is a sign the activity is worth continuing. A behavior can be persistent and still not good for the person doing it.

Loosening the grip

  • Notice the cue. Identify what triggers the urge: boredom, a time of day, a notification, or a feeling.
  • Change the environment. Turn off notifications, remove shortcuts and put friction between you and the action.
  • Replace the routine. If a cue triggers the habit, plan a different response, such as a walk or a call to a friend.
  • Limit exposure. Use time and deposit limits, and take regular breaks.
  • Track frequency, not only outcomes. Counting how often you play can reveal a habit that outcomes alone hide.
  • Be patient. Habits take time to fade, and urges tend to come in waves that pass.

If you feel unable to stop, or if gambling is affecting your finances, relationships or mood, a national problem-gambling helpline or local health service can offer confidential support. Gambling is age-restricted or illegal in many places, and rules vary by country.

The complete guide to behavioral psychology

Behavioral Psychology of Risk and Reward
Read the guide

Keep exploring

Guides from other topics

All topics →