Variable ratio reinforcement
If a machine paid out every ten spins, you would stop as soon as you got it. If it never paid, you would stop too. What hooks you is not knowing when.
This pattern is called variable ratio reinforcement and it is by far the most resistant to extinction of any known in the psychology of learning. It is the same mechanism that keeps you checking your phone: unpredictable reward sustains behaviour long after the reward stops arriving.
Slot machines are the purest application of this principle in existence. That is not an accident: it is the result of decades of optimisation.
Dopamine arrives before the prize
Intuition says the rush comes from winning. Measurement says otherwise: the dopamine peak occurs during anticipation, while the reels spin and the outcome is still unknown.
This has an uncomfortable consequence. If what reinforces is the wait rather than the result, then losing also reinforces, provided the wait was intense enough. That is why you can leave a losing session wanting to come back.
Near misses — two matching symbols and the third one short — activate circuits similar to winning, despite being losses. Machines generate them more often than chance alone would.
The environment is designed too
No clocks, no windows, constant lighting, carpet that absorbs sound and exits that are hard to spot: the design of a gambling venue works to remove your sense of time.
Online, the equivalents are notifications, bonuses with expiry dates, one-tap deposits and the total absence of friction. Every element removed from the process is a moment at which you could have stopped.
The takeaway: None of this means you cannot get out. It means getting out is not a matter of resolving harder, but of changing the conditions. That is what the following steps are for.