Before 1961, the dominant account of how people learn came from behaviourism: you learn a behaviour because you were personally rewarded or punished for it. No direct reinforcement, no learning. Albert Bandura's Bobo doll experiment, run at Stanford that year, set out to test whether that was actually the whole story — whether watching someone else's behaviour, with no reward or punishment involved for the child at all, was enough on its own to teach something new.
Watch an adult hit a doll, then hand a child the same doll
Preschool-aged children individually watched an adult model interact with a large inflatable "Bobo doll" — either aggressively, punching it, kicking it, striking it with a mallet while shouting specific phrases, or calmly, ignoring the doll in favour of other toys in the room. A third group saw no model at all. Afterwards, each child was mildly frustrated — shown appealing toys, then told they couldn't play with them — and led into a room containing the same Bobo doll along with other toys, and observed through a one-way mirror.
Kids didn't just get aggressive. They copied specific moves and specific words
Children who'd watched the aggressive model were significantly more aggressive toward the doll than either other group — and, tellingly, the aggression wasn't generic. They reproduced the model's specific actions and specific phrases, the ones they'd watched minutes earlier, almost verbatim. Nobody rewarded them for it. They'd simply seen it done. Bandura called this observational learning, part of what became social learning theory: the claim that behaviour can be acquired purely by watching another person, without any direct reinforcement to the observer at all.
What we're still unsure about
The study has drawn lasting methodological criticism alongside its influence. The doll itself was built to be struck — weighted to bounce back upright — which makes it a genuinely ambiguous object: some of what looked like acquired hostility may partly reflect what the doll's shape invited, rather than aggression in a fuller sense. The lab setting — a frustrating task, a strange room, unfamiliar adults filming — is also highly artificial next to how children actually pick up behaviour from people they know over months and years. Later research into whether observed violence, particularly on screens, produces comparable real-world aggression has been far messier and more contested than Bandura's original, clean result, and that debate hasn't settled.
This sits inside Observational Learning (Bandura), one of seven topics in Behavioral Psychology, one of four domains in Psychology, one of seventeen subjects the app can quiz you on.