Forgetting is not a failure of memory — it is, in many cognitive systems, an active process of selection. The psychologist Frederic Bartlett showed in his 1932 experiments with repeated storytelling that people don't retrieve memories so much as reconstruct them, filling gaps with whatever schema fits the situation. What he found — and what rarely makes it into business or technology conversations — is that the errors weren't random. People systematically replaced unfamiliar details with familiar ones, smoothing the strange into the recognizable. The distortion had a direction. This matters enormously for how humans update their understanding of other people. When you form a view of a colleague, a customer, or a competitor, subsequent information doesn't land on a blank surface — it gets absorbed into the existing story, quietly reshaped to fit. The implication isn't that memory is unreliable in a way that should make you anxious. It's that your current model of another person is probably more coherent than the actual person — and that coherence is the distortion. The corrective Bartlett implies is deliberate: when you revise your view of someone, notice whether you're genuinely updating or just smoothing the new data into the old shape.

Think of someone whose behaviour recently surprised you. Did you revise your model of them — or find a way to make the surprise fit the model you already had?

Drawing from Cognitive psychology / Experimental psychology — Frederic Bartlett