An older person says with confidence that the word "sunflower" was on the list, even though only "almond" was there. Both words are tied to oil, and that semantic thread was enough for the brain to fill in the memory. Memory works like a compass: one needle points to meaning, the other to the order of events. With age, one of the needles starts to outweigh the other.
In a study of 90 people aged 18 to 85, researchers examined lists of 16 words. Each list contained groups of words linked by meaning, for example "actor," "film," "screen," "cinema." These groups appeared either in a row or separated by four other words. That way the scientists teased apart two axes: semantic closeness and temporal closeness. After studying the lists, participants decided which words they had actually seen and which only felt familiar.
The difference showed up when participants recognized real words. Young people remembered better those words that appeared next to each other and were linked by meaning. Older adults, by contrast, remembered well only those that were linked by meaning, and order did not help them. So the young brain uses both needles of the compass, while the older brain increasingly consults only the semantic one.
This does not mean that older adults' memory is simply weaker. Sensitivity to the differences between old and new words was the same in everyone. What changed was the strategy: instead of the order of events, the brain began to lean on their meaning. This shift explains why a grandmother mixes up what happened yesterday and what happened a year ago, yet remembers perfectly well that sunflower oil and almond oil are one and the same thing.
Previously it was unclear why older adults more often confuse events that are similar in meaning: because of a general weakening of memory or because of a change in strategy. Now it is clear that false memories depend on meaning equally in everyone, while the reliance on order disappears in older adults. This finding matters for memory training: if older adults are taught to pay attention to details and sequence, false memories may become less frequent. And in the clinic, such false recognitions are already used as an early marker of memory disorders.
Participants scoring below 26 on the Montreal Cognitive Assessment were excluded, so the sample was cognitively healthy. Even in such people, false memories reached 30% under some conditions.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.