Hacker News Reader

Comment by feanaro | Hacker News Reader

You can prove the limitations of your memory quite easily to yourself -- but don't cheat!

Find a so-called DRM word list (acronym originating from the researchers' names Deese, Roediger and McDermott) and have someone test you: They present you with the word list for a predetermined amount of time during which you attempt to memorize it. Then they ask you to recall which words were present. Along with recalling (some of the) actual words, you are in addition likely to recall a so-called critical word which was not actually present in the list.

A collection of such word lists can be found at https://www3.nd.edu/~memory/OLD/Materials/DRM.pdf

View on HN

I actually did one of these in college (like 20 years ago), and got it perfect. Not sure if it was a DRM word list specifically, since it was just a psych major room mate testing me, but it seemed very similar.

But even if I didn't, that's not really the kind of thing I would ever call upon my memory for. Usually it is something like (real, recent examples from my life): "Who said what, when, how, and to whom", "who was here and who left first", "did this scene happen before or after this other scene in the movie", "who threw the first punch", etc.

Well, congratulations, you're the odd one out then. It's certainly not logically impossible for a particular human to be less susceptible to false memory than others. It does nothing to dispel the fact that human memory is still fundamentally fallible in the general case. (And how could it not be? It is after all just a lossy encoding of a much larger amount of physical information.)

> But even if I didn't, that's not really the kind of thing I would ever call upon my memory for.

But it's the same type of thing! If a DRM list produces false memories, why should these other situations not? You don't even have to reach for the DRM paradigm; it's well documented that witness reports in courts (i.e. exactly the kind of thing you mention) are unreliable.

I've done that test in a real study, I scored perfectly. As long as you have reasonable understanding of the limitations of your own memory then you wont over or under estimate the accuracy of it, this isn't particularly hard to do for some people.

Just because many people makes that mistake doesn't mean that such mistakes are common in everyone.

See my [other response](https://news.ycombinator.com/item?id=31913649). The claim is certainly you cannot get a DRM-style test "right". Just because someone somewhere gets it right, that doesn't mean you can rely on your memory to be infallible.

Reply on news.ycombinator.com