Posted by: structureofnews | August 27, 2013

The Persistence of Data

3salvador-dali-persistence-of-memoryI’ve just finished reading a really interesting book on memory.  If only I could remember the key points.

Which is not that strange, actuallyPieces of Light: How the New Science of Memory Illuminates the Stories We Tell About Our Pasts, by Charles Fernyhough, is really about how fragile the process of remembering is.

We like to think of our memory as some kind of mental DVR, recording key events that we can – with greater and lesser success – play back in whole or part.  But in fact, memory is a much more fragile and mutable thing – recollections are created (and changed) each time we call on them; they’re affected by our state of mind at the time of the event, and on our state of mind when we try to remember them.  And they’re notoriously inaccurate.

Which should tell us something about one of the core processes of journalism – asking people about things they remember – and about the importance of records and data.

As Fernyhough notes, we’re wired to remember not details but meaning.  What we recall aren’t the exact words of a speech we’ve heard, but what we think it means.  That makes sense: We’d be overloaded with information otherwise, and this lets us sift through a ton of experiences effectively.

In other words, our memories mostly fulfill the tasks they are charged with, skipping the details and drilling down to the real, useful meaning of the information we are trying to store.  We remember what we need to remember, and forget the rest.

But meaning is colored by our bias, mood, hindsight, and a host of other factors, and those all play into how we often create new, usually vivid, but false, details into what we remember.

Other memory errors, such as bias (allowing your memories to be colored by your present-day attitudes) and suggestibility (claiming memories for events you never experienced), reflect the operation of a combinational system that can stick together information from many different sources in re-creating an event.

In fact, he notes, it’s actually not that hard to give people memories of events they never had – one experiment involved showing manipulated images of childhood photos to subjects, who subsequently “remembered” the events that were shown.

Other, more obvious, biases – like wanting to look good – kick in as well.   Joseph T. Hallinan, in Why We Make Mistakes: How We Look Without Seeing, Forget Things in Seconds, and Are All Pretty Sure We Are Way Above Average – another good book on human frailty – cited experiments that show how people recall doing better in school than they actually did, or being more accurate in predictions than they actually were.

There’s lots more in both books – go read them – but to be honest, I don’t remember the details.  (Just as I didn’t recall touching on this topic in an earlier post while writing this one… but I did.)

What I do recall – the meaning I take away from the books – is that we as journalists need to much more careful about accepting people’s recollections, even when vivid and full of detail.  (The classic example is John Dean’s incredibly detailed – and incorrect – testimony about conversations with Richard Nixon in the Oval Office, which could be later checked against Nixon’s secret recordings.) There’s a reason there’s a two-source standard – it’s much harder for two people to misremember the same details.

But it’s also why we should be looking as far as possible for documents and data to back up our findings.  Those are flawed as well – they could be inaccurate records of events, and even where accurate, often have some kind of bias built into them.  But they have one advantage over memory – at least those flaws are fixed in time, and don’t change each time we consult them.

And as the ubiquity of recording devices grows – cellphone cameras, Google Glass, and so on – we may well have access to even more records taken at the time of an event.  How that will change our behavior is an open question, as a recent piece in the NYT notes.  But we’ll be relying less and less on memory and more and more on memory chips when it happens.


  1. […] an aside, this desire for narrative also explains why memory is so fallible – at least at remembering facts accurately; its real purpose may be to help us construct and inhabit a carefully edited life story that makes […]

  2. […] the mindbug that virtuoso = male becomes. We can depend less on human testimony – which is often flawed anyway – and expand out to look at documents and data – also flawed in their own way, but at […]

  3. […] is tricky for journalists, given how much we rely on people telling us what they remember of events.  And so the more there are contemporaneous reports of any event – other eyewitnesses, […]

  4. […] weaker the mindbug that virtuoso = male becomes. We can depend less on human testimony – which is often flawed anyway – and expand out to look at documents and data – also flawed in their own way, but at least in […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: