For all the times I curse the Internet, I also see its value. I recently heard a report on National Public Radio that interested me, but the next day I couldn’t remember any detail. So I went on NPR’s website, re-heard the broadcast and downloaded a transcript.

The broadcast I couldn’t remember was about our faulty memories.

This wasn’t some run-of-the-mill story about how poorly we remember past events. (Not much news there.) Rather, NPR reported on a study that showed how our memories are so bad that at times we clearly remember events that — get this — never happened. That’s right, we become convinced that something happened that really didn’t, and not because we’ve been repeatedly told of it through some organized disinformation campaign. With very little prompting, we simply remember it that way.

In the study, subjects viewed altered photographs that suggested the occurrence of an event that in truth never occurred. For instance, in one photograph, former President George W. Bush was shown to be vacationing with Roger Clemens, while an accompanying caption noted that the vacation took place at the same time that Hurricane Katrina hit the Louisiana coast. In another “never really happened” photo, President Obama was shown shaking hands with the Iranian president during a visit to the United Nations.

A remarkably high number of people –- one out of every two — clearly remembered such events happening. And one of out of four people could provide specific details about where and when they had seen these photographs before. (Remember, these photos never existed prior to this study — they were photoshopped for this purpose.)

The study focused on political figures, and the NPR story went on to describe how our political biases reinforce our fantasy memories. Conservatives were more inclined to remember, as factual, a false image that puts Obama in a bad light, while liberals were more inclined to remember a false and negative image about Bush.

I find that ideological dimension fascinating. But I’m more interested in the idea underlying this work: that we are capable of feeling so strongly that something happened when it never did.

I don’t find this idea startling at all. It seems to fit with my own experiences. I’ve grown to think of memory differently from the popular view as some kind of unalterable video recording of the past. It seems to me that at least some memories change over time. I’m not speaking about the change that comes from memories fading. Rather, I’m talking about memories that remain clear, but now differ in their details from a few years before.

I sense it most when talking to old friends or family members, with whom I’ve had conversations over many years about something from the distant past. For instance, I may notice when I talk with one of my sisters about an event from my youth, that she talks about that event differently this year from how I remember her talking about it 10 years ago. Now perhaps I’m simply not remembering the conversation from 10 years ago correctly. That’s possible in some instances — but unlikely in all. It’s more likely — at least at times — that my sister’s memory of an event has changed.

I have even noticed how my own memory has changed. I recently read through a number of my old letters and diary entries, some of which recorded prior events as I remembered them on that day. In more than one instance, my past memory conflicted with my current memory. Let me be clear: This is not a case of me forgetting an event, or of me remembering an event differently than someone else. Rather, these are instances of me remembering an event differently than myself: my own memory of a past event had changed over time.

There’s a researcher at McGill University — Karim Nadir — who posits that the very act of remembering a past event changes the underlying memory of that event. In fact, his research suggests that the more we think of a past event, the more likely our memory of it changes. Ironically, the things we feel we remember most clearly may be what we remember most inaccurately.

The NPR report raised a somewhat similar point: the folks most likely to think an event really happened when it didn’t are those who pride themselves on staying abreast of public affairs.

The mind is a funny thing.

To reinforce that point, I’ll share with you what has been running through my mind since first hearing that NPR report. Yes, I’ve been thinking about our faulty memories — as my writing above attests. But my thoughts keep coming back to the power of images, even fantasy images. I’m not now talking about altered photos of Obama and Bush, but altered images of a different kind — the kind that form in your brain. I’ve been thinking about Karen Wood and Donald Rogerson.

You may remember the tragic story. In 1988, Hermon resident Karen Wood was accidentally shot by Donald Rogerson. Wood, upon hearing gun shots in the neighborhood, put on her coat and white mittens, and stepped into her backyard to warn hunters to stay clear. Rogerson — by all accounts an experienced and careful hunter — fired at Wood and killed her.

Rogerson swore he saw a deer when he aimed his rifle. I suspect he did.

Since grade school we’ve been taught that the eye works like a camera, and certainly there are similarities. But our eyes do not send photos to our brain. Rather, our eyes send just bits and pieces of data that our brain then forms into an image. Our brain completes the picture. When data from the eyes is sparse, the brain may rely as much on what we expect to see as what is in front of us at that moment. But to the viewer, it all looks the same. There is no way to know when the brain is filling in a little or a lot, whether the image is computer-enhanced or computer-generated.

Consider what happened behind Karen Wood’s house that day. Rogerson saw signs of deer all around him — a puddle cloud, a hoof print. He expected a deer, then saw one, then shot. There may or may not have been a deer in front of him at that moment; but regardless, he saw it.

We can’t fully trust our eyes. And we can’t fully trust our memories. Where does this leave us?

The NPR report concluded that we all need to show more humility, that we should never be so adamant that we are right. I’d submit that we also need to show more caution — in so much of what we do. Aiming a gun is just one small example.

Why are humility and caution so important? Because we may well be wrong — even with what we once thought of as rock solid fact.

John Piotti of Unity is executive director of Maine Farmland Trust and this newspaper’s newest columnist. “Cedar and Pearl” appears every other week.