Ask someone what they think of your photograph and you’ll get a sentence. Watch where they look instead and you’ll get the truth.
In 2015 Canon sat three people in front of a single print and tracked their eyes. One was a non photographer. One was a photography student. The third was Joel Grimes, the professional who made the picture. It was an advertisement for a printer, and it accidentally produced the most useful thing anyone has published about looking at your own work.
The non photographer settled on about ten points and made 212 eye movements. The student made roughly twice as many. The photographer covered almost every spot on the print, his eyes bouncing around five and a half times more than the first viewer and two and a half times more than the second. You can see the numbers and the heat maps here.
That headline finding is the least interesting part of it. Of course the photographer looked hardest. He made the thing, he was being filmed, and he had every reason to perform his own attentiveness.
The part nobody quotes
Buried in the results is a detail that should bother every one of us. The student spent almost no time on the two areas where the non photographer looked the most.
Read that again, because it isn’t a story about photographers looking more carefully than other people. It’s a story about photographers looking somewhere else entirely. The student hadn’t just added attention, he’d moved it. A year or two of learning had already pulled his eyes away from the places an ordinary viewer goes first, and toward the places that only matter if you know how pictures get made.
If that’s true after a couple of years, consider what’s happened to you after ten. You’re not a more attentive version of your audience. You’re looking at a different photograph than they are.
Why asking doesn’t work
This is the mechanism underneath something we’ve written about before, the gap between the photograph you love and the one everybody else picks. The usual explanation is information. You know what it took to get there, the cold, the waiting, the person who nearly said no. The viewer has none of that and only has the image.
The eye tracking adds something harder. It isn’t only that they know less than you. It’s that their eyes physically go somewhere else, in the first second, before anyone has formed an opinion to report.
Which is why asking people what they think is close to useless. By the time someone says “I like the mood” they’ve already looked, decided, and translated all of it into something polite enough to say out loud. You get the summary, after the editing, with the awkward parts removed. You don’t get the four seconds that actually happened.
Watch the face, not the photograph
Here’s the practice, and it costs nothing. Show someone a print. Say nothing at all. Then watch their face instead of the picture.
You’re looking for three things. Where their eyes land first, which tells you what the photograph is actually about rather than what you intended it to be about. Whether they travel anywhere afterwards, because a picture that gets one look and no journey is a picture with one idea in it. And how long before they look away, which is the only honest measure of whether it holds.
None of that can be faked and most of it can’t even be articulated. Ask afterwards if you want, but you’ll already know. The gap between what their eyes did and what their mouth says is itself worth noticing.
It works better with a print than a screen, partly because you can see their face while they hold it, and partly because people scroll past a screen without ever deciding to.
When there’s no stranger available
Most of the time you’re editing alone at night, and the whole problem is that you’ve looked at the frame so often you can no longer see it. Everything below is a way of borrowing a stranger’s eyes when no stranger is in the room.
Flip it horizontally. The cheapest and most effective one. Your memory auto completes a composition you’ve stared at, and mirroring it breaks that instantly. Imbalance you’d stopped noticing becomes obvious, and so does the fact that the eye has nowhere to go.
Make it small. Shrink it to a thumbnail, or prop it up and walk to the far side of the room. At that size the detail you spent an hour on disappears completely and only the structure survives, which is roughly what a viewer gets in their first second. If it doesn’t hold in two tones from across a room, no amount of local work is going to save it.
Leave it a week. Time is the only thing that reliably restores your ignorance. The version you were certain about on Sunday night is frequently a different picture by the following Saturday.
Watch your own first second. Open a folder of your work at thumbnail size and notice which frame your eye goes to before you’ve thought about any of them. That reaction is the closest thing you have to an outside opinion, and it arrives before the memories attach themselves.
This is not an argument against care
It would be easy to take all this as permission to stop bothering. Nobody notices the dust spot, so why clean it. That’s the wrong conclusion and it’s worth saying so directly.
Craft still matters, and the accumulation of small decisions is most of what separates a photograph that works from one that nearly works. The argument isn’t against care. It’s against spending care in the wrong place. An hour on a corner nobody’s eye will ever reach is an hour not spent on the thing their eye hits first and leaves unsatisfied.
We’ve argued before that technical precision gets mistaken for quality, and this is the practical face of that. Perfectionism isn’t a virtue or a vice on its own. It’s a budget, and most of us are spending it where the audience isn’t looking.
What it’s worth
Three people, one print, no controls, and an advertisement for a printer. It isn’t science and it shouldn’t be quoted as though it were. Someone pointed out at the time that an involuntary eye movement condition would have wrecked the numbers, and nobody has said whether the three viewings were even the same length.
It doesn’t need to be science. It only needs to be a demonstration of something you can verify yourself this weekend with one print and one friend who hasn’t seen it. Hand it over, say nothing, and watch their eyes.
You’ll find out in about four seconds whether the picture is doing what you think it’s doing. That’s faster than any critique, and considerably more honest.
Featured photograph by Victor Freitas on Unsplash, converted to black and white.