How One Man Beat The Mackworth Clock

The Mackworth Clock is a famous test that was used to assess the vigilance of radar operators during World War II. Can you notice when a clock jumps one second ahead? but one man "hacked" the test, getting an unheard-of score. Years later, he explained his hack — and brought up a major flaw in most psychological studies.

Top image: Helder Mira.

The Mackworth Clock

The Mackworth Clock was design to occasionally click forward two seconds instead of one, at the rate of about 12 per half-hour. Whenever the person watching the clock saw the double-click, they would press a button, indicating they had seen the anomaly. Clock-watchers usually got two-hour shifts, during which they did nothing but watch seconds tick by. The experiment was meant to see how attention falters over a prolonged period of time.

How One Man Beat The Mackworth Clock

The Mackworth clock is still used to test attention, though through the years, scientists have put different spins on it. One famous variation happened in 1952, when participants were asked to take the clock test during consecutive nights of sleep deprivation. The study wasn't remarkable for itself, nor did it turn up very remarkable overall results. It was made famous because of its one anomalous result — Brian Shackel. Shackel's results didn't get any worse, even after days of sleeplessness.

It was years later that Shackel wrote about why his attention never wavered. He hadn't really been paying attention in the first place. On the first night, he started pacing the room, and counting the clicks of the clock. Once, after counting out 25 clicks, he noticed that the clock had gone an extra space forward. He hit the button. Since the test gave the subject a 15-second grace period to hit their button, most of the time he pressed the button in time to correctly indicate a double click. Although keeping up the counting was difficult, it wasn't as difficult as silently sitting and watching a second hand click off tiny increments on a clock.

Our Best Performance

Shackel wrote about his experience because as a psychologist, he realized how odd it was that nobody asked him about the technique he'd used to get such outstanding results. He wrote about his hacking the Mackworth clock as a way to explain how important follow-up interviews were in psychological studies. He also wrote, "All subjects think they are being personally evaluated, so will find any way to give their best performance."

Shackel wrote that to stress the need for investigators to watch the tests as subjects undergo them (although I imagine watching some guy watch a clock may be even more boring than watching the clock itself), but there is something more in this. One of the greatest stumbling blocks to psychology is the human ego. Everyone takes the test personally, and no one wants to be on the left hand of the curve. (If you're very unlucky, you'll get a Stanford Prison experiment or a Milgram experiment, and represent humanity's capacity for evil for all time.) Everyone tries to be "right."

This makes scientists go to extraordinary lengths to make sure people don't know what the "right" way to behave is. In some experiments, like the one mentioned above, this itself is the problem. The experiment's most outstanding test results came from someone who wasn't even participating in the experiment. Less dramatically, it pushes subjects into a paranoid state that might also skew the results.

I have only participated in one experiment. It involved a photo of me, and a math test. According to the interview afterwards, I did well on the test. I never got to look at the photo. All I remember was being hyper-aware that I was being evaluated. I've rarely behaved less naturally. (If anyone reading this has participated in an experiment, be sure to mention what you had to do, and whether you felt you behaved normally while doing it.)

It seems to me that, in experimental situations, scientists are in a double bind. Give people a goal to shoot for, and they'll shoot for it, skewing the results. But in some situations, like the Mackworth Clock test, perhaps it's more productive to tell subjects the goal of the test (in this case, to measure the real effects of sleep deprivation on detailed attention), instead of letting them make up a goal as they go along.

[Via How I Broke the Mackworth Clock Test (And What I Learned), The Mackworth Clock Test: A Computerized Version.]