For the record, I got it first try without cheating. I don't know how proud I am of that.
That was a perfect example of worthless human-science research which gets published but which adds exactly zilch to our knowledge of things that matter. It does the usual trick: it doesn't try to understand anything, all it does is try to trick the subjects into being wrong. Furthermore, it's definition of "being wrong" is pretty absurd.
It's been well documented forever that recall memory (being able to come up with the logo) is way worse than recognition memory (knowing it when you see it). So, this is kinda like an experiment to see gravity is real by dropping an apple on somebody's head. Now, I suppose they would claim that by comparing correctness with confidence, they've added to our store of knowledge... but exactly what did they add that is applicable to anything other than old Apple logo's? Pretty much nothing.
The outcomes could have easily been manipulated to achieve different results by playing with (a) the judging criteria on the recall test and (b ) the degree of similarity between the choices on the recognition test. For the recall test, they show various drawn images, nearly all of which show the main feature of the old logo (an apple with a bite out of it)... yet, by their criteria, nearly all of those were deemed to be incorrect.
When you read the study, they explain what they did, but they don't in any way justify why they did it. For example, why should somebody who gets 10 of the 14 features of the logo correct be judged to have not remembered it? Sounds to me like they remembered it pretty damn well... but only those who got 11 or more of the 14 features right were judged to have remembered it. And those 14 criteria included whether or not they had a vein in the little leaf at the top. Why, what difference does the vein the leaf make?
This is the kind of crap that gives Psych research a bad name among other scientists. IMO, the people who did it are just pretend scientists, and all they did was find stupid ways to deploy the "scientific method" to prove that subjects can be wrong, regardless of whether it matters or not. They didn't do any real science, all they did was misuse the scientific method to boost their own publication list.