Last year, the Journal of Personality and Social Psychology raised quite a few eyebrows when it published a study that supported precognition — the idea that people know, or can predict, events before they happen. The same journal is once again causing as much of a stir, now refused to publish other studies that repeat the experiments outline in original study.
Social psychologist Daryl Bem started the whole thing with his paper on precognition. In it, he adapted several standard psychological tests and ran them in reverse. From the New Scientist:
In Bem’s back-to-front version, participants were shown a list of words and then asked to recall words from it. Later Bem showed them words randomly selected from the same list, and it turned out that they had been better at recalling these words in the prior test.
After Bem’s publication, many other scientists sought to repeat his findings, as is the standard process why which a scientific conjecture gains a body of supporting evidence. But the response from the Journal of Personality and Social Psychology has been less than welcoming.
One such paper, from the University of Indiana, replicated Bem’s experiments and found no evidence of precognition. When they went to submit it, they were told by editor Eliot Smith that “this journal does not publish replication studies, whether successful or unsuccessful.” Indeed, the journal has rejected other replications of Bem’s findings, one of which supported them. Smith’s concerns are perhaps understandable, as many would likely rush to prove or disprove such an attention-grabbing claim. “We don’t want to be the Journal of Bem Replication,” he is quoted as saying.
In their coverage of this feather-ruffling the New Scientist quotes Richard Wiseman, who has a very different view of how journals like the Personality and Social Psychology should operate. His believes that these policies make it unclear if the results published are truly genuine. In his statements, he calls for scientific journals to carry at least the abstracts of replication studies, with the full text online. This could at least remind those reading the journals of the ongoing dialogue surrounding a particular study, and perhaps serve as a rundown of how a scientific notion is fairing.
While many journals refuse to publish replication studies, this issue underlines the fact that journals are repositories of current research and not established scientific fact. Often in the media (and sometimes, on this website as well, though we try to put such work in context) if it appears in a journal it’s good enough to be disseminated. Of course, this is not how science really works. Having an article peer-reviewed and accepted for publication by a relevant journal is certainly a major step for any research, it is by no means the end. As mentioned before,  replication of experiments with consistant results is what “proves” a theory.
In truth, no publication is morally obligated to provide each and every step in the life of a story or a study. News outlets do not provide endless follow-ups to their stories, instead waiting for either large developments or total refutations of their previous reporting. Likewise, journals have been known to throw out studies if later evidence arises that the work was somehow tainted. Science, like stories, is often a discourse between many people all over the world. This furor and confusion surrounding the decision of one journal is a careful reminder that what is printed is often only a part of what has happened.
(via The New Scientist)
