​Science And Technology

Truth in Science Publishing: A Personal Perspective

 

Scientists, officials and patient representatives alike questioned more and more the validity of published scientific results, endangering the public's acceptance of science. Here I argue that the emerging flaws in the integrity of the peer review system are largely responsible. Peer-reviewed distortions are driven by economic forces and activated by a lack of accountability of magazines, publishers and authors. An approach to restore confidence in the validity of published results would be to establish ground rules that make it more transparent by peers, such as the publication of notices (a practice already adopted by certain magazines) and non- Only the record of authors, but also publishers and magazines.

 

As scientists, we are part of society. So, you can expect that questions about the reproducibility of published studies, which are, in fact, questions about the veracity of these studies, are a fair game. After all, a lot of science is funded publicly, and the public has the right to ask themselves if their funds are well spent. Moreover, at a time when many of the opinions held and publicly expressed are clearly false (although it is not always clear whether those who express these views are aware that they are lies), Is important for a citizenship engaged in seeking evidence in support of the claims. For too long, elected representatives in the Member States have not been questioned when they say, for example, that there is no global warming, that tax cuts increase tax revenues, Government-run health insurance is less effective than performing private health insurance, although in all cases the evidence indicates otherwise.

 

But scientists, unlike politicians, should be guided by the facts. Scientists must be isolated from the disregard of patents of facts in public discourse. Why, then, are a growing number of scientific studies increasingly suspect? In my opinion, we as scientists must accept at least some of the blame for failing to ensure that studies report facts and not imaginary. The growing problem is not here because of the outright fraud, which continues to be rare. Instead, the problem is due to distorted interpretations of the experimental results. These interpretations lead to exaggerated statements of fact (findings also known as), which are often distantly related to actual data in a document.

 

As a concrete example, the last decade has seen hundreds of neurosciences in which the behavior of an animal is analyzed after maps of defined populations of neurons were excited or inhibited with optogenetic methods. These documents provide valuable information, but commonly conclude that manipulated neurons play the physiologically studied behaviors that change during manipulations. However, these manipulations induce massive changes in neuronal activity that do not replicate the normal functioning of affected neurons. In addition, major changes in activities that are induced spread throughout the brain, and thus probably cause a myriad of downstream effects. As a result, the results of these experiments on the normal functions of manipulated neurons are difficult to maintain without one of the independent additional tests [3]. Indeed, sometimes the same population ablation of neurons whose optical manipulation produces important effects does not have behavioral consequences [4].

 

Two checkpoints are intended to preserve scientific truth and avoid unwarranted conclusions: peer review and reproducibility. Peer review examines the validity of the experiments and conclusions presented, and reproducibility ensures that experimental results and conclusions can be replicated. The two checkpoints are threatened mainly due to economic factors.

​

Compromise peer review

 

Peer review of scientific results occurs at several levels, including the review of scientific manuscripts, is probably the most important because the published works provide the benchmark for all other types of peer review Selection of grants, promotions, invitations and rewards). Given the large amount of data produced, magazines play an important role in identifying results and conclusions that merit attention, no one can read all the documents in a field! In recent times, the role of magazines in selection studies has become omnipotent. In some magazines, the publishers who manage the selection process have become similar to the high priests and priestesses of science, and here a puff of ancient Egypt with the Pharaohs who control access to wisdom can permeate the process of " examination.

 

As a result, three problems emerged in the peer review that corrupted the process, which diminishes its value. First, magazines and their listeners often have a conflict of interest that is hidden. Profits Journal depends on the width of readers and advertisers, leading to geographical bias (articles from economically important countries are preferred) and content bias (articles on trend topics are selected), and auditors may have Other programs (eg, friends who support or have an economic or professional interest in the results). Sometimes magazines and reviewers may not even be aware of the corrupting influence of commercial interests. Attacking this problem requires more than just a statement of advice and ownership. In an ideal world, a newspaper should not be funded by advertisers or subscribers, but fees and verifiers of authors should challenge in the event of commercial and personal conflicts. As noted below, at a minimum, newspapers should be held accountable not only by their owners for the money they make, but also by the public for the value they bring, just as a pharmaceutical company can not Simply selling a drug, but must prove that the drug is safe and effective, a newspaper should not be allowed to "sell" their products without being responsible for its contents.

​

This brings me to the second problem with the peer review: there is little responsibility for newspapers and users. If a newspaper repeatedly publishes maps that draw unsustainable conclusions, at the end of the paper authors can be blamed, but editors and reviewers who are probably responsible for gross negligence, are not held liable. There are insufficient controls and balances in the publication system; When high-ranking magazines repeatedly publish documents that are then considered unreliable or even hijacked, magazines seem to face the primary consequences their status remains intact.

 

The third peer-review issue, finally, is that there is no real competition between magazines as a channel for communicating science. Capitalism lives and depends on competition. Just like in many other industries today, however, authors are not really choosing between magazines. Most high-level journals are run by companies with significant profits, and it is very difficult for newcomers to break into this system. Lack of competition means the magazine that authors have limited choices in peer-reviewed journals that are better chosen, lessen the economic pressure on investment in peer review of high quality reviews.

​

​

In danger reproducibility

 

The other pillar of scientific truth, reproducibility, means that another scientist can repeat an experiment and get the same results, or, on the contrary, show that the results are not reproducible. As with peer review, many problems still threaten reproducibility. For example, it is common for a first large-scale study to report amazing results with a surprising conclusion. Then, when the experiments are repeated, only trends with the same conclusion can be observed with increasingly smaller effect sizes. This result does not contradict nor confirms the original study, but it is a dead end, and the original map is slowly forgotten. As noted earlier, the problem does not arise that the initial map is fraudulent, but the results have been "pinched" or selected, or represents a statistical aberrant value, leading to a misleading conclusion.

 

A second emerging problem of reproducibility is that many experiments are by design impossible to repeat. As formalized by Karl Popper [5], scientific truth requires interpersonal reproducibility. On the basis of this postulate, no conclusion can be falsified because the experience below can not be repeated exactly in the same way is not a scientific conclusion. Many current experiments are so complex that the result of differences can always be attributed to differences in experimental conditions (as in many recent neuroscientific studies due to the complexity of the nervous system). If an experiment depends on many variables that can not be conserved consistently and reliably, the scientific community must not accept the conclusions of an experiment as true or false. These findings are simply non-scientific, even if it is based on experience.

 

A third problem is validation of the reproducibility of reagents and methods. Too often, articles in first reviews are published without sufficient experimental controls that take up too much space in the precious real diary! -o with reagents which are not verified after being acquired. In addition to these problems of reproducibility is the almost impossibility of publishing negative results, due to the reluctance of newspapers largely motivated by economic pressures to devote valuable space to these documents, and the reluctance of authors to recognize the Errors.

​

​

For more reliable results

 

So we, the scientific community face major problems to ensure the legitimacy of science. Although correcting these problems will not be trivial, simple steps could increase scientific veracity. In my opinion, there is no alternative to magazines, we need magazines as a filter, now more than ever, and magazines need to be economically viable. However, given the strong profit margins of many magazines, I think it is reasonable to insist that scientific journals adhere to a minimum set of rules. For example, exams should be published, not hidden. Publishers must be appointed as part of the journals displayed and must be held liable if the cards can not meet the basic quality and reproducibility standards. Contributions should be evaluated historically, not for quotes (which may mislead), but keep track of tracking those cards. At least, the most important journals should systematically monitor the next task (or lack of it) that emerges from large studios. The submitted documents must be evaluated by a checklist that ensures that appropriate reactive checks and validations are present, and these validations must be retained for additional materials. It is amazing how many important documents show immunoblot where the presumed target proteins have the right size! Editors need to have the qualifications required to judge the overall technical validity of the experiments, but not able to evaluate the specific details (which is the job of the auditors). In addition, publishers must have the time to read the documents carefully and understand the methods and experiences, and should be better paid for the large amount of work they are asked to do. More importantly, as evaluators, it should be emphasized how exciting it is less the result, although it can not be true and focus more on the fact that the result is actually solid (true).

 

A more demanding, but perhaps necessary change to ensure scientific truth is to show immediate reproducibility. A conclusion should not be based on a single measure, but on the different parallel approaches. Ideally, scientists should recruit other groups to independently replicate the main results. More pressing, however, to ensure the validity of scientific studies may be what I would call the common sense rule: the more a document raises the question why it seems to contradict the meaning and / or because they get diametrically different conclusions from the studies (Also known as a "novelty"), it should be required more evidence. From time to time, she studies that most challenge credulity are published with less real experimental support because these studies exude excitement-however, these are the studies that require most experimental support!

 

Many of these ideas have been expressed several times before. But never has the need to act was more urgent than now, when all of our society is increasingly threatened by lies, with the science of being only a small part of it. Given that the factors that determine the threats to scientific practice are economic and political, we should speak and express our concerns. As it seems unlikely "voluntary" action, we should demand rules that inject responsibility into the system, unrestrained capitalism seems to become self-destructive and leads to self-sufficient and selfish monopolies that hinder progress.