Neuroscience journal retracts paper for lack of “scientific soundness”

An unusual article that considered the concept of change from a systems perspective — including change in medicine, economics, and decision-making, for instance — has, well, changed from “published” to “retracted.”

After commenters on PubPeer called the 2014 paper “gibberish” and even suggested it might be computer-generated, Frontiers in Computational Neuroscience retracted it, noting it “does not meet the standards of editorial and scientific soundness” for the journal, according to the retraction notice. The paper’s editor and author maintain there was nothing wrong with the science in the paper.

Here’s the full note for “Sensing risk, fearing uncertainty: systems science approach to change:”

Following a formal complaint concerning the publication cited above, the Specialty Chief Editors of Frontiers in Computational Neuroscience conducted an assessment of the article, according to the Frontiers complaints protocol. The Specialty Chief Editors concluded that the publication should not have been accepted in its published form, as it does not meet the standards of editorial and scientific soundness for Frontiers in Computational Neuroscience. This assessment was conducted in consultation with the Handling Editor, Dr Tobias A Mattei, who agreed to this conclusion. The author agrees to the retraction, commenting that the article was inappropriate for the Journal and its audience.

The article has not been cited, according to Thomson Reuters Web of Science.

It’s an unusual article, to be sure. Here’s an excerpt from the introduction:

The purpose of this study is to make change more manageable; the path, to reach this goal, is to differentiate it into what is potentially quantifiable and what is not. Risk, as defined in this study and with applicability to human bodily system, is considered quantifiable through sensory input which also generates a degree of experiential confidence and predominantly rational optimism; what is outside of sensory spectrum is regarded as unquantifiable uncertainty which evokes emotions such as fear, greed, irrationality, etc. The advantage of separating sensory-defined risk from the remaining uncertainty is that it isolates a portion of change whose outcome has a relatively high degree of predictability; uncertainty has none as it has random outcome.

A large portion of the paper consists of a list of articles that include instances of change, for instance:

Global temperatures highest in 4000 years (Marcott, 2013).

Uncertainty—the other side of prognosis (Smith et al., 2013a,b).

Every time our bodies record an experience… the essence of who you are is stored as synaptic interactions in and between the various systems of your brain… activity induces growth (LeDoux, 2003).

Between 3 and 300 milliseconds is the range of time [that] is involved in many aspects of cognition… [For example] the response time while driving a car is around 300 milliseconds (Nolte, 2001).

The conclusion reads, in part:

Change is never an isolated event; its understanding requires context; it is also not homogeneous but is composed of innumerable changes arising from other systems.

Last year, a commenter on PubPeer raised the possibility it had been generated by a computer:

My deepest apologies to the author if this isn’t the case, but this paper reads as though it is computer generated. Our group was cited by it, but I can’t make any sense of why.

Please carefully look at the figures and read some text. At first I thought that I just didn’t understand it, but the more I read the more I think there is something strange going on.

Tobias Alecio Mattei, both the editor and a reviewer of the paper, told us that he did not agree that the article was scientifically unsound:

In my view, there was not any scientific reason for its retraction. There was no identified ethical or plagiarism issues. As the article is essentially theoretical in nature,  there was also no clear methodological concerns or statistical problems with any empirical data. I believe the decision for retraction was essentially based on aesthetic/stylistic purposes and not on scientific concerns.

Mattei — affiliated with InvisionHealth in Buffalo, NY — explained to us that he told the editors in chief:

If they would prefer to retract the paper (as it became the case) I would accept the decision. In my personal opinion the paper had value enough to be kept, but I knew that the impact upon Frontiers profile, taking into account the online criticisms in the lay media and online blogs, especially due to its non-standard format, might ultimately lead the chief editors of Frontiers to decide to retract it.

We asked Misha Tsodyks, one of two editors in chief at Frontiers in Computational Neuroscience, about the discrepancy between Mattei’s view and the note:

The paper was deemed below the standards as a result of a separate review, in which the original editor was not involved. There can always be disagreements.

The sole author on the paper is Ivo P Janecka, who used to work at Harvard and says he’s now affiliated with a private foundation. He received coverage in the New York Times in 1997 for a surgery called “facial translocation:”

…a systematic method of swinging whole portions of the [patient’s] face out of the way to expose a tumor in the middle of the head.

Janecka told us that he agreed to the retraction — but, like Mattei, not because he thinks the paper scientifically unsound:

When the journal Frontiers in Computational Neuroscience decided to go ahead with the retraction of my paper, Sensing Risk…, it was clear to me that neither some readers, who thought that this paper might have been written by a computer, nor the publisher, who then wanted to disengage from this article ASAP, were interested in anything rational; I sent them the following note that should have been included, verbatim, with their retraction statement:The author agrees to the retraction of the article from this specific journal, as the selection of this journal, though based on editorial invitation and acceptance, was misplaced.

Janecka explained that, in his view, the study does include data:

References in this paper represent the “data” used in this study, describing fragmentary observations of various biologic systems, small and large, just waiting to be connected, as was done in this study. With the help of systems science, the goal of this paper was to construct an inter-related “whole” from these “fragmentary” observations, listed in the reference, in order to show the linkage through systems science principles, and, ultimately, come up with the meaning of the “whole”.

He added:

A good question to ask, within the context of this discussion, is: What is science and who is to judge what science is? No simple answer is available and anyone who feels to be the final arbiter is likely mistaken.

Of the assertion that the paper is computer generated, Mattei said:

To anyone that read the paper (in its fullness) it is obviously that there is no computer algorithm which could have generated it.

After seeing the paper on PubPeer, Guillaume Rousselet, a cognitive scientist at the University of Glasgow, launched one of the complaints mentioned in the retraction notice, as Leonid Schneider reports. Rousselet was an editor at Frontiers in Human Neuroscience and is editor for Frontiers in Perception Science, but said he’s resigning due to the “painful process” of dealing with the reviewing forum, in which editors, authors, and reviewers interact during the review process.

He forwarded us his email to the journal, in which he says:

There is a problem with this paper, which reads as if it had been computer generated…Can you investigate?

The journal told him they were investigating; eventually, the journal said it did not find any kind of misconduct in the paper.

After follow-up emails from Rousselet and another researcher (on which Rousselet was cc’d, and forwarded us), the journal retracted the paper.

David Kronemyer, one of the reviewers of the paper and a postdoc at the University of California, Los Angeles, told us his thoughts on the paper:

My initial reaction on reading this paper was that while its organization was unconventional, it did have a number of interesting ideas. It was not, however, in shape for publication. As a reviewer, I suggested a number of changes over several drafts, some of which were made, many of which were not made. To put the paper in context, I also looked at other publications in the field, as well as some of Dr. Janecka’s earlier work. I would have continued recommending changes to subsequent drafts, however, there came a time when the journal’s editorial staff decided the paper was sufficient.

Mattei told us why he took on a role both as editor of the paper and reviewer:

The way Frontiers special topics works involve the proposal of a special theme (in my situation ‘Non-linear analysis in neuroscience and cognition'[)] with a list of invited authors who had been previously contacted and agreed with submitting proposals. Dr. Janecka was in the initial list of contributors. He  submitted his work on time, dedicated a significant amount of his personal time to research on the issue, followed the deadlines and was very willing in revising and improving his paper after the initial reviewers suggestions. Additionally, I saw a significant value in the addressed issue, despite the very confuse and chaotic format of the initial submission.

As the only two reviewers who accepted to review such paper were not able to improve it to the point where it would become publishable, I decided to invest my time and efforts in order to assist Dr. Janecka in his work. Ultimately, as editor-in-chief for that special issue I am responsible for supervising all the submitted articles as well as the work of the involved reviewers. I am also a reviewer for Frontiers Neuroscience, providing inputs and comments for papers submitted to other special topics when invited to do so. As there is no conflict of interest between my role as reviewer and editor (as I had no involvement with Dr. Janecka or any participation in his initial article) and, by the Frontiers policy, I am encourage to actively participate in the selection, supervision and improvement of the articles in my special topic, I decided to assist Dr. Janecka as a reviewer.

Frederick Fenter, the executive editor of Frontiers, sent us a statement on the retraction:

Frontiers is putting publishing back into the hands of researchers, where editorial responsibility is deeply embedded into the academic communities. In fact, we have over 8’000 highly qualified Associate Editors with the authority to accept manuscripts, following the endorsement of at least two reviewers. The system has made Frontiers a leader in publishing  but there will always be occasions when controversy surrounds a specific acceptance decision.  For these situations, Frontiers has one of the most transparent community complaints policies  in publishing, which we invite all to read.  The protocol  is invoked when a valid concern is sent to a Frontiers journal, and can lead to retraction in those cases where the external editorial boards deem this to be the correct action to take in order to safeguard the academic literature.

Of note: Frontiers is on University of Colorado librarian Jeffrey Beall’s list of “potential, possible, or probable” predatory publishers.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.