Recently, a group of 124 researchers wrote a public statement to express some concerns about the integrated information theory (IIT) of consciousness, especially regarding how it has been falsely promoted as a scientifically established theory. Some points made in the statement, maybe especially the use of the word “pseudoscience”, have caused what has been described as an “uproar” on social media (specifically, X/Twitter), with the authors of the public statement likened to mass murderers. A follow-up will be published online on Mar 10, 2025, outlining the arguments behind the initial letter. Because of journal requirements, space and reference count are limited in the new piece. This document provides some further background to this controversy, and links to other relevant articles.
This document was prepared with substantive input from various members from the IIT-concerned consortium/group, but it does not mean that every member agrees with everything said here.
We wrote the statement (also sometimes referred to here as the ‘letter’) because we think that scientists should not falsely promote their unsubstantiated personal worldviews as well established science. The fact that this is happening with IIT is worthy of recognition and further discussion, as the said false promotion has been relatively intense, and has made considerable impact. There are also various ethical and policy implications. No matter whether one agrees with the statement or not, it is also an undeniable fact that a sizable group of researchers, from both within and outside the field, consider IIT as pseudoscience, in part because of the said false promotion. The release of the letter reflected a long-standing problem within the field, and brought new attention to it. We believe that the general public, including potential new trainees, funders, journalists, and editors, will benefit from being informed of these facts.
Yes, but that is because the statement was intended to be a letter-to-editor type of short piece, i.e. an expression of concern, not a research paper.
Several co-signers of the statement have independently written to explain the arguments and reasoning behind. See e.g.:
Others who did not have a chance to sign the first letter also expressed some support, e.g.
Well before the events leading to the writing of the statement, the various problems of IIT have already been brought up many times over the years, in both informal and formal settings. See e.g.:
(This list is by no means exhaustive.)
In other words, most of the concerns leading to the statement are not new at all. The discussion of IIT as pseudoscience also isn’t new. Although there have been some replies on some technicalities, we do not find them satisfactory, and importantly, the fundamental issues remain either unresolved or unacknowledged.
However, we understand that the reaction in defense of IIT has now drawn the attention of new readers to the problem, who may not be familiar with the literature. In the light of this, members of the group have since co-authored a new version of the statement that is expanded into a longer article (online from Mar 10, 2025), with the core concerns summarized and laid out.
The statement has not been retracted. The group as a whole continues to stand by it and the considerations behind. This overall group position was established through extensive email discussions and voting. Everyone has been given opportunities to reconsider their positions and to withdraw from the project if they wished to, and an absolute majority of them chose to remain.
However, because the new version of the paper is going to be published in an academic journal, the requirement for ‘co-authorship’ needs to be made clearer than the mere co-signing of a public statement of concern. Although the original statement has 124 co-signers, only 10 authors were responsible for drafting it. In the new version, authors are those who considered themselves having contributed significantly. They are those who have been given the opportunity to participate in actual writing, voting, and choosing the title of the new paper, etc. This number has substantially increased (see also Q5 & Q6 below). There are also new authors in the new version of the paper, who did not have a chance to sign the first statement.
As noted above in Q2, we have known for some time that IIT is popular among non-researchers, especially on social media, likely in part due to over a decade of hyperbolic promotion. Also, as IIT is now featured in several well-funded projects by the Templeton World Charity Foundation, we understand that some researchers involved may also feel strongly about open criticisms of the theory to the effect of dismissing the merits of those projects altogether. Many influential scholars have also been involved in promoting IIT in various ways, and they may feel the need to defend their own past actions. However, the co-signers of the letters are also a sizable part of the research community. In general, heated exchanges on social media tend not to unequivocally resolve academic debates. We hope that the new version of the paper will move these discussions forward.
The initial statement was drafted together by the first 10 authors as listed in the letter; Steve Fleming was listed as first-author because of the order of alphabetization within that group. The co-authoring and co-signing of the letter, involving as many as 124 researchers, were initially organized by Alan Lee and Hakwan Lau, with help from others. Accordingly they assumed the role of corresponding authors in that letter.
To ensure a broader participation from the group, in the new version (the more fleshed out academic paper), dozens of authors contributed to the writing. This process is complicated and intensive. The new corresponding authors are those who volunteered to take up the responsibility to decide to approve or reject suggested edits made by the others. There was no selection or competition for the roles of corresponding authorship. They are just the authors who were willing to take up that task.
In the Postscript section of this article, Hakwan Lau has explained why he has personally decided not to be a corresponding author in the new version. Alan Lee likewise doesn’t feel the need to continue to assume this role. This has always been a group effort, and a time-consuming one. Different members took turns to get the necessary work done.
The title of the initial statement was decided by the group of first 10 authors through voting; additional members were consulted. It was explicit in the draft passed around to others, at every stage while it was co-signed. In the email sent out for inviting researchers for co-signing, the issue of why we felt we had to use the word ‘pseudoscience’, despite the severe nature of the charge, was also explicitly discussed. The issue of potential funding impact was also raised in that email.
In the new version, the decisions to continue to use the word ‘pseudoscience’ in the paper, and whether to publish it as a preprint or a journal article, were made by the entire group through several rounds of discussion and voting procedures. We also took into account suggestions from the journal editor.
It is challenging to write a long article involving as many as 124 authors. Hence the original letter sought to function as a short statement of concern rather than a detailed review of the scientific arguments behind. As mentioned in Q2, most of the arguments are not new. The initial statement was also triggered by some recent events of a public nature, which elevated the problem to a new level, requiring a timely (and therefore relatively brief) response.
The event was not organized and endorsed by ASSC per se. It was a satellite event of ASSC’s annual conference (2023), and individual members of the local organizing committee were responsible for its organization and curation. As pointed out here (see section 7), one concerned member, David Chalmers, is also part of Cogitate. To those interested in the details of this matter, please contact the relevant members or ASSC for direct clarification.
Along with other events at the conference, there is a video of the concerned public event. While videos of the other events have already been posted online, the video for the public event remains undisclosed at the time of writing. We understand that this is because some involved in the organization of the event, and are part of Cogitate, do not want the release of the video to influence the peer review process of the Cogitate paper. We hope and expect that eventually the video will be released, so that others can see for themselves what was the nature of the event. Meanwhile, journalists have covered it in some detail.
We disagree. While some influential work in the 1980s (e.g. by Larry Laudan) has expressed similar views, highly regarded scholars within philosophy of science continue to meaningfully use the word with justification, e.g. to refer to problematic views such as creationism and astrology (e.g. this and this by Philip Kitcher, and this by Paul Thagard). Meanwhile, new philosophical work also explicitly addresses that position from the 1980s (e.g. this, and this). In many other academic disciplines, as well as in other areas of public discourse, the term continues to be used in important and meaningful ways (e.g. this, and see also references mentioned in the opening sections of this).
Tim Bayne, who does not think that IIT is pseudoscience, also offers his own solution to the demarcation problem regarding what qualifies as “genuinely scientific”, thereby implicitly rejecting Laudan’s position that such demarcation isn’t possible. (Kayuet Liu discusses various problems of Bayne’s article here.)
Regardless of one’s own position on this philosophical matter, we urge the reader to not let a simple matter of word choice become a distraction; the arguments against the untruthful and unjustified promotion of unscientific claims to the general public, with the pretension that they are established good science, remain the same, however we call the phenomenon.
Also, as we mentioned above in Q2, the discussion of whether IIT is pseudoscience is not new. People within the community should understand that there are genuine concerns, regardless of whether they agree with them. They should not contrive to give outsiders the false impression that the word was being brought up all of a sudden.
To the extent that pseudoscience is a real phenomenon, which we think it is, disallowing the use of the term may amount to nothing but giving pseudoscientists a free pass, and could therefore be generally harmful to society.
We think that the situation there is entirely different. With theoretical physics, most theories make precise quantitative predictions that are clearly entailed logically by the theories. Even for string theory, which arguably makes no testable empirical predictions that could be used for distinguishing it from some alternatives, the theory itself is constrained and (at least partially) supported by other empirically verified assumptions from physics. For IIT, however, the so-called axioms are based on nothing but the introspection of the proponents. Although the proponents claimed that they are self-evidently true or irrefutable, these axioms are controversial and have in fact changed over time. The proponents of IIT also acknowledge that the theory may well be incompatible with physics as we know it. Without further assumptions, many of which are vague and implicit, the theory’s postulates just aren’t logically derived from the axioms.
As others have pointed out, these considerations ultimately depend on whether IIT is in fact pseudoscience. We believe that it is, according to common dictionary definitions, as well as recent work in the history and philosophy of science (see Q9 above). As such, it is unfortunate that the field has gotten to such a state to allow a pseudoscientific theory to be promoted repeatedly as established good science to the general public. To fix the problem, there may in fact be some short term negative impact, as the problem gets exposed. However, we do not see how exercising willful ignorance and letting the problem grow bigger would be good for the future of the discipline.
There is also the external cost to other disciplines, as well as to society at large to consider. These issues have been discussed here, here and here by some authors.
As to who is responsible for this unfortunate situation, an analogy may help: suppose a group of citizens called out the evidently corrupt behavior of certain politicians. Should we blame these citizens for drawing attention to the problem, rather than holding the politicians themselves accountable? And if we openly persecute these citizens for making the country look bad, how would it actually make the country look?
Some colleagues are in fact accusing us of harming the reputation of the entire field. Ironically, in doing so, some resorted to using a language so “disproportionate” that it ended up attracting more media attention than we had sought. In other words, the accusations amplified the very issues that these colleagues apparently wished to suppress.
It has not been our primary intention to criticize empirical variants of IIT (contra the metaphysical one which is what is the most prominently promoted version; see this for the distinction). In the new version of the paper we have a dedication section defending some of these works, stating that they are not unscientific. However, they do not support IIT per se, and face other challenges.
Specifically, if the complexity measures ‘inspired’ by IIT are taken to reflect consciousness in general, some recent evidence would have to be interpreted as suggesting that fetuses are probably the most conscious when they are premature, especially in males. We therefore caution against such generalization even within humans, let alone inanimate objects (as intended by IIT).
Researchers who promoted and defended IIT often encouraged others to consider thorny issues such as consciousness in fetuses, in light of the theory. In fact, in an article on fetal consciousness, the above mentioned study (Q13) about fetuses showing higher brain complexity (and hence higher degree of consciousness, according to standard interpretations of IIT) when they were premature, was also cited.
Meanwhile, another controversial paper within the field (for the counter arguments please see this) has also been cited in a document used in the US Supreme Court case about abortion.
Proponents of IIT are also vocal about the status of organoids and their research.
Therefore, it would be naive to think that in general, our research and its public promotion have no practical implications to these issues.
As to whether IIT should be evaluated based on its scientific merits alone, our position is that its scientific merits are limited. We also believe that extraordinary claims require extraordinary evidence. When the ethical implications are potentially far-reaching, they need to be evaluated with extra care.
We try to make sure that their efforts are correctly acknowledged, and not unfairly criticized. However, it is also part of scientific training for all of us to recognize that our ideas, or ideas that our advisors have subscribed to, can be totally wrong or misguided. It is unfortunately true that participating in projects that are pseudoscientific in nature can turn out to be a career setback to some extent. But this is also exactly why we feel responsible for helping potential new trainees see this reality. As pointed out in Q2, it is not a new fact that IIT is considered pseudoscience by many, including some potential future employers of the relevant trainees. If we keep this information suppressed, it would only mislead trainees, especially those not fortunate enough to be able to know the insider gossip. They deserve to know the reality as it is, in order to make informed decisions about what projects and labs they should join. They are also encouraged to participate in current debates on this matter, as many of them do.
Yes we should be open to unconventional ideas. But this does not mean that we should allow them to be prematurely and untruthfully promoted to the general public as established good science, and to consume tens of millions of dollars of research funding in vain. Nobody is trying to ‘ban’ or ‘cancel’ these ideas. One can have them, or discuss them, just as we are doing now. Our intention is merely to have them characterized correctly so as to avoid misinformation to the general public.
These issues have been addressed elsewhere in detail, in a piece that responds to Anil Seth’s plea for the right to (be) wrong.
Depending on how it is applied, the appeal to the freedom of expressing unconventional ideas could also be self-defeating in this context, because if one demands such freedom one should at least also grant the same to critics.