Skip to main content

More Puffery Regarding Animal Models

Michael E Goldberg of Columbia University is the immediate past president of the Society for Neuroscience. He is also the author of Animal research and medical progress published 12-16-2010 by the San Diego Union-Tribune. His essay starts in the usual way:

If you have ever taken medication, received a flu shot, taken a pet to a veterinarian, undergone minimally invasive surgery or been helped by countless other medical interventions, you are a beneficiary of health advances made possible through decades of responsible animal research. Carefully regulated, humane animal research has been the foundation of nearly every major medical advance in the last century, and it will be equally essential to the next century’s progress.

These claims are standard fare for vivisection activists and I addressed them in my three part series “Argument From Authority.” Part I, Part II, and Part III can be accessed by clicking. Rhetoric is not the same as evidence and merely repeating the same rhetoric over and over again is not evidence either.

Goldberg then goes on to promise future treatments for today’s horrible illnesses if only society will continue to fund animal experiments. All of this is very similar to the puff piece by Liz Hodge that I addressed in Another Puff Piece from the Vested Interest Groups. For more on the fallacies in these kind of articles please read the blogs linked to above.


When it comes to animal research, the stakes for medical progress are high. Discussions should be based on facts, not misinformation campaigns and a serious lack of public awareness regarding medical progress.

I strongly agree! I wonder when Goldberg will get in touch with me to schedule a debate so public awareness can be increased? The complete disingenuousness of these vivisection activists is exposed by the fact they simply will not debate this issue in front of a live audience despite endlessly making claims like Goldberg’s. They are right, they assure us, and they really want society to understand all this but debating the topic in public (or in private and taped) is just not possible. Gosh darn it! Neither will they participate in a point counterpoint in the peer-review literature. That is what we in science call an unfalsifiable position. And things that are not falsifiable are not science.

I recently participated in a series of responses to an editorial in The Scientist where several scientists assured me animal models were invaluable. But when they were asked to defend their position, as always, they refused. One does not need to be a scientist to see through this farce.

In the November issue of Nature Medicine, Jucker wrote a similar article titled The benefits and limitations of animal models for translational research in neurodegenerative disease. Jucker:

Age-related neurodegenerative diseases are largely limited to humans and rarely occur spontaneously in animals. Genetically engineered mouse models recapitulate aspects of the corresponding human diseases and are instrumental in studying disease mechanisms and testing therapeutic strategies. If considered within the range of their validity, mouse models have been predictive of clinical outcome. Translational failure is less the result of the incomplete nature of the models than of inadequate preclinical studies and misinterpretation of the models. This commentary summarizes current models and highlights key questions we should be asking about animal models, as well as questions that cannot be answered with the current models. (Jucker 2010) (Emphasis added.)

Note that Jucker states that: If considered within the range of their validity, mouse models have been predictive of clinical outcome. I suspect what he means is that a mouse model gave results that correlated with humans. But this is not what predictive means in science. It is what predictive means in astrology.

In the next sentence Jucker states:

Age-related neurodegenerative diseases are largely human-specific diseases. Although aged nonhuman primates and some other higher-order animal species show aspects similar to those of human brain aging, these animals generally do not readily develop the full neuropathological or clinical phenotypes seen in humans (Gerlach and Riederer 1996; Walker and Cork 1999) (with a few possible exceptions, such as mutant SOD-linked canine degenerative myelopathy, resembling amyotrophic lateral sclerosis (ALS)) (Awano 2009). (Emphasis added.)

This further supports my suspicion that Jucker is misusing scientific terms. While causal disanalogies can exist between the model and the humans being modeled, the fact that such disanalogies exist means that the model will not in all likelihood exhibit a high positive predictive value and negative predictive value.

Jucker states:

Phenotypic similarities between genetic and idiopathic variants, in combination with a paucity of natural animal models, have driven the development of genetically modified animal models based on familial disease mutations.

Artificially induced diseases do not have a good track record for predicting human response to drugs and disease. The fact that such models are needed is not a good prognosticator for the animal models being predictive, as Jucker has stated that they are.

Jucker then discusses mouse models of Alzheimer’s and Parkinson’s disease. As would be expected, there are similarities to and differences from the human versions. With such differences, we would expect the model not to predict human response. They may be heuristics but they are not predictive.

It appears that Jucker is using the word predict in two incorrect, yet commonly used ways.

1. Retrospective correlation. If a model was developed that reproduced findings from humans, many animal modelers will say the animal model predicted the human response. This is simply nonsense; literally making no sense as predictions are not retrospective. (Note there is a difference between retrodiction and retrospective prediction.)

2. If a model and humans share a trait this does not mean the model predicted the trait. In order for a model to be said to be predictive, risk must be involved and there must be a track record. There is no risk when one simply cherry picks the data until one finds what one is looking for, and N=1 is not a track record. When track records have been studied animal models have been consistently shown to fail to predict human response to drugs and disease. This is to be expected based on our current knowledge of complex systems and evolution.

After giving us excellent reasons to suspect that mouse models cannot in fact predict human response, Jucker goes on to claim that the differences between the model and humans are very important and that this is why they should be used in research. Again, this is typical for the vivisection activist. First they claim that animals and humans are similar enough for the animal models to be predictive for humans but later they acknowledge all the differences which means the animal models cannot predict humans response and attempt to rationalize why this is actually a good thing.

From a scientific perspective, differences among species do yield some very interesting research. And, if your goal in performing research is to outline such differences, then animal-based research is fascinating. However, if you are taking money under the guise of curing human disease by studying animals, then you are committing fraud.

Finally Jucker states that when the mouse models give different outcomes from humans, the mouse models are right. While this is mindboggling, it is a common response for the vivisection activist when asked to explain why animals give different results from humans. “The clinicians got it wrong.” I heard that refrain more times than I can count when I was in my first two years of medical school being taught by PhD basic researchers.

The fact that the scientific community per se does not jump all over this is yet more evidence that individual scientists value their jobs over truth or are simply too lazy to be bothered. Neither is a flattering representation of their character.

To be fair to Jucker however, I must point out that near the end of the article he also states the following:

The external validity of a model refers to what a model is good for. However, the limitations of models are rarely acknowledged, and the predictive validity of these models for humans is often overstated or misinterpreted, raising false or premature hopes for clinical efficacy. APP-transgenic mice are models of cerebral β-amyloidosis and not of Alzheimer's disease. Furthermore, a treatment that is effective in female mice may not necessarily predict efficacy in male subjects. Similarly, a treatment assessed in one mouse model may not be replicable in another model, and the outcome measure of a treatment in a mouse model may be of no use for clinical translation (Box 1).

Truth at last.

As I stated above, if you want to do neat science you can use animals and learn things. But if you sell to society the notion of using animals as predicting disease and drug response in humans, then you are committing an immoral act, namely fraud or at least disingenuousness in an area where lives are at stake.

This paper is typical of its ilk in that it simultaneously states how wonderful and predictive animal models are and how flawed they are. It concludes by saying that more mouse research is needed. If you are of a scientific disposition, I encourage you to read the entire paper in order to appreciate the overall tone. I can understand why mouse modelers say these things. I cannot understand why the scientific community stands by and lets them get away with it. The scientific community cannot complain about homeopathy, astrology, creationism, and the consequences of society’s overall neglect of real science while allowing its own kind to get away with such nonsense.

For real science please see Animal Models in Light of Evolution. It is not a page-turner but it is science and not puffery.


Awano, T. 2009. Proc. Natl. Acad. Sci. USA 106:2794-2799.

Gerlach, M., and P. Riederer. 1996. J. Neural Transm. 103:987-1041.

Jucker, Mathias. 2010. The benefits and limitations of animal models for translational research in neurodegenerative diseases. Nat Med 16 (11):1210-1214.

Walker, L. C., and L. C. Cork. 1999. Alzheimer Disease. Nature Publishing Group, a division of Macmillan Publishers Limited. 


Popular Video