Talk:Twin study/Archive 1

Comment
Can anybody verify or give more information about this claim that was in the article?
 * "Additionally, there are different ways of calculating concordance (the presence of a similar disease phenotype in twins) which can give markedly different results."

Can anybody verify the prenatal imprinting argument? It seems like it would be easy to test by comparing trait concordance between non-identical twins reared apart with normal siblings reared apart. --Nectarflowed T 09:44, 15 July 2005 (UTC)

private language
Are there any studies about twin idioglossia, or secret language like Poto and Cabengo had? That seems like a good thing to include in the article Phoenix-forgotten 03:18, 20 September 2006 (UTC)

---

There was a programme aired here in the UK (can't remember if it was an Equinox or a Horizon one) relating to a social experiment that took place in the US re: Nature vs Nurture. A set of twins was split up and adopted by two different social sets of parents and I THINK (!!) nature won out. I might be wrong on this, though, as it was about 10 years ago.

Can anyone help me out with the title of this or where I might be able to buy/loan a copy?

Many thanks in advance!

Organization
Hi. I think the readability of the article would be improved if the article was written under the following outline 1 Overview (pretty much as is - why twins are potentially helpful, some history. Note what the potential findings are (ranging from no effects of DNA on differences between people, to complete control, with some examples of. And a note )

2 Method overview (pretty much as is - just the logic of the design)

3 Assumptions (pretty much as is - what assumptions does the logic entail? This should link down to the detailed method section on designs aimed at answering more complex and realistic questions about g*e and ge corr etc.)

4 Criticism (Content similar to now, laid out in a list of critiques linked to the assumption they dispute, the criticism described, and with a rebuttal paragraph beneath it) i.e:

4.1 Sampling 4.1.1 Assumption needed 4.1.2 Evidence and impact of actual variace from assumed value 4.1.3 Brief rebuttal

5 Detailed Methodology (Modeling, terminology, complex models)

FYI, the initial structure is this: 1 History 2 Criticism 3 Methods 3.1 Modern Modeling 3.2 Assumptions

We also need to find a way of not coming back to a dozen edits claiming that "the method is proven invalid" and similar remarks. I think the new structure will help obviate the need for this, allowing readers to decide for themselves. There are several points of view, and we should be able to make this page neutral: putting the substantive points of view in an articulate fashion, allowing intelligent readers to reach conclusions and understand what is uncertain.

Ideally, we would arrange all the pro- and con- points around the assumptions which they entail, and also note what the likely effect of the criticism, if true, is on our theories of human behavior. i.e., what sort of confidence intervals should we put on conclusions, based on reasonable criticisms of the designs.

We also need to find a way of stopping this reverting to a Peter Schonemann hagiography. Can we agree that he is one of several people who don't like genetic research?

Also, given that this and related pages are going to be edited on an active basis in the future, can we please all 1. log in before editing 2. talk before making substantive changes 3. consolidate edits (rather than a dozen insertions which are hard to track but completely change the intent of sentences substituting "might" for "is" and vice versa? cheers, Timothy Bates.  —Preceding unsigned comment added by Tim bates (talk • contribs) 10:07, 11 September 2007 (UTC)

Hi Tim, this sounds very reasonable to me and your outline seems to do justice to all aspects of the issue. Good plan, go ahead! --Crusio 10:11, 11 September 2007 (UTC)

Pairwise concordance
In this section an example is given, saying "become affected". As far as I see, they can be affected from the start. I think the procedure is to select a group of affected patients who are twins and then see how many of the twin pairs are concordant or discordant. But I may be mistaken and currently have no time to look it up. Anyway, it seems unlogical as it is defined now, because concordance rates would depend on the duration of the study: the faster the study is conducted, the lower concordance rates will be... --Crusio 12:56, 25 September 2007 (UTC)

Ghastly Example
Mention (without advocating!) Joseph Mengeles twin studies at Auchwitz Concentration Camp? Hugo999 (talk) 01:20, 18 November 2007 (UTC)

I agree. Though his methods were barbaric and malevolent, he was still a twin studies researcher. He research may be inadmissible and much of the time it was psuedo-scientific, but it deserves mention in this article. Napkin65 (talk) 15:52, 11 June 2010 (UTC)

"...identical twins share 100% of their genes"
This is not true. Genomes of identical twins may differ in copy number variation. The article also doesnt mention epigenetic changes in the twins genomes. --84.189.119.111 (talk) 17:18, 6 October 2009 (UTC)
 * I agree with you on CNVs, which would have to be caused by de novo mutations, in this case. However, "epigenetic changes in the twins genomes" is an oxymoron; "epigenetics is the study of inherited changes in phenotype (appearance) or gene expression caused by mechanisms other than changes in the underlying DNA sequence", so their genomes cannot change. It's just like how your white blood cells are genetically identical to your neurons, save for any possible mutations (like tumours). MichaelExe (talk) 18:19, 4 July 2010 (UTC)

Citations List Including Citations to New Twin Study Papers
I see the references for this article at present miss out on some of the best recent review articles on twin studies by experienced researchers on human twins. You may find it helpful while reading or editing articles to look at a bibliography of Intelligence Citations, posted for the use of all Wikipedians who have occasion to edit articles on issues related to human intelligence, including heritability studies based on MZ and DZ twins. I happen to have circulating access to a huge academic research library at a university with an active research program in these issues (and to another library that is one of the ten largest public library systems in the United States) and have been researching these issues since 1989. You are welcome to use these citations for your own research. You can help other Wikipedians by suggesting new sources through comments on that page. It will be extremely helpful for articles on human genetic studies to edit them according to the Wikipedia standards for reliable sources for medicine-related articles, as it is important to get these issues as well verified as possible. -- WeijiBaikeBianji (talk) 05:14, 4 August 2010 (UTC)

Sources for images on twin concordance?
As usual now, when I see an image on Wikipedia, I look at the full image file (which by best practice should be hosted on Wikimedia Commons) to see if there is any source for the image cited. I don't see any source cited for the images for this article on twin concordance and heritability estimated from twin concordance; indeed I see a notation on the image file metadata that the image comes "(from unknown source)." On that basis, I am about to delete the images, which disagrees with some sources that I have at hand and hope to add to the article soon. Any sourced image on the issue available to another editor would be much appreciated. -- WeijiBaikeBianji (talk) 01:28, 29 September 2010 (UTC)

ACE and ADE models
A good explanation of ACE and ADE models are needed, and how they should be interpreted —Preceding unsigned comment added by 192.38.117.2 (talk) 20:54, 2 November 2010 (UTC)

Criticisms of correlation model and response
It does not matter - single/multi-variete correlations. Still being correlations. In "extended twin designs" and "simplex models" papers is nothing about that it is not correlation only (It is more sophisticated ("ACE"-like) correlation model and it can not imply causation). —Preceding unsigned comment added by 194.228.230.250 (talk) 14:35, 5 November 2010 (UTC)

Adopted Twins
I just wanted to add that twins who were adopted out already have suffered early trauma, that of being adopted out, which separates them from the rest of the population. Also any assumption that twins raised together share the same parenting environment is false, no parent can be the same to her children and even if she could, children (even twins) have different dispositions which mean they experience the parent in a completely different manner. WjtWeston (talk) 22:24, 6 March 2011 (UTC)

Equal environments assumption
Can I recommend this change to clarify the current section on EEA?

From

Equal environments. It can be seen from the modelling above, that the main assumption of the twin study is that of equal environments.

To

Equal environments. It can be seen from the modelling above, that the main assumption of the twin study is that the amount of environment shared by individuals within twin pairs is same for both types of twin pairs.

I know it confused me for one until I looked it up! TP21 (talk) 08:57, 11 April 2008 (UTC)

I think that it needs more changes. Equations (in: Twin studies - method)

Rmz = A  + C

Rdz = A/2 + C

A = 2*(Rmz - Rdz)

must be wrong. Because human gender heritability means: Rmz=1 (perfectly correlated twins - same gender), Rdz=0 (uncorrelated in gender - random as siblings (see: Response to representativeness of twins); it is NOT 0.5 as average of 2 perfectly correlated cases with r=1 and 2 uncorrelated with r=0 but anti-correlated with r=-1) and C=0 (approximately). Thus A=1 and factor 1/2 must be 0. (in the case of equations above A=2 (200% is not possible) and C=-1 (anti-correlated ... why?))

Rmz = A  + C

It is not the case that rDZ = 0 just because the genders are not correlated. DZs share on average half their autosomal genes. Many traits appear not to be sex-limited, but this is another assumption. If rDZMM == rDZFF == rDZFM, then sex appears not to matter for covariance. If it does, then sex limitation models can be used. Covering all of this is a longer term goal for the article, so added to the todo list Tim bates (talk) 13:57, 3 April 2011 (UTC)

Rdz =      C

A = (Rmz - Rdz)

(heritability estimate become half in values that were presented) More exact (not simply linearly additive) solution such as correlation between monozygotic and dizygotic correlations is needed. —Preceding unsigned comment added by 78.102.150.18 (talk) 12:23, 26 December 2008 (UTC)

Suggested restructuring of article
I don't think it's conducive to peoples' learning for there to be a back and force of sections, so that it looks like
 * Assumptions
 * Twin Study Assumption A
 * Criticisms
 * Criticism of Assumption A
 * Response to the Criticism of that Assumption

I think we should restructure the Assumptions/Criticisms parts simply around the assumptions of twin studies. Within each assumption's section we could give full voice to the criticisms and defenses of it. Otherwise the article is very much a confusing and dispersed dialectic, and doesn't help people understand how twin studies work (and don't). To learn what scholars have to say about the equal environments assumption, you need to go back and forth between separate parts of the article. Additionally, a scientific subject is converted into a defenders vs criticizers war, when in fact most psychologists/behavioral geneticists/etc hold a nuanced view on each assumption. Here is what I propose


 * Assumptions
 * Assumption A- Section would include description, criticism, and defense of assumption A.

Any thoughts on this proposed restructuring?--Babank (talk) 23:12, 2 April 2011 (UTC)

This was the old arrangement: what is lead to was edit wars and promotion of disputatious statements to the foreground, without ever letting people understand the (pretty simple) basis of the design. If the "criticisms" are brief, then what you say would make sense. However the criticisms are not reasonable cautions about making sure the assumptions are met, but disputes over whether twin studies are ever valid: hence placing them in a separate section. Much like one might do on the Natural Selection page for instance. I agree that it would be helpful to raise the need and method of checking each assumption in the main article Tim bates (talk) 13:46, 3 April 2011 (UTC)


 * Wouldn't it be possible to describe the assumptions in the first paragraph or two, then in later paragraphs discuss how these might be violated? I believe we can include criticisms that question even the complete validity of twin studies in the same section.


 * I looked at the natural selection page and it doesn't have this back and forth criticism structure. I really believe we can simplify the organization, without substantially changing the content.  We just need to give the first paragraph or two room to breathe before the assumption is criticized.


 * Lastly, as I understand it, section structuring should not be based on what will lead to edit wars, but rather to what will educate readers most efficiently on various reliably sourced views. Your thoughts?--Babank (talk) 19:10, 3 April 2011 (UTC)


 * Agreed the natural selection page doesn't have this back and forth criticism. It also doesn't open by saying "Lots of creationists and lay people have provided substantial evidence showing that natural selection is completely invalid" I suggest you just have a go at editing the first couple of paragraphs to list the real assumptions. Perhaps the too-and-fro part can just be absorbed.


 * I think we do cover assumptions in the first paragraph or two (it even has its own heading) and then in later paragraphs discuss how these might be violated? Perhaps that is all that we are needing. I really doubt that there are referenceable criticisms that question "the complete validity of twin studies". What is it you want to include?  Tim bates (talk)


 * I don't want to add much to the article, just restructure it for simplicity as described above.. As for the intro I believe it will be relatively straightforward to summarize the entire article (including the assumptions) once the sections or organized.--Babank (talk) 16:08, 6 April 2011 (UTC)


 * fire way buddy :-) Tim bates (talk) 10:28, 8 April 2011 (UTC)

Use of language in Behavorial Genetics articles
Just a short note to say that these articles have been written as though it were speech. Someone needs to clean up all the articles about Behaviourial Genetics and write them in a literary way rather than the current trend of writing them as if the author is speaking.

Phrases such as "Fidget and squirm in their seats" (from the ADHD article) are not professional. A better term would be "Appear restless when still"

In the Twin Studies article, much bias is made such as "Twin research is a key tool in behavioral genetics and related fields" The word considered is needed after is to create an unbiased article.

I have many more rants about related articles but I shall not bore you. — Preceding unsigned comment added by 58.169.184.246 (talk) 07:17, 11 August 2011 (UTC)

Review of new book on twin research
Why do identical twins end up having such different lives?

The popular book reviewed in this article looks interesting. I live in Minnesota and know many of the researchers on the Minnesota twin study, including one who is a distant cousin of mine. There is always something new to read about twin research, which can perhaps help with updates to this article. -- WeijiBaikeBianji (talk, how I edit) 00:55, 2 June 2013 (UTC)

POV
In the last few edits, someone has been making non-neutral edits with POV and excessive citation of a single scholar. Criticism of twin studies belongs in the Criticism section, and folks should avoid mis-characterizations of the literature. While there has been some criticism of twin studies, it is not correct to characterize this criticism as the majority opinion. —Preceding unsigned comment added by Twinster (talk • contribs)


 * Indeed. And the Criticism section is rather rambling. The last phrase contains some errors of logic, for example. Someone with more time than me, please clean this up.... And although the inclusion of a criticism section is certainly justified given past controversies, others have provided much more eloquent criticisms than Schonemann. --Crusio 14:48, 10 September 2007 (UTC)

The lede of the article describes twin research as well-accepted and -regarded by the academic community. I believe this is true, but these are very powerful statements with regard to giving the reader the impression that these studies and their implications are "established science". I think the article would be well-served by sources backing up the claims that a consensus exists in the academic community. Jess (talk) 17:21, 8 July 2013 (UTC)

Separation studies
As I understand it (and no, I can't supply a reference, but I think I saw this on Pharyngula quite recently), the overlap between (the era of having orphaned twins raised separately) and (the era of proper scientific study) is quite, quite small. As in, the total number of such pairs known is around 80 (or maybe that's the number of individuals) -- and no new pairs are being created. Once a proper ref. to support this can be located, I think this should definitely be added to the article. DS (talk) 15:19, 7 April 2012 (UTC)
 * And what are the findings of the studies on these approximately 80 pairs/individuals? As these would presumably be the highest quality studies, this is of great interest.--greenrd (talk) 08:29, 14 June 2015 (UTC)

Strong influence versus Intermediate influence
In the last paragraph before the 'Contents' box, there is a sentence that says "Modern twin studies have shown that almost all traits are in part influenced by genetic differences, with some characteristics showing a strong influence (e.g. height), others an intermediate level (e.g. intelligence quotient) and..."

However, the general consensus on the heritability of height is that it is .8 (http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.0020041; and the general consensus on the heritability of IQ is that it is .75 to .85 (the comprehensive APA study "Intelligence: Knowns and Unknowns" http://psycnet.apa.org/?&fa=main.doiLanding&doi=10.1037/0003-066X.51.2.77 and the meta-study "Genetic Influence on Human Psychological Traits" http://cdp.sagepub.com/content/13/4/148). So They both have the same level of heritability.  Therefore, I'm going to change the sentence to something more accurate.  I'm going to change "intelligence quotient" to "personality traits".  The general consensus on the heritability of personality traits is that they are around .5 heritable.  That fits the term "intermediate" far better.  I think I've laid out a very sound a logical reasoning for this change, so if anyone has any objections, please don't just change it back, but lets talk about it here first. — Preceding unsigned comment added by 68.80.141.131 (talk) 16:15, 8 July 2015 (UTC)


 * There are a lot of sources that disagree with your proposed edit, and they are better sources (secondary sources rather than one primary source), so I've reverted the edit. What do standard textbooks on human behavior genetics say on this point? -- WeijiBaikeBianji (talk, how I edit) 17:14, 8 July 2015 (UTC)

Textbooks absolutely say that the heritability of IQ is between .75 and .85. The APA's study is one of the most comprehensive studies done on the subject, and the meta-study I cited is also extremely detailed and cites a large number of studies. Not to use the argument from authority, but I have a masters in psychology and this is taught, at least at the University of Vermont. I don't recall the name of the textbooks used, but it's the general consensus. — Preceding unsigned comment added by 68.80.141.131 (talk) 17:49, 8 July 2015 (UTC)

Also, one last comment - even if you disagree on the consensus of the heritability of IQ (which I don't see how you could), the fact that the American Psychological Association's most comprehensive study and a respected meta-study both say that heritability is between /75 and .85 suggests that we should at least avoid using it as an example since there is, at the very least, a disagreement. — Preceding unsigned comment added by 68.80.141.131 (talk) 17:57, 8 July 2015 (UTC)