Q18 - A large study of people

[dictronic110]

Can someone help to break down this question? Thanks!!

[ohthatpatrick]

Sure thing!

Question Type: Flaw

Argument Core:
Conc - Pineal gland produces less melatonin as it ages
(why?)
P1 - Large study of 65-81 insomniacs showed that more melatonin helped
P2 - Melatonin is produced by the pineal gland

Okay, where did this conclusion come from? Where did we talk about how much melatonin the pineal gland produces?

How did the author get from the results of the study to this claim?

He must be assuming that "since giving insomniacs melatonin HELPED their insomnia, their insomnia must have been caused by insufficient melatonin."

Does that have to be true?

If I give a sad person $1000 and make them happier, does that mean their sadness was caused by not having enough money?

Of course not! Maybe they were say because their favorite NBA team just got eliminated from the playoffs but a sudden gift of $1000 is enough to distract them and perk them up.

So we could say that a flaw the author makes is "assuming that something that TREATS a certain problem was CAUSING the problem in the first place".

Another way to approach Flaw (and other Assumption Family questions) is simply to debate the conclusion.

The author claims the pineal gland produces less melatonin as it ages.

So we have to argue that the pineal gland produces the same (or more) as it ages.

AUTHOR: oh yeah? well then how come insomniacs got better when we gave them more melatonin? doesn't that suggest that they didn't have enough to begin with?

US: Maybe. But if Tylenol helps you get rid of your headache, does that mean your headache was caused by too little Tylenol in your bloodstream? We don't know the insomnia is caused by too little melatonin. But let's assume for a second that the problem IS caused by too little melatonin -- do we know that the pineal gland produces less as it ages? Did these people have insomnia when they were young? Maybe they've ALWAYS had too little melatonin. They might just have bad pineal glands, and this study has nothing to do with the effects of aging.

Let's check out answers:

(A) This sounds kinda tempting. It has causal wording. What "effect of an action" was there? The action of ingesting melatonin had the effect of helping insomnia. Did the author illicitly conclude that we had people ingest melatonin with the INTENT of helping their insomnia?

This answer deals with whether or not the experimenters intended to help insomnia by giving melatonin. Who cares?

We're debating whether or not the pineal gland produces less melatonin as it ages. Eliminate.

(B) This would tempt people because the conclusion mentions the manufacturers of melatonin supplements. But to RELY on their opinion means that their opinion was THE premise.

That's not at all accurate. The premise was the large study. We don't know who conducted the study. And the results of the study are not opinions. The author's argument relies only on the results of the study and knowledge that the pineal gland produces melatonin and relates to the sleep cycle. Eliminate.

(C) This old classic flaw? Next thing you know they'll offer me the ol' Circular Reasoning. Two Different Meanings and Circular Reasoning pop up in answer choices all the time and are almost always wrong. Did we identify a term that was used in two different ways? Nope. Eliminate.

(D) Causal language again ... did we have an effect of a phenomenon? Yes, the phenomenon of "taking melatonin" had the effect of "helping insomnia". Did the author get those backwards? Did "helping insomnia" really have the effect of "taking melatonin"?

That's crazy talk! I don't even know what that means. (D) is code for "reverse causality". That comes into play with correlation -> causality arguments.

Our causal issue here was with the reasoning of, "Since more X helps the problem, too little X must be causing the problem."

(E) Unrepresentative sample? Hmmm. Didn't think that was the problem. It IS a "large study". But wait --- it's a study of 65-81 year olds and the conclusion is about the pineal gland "as it ages".

Maybe we could object that this study doesn't help us analyze the pineal gland "as it ages" since it only looks at "aged people".

That's a bit weird, but it's kinda like the objection we had when we said "What if these people have ALWAYS had insomnia? How does this study show us that nowadays they have less melatonin in their system than previously?"

In order to draw a stronger conclusion about the pineal gland "as it ages", we would want to do a study that looks at insomniacs of all age groups.

If melatonin was only helping the older people, THAT would suggest that as you get older you need some more melatonin.

If melatonin was helping ALL age groups, then we wouldn't draw any conclusion about the pineal gland producing less M as it ages. We would probably just think "melatonin is an effective treatment for insomnia, and it looks like MAYBE insomnia is caused in part by having too little melatonin."

This is a tricky correct answer. Note that what will screw up most of us on this problem is that we are very likely to see the initial problem with this logic: "just because more X makes the problem go away doesn't mean that too little X was causing the problem in the first place".

We might call this the Primary Flaw. However, if you pretend for a second that we DO know that too little melatonin causes the problem, you STILL can find a Secondary Flaw, which is "how do we know the problem gets worse with age?"

This is typical of really hard Flaw problems -- the correct answer doesn't reward the Primary Flaw most of us initially notice ... you have to go along with some of the author's sketchy moves to get to ADDITIONAL problems with his thinking.

[dictronic110]

Hi ohthatpatrick, thanks very much!! This was my first actual LSAT and I had absolutely no idea about this question on the exam. This question has been lurking in my head for several months. Before I take my third (hopefully last!) LSAT in a few days, I want to convince myself that I understand it properly. Your thorough analysis has inspired me to think about this question from a new perspective!


But I'm not sure that the author has to assume insufficient melatonin caused the insomnia in the first place. The premise is that those old folks with insomnia have been helped by more melatonin. The conclusion is a correlation between aging and decreasing melatonin. The author doesn't need to know what caused the insomnia in the first place. The only issue about the old folks' insomnia/ melatonin for her is that these old folks don't have a ton of melatonin, which can be inferred from the premise. Because had they have enough melatonin, their insomnia wouldn't have been helped by more melatonin.


So I don't think the primary flaw is that since X helps Y, insufficient X caused Y. But once we zero in the conclusion, there is a huge gap in the reasoning. Just because folks are old and have little melatonin, we can't infer a correlation between aging and amount of melatonin. Since we don't know what amount of melatonin they have when they're 20s, 30s, 40s and so on. We can't infer from a point of affair any comparison, not to say correlation.


Another potential issue with the sample is that the conclusion is sweeping for all people, but the sample only have old people suffering from insomnia. So in order to justify the conclusion, the study not only needs to know the amount of melatonin when these old folks were young, but also the amounts of melatonin of other folks who don't suffer from insomnia and those of when they were young.


Since the LSAT typically gives us correlation in the premise and gets to causation in the conclusion, the difficulty for this kind of question is that I always think they have some implied causation that goes wrong while in fact the only fault is about the correlation in the conclusion.

The nearest example I can think about is PT59 S3 #20, where the premise gives one point of affair and concludes another point of affair in the other end of the spectrum. It doesn't need to have anything to do with causation but it definitely needs to assume a correlation which wasn't legal. And that's what the right answer choice exploits.

Another example with correlation in the conclusion is PT 56 S2, #14, where there seems to be something about causation going on while that's actually not the case.

I think in order to get these question right in a real exam setting, we need to be rock solid on the foundation about correlation/causation to know when the author is talking about causation and when she's not, and focus on the conclusion.

Please let me know where my analysis goes wrong. Thanks!!

[ohthatpatrick]

I think you made a lot of great points and distinctions (I particularly like the one about our conclusion being about the whole population, not just insomniacs).

But I disagree with one big thing:
"The only issue about the old folks' insomnia/ melatonin for her is that these old folks don't have a ton of melatonin, which can be inferred from the premise."

You CAN'T infer that from the premise. That was the whole point of my 'primary flaw' discussion.

You can't infer, just because Tylenol helps a headache, that the person suffering from the headache had lower-than-normal amounts of Tylenol in their blood when the headache started.

If you give a sad old person a free Tesla sports car and it makes them happier, can you infer that the old person USED to have a Tesla and lost it?

Of course not.

Benefiting from the addition of something doesn't allow us to infer that someone previously suffered from a deficiency.

And the author is definitely assuming that the old insomniacs suffered from a deficiency of melatonin ... his conclusion is offering a causal explanation for why they would have less melatonin.

Problem 1: we don't know they have less melatonin
Problem 2: even if we DID know they have less melatonin, we don't know that reduced-pineal gland functionality is the cause (maybe these old people are taking a heart medication whose side effect is removing melatonin from the bloodstream).

I would just caution you against thinking Flaw questions need to only have one real flaw. Many of them have several attack points, and our flexibility and willingness to consider those various attack points can be crucial if the correct answer choice doesn't seem to be going after the attack point we considered.

[dictronic110]

I kind of get it. Thanks!

[jwms]

Correct me if I misread ohthatpatrick's explanation, but the flaw that I thought this question was honing in on was the fact that the study was of people aged 65-81 and suffering from insomnia.

That narrow scope of study is even more problematic than merely the age thing. What if these insomniacs had melatonin issues/pineal gland issues caused by something else? It might not be age-related at all. Because we haven't compared like with like (meaning: 65-81 years olds that are insomniacs and 65-81 year olds that are NOT insomniacs), we have no way to isolate aging as a causal factor here.

[layamaheshwari]

Sure thing!

Question Type: Flaw

Argument Core:
Conc - Pineal gland produces less melatonin as it ages
(why?)
P1 - Large study of 65-81 insomniacs showed that more melatonin helped
P2 - Melatonin is produced by the pineal gland

Thank you for your detailed explanation, Patrick, but I feel we sidestepped the real, basic, and quite apparent flaw in this stimulus.

Conclusion: It is correct that the pineal gland produces less melatonin as it ages.
Evidence: A study of 65-81 year olds suffering from insomnia were supplied with extra melatonin and that alleviated their insomnia.

Wait a minute. These were old people suffering from insomnia! That makes them different from old people in general. Perhaps old people without insomnia have perfectly fine pineal glands that don't produce less melatonin. Maybe it's the insomnia that's reducing melatonin, not ageing pineal glands. We can't isolate the cause, and then extend that to the whole population from this sample.

Thus, the argument relies on a sample that does not represent the general population. (E) is correct.