A puzzle on the ASVAB

post by Hul-Gil · 2011-05-30T04:01:00.757Z · LW · GW · Legacy · 17 comments

 I was linked to this on another forum. No instructions were given, apparently - just this picture. What's the deal?

It seems to me the answer is clearly C, not A as the test indicates; and the members in the original thread appear to agree. However, attempted justifications of A have been made, none of which are very convincing to me - mainly because if there are no instructions and an obvious answer, there's not really any benefit for them to reward a different interpretation, which would almost certainly involve arbitrary assumptions regarding the rules they really want you to apply.

 Trick questions on exams seem to rely on failure to pay close attention to instructions, or insufficiently rigorously apply rules; when there are no instructions, what justification would anyone have for not choosing the most obvious interpretation? Any could be right!

What do the geniuses here at MoreRight think?

17 comments

Comments sorted by top scores.

comment by Miller · 2011-05-30T05:05:41.116Z · LW(p) · GW(p)

Kobayashi Maru!

C is almost certainly the correct answer. This image has been photoshopped or the 'correct answer A' references the next question, or some other similar mishap.

sample questions all point to C being the straightforward choice.

comment by j_andrew_rogers · 2011-05-31T07:20:25.721Z · LW(p) · GW(p)

The ASVAB is not an exemplar of careful correctness and it is not targeted at people for which that would be beneficial. When I took it many years ago there were a few questions with glaring ambiguities and questionable assumptions; I simply picked the answer that I thought they would want me to pick if I was ignorant of the subject matter.

I maxed the test.

The test is not aimed at intelligent, educated people. It is designed to filter out people of low intelligence. I've met many people that struggled to achieve 50%, something I used to find shocking. If there are a few technical ambiguities then that is of little consequence for its intended purpose. While there is some basic occupation recommendations based on the ASVAB, it is not designed to identify the significantly above average -- quite the opposite.

comment by Sniffnoy · 2011-05-30T04:21:08.775Z · LW(p) · GW(p)

...huh?

comment by knb · 2011-05-30T05:44:55.993Z · LW(p) · GW(p)

I agree the answer seems to be C. I assumed this was a mental rotation style problem, in which case the answer would be C.

Maybe the A and B are not supposed to represent specific points on the objects but the whole object? The question seems bizarre, though the ASVAB is highly regarded amongst psychometricians.

comment by Normal_Anomaly · 2011-05-30T21:52:43.272Z · LW(p) · GW(p)

Looking at it without either looking at your answer or noticing the so-called "Correct Answer," I figured it was C. It is C.

comment by komponisto · 2011-05-30T04:23:37.855Z · LW(p) · GW(p)

It seems to me the answer is clearly C, not A

I don't understand what the question is.

EDIT:

when there are no instructions

Oh, I see, so that was intentional... well, yes, I suppose I fail then.

What is this test, exactly?

comment by Manfred · 2011-05-30T04:17:54.631Z · LW(p) · GW(p)

Yeah pretty much.

EDIT: "Yeah pretty much" here means "yes, I agree that the answer is clearly C."

comment by Hul-Gil · 2011-05-30T04:33:30.743Z · LW(p) · GW(p)

Why did I get a -1 for this? I didn't make the puzzle. I'm just as confused by the lack of instructions.

This test, as I understand it, is the U.S. military "entrance exam". This may be from a practice version, unless they let you take it online. I'm guessing you're supposed to pick the answer that shows a line going from point A to point B.

Replies from: komponisto
comment by komponisto · 2011-05-30T04:43:51.147Z · LW(p) · GW(p)

I'm just as confused by the lack of instructions.

You should probably have made that clear at the very beginning of the post; when I read it I first assumed you had simply neglected to tell us the instructions.

This test, as I understand it, is the U.S. military "entrance exam"

In that case, I suspect it's probably supposed to be a test of physical intuition -- or more to the point, of whether you're the kind of person who spends enough brain cycles on intuitive physical modeling in order to suspect that's what's being tested.

In which case, A probably does make the most sense, since it seems to represent the most "equilibrium-like" (lowest potential energy) situation, assuming what is being represented is a box with an object tied to it via a string, viewed from the side (not above or below).

Replies from: Manfred, Hul-Gil
comment by Manfred · 2011-05-30T05:14:44.764Z · LW(p) · GW(p)

Why assume that the rectangle is stationary and the star moves? Why ignore how things are attached? Why not think of them falling out of an airplane and choose D as clearly superior?

I don't think "it's probably supposed to be a test of physical intuition" is all that plausible. So I just googled, and it looks like I'm right. It seems to fall into the "assembling objects" category.

Replies from: Unnamed, komponisto
comment by Unnamed · 2011-05-30T05:54:31.994Z · LW(p) · GW(p)

Googling "assembling objects" turned up another sample problem of the same format which includes these instructions:

Which figure best shows how the objects [...] will touch if the letters for each object are matched?

In other words, there are 3 objects (a star, line, and rectangle), and you are supposed to put them together at the places with the same label.

comment by komponisto · 2011-05-30T05:49:22.952Z · LW(p) · GW(p)

Why assume...? Why ignore...? Why not think...?

I don't know! Lord knows, I didn't design the test, and the burden certainly isn't on me to explain it. All I did was explain my best guess about what the question was. You don't have to agree. But I could ask an a similar sequence of questions in response to whatever interpretation you might propose -- as indeed, either of us could do in response to whatever the "correct" interpretation is (if it's something different).

I don't think "it's probably supposed to be a test of physical intuition" is all that plausible

Why not? It's a military exam, for goodness' sake. Intuition about how mechanical objects act under the influence of gravity seems to me like a perfectly natural thing for the military to be interested in testing.

So I just googled, and it looks like I'm right. It seems to fall into the "assembling objects" category.

Your link doesn't specifically support that. Maybe you meant to link somewhere where that particular question was specifically classified as falling into that category? If not -- if you were just making an assumption to this effect -- then I could equally well put it under "mechanical comprehension".

But even if it's actually in the "assembling objects" category, how does that refute my interpretation? "What happens when you put these two things together?" sounds like a perfectly good "assembling objects" question to me.

(And surely "assembling objects" requires some "mechanical comprehension" in any case?)

The question is awful. In no way should my revelation of the particular interpretation that happened to suggest itself to me be construed as an endorsement of the reasonableness of this question or any other part of the test.

Replies from: Manfred
comment by Manfred · 2011-05-30T06:25:57.991Z · LW(p) · GW(p)

But I could ask an a similar sequence of questions in response to whatever interpretation you might propose.

Agreed. But I asked those questions not because I think your reasoning was inherently wrong, but because I felt like you were making a post-hoc story to justify specifically the answer A. While some sort of problem-solving logic that happens to generate A provides fairly easy answers for those kinds of questions, generating A post-hoc does not help answer them.

I don't think "it's probably supposed to be a test of physical intuition" is all that plausible

Why not?

It would be a pretty bad test of physical intuition of the sort you describe, but a pretty good test of spatial reasoning, and I assume that the army people are competent at designing tests. If it was a test of physical intuition, why change how the objects were attached to each other in each answer? And so on and so forth.

Note: If you asked me "why would they change the relative positions in each answer if it was a test of assembling objects?" I would say "Because changing things around increases the need for spatial reasoning."

Your link doesn't specifically support that.

I thought the evidence was strong enough to say that pretty confidently, but you're right it's not overwhelming.

Additional note: if you look at the page Unnamed linked to, you'll see that the picture is all one piece, and the instructions are down below in text. If the question we got this picture from was similar, that may explain the disappearing instructions!

Replies from: komponisto
comment by komponisto · 2011-05-30T06:48:08.910Z · LW(p) · GW(p)

I asked those questions not because I think your reasoning was inherently wrong, but because I felt like you were making a post-hoc story to justify specifically the answer A

Yes, that's exactly what I was doing, openly and unashamedly. Recall that I had no idea what the question was until informed that it was a military exam. That information made A a more plausible answer than it would otherwise have been, for the reasons I explained.

I was proposing my best guess at the answer to the question "why is A the correct answer?" As such, my comments were not intended to strongly dispute the idea that there might be a better answer. (P(X|Y) vs P(Y|X) and all that, you know.)

comment by Hul-Gil · 2011-05-30T04:48:26.161Z · LW(p) · GW(p)

You should probably have made that clear at the very beginning of the post; when I read it I first assumed you had simply neglected to tell us the instructions.

Gotcha. I'll edit.

In which case, A probably does make the most sense, since it seems to represent the most "equilibrium-like" situation, assuming what is being represented is a box with an object tied to it via a string, viewed from the side (not above or below).

Really? I don't see that at all. Surely B would be more reasonable, in that case? On A, it'd sort of be hanging off to the side.

Replies from: jasonmcdowell, komponisto
comment by jasonmcdowell · 2011-05-30T05:37:21.823Z · LW(p) · GW(p)

C is the only answer where the line segment is touching the same spots indicated on the both objects. Point A is on the point of the star, point B is near the little box on the rectangle thing.

The rectangle thing is flipped vertically though (as if in 3D), rather than being rotated in the plane of the 2D drawing.

comment by komponisto · 2011-05-30T04:56:33.126Z · LW(p) · GW(p)

Surely B would be more reasonable, in that case? On A, it'd sort of be hanging off to the side.

In B, the pointed object is higher (i.e. farther from the ground), so the potential energy due to gravity (U = mgh) is greater.

ETA: In intuitive terms, the question seems to be: "Imagine you tie this pointed object to this box, and then release. What happens? I.e. to what state does this system you just created most 'naturally go'?"