Nice to have a visual challenge for a change although this one is perverse. Another possible solution:

Collect all the straight lines segments and all arc segments from any intersecting circles in the first two columns. Then in the third column connect them together. Under this rule No. 3, triangle, would be the correct answer.

You know a winter in Norway can do strange things to the mind.

The answer is 1.

My reasoning (in spoiler) is incomprehensible, so seriously, don't read it....

the correct answer is the number 2. the first figure of every line contains the elements that the figures of the second and third column (for every line) don't share.

i'm not really seeing it. what is the full set of elements? can you walk through applying your rule to rows 1 and 2 in more detail?

After seeing it again 3 years later i can now see another pattern as well. Consider the number of regions in each figure (ie a circle creates 2 regions in the plane, the inside the circle and the outside. A set of lines crossing does not produce any additional regions (with area) so only one exists. A circle that has a cross inside that doesnt touch the circumference is still 2 areas, the outside of the circle and the inside same as a circle alone.

According to this property going left to right we have on rows 1,2

4 regions, then 2 regions and finally 1.

8 regions, 4 regions and finally the square 2.

(the pattern is declining number of regions defined left to right)

the last row now is

3 regions, then 2 regions and the last now has to be 1 region less than 2 so only choice 1 is consistent, the others all define more than 1 regions.

I still have to see a simpler explanation but i am sure there is one. However in my mind the just presented pattern is also simple enough.

This is why its all bs without explaining your arguments.


That said i still insist that the one that spots the easiest pattern faster than the rest of us is exhibiting (if it happens over many similarly tough questions consistently) a higher form of intelligence in that type of projects/pattern recognition etc.

So yes the IQ test may not be accurate but it still will prove true that over many questions those that score very well are no way in hell anything other than very smart and a very smart person will be almost impossible to score bad as well but may not score as high as they deserve because of issues like above in some of the questions where their brain focuses on other patterns that may not be the standard easiest ones. For me even paying attention to complicated enough patterns but which remain simple in their definition is also equal respect kind of intelligence than the one that spots the after the fact most straightforward pattern. And of course eventually the only thing that matters long term (from a practical perspective of the value of that intelligence) is the one that goes after as many patterns as possible and registers many of them and the path their brain takes to higher complicated patterns that ultimately are easy to define is also an indication of good intelligence and persistence/curiosity.


PS: I also notice the first response by FBandit talked also about closed areas which is related to what i said its the same plus 1 basically.

There are 3 with 0 circles and 3 with 2 circles, so there must be 3 with 1 circle, so the answer is choice 2. Easy game.

I think the idea is to use both dimensions, but if you're going to ignore one dimension, I might as well ignore both as I did. But I can use both dimensions if I note that the number of circles decreases or stays the same going across, and the number of lines increases or stays the same going down, so choice 5 would maintain that pattern. That would also maintain the pattern that the number of lines increases or stays the same going across.

Yeah i also didnt like the left to right row by row de facto choice to see patterns and wanted for a while to see more global (or both dimensions) patterns but i gave up. Since you spotted one though that makes reasonable sense (and usually has to be the correct because the tests are addressed to a broad population of people that are not necessarily mathematically inclined to see complex patterns, so the easier to be accepted by a broader public is their intended choice likely) i asked myself if there is something else too that can pass the global test but still fail the most people argument (just to maintain my light criticism of IQ tests in that regard lol) and i came up with the axis of symmetry thing (hopefully i didnt miss anything);

I can separate the patterns according to how many different axes of symmetry they each have and i see there are 3 with 0 axes (figures with eg matrix coordinates (1,2),(1,3),(2,2)), 2 with 1 axis ((1,1) and (2,1)), 1 with 2 axes (3,1), 2 with 4 axes ((2,3) and (3,2)) and so we are missing an example of 3 axis (we have so far examples for 0,1,2,4) and this is provided by the equilateral triangle choice 3. Now all 9 cover examples from 0 to 4 axes.

Did you notice, masque, that I didn't piss you off about this sort of thing about 3 years ago?

Thats because you talked about me not liking carnage. And you were not architecturally designed rejectionist for the sport of it lol. I dont like to piss off people at all either. Not because i am afraid of anything. Fack them all if they are aholes and they dont like me and they prefer bs games and eternal teasing and sarcasm. I like my jokes to be neutral and to never involve the other person in a negative manner. I will never start the attack. I dont want a world where we do these things to each other. I prefer a world where we agree on 7 things and disagree on 3 and when we talk about those 3 we do it as friends not as punk teens at high school or m/fers politicians or people on cable news shows, and we build on those 7 and the 1.5 that result out of the constructive disagreement.

You will never see me gang up on anyone about math or physics in these threads.

I still insist that IQ tests and IMO in particular type of problems will always find amazing brains at the top of the scores. Also the smart people will not score badly in these tests but may not get a clear read on certain cases and may require further testing. Ultimately real hardcore problems is where you see who is the boss, the fighter or the little kid that could (probably all the same person at different stages of their lives). (ok IMO is very very different and they will score bad there, even the decent brains, if they have no experience but still 7/42 is a good start lol, just different scale but 20+ is more like what a 140+IQ with interest in math would do).

My point has always been that smartly designed tests and puzzles and math problems will always be able to find the smart people if given multiple chances. If the math ones dont its because someone has messed them up as kids/students to not like math or to have had the chance to like it on their own with proper stimulation of their own curiosity.

I never offered IQ tests more respect than their limitations deserve but they still can perform some function that correlates well with ability of people to handle complex problems later when properly educated. And of course intelligence is super complex. And sometimes even complex deep thinking may fail simpler tests because it follows different routes. After all Einstein wasnt the smartest of his generation. Far from it. But he had the secret to synthesis. And so did Feynman. In the end if you do not have enough IQ its extremely hard to do anything in sciences or math even if you are hard worker and love them. But sure enough it takes persistence and other qualities and IQ alone, however you measure it, will not be the only factor and often will prove irrelevant without proper education and guidance to engage proper projects with potential for creativity.

You can max out getting this wrong, this one was out of the rest of the test level so it's hard to know how obvious the answer is and you can end up with a too complex answer easily.

It could be that that the rule is that in each box the number of crossed line segments must equal the number of curved figures in each box. So the only answer that satisfies that rule is 3.Triangle

Row 1 Col 1 two circles so two crossed lines
Col 2 one circle so one crossed line
Col 3 four semicircles so four crossed lines

Row 2 Col 1 two circles and two ellipses=four curved figures so four crosses
Col 2 two curved arcs so two crossed lines
Col 3 zero curved figures so zero crosses

Row 3 Col 1 two circles so two crossed lines
Col 2 one circle so one crossed line
Col 3 Must be the triangle (zero crosses) since all other choices fail the rule

the figure of the first line contains elements that are NOT present in the second and in the third lines. at the same time, it doesn't contains elements that, instead, are present (even if in decomposed way) in the figures of the second and third column.
for instance, in the first figure of the first line there are a great circle and a small circle, since they are not shared from the two following figures, but there are not the oblique line that is shared instead. in the first figure of the second line there is not the square, since it is shared, but there are two arched lines because they are present only in the second figure.
sorry for my english. i'm from italy.

Lol. It is a timed test.

Choice 4 is one circle as well.

I think I finally got it.

It is a "simple" addition problem.

Only straight lines count, so:

0 + 1 = 1
0 + 4 = 4
1 + 2 = 3

So the correct answer is 3.

The pattern is that there is no pattern. Thus the answer is 1 because it doesn't make any sense.