Q21 - In an experiment, subjects were shown a series

[ohthatpatrick]

Question Type:
Inference (must be true)

Stimulus Breakdown:
By guessing according to a pattern they thought they saw, subjects had a lower rate than if they had just guessed "top" every single time, since the image appeared at the top most of the time.

Answer Anticipation:
Can we combine any of these facts to derive some new idea? Inference answers usually reward our understanding of CONDITIONAL, CAUSAL, or QUANTITATIVE ideas.

It seems like Causality and Quantitative are on the menu. People tried to guess a pattern, which CAUSED them to have a QUANTITATIVELY worse outcome. We could also infer that, since the actual pattern was "it's on top most of the time" and since their guess at the pattern caused them to have worse-than-that accuracy, they clearly did not guess "it's on top most of the time" as their pattern.
Correct Answer:
D

Answer Choice Analysis:
(A) Very tempting, but there's no way to infer what people BELIEVED they saw. If you removed "they believed" from this answer choice, it's pretty compelling.

(B) No, this answer is trying to generalize from this experiment to life in general. There's no way we can take anything from this experiment and extrapolate a generalization that applies to all situations in which we're basing our guess on a perceived pattern vs. on the prior data point.

(C) Too extreme. "There was NO predictable pattern?" We know that it was on top most of the time. There may have been a discernible pattern, like "top, top, bottom". We don't know.

(D) YES! It's super weakly worded, so it deserves a lot of loving attention. We just have to prove that "at least once" someone guessed "bottom" but it was actually "top". What LSAT actually (masterfully) did is disguise one of the oldest Inferences in the book (Most A's are B + Most A's are C = Some B's are C). We know "most times were wrong guesses" and "most times were top", so we can derive "at least one time there was a wrong guess, it was at the top". That proves that at least one time someone guessed "bottom" but was wrong.

(E) Too extreme. We can't rank "the most rational" strategy. We only know that this would have resulted in more accuracy, not that it was therefore the most rational strategy.

Takeaway/Pattern: They snuck this one past me on a first read, but if we're reading for CONDITIONAL / CAUSAL / QUANTITATIVE wording on Inference questions, we might see the "less than half the time" and "most of the time" telltale ingredients of the "MOST + MOST = some overlap" inference.

#officialexplanation

[tw4jp]

Not sure about the explanation of answer choice A. The fourth sentence in the stimulus did mention they based on their guesses on patterns "they believed" they saw in the sequence. In answer choice A, is the word "any" wrong? Because sometimes the subjects guessed correctly about images at the top of the screen, and all of the guesses are based on what they believe.

[ohthatpatrick]

The answer is meant to be appealing because we know they were ...
BASING THEIR GUESSES ON PATTERNS THEY BELIEVED THEY SAW
and
NOT GUESSING THAT THE NEXT IMAGE WOULD APPEAR AT THE TOP

This answer tries to put those two ideas into a timeless conditional relationship:
"If they had been guessing next image at top, they would not have been basing their guesses on a pattern they believed they saw."

We can't judge a hypothetical about what they would or wouldn't have believed.

Consider: A bunch of subjects were told to be nice to the next person they met. Most of them complimented the person's clothing. They reported that they based their compliment on something they believed the person would appreciate hearing.

(A) would be saying
"If the subjects had always complimented something other than clothing, then they would not have been basing their compliment on something they believed the person would appreciate hearing."

We can't infer that. If some other possible source of complimenting had occurred to those people, they might still have thought they were basing it on what they thought the person would like.

Had the idea of guessing "always at the top" occurred to any of the subjects, they may well have still been basing it on a perceived pattern.

[peterbobolis]

Why must they choose from among only top or bottom appearances? What if we had occurrences and guesses on the right/left side of the screen? Then D need not be true, no?

[ohthatpatrick]

I think you're right.

But I don't think they were allowing for any left/right occurrences when they said,
"Subjects were shown a series of images, appearing usually at the top but occasionally at the bottom".

That was intended to be a binary choice, even if they didn't technically secure that with the wording they used.

[LolaC289]

(A) is really tempting! I think if the "...any pattern they believed they saw" is changed to "...any pattern they reported to be believing seeing", it'll be correct.

Because people can believe to see many different sequences every time they were shown a series of images. What's believed does NOT have to be the same each time. However, the pattern that they have reported believing seeing is certain, as reported specifically for this time.

Patrick's example is a good one. I'll add another one. In a sleeping experiment, subjects were shown monochromatic (single-color) pictures after they were asleep. They all reported seeing world famous paintings during sleep because they said they have dreamt about those paintings during their sleep.

(A) would be like saying: if they did not report seeing world famous paintings, they would not have been basing their reports on their dreams.

It doesn't have be true. People could have dreamed about different things every time they dream. They can be reporting seeing children paintings and still be basing on their dreams.

(D) however, is specific to this time. It would be like saying: some subjects reported seeing world famous painting, but were not correct.