LR: recognizing and mastering pattern recognition

dkb17xzx
Posts: 405
Joined: Wed Jan 20, 2010 6:25 pm

LR: recognizing and mastering pattern recognition

Postby dkb17xzx » Sat Mar 24, 2012 11:14 pm

So I've been browsing through quite a few threads on LR prep, and I keep coming across "pattern recognition." Anyone care to explain what exactly it means and how do you go about mastering it? What am I supposed to be looking for while doing this?

Thanks

collegebum1989
Posts: 323
Joined: Sat Jan 07, 2012 8:03 pm

Re: LR: recognizing and mastering pattern recognition

Postby collegebum1989 » Sun Mar 25, 2012 12:44 am

I believe people are referring to repeating logical patterns. Although the stimuli topics vary considerably, the underlying argument structure is consistent.

Take for example, a traditional Sufficient Assumption question, which usually involve phrase shifts since a sufficient assumption leads the premise-conclusion relationship to be valid under all cases. Ex. Premise: if A then B; conclusion: if A then C, a sufficient assumption is if B then C.

This pattern recognition translates well to detecting and anticipating incorrect answers as well. In Must Be True questions, there will commonly be mistaken reversals and mistaken negations, when the correct answer is the contrapositive of a conditional statement in the stimulus.

Finally, the pattern recognition also includes the different types of stimuli: fact set, argument, or two speakers. Each type has specific question types associated with them respectively. So then, after rounds of practice, you will be able to anticipate answers based of reading the stimulus and question stem. This makes choosing an answer easier and known as prephrasing.

User avatar
NoleMatt
Posts: 61
Joined: Tue Oct 25, 2011 4:48 pm

Re: LR: recognizing and mastering pattern recognition

Postby NoleMatt » Sun Mar 25, 2012 12:00 pm

collegebum1989 wrote:I believe people are referring to repeating logical patterns. Although the stimuli topics vary considerably, the underlying argument structure is consistent.

Take for example, a traditional Sufficient Assumption question, which usually involve phrase shifts since a sufficient assumption leads the premise-conclusion relationship to be valid under all cases. Ex. Premise: if A then B; conclusion: if A then C, a sufficient assumption is if B then C.

This pattern recognition translates well to detecting and anticipating incorrect answers as well. In Must Be True questions, there will commonly be mistaken reversals and mistaken negations, when the correct answer is the contrapositive of a conditional statement in the stimulus.

Finally, the pattern recognition also includes the different types of stimuli: fact set, argument, or two speakers. Each type has specific question types associated with them respectively. So then, after rounds of practice, you will be able to anticipate answers based of reading the stimulus and question stem. This makes choosing an answer easier and known as prephrasing.


This was a very helpful post, thank you.

The only patterns i'm really comfortable with are how the stimuli are set up, for example if it is an ascriptive stimulus, i know the flaw is going to be in the ascription, if it's a prescriptive stimulus, the flaw is going to be in the prescription. Another small one is on Inference questions, a majorty of the answers are limited in scope (a lot contain the word "most").

I'm interested to see what other patterns are explained ITT.

User avatar
Br3v
Posts: 4174
Joined: Mon Jun 13, 2011 7:18 pm

Re: LR: recognizing and mastering pattern recognition

Postby Br3v » Sun Mar 25, 2012 12:03 pm

NoleMatt wrote:
collegebum1989 wrote:I believe people are referring to repeating logical patterns. Although the stimuli topics vary considerably, the underlying argument structure is consistent.

Take for example, a traditional Sufficient Assumption question, which usually involve phrase shifts since a sufficient assumption leads the premise-conclusion relationship to be valid under all cases. Ex. Premise: if A then B; conclusion: if A then C, a sufficient assumption is if B then C.

This pattern recognition translates well to detecting and anticipating incorrect answers as well. In Must Be True questions, there will commonly be mistaken reversals and mistaken negations, when the correct answer is the contrapositive of a conditional statement in the stimulus.

Finally, the pattern recognition also includes the different types of stimuli: fact set, argument, or two speakers. Each type has specific question types associated with them respectively. So then, after rounds of practice, you will be able to anticipate answers based of reading the stimulus and question stem. This makes choosing an answer easier and known as prephrasing.


This was a very helpful post, thank you.

The only patterns i'm really comfortable with are how the stimuli are set up, for example if it is an ascriptive stimulus, i know the flaw is going to be in the ascription, if it's a prescriptive stimulus, the flaw is going to be in the prescription. Another small one is on Inference questions, a majorty of the answers are limited in scope (a lot contain the word "most").

I'm interested to see what other patterns are explained ITT.


Big thing with that is to look for the (most likely incorrect) answer that is nit narrow in scope. Generally it uses wording to strongly (al, every, none, etc) than the stimuli allows for

dkb17xzx
Posts: 405
Joined: Wed Jan 20, 2010 6:25 pm

Re: LR: recognizing and mastering pattern recognition

Postby dkb17xzx » Sun Mar 25, 2012 1:00 pm

collegebum1989 wrote:I believe people are referring to repeating logical patterns. Although the stimuli topics vary considerably, the underlying argument structure is consistent.

Take for example, a traditional Sufficient Assumption question, which usually involve phrase shifts since a sufficient assumption leads the premise-conclusion relationship to be valid under all cases. Ex. Premise: if A then B; conclusion: if A then C, a sufficient assumption is if B then C.

This pattern recognition translates well to detecting and anticipating incorrect answers as well. In Must Be True questions, there will commonly be mistaken reversals and mistaken negations, when the correct answer is the contrapositive of a conditional statement in the stimulus.

Finally, the pattern recognition also includes the different types of stimuli: fact set, argument, or two speakers. Each type has specific question types associated with them respectively. So then, after rounds of practice, you will be able to anticipate answers based of reading the stimulus and question stem. This makes choosing an answer easier and known as prephrasing.



This was incredibly helpful. Thanks a lot collegebum. I get the feeling (and maybe this is obvious) extensive drilling will lead to pattern recognition.

NoleMatt, I think we should start a thread that discusses pattern recognition for specific question types. I think it'll be incredibly helpful not just for us but for other taking the LSAT in the future.

B3V, thanks for your tip as well. It is much appreciated.


For any other posters - would you care to share pattern recognition strategies for specific question types? It'd be great to get this threat going.

User avatar
Clearly
Posts: 4166
Joined: Sat Feb 11, 2012 4:09 pm

Re: LR: recognizing and mastering pattern recognition

Postby Clearly » Mon Mar 26, 2012 12:55 am

Pattern recognition isn't a tip for the lsat... It IS the lsat.
The time needed to approach every argument on LR as it's own, individual process, requiring analysis would be greater then 35 minutes. I'm convinced those that struggle with LR timing do so strictly because they haven't yet gotten a hang for LR, and they approach each question as unique. Granted I suck at LG and am only average at RC, I'm consistently -1 per LR section, and I usually finish completely around 28 min. It's deeper then assumption questions. I'd be willing to bet that consistent 170+ scorers could read nearly every stimulus and tell you the question type without reading it...That's how repetitive the logic is. There will always be a handful of questions in every section in which the pattern doesn't immediately jump out at you, the beauty of recognizing all the other patterns is that you'll have 3 minutes or so to knock off each of these remaining questions.

Developing this pattern recognition takes practice and exposure to a significant enough sample size of questions. It is simply a matter of experience (provided you're studying properly). At the end of the day the pool of logical structures that make up stimuli, correct answer choices, and incorrect answers choices is only so big. In other words, there are only so many ways for a right answer to be right, and a wrong answer to be wrong. You (most likely) won't recognize every one of them, but when you get to that last parallel reasoning question that makes no sense...and you have 8 minutes left to fuck with it, you're pretty relieved.

Also, upon rereading the post, I have this to say...Please disregard the authoritarian nature of the rant, I'm no expert, I just get awfully fucking psyched about LR.

doclover
Posts: 36
Joined: Wed Feb 29, 2012 12:34 pm

Re: LR: recognizing and mastering pattern recognition

Postby doclover » Mon Mar 26, 2012 12:02 pm

Clearlynotstefan wrote:Pattern recognition isn't a tip for the lsat... It IS the lsat.
The time needed to approach every argument on LR as it's own, individual process, requiring analysis would be greater then 35 minutes. I'm convinced those that struggle with LR timing do so strictly because they haven't yet gotten a hang for LR, and they approach each question as unique. Granted I suck at LG and am only average at RC, I'm consistently -1 per LR section, and I usually finish completely around 28 min. It's deeper then assumption questions. I'd be willing to bet that consistent 170+ scorers could read nearly every stimulus and tell you the question type without reading it...That's how repetitive the logic is. There will always be a handful of questions in every section in which the pattern doesn't immediately jump out at you, the beauty of recognizing all the other patterns is that you'll have 3 minutes or so to knock off each of these remaining questions.

Developing this pattern recognition takes practice and exposure to a significant enough sample size of questions. It is simply a matter of experience (provided you're studying properly). At the end of the day the pool of logical structures that make up stimuli, correct answer choices, and incorrect answers choices is only so big. In other words, there are only so many ways for a right answer to be right, and a wrong answer to be wrong. You (most likely) won't recognize every one of them, but when you get to that last parallel reasoning question that makes no sense...and you have 8 minutes left to fuck with it, you're pretty relieved.

Also, upon rereading the post, I have this to say...Please disregard the authoritarian nature of the rant, I'm no expert, I just get awfully fucking psyched about LR.


This is very helpful!

User avatar
Br3v
Posts: 4174
Joined: Mon Jun 13, 2011 7:18 pm

Re: LR: recognizing and mastering pattern recognition

Postby Br3v » Mon Mar 26, 2012 12:08 pm

dkb17xzx wrote:
collegebum1989 wrote:I believe people are referring to repeating logical patterns. Although the stimuli topics vary considerably, the underlying argument structure is consistent.

Take for example, a traditional Sufficient Assumption question, which usually involve phrase shifts since a sufficient assumption leads the premise-conclusion relationship to be valid under all cases. Ex. Premise: if A then B; conclusion: if A then C, a sufficient assumption is if B then C.

This pattern recognition translates well to detecting and anticipating incorrect answers as well. In Must Be True questions, there will commonly be mistaken reversals and mistaken negations, when the correct answer is the contrapositive of a conditional statement in the stimulus.

Finally, the pattern recognition also includes the different types of stimuli: fact set, argument, or two speakers. Each type has specific question types associated with them respectively. So then, after rounds of practice, you will be able to anticipate answers based of reading the stimulus and question stem. This makes choosing an answer easier and known as prephrasing.



This was incredibly helpful. Thanks a lot collegebum. I get the feeling (and maybe this is obvious) extensive drilling will lead to pattern recognition.

NoleMatt, I think we should start a thread that discusses pattern recognition for specific question types. I think it'll be incredibly helpful not just for us but for other taking the LSAT in the future.

B3V, thanks for your tip as well. It is much appreciated.


For any other posters - would you care to share pattern recognition strategies for specific question types? It'd be great to get this threat going.


sorry about the typos lol *nit=not and al=all*

User avatar
Micdiddy
Posts: 2190
Joined: Tue Mar 20, 2012 3:38 pm

Re: LR: recognizing and mastering pattern recognition

Postby Micdiddy » Mon Mar 26, 2012 2:10 pm

One pattern I have recognized in LR:

When it asks which of the following most weakens the argument, most of the answers will actually strengthen the argument and vice-versa. They may do this for two reasons: 1. To prevent ambiguity. If one answer definitively most weakens the argument, then there can be no case that another answer is better, therefore the other answers must help the argument, be tangential, or I guess it's possible one could weaken the argument in a severely less effective way. Also 2: They want to trick people who misread the question and may pick the answer they think most strengthens the argument (though even if you did misread the question, the fact that 3 or 4 answers could be right should clue you in that you missed something).

dkb17xzx
Posts: 405
Joined: Wed Jan 20, 2010 6:25 pm

Re: LR: recognizing and mastering pattern recognition

Postby dkb17xzx » Mon Mar 26, 2012 3:55 pm

Clearlynotstefan wrote:Pattern recognition isn't a tip for the lsat... It IS the lsat.
The time needed to approach every argument on LR as it's own, individual process, requiring analysis would be greater then 35 minutes. I'm convinced those that struggle with LR timing do so strictly because they haven't yet gotten a hang for LR, and they approach each question as unique. Granted I suck at LG and am only average at RC, I'm consistently -1 per LR section, and I usually finish completely around 28 min. It's deeper then assumption questions. I'd be willing to bet that consistent 170+ scorers could read nearly every stimulus and tell you the question type without reading it...That's how repetitive the logic is. There will always be a handful of questions in every section in which the pattern doesn't immediately jump out at you, the beauty of recognizing all the other patterns is that you'll have 3 minutes or so to knock off each of these remaining questions.

Developing this pattern recognition takes practice and exposure to a significant enough sample size of questions. It is simply a matter of experience (provided you're studying properly). At the end of the day the pool of logical structures that make up stimuli, correct answer choices, and incorrect answers choices is only so big. In other words, there are only so many ways for a right answer to be right, and a wrong answer to be wrong. You (most likely) won't recognize every one of them, but when you get to that last parallel reasoning question that makes no sense...and you have 8 minutes left to fuck with it, you're pretty relieved.

Also, upon rereading the post, I have this to say...Please disregard the authoritarian nature of the rant, I'm no expert, I just get awfully fucking psyched about LR.



Helpful post...thank you. I have the opposite problem. I am able to pick out patterns in LG and I consistently get -1 and -0 on. It also has to do a lot with the fact that I do a lot of games and review them very thoroughly. I actually enjoy them.

Reading the responses above and other threads, I'm thinking pattern recognition comes intuitively after doing tens, if not hundreds, of questions of the same type and reviewing those carefully.

dkb17xzx
Posts: 405
Joined: Wed Jan 20, 2010 6:25 pm

Re: LR: recognizing and mastering pattern recognition

Postby dkb17xzx » Mon Mar 26, 2012 3:58 pm

Micdiddy wrote:One pattern I have recognized in LR:

When it asks which of the following most weakens the argument, most of the answers will actually strengthen the argument and vice-versa. They may do this for two reasons: 1. To prevent ambiguity. If one answer definitively most weakens the argument, then there can be no case that another answer is better, therefore the other answers must help the argument, be tangential, or I guess it's possible one could weaken the argument in a severely less effective way. Also 2: They want to trick people who misread the question and may pick the answer they think most strengthens the argument (though even if you did misread the question, the fact that 3 or 4 answers could be right should clue you in that you missed something).



Thanks for sharing. I definitely think it's a trick aspect - under time constrains and pressure, most of us are looking for answer choices that tie back to the argument core.


I don't know if you've gone through these videos, but they are very helpful: http://www.velocitylsat.com/online-course

User avatar
Clearly
Posts: 4166
Joined: Sat Feb 11, 2012 4:09 pm

Re: LR: recognizing and mastering pattern recognition

Postby Clearly » Mon Mar 26, 2012 9:36 pm

Speaking of pattern recognition I got fucked by assuming one today. Granted I went -2 on the LR section, one of the two I got wrong was the second question. I couldn't believe it. It was a point of disagreement question (I thought). Read the two arguments, figure out what their beef is, was thrown off by several decent answers, picked the one most likely to be the point of disagreement, and proceeded through the section. Come to find out upon grading it, it was a point of AGREEMENT question. It took me 5 min of staring at the question going "There is no way they disagree about that, it must be a typo" before I saw the word agree.

Point being, follow patterns, but be careful!

dkb17xzx
Posts: 405
Joined: Wed Jan 20, 2010 6:25 pm

Re: LR: recognizing and mastering pattern recognition

Postby dkb17xzx » Tue Mar 27, 2012 12:11 am

Wait...so you did not read the question? I remember you mentioned this:

...consistent 170+ scorers could read nearly every stimulus and tell you the question type without reading it...


is this what you attempted?

User avatar
Clearly
Posts: 4166
Joined: Sat Feb 11, 2012 4:09 pm

Re: LR: recognizing and mastering pattern recognition

Postby Clearly » Tue Mar 27, 2012 3:23 am

HAHA Nope. Just a stupid oversight! To be clear, I didn't say good scorers don't read the question, just that they could likely guess the question type based on the stimulus, as a way of establishing that patterns of arguments and logic are so often repeated in LR that they're easily detected with experience.

As for my oversight today, in my defense, I was in a hospital waiting room, chaos everywhere. What had thrown me off was the relatively argumentative structure of the argument/counter-argument... amidst the chaos I just scanned over "agree" and saw "disagree" :|
Oh well, -2 aint bad, but I thought coming out of it (even with distractions) it was easily a -0 or -1 section.

Now if I could just get consistent with LG and RC, I'm set!

User avatar
PDaddy
Posts: 2073
Joined: Sat Jan 16, 2010 4:40 am

Re: LR: recognizing and mastering pattern recognition

Postby PDaddy » Tue Mar 27, 2012 4:05 am

dkb17xzx wrote:So I've been browsing through quite a few threads on LR prep, and I keep coming across "pattern recognition." Anyone care to explain what exactly it means and how do you go about mastering it? What am I supposed to be looking for while doing this?

Thanks


The LSAT throws patterns in a variety of ways: topical (the recycling of topics such as global warming, the Amazon basin, the Michelangelo and the Sistine Chapel, city government expenditures, the ice age, etc.), logical (in terms of the structures of the arguments the LSAC employes on certain question types), vernacular, sentence structure and vocabulary, and wrong answer types. Certain wrong answer types re-occur in certain question types. Equally important is that the answer types correspond to the type of logic employed in the stimuli. Familiarizing yourself with these wrong answer types will allow you to answer questions with great speed and accuracy.

I know my TLS bretheren will expand on this, so I will refrain from delving deeper at this point. But doing many many tests will acclimate you to the various tricks employed by LSAC.

User avatar
Clearly
Posts: 4166
Joined: Sat Feb 11, 2012 4:09 pm

Re: LR: recognizing and mastering pattern recognition

Postby Clearly » Tue Mar 27, 2012 5:10 am

PDaddy wrote:
dkb17xzx wrote:So I've been browsing through quite a few threads on LR prep, and I keep coming across "pattern recognition." Anyone care to explain what exactly it means and how do you go about mastering it? What am I supposed to be looking for while doing this?

Thanks


The LSAT throws patterns in a variety of ways: topical (the recycling of topics such as global warming, the Amazon basin, the Michelangelo and the Sistine Chapel, city government expenditures, the ice age, etc.), logical (in terms of the structures of the arguments the LSAC employes on certain question types), vernacular, sentence structure and vocabulary, and wrong answer types. Certain wrong answer types re-occur in certain question types. Equally important is that the answer types correspond to the type of logic employed in the stimuli. Familiarizing yourself with these wrong answer types will allow you to answer questions with great speed and accuracy.

I know my TLS bretheren will expand on this, so I will refrain from delving deeper at this point. But doing many many tests will acclimate you to the various tricks employed by LSAC.


exactly.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: Yahoo [Bot] and 4 guests