Question for those who've used Atlas LG's Logic Chain System

77to101
Posts: 22
Joined: Wed Sep 23, 2009 2:53 am

Question for those who've used Atlas LG's Logic Chain System

Postby 77to101 » Sun Jan 03, 2010 7:29 am

How do you go about mapping out compound conditional statements? The book lets us know that there are two special compound conditional statements that can't be broken up ( "if x and y then z" and "if x then y or z"), but then ignores them when explaining how to map out conditional statements. If there's a game with one of those, are we to just abandon the logic chain system? Thanks!

aether
Posts: 57
Joined: Wed Dec 09, 2009 12:25 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby aether » Sun Jan 03, 2010 1:53 pm

77to101 wrote:How do you go about mapping out compound conditional statements?

Every time I tried to map conditional statements into the logic chain, or build multiple logic chains, it always turned into a disaster. My advice: build the chain using all the non-conditional information that's given, but just write the conditional logic statement BESIDE the chain diagram.

If there's a game with one of those, are we to just abandon the logic chain system?

Don't abandon the whole system -- you still use it for all the non-conditional logic. But don't try to build conditional "switches" into your logic chain diagram. You'll regret it! 8)

User avatar
blhblahblah
Posts: 170
Joined: Mon Aug 27, 2007 10:54 pm

Re: Question for those who've used Atlas LG's Logic Chain System

Postby blhblahblah » Sun Jan 03, 2010 2:12 pm

aether wrote:
77to101 wrote:How do you go about mapping out compound conditional statements?

Every time I tried to map conditional statements into the logic chain, or build multiple logic chains, it always turned into a disaster. My advice: build the chain using all the non-conditional information that's given, but just write the conditional logic statement BESIDE the chain diagram.

If there's a game with one of those, are we to just abandon the logic chain system?

Don't abandon the whole system -- you still use it for all the non-conditional logic. But don't try to build conditional "switches" into your logic chain diagram. You'll regret it! 8)


This advice is only good for those who havn't mastered conditional logic. Linking conditional statements in undefined grouping games can sometimes yield huge inferences. At times, the whole game can be linked into one neat and tidy chain, allowing you to effectively steamroll the game in under 4 minutes.

Just keep track of the variables that are not in the chain and the minimum or maximum number of variables that must be included. The rest is cake, trust me.


FWIW, X-->A and B should, IMO, be drawn as two separate statements. The negation of either (or both) of A and B will yield the negation of X. Be careful with X-->A or B, however, as the negation of either (except for both) will not necessarily lead to the negation of X. For these statements, draw them out as is. If the game contains them, keep them separate from the main chain, unless you're absolutely sure.

aether
Posts: 57
Joined: Wed Dec 09, 2009 12:25 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby aether » Sun Jan 03, 2010 2:35 pm

blhblahblah wrote:
aether wrote:don't try to build conditional "switches" into your logic chain diagram. You'll regret it! 8)

This advice is only good for those who havn't mastered conditional logic.

Then perhaps you will help me "master conditional logic" by explaining how to represent a compound conditional statement on an Atlas-style logic chain diagram. I don't know how to do it, and neither did the Atlas instructors when I asked them during class. They are the ones who recommended writing it next to the logic chain diagram, which is the answer I gave to the OP.

Just to make sure that we're all on the same page: we are talking about how to represent a statement like "if the seminar is held on Tuesday, then either Sally is teaching Economics or else Bob is teaching Writing." I believe compound conditional statements like that one simply do not fit into an Atlas-style logic chain diagram.

Blahblah, how would you add this to the logic chain in a manner that made the problem easier to solve, rather than harder?

User avatar
blhblahblah
Posts: 170
Joined: Mon Aug 27, 2007 10:54 pm

Re: Question for those who've used Atlas LG's Logic Chain System

Postby blhblahblah » Sun Jan 03, 2010 2:48 pm

As noted earlier, compound statements do not mesh well into a big, master diagram.

Keep them separate.

My advice was mainly a response to this:

But don't try to build conditional "switches" into your logic chain diagram. You'll regret it! 8)


which is ill-founded, especially when there have been countless undefined grouping games in the past which've lent themselves to mass linkage.

Take game 3 in PT 45, for example:


Code: Select all

                -> ~T
              /       
U-->S-->W
              \
                ->~R-->Y


Now look how easy this game is. All of the variables listed in the rules have been neatly chained together. The rest of the game is cake.

~Y will yield R, ~W, ~S, ~U, for instance

User avatar
bluejayk
Posts: 181
Joined: Tue Jun 17, 2008 3:06 pm

Re: Question for those who've used Atlas LG's Logic Chain System

Postby bluejayk » Sun Jan 03, 2010 2:49 pm

It's now called "Kaplan's LG Logic Chain System".

aether
Posts: 57
Joined: Wed Dec 09, 2009 12:25 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby aether » Sun Jan 03, 2010 3:34 pm

blhblahblah wrote:My advice was mainly a response to this:
But don't try to build conditional "switches" into your logic chain diagram. You'll regret it! 8)

which is ill-founded, especially when there have been countless undefined grouping games in the past which've lent themselves to mass linkage.

Take game 3 in PT 45, for example:

Code: Select all

                -> ~T
              /       
U-->S-->W
              \
                ->~R-->Y

Now look how easy this game is. All of the variables listed in the rules have been neatly chained together. The rest of the game is cake.

Blahblah, I am trying to figure out whether you are honestly misunderstanding this entire thread, or whether you are pulling my leg. You seem sincere in terms of the information you present, but I'm having trouble believing because your posts are way off the topic and your tone is condescending ("ill-founded advice... good only for those who haven't mastered conditional logic.")

To recap: this thread is about using Atlas logic chain diagrams. That's the thread title, and the OP's whole question is how to fit compound conditionals into such diagrams. Right?

Your diagram (above) is not an Atlas logic chain, and has absolutely no resemblance to one. Do you understand the Atlas system? If so, then let's talk about it, since it's the topic of the thread. If not, then what are you doing trying to educate us on a topic you don't even understand?

You mentioned PT 45 G 3. That's a very poor example, because (1) it contains no compound conditionals and (2) it fits perfectly into the Atlas logic chain system. For your reference, here is the PDF for an Atlas logic chain diagram that solves that game: http://www.atlaslsat.com/forums/download/file.php?id=6

It feels to me (and please correct me if I'm wrong!) like you aren't talking about the Atlas logic chain system at all.

77to101
Posts: 22
Joined: Wed Sep 23, 2009 2:53 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby 77to101 » Sun Jan 03, 2010 4:17 pm

Thank you for your replies. I had a sneaking suspicion that mapping out the logic chain without the compounds, then just keeping special note of the compounds on the side was the way to go about it, but I was hoping there would be some magical way to fit them in. Guess not! In that case, I'm guessing that none of the mapped out logic chains will ever contradict the compound conditional statements that we're not including in the chains?

aether
Posts: 57
Joined: Wed Dec 09, 2009 12:25 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby aether » Sun Jan 03, 2010 4:29 pm

77to101 wrote:I had a sneaking suspicion that mapping out the logic chain without the compounds, then just keeping special note of the compounds on the side was the way to go about it, but I was hoping there would be some magical way to fit them in. Guess not! In that case, I'm guessing that none of the mapped out logic chains will ever contradict the compound conditional statements that we're not including in the chains?

The logic chain is just a visual way of representing the game's rules? If the rules (as visually represented by the logic chain) conflict with another game rule, there are two possibilities: either (1) LSAC made an unprecedented mistake by putting an obviously broken game into the LSAT exam, or (2) you made an error in drawing your logic chain.

I have great confidence in you, so I'll vote for option #1! :mrgreen:

User avatar
blhblahblah
Posts: 170
Joined: Mon Aug 27, 2007 10:54 pm

Re: Question for those who've used Atlas LG's Logic Chain System

Postby blhblahblah » Sun Jan 03, 2010 4:31 pm

No. Let's not talk about it. I don't care about the "Atlas Grouping System" or whatever gimmick name is ascribed to the notion of connecting conditional statements.

The fact of the matter is, connecting them, when possible, will yield a better setup and hence a more powerful command of the game.

I have agreed that compound conditionals do not lend themselves well to mass linkage, but all other sorts of conditionals do.*

Your advice to avoid mass linkage in fear of turning things into a "disaster" is something a logical weakling would say. If, however, your statement was limited to compound conditionals, then I'll agree for reasons mentioned above.


ed.
*Actually, now that I think about it, if you have a compound conditional like this:

If X, then either A is before B or B before C, but not both

C\
A-B

B-C
\A

And, if there are several other variables that lend themselves to mass linkage, then I would draw out the two above hypotheticals, and, within each, conjoin the mass chain, wherever possible.

A-E-F

C\
A-B
\E-F

B-C
\A-E-F

aether
Posts: 57
Joined: Wed Dec 09, 2009 12:25 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby aether » Sun Jan 03, 2010 4:35 pm

77to101 wrote:Thank you for your replies. I had a sneaking suspicion that mapping out the logic chain without the compounds, then just keeping special note of the compounds on the side was the way to go about it, but I was hoping there would be some magical way to fit them in.

Let me clarify one other point. While Blahblah and I went off on a tangent, he made one excellent point that shouldn't be overlooked.

He is correct that A --> B and C isn't really a compound statement, and should be broken up before you build your logic chain. That statement is exactly equivalent to A --> B plus the statement A --> C, and it's much more useful to list and diagram them separately.

Here are the two formats that should be split up:

A --> B and C (split into A --> B plus A --> C)
A or B --> C (split into A --> C plus B --> C)

The following two formats cannot be split up, and must be handled as compound conditionals:

A and B --> C (no split possible)
A --> B or C (no split possible)

Make sense?

aether
Posts: 57
Joined: Wed Dec 09, 2009 12:25 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby aether » Sun Jan 03, 2010 4:49 pm

blhblahblah wrote:No. Let's not talk about it. I don't care about the "Atlas Grouping System" or whatever gimmick name is ascribed to the notion of connecting conditional statements.

It's not a "gimmick name". It's a different method of diagramming, exactly as a pie chart is different from a bar chart (and each used for different things).

Someone who entered a thread labeled "questions about bar charts" and promptly posted a bunch of condescending-toned material about pie charts instead might just deserve some flamage... 8)

I have agreed that compound conditionals do not lend themselves well to mass linkage, but all other sorts of conditionals do.*

Then we agree. /handshake/

User avatar
Atlas LSAT Teacher
Posts: 283
Joined: Tue May 12, 2009 10:18 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby Atlas LSAT Teacher » Mon Feb 22, 2010 11:58 am

77to101 wrote:How do you go about mapping out compound conditional statements? The book lets us know that there are two special compound conditional statements that can't be broken up ( "if x and y then z" and "if x then y or z"), but then ignores them when explaining how to map out conditional statements. If there's a game with one of those, are we to just abandon the logic chain system? Thanks!

Yeah, these types of statements do stretch the chain bit (pardon the pun, if there was one). As you noted, we generally recommend writing those conditions to the side, and for two reasons: 1, they can make the diagram pretty messy -- and every diagram has its strengths and weaknesses, and messiness is the weakness of this one -- and 2, the LSAT rarely employs the contrapositive of such statements, and adding in those can be quite a burden.

That said, if there's only one of those statements and it doesn't seem that drawing it in will make your diagram a catastrophe, you can draw them in by connecting the two items that connect (i.e., with "If J then K or L", I'm talking about connecting K and L, but with "If M and Q then Z", we're talking about connecting M and Q). Then, where they connect, putting a circle with:
- "=1" for only 1,
- "1+" for an either (though not an "either, but not both", which would require "=1").
- or a "2" if both are required

If the game is all about compound conditionals, you can write in your elements in groups. So, if a rule says "If M and K then R", you could write "M + K" as an element on the In side of your diagram. Similarly, you could write something like "K or P" on the out, if only one of those is triggered. Here's an example where that comes in handy, http://www.atlaslsat.com/forums/june-20 ... -t209.html though it's the diagram that causes the most uproar against the chain!

I hope that helps.

JasonR
Posts: 421
Joined: Fri Jan 01, 2010 9:09 am

Re: Question for those who've used Atlas LG's Logic Chain System

Postby JasonR » Mon Feb 22, 2010 9:56 pm

blhblahblah wrote:No. Let's not talk about it. I don't care about the "Atlas Grouping System" or whatever gimmick name is ascribed to the notion of connecting conditional statements.

The fact of the matter is, connecting them, when possible, will yield a better setup and hence a more powerful command of the game.

I have agreed that compound conditionals do not lend themselves well to mass linkage, but all other sorts of conditionals do.*

Your advice to avoid mass linkage in fear of turning things into a "disaster" is something a logical weakling would say. If, however, your statement was limited to compound conditionals, then I'll agree for reasons mentioned above.


ed.
*Actually, now that I think about it, if you have a compound conditional like this:

If X, then either A is before B or B before C, but not both

C\
A-B

B-C
\A

And, if there are several other variables that lend themselves to mass linkage, then I would draw out the two above hypotheticals, and, within each, conjoin the mass chain, wherever possible.

A-E-F

C\
A-B
\E-F

B-C
\A-E-F



Good God, shut the fuck up already, you tool. You should have just disappeared after his last post.

He's not arguing against mass linkage. He's arguing against trying to map a specific kind of rule into a proprietary set-up structure advocated by Atlas.

Your condescending babbling is entirely off-topic, given that the fucking thread is so obviously specifically concerning the suitability of Atlas's system to the types of rules in question.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: Baidu [Spider], cherrygalore, jen203 and 10 guests