attr_reader question

I'm a bit confused on how ruby makes the jump of replacing a function

def [method] ;
   '@arg1' ;
end 

to

just attr_reader :arg1

In the first, you have the option of naming your method whatever you want. And while the examples usually read

def ['method_name'];
 @'methodname' ;
end

It seems a bit of a jump to use att_reader to make that assumption that your method matches the argument name.

Can anyone clarify this for me? Apologies if the logic seems a bit off, but I'm confused on what would seem like a potentially hazardous jump in code. Feel free to ask me to clarify.

Best, Andrew Carr

Sryz for the odd formatting. I put in the semi-colons to show end of lines since the text viewer changes my original input's formatting.