The data stored in the table depends on the collation, as well as the type of column,
especially if the type is VARCHAR,
especially for some letters of Persian and Arabic languages.
personally, for Persian language, in older versions of sql-server, i prefer the "SQL_Latin1_General_CP1256_CI _..." collation.
the character 'ی' is used in this word 'زیست'
the character 'ي' is used in this word 'زیست'
in stand alone letter, you see difference but when it used in a word, visually it isn't appear
for a temporary solution without review,you must try something like this :
select * from tblLesson where Replace(Lesson,'ي','ی') like '%زیست%'
for clear view of characters see the picture :
clear view of characters