The study compared two methods, Ridge and Lasso, to deal with problems in linear models used in biology. These methods help when variables are related, leading to inaccurate results. The Lasso method was found to be more effective in reducing errors compared to the Ridge method, especially with more variables and larger sample sizes. Ridge is better when there are more samples than variables, but it can't completely eliminate errors like Lasso can.