Most methods currently available to measure intellectual development in college students are either unreliable or prohibitively expensive and time consuming. To circumvent these difficulties, we have developed Cogito, a software package which used a neural network to find patterns in "noisy" paper-and-pencil data and relate them to the Perry or Reflective Judgment models of intellectual development. If we are successful, Cogito will provide an inexpensive and reliable alternative to current interview or paper-and-pencil tests. Our work to date shows a high correlation between interview and Cogito results.