The purpose of this study is to examine repertory grid methodology as a way of capturing the cognitive levels and contradictions of teacher candidates about the limit subject in mathematics. For application, 5 prospective math teachers were chosen from the second-semester calculus classes in the department of mathematics education at a metropole university. Four of them participated in the study from start to finish and the data were collected in four sessions. Repertory grids were elicited in the first and last sessions by interviewing individually. The relationships between constructs were determined by calculating difference scores and correlation scores (Cohen, 2000; Bannister, 1968). The relationship figures were formed by cluster analysis to show the relationships between constructs. The report focuses on affects of experimental sessions to 2 participants’ informal models which involve notions like “reachability of limits”, “closeness” etc. The results of the study revealed that the repertory grid methodology is successful on bringing up the cognitive levels and contradictions of paricipants and determining the critical points of the subject. Their informal models about limit subject were affected by the sessions designed to alter them.