Objective: According to the Common Sense Model (CSM), cognitive and emotional representations influence recovery. We used groups of individuals grouping multiple cognitive representations (schemas) to predict the recovery process after total hip arthroplasty (THR). Methods: The aim of this prospective cohort study with three collection time points was to examine the significance of these schemas for functionality three and six months after THR. We assessed illness perception with the Illness Perception Questionnaire-Revised and the functionality with the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). Data were collected four weeks before (n = 317), three (n = 268), and six months after (n = 292) primary THR. Groups of individuals with the same schemas were identified using a two-step cluster analysis across cognitive representations. Controlling for WOMAC variables (functionality, stiffness, pain, and activities of daily living) before THR, we calculated the regression of schemas and emotional representations before THR on WOMAC variables after THR. Results: Before THR, two cognitive schemes were found: Schema One: medium identity, long duration, many consequences, low personal and treatment control, and low coherence; Schema Two: low identity, short timeline, low consequences, and high personal and treatment control. Patients with Schema Two had better functionality and lower pain and stiffness three months after surgery compared to those with Schema One. After three months, the influence of cognitive schemas was stronger on functionality than that of emotional representation (f²: 0.04 /0.02). Conclusion: Further study is needed to determine whether a psychological intervention can change Schema One.