The self-healing behavior of asphalt materials is increasingly recognized as a promising strategy to extend pavement life and reduce maintenance costs. However, research in this area remains fragmented, lacking a unified, critical synthesis of fundamental mechanisms and evaluation methods. This review paper, the first part of a two-part study, addresses this gap by consolidating current knowledge on self-healing in asphalt materials. The article first explores the theoretical principles that govern healing phenomena, including capillary flow and molecular diffusion, as well as viscoelastic phenomena (thixotropic recovery and steric hardening) that can inflate apparent healing if not controlled. It then examines how internal factors (e.g., binder chemistry, oxidative aging, and air-void content) and external factors (e.g., temperature, rest periods, and traffic loading) influence the healing potential. Special focus is given to multiscale evaluation methods, ranging from rheological recovery tests at the binder level to assessments of stiffness recovery, fatigue resistance, and fracture at higher scales. Emerging nondestructive methods, including computed tomography and acoustic emission, are also reviewed. By integrating dispersed findings into a coherent framework, this work contributes to the development of more reliable and standardized healing assessment methods, laying the scientific groundwork for Part 2, which will address advanced self-healing strategies, practical implementation, and environmental and economic evaluation. • Maps the dominant wetting-flow-diffusion mechanisms governing asphalt self-healing across scales. • Quantifies how temperature, rest time, prior damage, and aging influence the healing rate and recovery. • Benchmarks mechanical and imaging methods, clarifying what each measures and how to compare indices. • Separates true healing from viscoelastic, thixotropic, and steric-hardening effects that inflate gains. • Provides reporting guidelines to harmonize protocols and improve reproducibility across studies.
Loureiro et al. (Wed,) studied this question.