Mild cognitive impairment (MCI) is an intermediate clinical state between normal aging and dementia, and MCI patients are the risk population for the development of Alzheimer's disease (AD). This article firstly introduced the international diagnosis criteria of MCI and its variations and modifications, and based on the previous findings, we proposed an ideal diagnosis model for MCI detection in which outmeasures included both patient's self-report, clinician's judgment and objective behavioral and neuropsychological tests as well as various biomarkers through advanced imaging techniques, cerebrospinal fluid analysis, or even urine endogenous formaldehyde analysis, both cross-sectional comparisons of an individual with the norms and longitudinal changes derived from multi-time point measurements were also involved. We reviewed the recent studies on the neuropsychological impairment characteristics of MCI and the corresponding measurement tools from both cognitive and neuropsychiatric aspects such as MMSE, MoCA, NPI and SCID, and various cognitive intervention methods for MCI were also reviewed. Future research is needed to extend the findings from experimental small samples into community-based large population, as well as from the cross-sectional studies into longitudinal studies, and to adopt multi-detection markers and multi-intervention methods to achieve the optimal effects of early detection and intervention of MCI.