A central tenet of rational decision-making is logical consistency across decisions, regardless of the manner in which available choices are presented. This assumption, known as invariance, which is a fundamental axiom of normal decision-making theory, is challenged by a wealth of empirical data. Kahneman and Tversky originally described this deviation from rational decision-making, which they termed as the framing effect. Framing effect refers to the phenomenon that people respond differently to the same problem with different but objectively equivalent descriptions, which is a hot area of research for behavior decision-making. Researchers put forward a series of terms to distinguish different kinds of valence framing effects, such as risky choice framing effect, attribute framing effect and goal framing effect. At present, advanced noninvasive neuroimaging techniques, like fMRI allow researchers to directly observe brain activities while subjects are finishing framing tasks, which contributed greatly to our understanding of brain mechanisms of framing effect. Results from neuroimaging indicate framing effect may be a joint result of emotional processing and cognitive processing. Furthermore, emotional processing is controlled by amygdala, while anterior cingulate cortex (ACC) and prefrontal cortex (PFC) play dominant roles in cognitive processing. PFC may incorporate inputs from the amygdala, which represents the motivational value of choices. This allows PFC to integrate and evaluate the incentive value of predicted outcomes in order to guide future behavior. Findings from various studies support a model in which the OMPFC evaluates and integrates emotional and cognitive information, thus underpinning more rational (i.e., description-invariant) behavior. Although neuroimaging data show an association between amygdala and the framing effect, this correlation does not imply causation. A lesion study found that framing effect persisted even in the patients who had bilateral amygdala lesions, indicating a complementary systems in generating framing effect. Here, we purpose that insula may partly contribute to the persistence of framing effect found in bilateral amygdala lesions patients. Results from different studies also suggest that functions of ACC and PFC in framing effect may vary by different context or frame type. Specifically, an fMRI study found participants reveal framing effect with activation in PFC when frames involve life-death decision problem. Others showed that PFC and ACC might partly contribute to framing effect rather than resist it when using different types of frames as experiment stuff. They found participants were susceptible to framing, and the levels of activity in ACC and PFC are linear related. Besides, results from lateralization showed that framing effect could be influenced by different thinking styles of brain hemispheres. Specifically, a significant framing effect could be found when right, but not left, hemisphere processing was initially enhanced. The results supported differential processing of contextual information, which could be regarded as the best explanation of lateralization on framed information. When it is presented to the right-contextual hemisphere, a significant effect of framing would be found; but when the same messages were presented to the left-inferential hemisphere, little or no framing effect would emerge. In order to get a full understanding of faming effect, further researches are expected to integrate neural mechanisms and theory explanations, explore its domain specificity and its origin.