In the modern era of psychology, researchers have used experiments to increase understanding of human memory, leading to the discovery of many memory phenomena. One example is the generation effect, where self-generating information often improves later memory for that information compared to reading. However, general assertions about memory such as the generation effect are often limited by boundary conditions, or situations where these effects no longer hold true (e.g., generating does not lead to better memory than reading). These boundary conditions indicate that memory is context sensitive, yet too often contextual experimental factors are neglected when researchers are designing studies and interpreting empirical data. In this article, we develop a contextual framework of the generation effect that describes experimental conditions that lead to increased or decreased memory performance by considering the interactions between four key experimental factors: encoding task, memory test, materials used in the experiment, and subject abilities. We present testable predictions generated by the framework, highlight the flexibility of this framework, and discuss avenues for future research. Overall, we argue that this framework can more fully account for a wide range of findings on the generation effect by expanding on existing multifactor theories attempting to explain the effect and its boundary conditions. This framework illuminates the need to investigate higher-order interactions between 4 key experimental factors to advance our understanding of the complex nature of the generation effect and human memory.