The aim of this paper is to give an overview of the techniques and principles for rendering the apparent source extent of sound sources in 3D audio displays. We mainly focus on techniques that use decorrelation as a mean to decrease the Interaural CrossCorrelation Coefficient (IACC) which has a direct impact on the perceived source extent. We then present techniques where decorrelation is varied in time and frequency, allowing to create tempo ral and spectral variations in the spatial extent of sound sources. Frequency dependant decorrelation can be employed to create an effect where a sound is spatially split in its different frequency bands, these having different positions and spatial extents. We fi nally present results of psychoacoustic experiments aimed at eval uating the effectiveness of decorrelation based techniques for the rendering of sound source extent. We found that the intended source extent matches well the mean perceived source extent by subjects.