The purpose of this work was to evaluate myocardial wall motion estimation methods for gated cardiac emission tomography using a dynamic, physical phantom. Wall motion has been increasingly used in 4D image reconstruction methods for improved image quality. A commercially available dynamic phantom was modified by attaching radioactive markers to the phantom myocardial wall. The markers were used to provide an independent measurement of the wall motion throughout the cardiac cycle during a gated SPECT acquisition. Then, without moving the phantom, and after the markers were allowed to decay to negligible levels, the myocardium was injected with 99mTc and a gated SPECT scan was acquired. Two wall motion estimation methods were applied to the gated SPECT data. The first method was a standard optical flow algorithm applied to OSEM reconstructions of the myocardial emission data. The second was a simultaneous image reconstruction/motion estimation algorithm. The error in the estimated motion fields was described by the average vector difference between the motion of the markers and the estimated myocardial motion. The simultaneous method gave more accurate motion estimates compared with the optical flow method, and this was more pronounced for larger motion magnitudes. The absolute differences were not great, and the practical implications of these findings are not certain. The physical phantom proved to be an effective tool for evaluating motion estimation methods.