Stable solutions of an integro-differential equation (known as “Amari equation”) have been proposed as a model of a neural population representation of remembered external stimuli.  In this talk I will present the study of the conditions that guarantee the existence and stability of multiple regions of high activity or ‘‘bumps’’ in a one dimensional, homogeneous neural field with localized inputs. These multi-bump solutions represent the core of an original dynamic field model of fast sequence learning that was developed and tested in a robotics experiment.