The present research infers aspects of spatial attention frommovement to targets (and preferably not to foils) of a mouse-controlled cursor on a computer monitor. The long-term goalis a data-rich and rapid assessment technique that can be usedto diagnose individual and clinical deficits of attention. Theaim of this present research is validating the approach usinga college population of subjects. In the experiment, partici-pants attempt to move a cursor toward three spatial positions atwhich targets appear rapidly but at irregular times, and attemptto inhibit movements toward foils appearing at those positions.We assume that cursor movements toward a position indicatesattention has been directed toward that position. A modifiedHidden Markov Model (HMM) uses five sources of evidence,all based on parameters to be estimated, to predict the timevarying movement of attention: four aspects of cursor move-ment and a probability that attention will move from one timeinterval to the next. Five minutes of data are used to estimateparameters for each subject that produce a predicted attentiontrajectory which best matches what the subject is instructed todo. These parameters are used to predict the attention trajec-tory for the remainder of the hour of testing. The predictionsof attention movements are then matched to the appearance oftargets and foils to infer such components of attention as abil-ity to respond to targets vs foils, times to do so, and changesin these components over time. The results illustrate a promis-ing approach to assessment of attention that could likely beemployed for both adults and children in clinical settings re-quiring short testing periods.