Aphasia is an impairment that affects the production or comprehension of spoken, written, or gestured speech. Effective prioritization and optimization of available treatment methods for people with aphasia (PWA) requires an understanding of how the damaged system responds to different kinds of learning experiences, i.e., a theory of learning. This proposal continues development of a theory of learning for aphasia rehabilitation based on retrieval practice (RP) principles, powerful learning principles derived from basic psychological research. This renewal submission builds on the following findings concerning the treatment of the debilitating and pervasive word retrieval deficit in aphasia: (a) retrieval practice-based naming treatment, in which the PWA attempts to name an object followed by feedback, confers greater benefit to later naming of the same items compared to errorless learning, in which the name is provided for the object and repeated by the PWA; (b) correct retrievals during retrieval practice confer greater improvement compared to failed retrievals; (c) greater spacing between an item’s trials confers greater improvement. Products of a basic phase of research, these observations were derived from study designs involving the administration of a fixed number of trials per item, regardless of the mastery achieved during training. This renewal submission takes important next steps to broaden the clinical relevance of the RP theory by examining a retrieval practice-based naming treatment termed criterion learning. In criterion learning, each item’s assigned criterion level dictates the number of times that item is correctly retrieved before it is dropped from further training within a session. Criterion learning optimally incorporates the most potent elements of the RP theory thus far—spaced retrieval practice with emphasis on correct retrievals during training. By examining scheduling and dosing parameters, and feasibility of at-home delivery, the present studies will provide critical observations for optimizing criterion learning for treating word processing deficits in people with aphasia. Study 1 will identify the optimal combination of criterion level and number of training sessions for efficiently attaining naming mastery of functional, personally-relevant vocabulary for PWA with word retrieval deficit. Using a multiple-case, within- participant, multiple baseline design, Study 2 will measure early efficacy and use of the optimal dosage- schedule combination for mastery of functional vocabulary identified in Study 1. The two modules of treatment in Study 2 will involve in-lab practice and unsupervised at-home practice. Brain imaging acquisition will permit individual differences explorations in response to treatment. Study 3 will examine whether practicing items in semantically related groups enhances naming treatment benefit in a multiple-session training regimen using criterion learning. Study 4 will investigate whether naming t...