Epilepsy and neuropsychology have enjoyed a very special relationship, with epilepsy being the first disciple to employ formal cognitive assessment when defining surgical effectiveness—postoperative seizure freedom associated with significant cognitive decline was considered a treatment failure (Loring, 2010). Following the development of magnetic resonance imaging (MRI), expansion of video-EEG (electroencephalography), refinement of intracranial monitoring, and implementation of functional measures including positron emission tomography (PET) and single photon emission computed tomography (SPECT), neuropsychology has transitioned from being employed primarily as a diagnostic procedure to one that in combination with other clinical variables is used prognostically to identify patients at risk of significant postoperative cognitive decline. Patients considered at increased risk for cognitive decline are identified based upon the degree to which their clinical variables (including neuropsychology) deviate from expected patterns associated with an established seizure onset. Baxendale and Thompson (2010) describe a “paradigm shift” in neuropsychology following Saling’s critical review of the evidence for material-specific memory function in temporal lobe epilepsy (Saling, 2009). Saling asserts that the verbal/nonverbal material-specific memory dichotomy has been historically conceptualized as fully independent constructs that localize to the left and right hippocampi, respectively. This view paints a picture with a rather large brush, but he correctly observes that there is increasing appreciation of the task-specificity inherent in neuropsychological assessment. Therefore, all verbal memory tasks are not equivalent or interchangeable measures of mesial temporal structure thought to be critical for “binding” associations. Hippocampal function, he argues, is best assessed using paired associate learning of unrelated word pairs, with prose passage memory or word list learning being more related to lateral structures that are related to semantic processing. The sensitivity of verbal list-learning tasks, a common approach to testing verbal learning, to left temporal epilepsy is negated when adjusting for language abilities (Hermann et al., 1992). One of the lessons of functional imaging is that cognitive abilities depend on an intact distributed network, and disruption at multiple levels in that network may be reflected in decreased neuropsychological function. This does not negate, however, that risk of impairment may differ depending on where that network is perturbed, especially when applied on an individual patient basis. Increasingly, neuropsychology will need to establish common metrics for clinical assessment rather than simply testing common cognitive constructs, which will facilitate data pooling across centers. Unlike for IQ, in which the Wechsler scales are considered the gold standard, there is no generally agreed upon approach to assessing memory, language, or other important neuropsychological constructs. Because of inconsistent clinical patient characterization (including neuropsychology), the National Institute of Neurological Disorders and Stroke (NINDS) embarked on the Common Data Elements initiative to develop data standards to “harmonize data collection across clinical studies” (NINDS 2010). This broadens clinical neuropsychology’s role beyond the surgical setting, and will help standardize the characterization of epilepsy syndromes, disease progression, or treatment emergent risks or benefits. Common metrics will also facilitate the determination of external validity of specific neuropsychological test results by providing large sample sizes on which meta-analytic approaches can be applied. As noted by the Baxendale and Thompson (2010), however, continuing development of new cognitive markers in focused areas should not be neglected. For example, Hamberger has demonstrated that although visual confrontation naming is often considered the clinical standard for assessing naming ability, there are meaningful distinctions between visual confrontation naming and auditory responsive naming (Hamberger et al., 2005, 2010). Therefore, there needs to be sufficient flexibility in assessment approaches to permit the development of novel techniques that eventually may be shown to be superior to existing clinical methods. Together, a common clinical core and novel assessment approaches will propel neuropsychology in the age of imaging toward a greater understanding of brain–behavior relationship in epilepsy patients and thereby improve clinical care. I confirm that I have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. I have no conflicts of interest to disclose. Salary support was provided, in part, by NIH R01038455 and the Epilepsy Foundation.