Existing methods may be characterized in several ways: according to the subjects studied (e.g., clinical population vs. healthy subjects); according to the functions studied (e.g., language, motor, etc.); according to type of tasks used (e.g., detection, identication, or discrimination); chronologically (from the oldest to the most recent); or according to the techniques used. Obviously, some overlap is unavoidable and, regardless of the criterion, one method may t into several categories. This is not an exhaustive list, but it includes brief descriptions of the methods most frequently used in neurolinguistic research on clinical and healthy subjects, as well as those that are, due to their invasiveness, costliness, or relative obsoleteness, used less frequently. Many methods used with healthy subjects are also applicable to clinical patients, such as neuroimaging methods. Since there are advantages and disadvantages to every single one of them, most authors agree that a combination of several different methods on the same subjects will yield the most reliable results. This chapter is limited mainly to the methods acceptable in work with humans, so different lesion methods, genetic and/or pharmacological manipulations, deprivation, and similar methods are not included. Their descriptions may be found in neuroscientic literature (Gazzaniga, Ivry, & Mangun, 2002; Judaš & Kostovic´, 1997; Purves et al., 2001; Sternberg, 2003).