Yetkin et al. (8) studied 28 patients with fMRI using motor tasks involving the Ÿngers, lips, and tongue, and also language tasks involving silent word-generation and counting.  ey reported agreement between sites of activity detected with fMRI and corresponding areas detected by means of intraoperative electrical stimulation, of 100% within 20 mm, and 87% within 10 mm. Petrella et al. (9) evaluated the impact of preoperative fMRI localization of language and motor

areas. Out of 39 patients with brain tumors, the fMRI results altered the therapeutic plan in 49% and enabled a more aggressive approach in 46%. Out of this group, 30 patients underwent surgery, and fMRI helped to shorten the surgical time in 60% of the cases. Fitzgerald et al. (10) assessed the sensitivity of fMRI compared with ECS for mapping of language areas and determined it to be 81% when the areas were required to be in contact, and 92% for the areas being within 2 cm of each other. In another example, Majos et al. (11) reported that the agreement in the location deŸned by fMRI and intraoperative mapping techniques was 84% for motor areas, 83% for sensory areas, and 98% when both kinds of activity are taken into account. Similarly, Arora et al. (12) state that in patients with epilepsy they found excellent agreement between language lateralization determined with fMRI and with the Wada test.  e agreement was 84% of patients (n = 40) with a reading task, 83% with an auditory task, 77% for a verbal ¥uency task, and 91% for the combination of tasks. However, they caution that it is important “to understand the neurophysiology of language processing, both in control subjects and in epilepsy patients, in order to better interpret the results from fMRI” (p. 2239). Bartos et al. (13) report that the fMRI sensitivity for localizing the primary motor cortex detected by ECS was only 71%, and therefore concluded that fMRI was not very reliable.  ey caution that fMRI should be used in combination with ECS, and this same recommendation is made by Xie et al. (14). Overall, it therefore appears from the scientiŸc papers that have been published to date that the use of fMRI for preoperative mapping shows great promise and has demonstrated beneŸts, yet should be used with caution and with a very good understanding of both the underlying physiology and the fMRI method.