INTRODUCTION: Cross-modal plasticity has been extensively studied in deaf adults with neuroimaging studies, yielding valuable results. A recent study in our laboratory with deaf-blind children found evidence of cross-modal plasticity, revealed in over-representation of median nerve somatosensory evoked potentials (SEP N20) in left hemisphere parietal, temporal and occipital regions. This finding led to asking whether SEP N20 changes are peculiar to deaf-blindness or are also present in sighted deaf children. OBJECTIVE: Assess cross-modal plasticity in deaf child cochlear implant candidates using neurophysiological techniques (visual evoked potentials and median nerve somatosensory evoked potentials). METHODS: Participants were 14 prelingually deaf children assessed in the Cuban Cochlear Implant Program. Flash visual-evoked potentials and SEP N20 were recorded at 19 scalp recording sites. Topographic maps were obtained and compared to those of control group children with normal hearing. Analysis took into account duration of hearing loss. RESULTS: Topographic maps of flash visual-evoked potentials did not show changes in deaf child cochlear implant candidates. However, SEP N20 from right median nerve stimulation did show changes from expansion of cortical activation into the left temporal region in deaf children aged >7 years, which was interpreted as neurophysiological evidence of cross-modal plasticity, not previously described for this technique and type of somatosensory stimulus. We interpret this finding as due in part to duration of deafness, particularly related to handedness, since expansion was selective for the left hemisphere in the children, who were all right-handed. CONCLUSIONS: Cortical over-representation of SEP N20 in the left temporal region is interpreted as evidence of cross-modal plasticity that occurs if the deaf child does not receive a cochlear implant early in life-before concluding the critical period of neural development- and relies on sign language for communication.