ABSTRACT CONTEXT AND OBJECTIVE: Patients undergoing the same neuromodulation protocol may present different responses. Computational models may help in understanding such differences. The aims of this study were, firstly, to compare the performance of aphasic patients in naming tasks before and after one session of transcranial direct current stimulation (tDCS), transcranial magnetic stimulation (TMS) and sham, and analyze the results between these neuromodulation techniques; and secondly, through computational model on the cortex and surrounding tissues, to assess current flow distribution and responses among patients who received tDCS and presented different levels of results from naming tasks. DESIGN AND SETTING: Prospective, descriptive, qualitative and quantitative, double blind, randomized and placebo-controlled study conducted at Faculdade de Ciências Médicas da Santa Casa de São Paulo. METHODS: Patients with aphasia received one session of tDCS, TMS or sham stimulation. The time taken to name pictures and the response time were evaluated before and after neuromodulation. Selected patients from the first intervention underwent a computational model stimulation procedure that simulated tDCS. RESULTS: The results did not indicate any statistically significant differences from before to after the stimulation.The computational models showed different current flow distributions. CONCLUSIONS: The present study did not show any statistically significant difference between tDCS, TMS and sham stimulation regarding naming tasks. The patients’responses to the computational model showed different patterns of current distribution.