Transcranial electric stimulation (tES) is a promising technique that has been shown to improve working memory (WM) performance and enhance the effect of cognitive training. However, experimental set up and electrode placement are not always determined based on neurofunctional knowledge about WM, leading to inconsistent results. Additional research on the effects of tES grounded on neurofunctional evidence is therefore necessary. Sixty young, healthy, volunteers, assigned to six different groups, participated in 5 days of stimulation or sham treatment. Twenty-five of these subjects also participated in MRI acquisition. We performed three experiments: In the first one, we evaluated tES using either direct current stimulation (tDCS) with bilateral stimulation of the frontal or parietal lobe; in the second one, we used the same tDCS protocol with a different electrode placement (i.e., supraorbital cathode); in the third one, we used alternating currents (tACS) of 35 Hz, applied bilaterally to either the frontal or parietal lobes. The behavioral outcome measure was the WM capacity (i.e., number of remembered spatial position) during the 5 days of training. In a subsample of subjects we evaluated the neural effects of tDCS by measuring resting state connectivity with functional MRI, before and after the 5 days of tDCS and visuo-spatial WM training. We found a significant impairment of WM training-related gains associated with parietal tACS and frontal tDCS. Five days of tDCS stimulation was also associated with significant change in resting state connectivity revealed by multivariate pattern analysis. None of the stimulation paradigms resulted in improved WM performance or enhanced WM training gains. These results show that tES can have negative effects on cognitive plasticity and affect resting-state functional connectivity.