An associative brain-computer-interface (BCI) that correlates a peripherally generated afferent volley with the peak negativity (PN) of the movement related cortical potential (MRCP) induces plastic changes in the human motor cortex. The aim of the current study was to compare the effectiveness of this intervention when the MRCP PN time is pre-determined from a training data set (BCIoffline), or detected online (BCIonline). Ten healthy participants completed both interventions in randomized order. The mean peak-to-peak motor evoked potential (MEP) amplitudes were significantly larger 30 min after (277 ± 72 µV) the BCI interventions compared to pre-intervention MEPs (233 ± 64 µV) regardless of intervention type and stimulation intensity (p = 0.029). These results provide further strong support for the associative nature of the associative BCI but also suggest that they likely differ to the associative long-term potentiation protocol they were modelled on in the exact sites of plasticity.