High-throughput electrophysiological recordings, using multichannel silicon probes, allow simultaneous registration of neural activity from multiple brain regions. However, the difficulty of determining the precise localisation of recording sites within neural subcompartments creates challenges for data analysis and the interpretation of experimental results. While electrode localisation can be achieved to some extent by post-mortem histological reconstruction of the probes, it can be improved by real-time discrimination of the electrophysiological features of the signal: evoked potentials triggered electrophysiologically display stereotypical depth profiles which can be used to discriminate the organization of the recording area relative to electrode position.To this end, we developed an automatic and unsupervised electrode localisation strategy using a customised Matlab user interface. Analysing the waveforms of evoked potentials, channel localisation is assigned by merging information of waveform clustering and template matching to standard potentials. Using the hippocampus as a model, we stimulated the Schaffer collaterals of adult, freely behaving male rats and simultaneously recorded evoked potentials from the retrosplenial cortex and dorsal hippocampus using silicon probes bearing 32 linearly arranged electrodes. Post-mortem histological probe reconstruction was then conducted to validate the accuracy of our algorithm-based electrode localisation. Our results confirmed that this approach offers a reliable means to determine electrode localisation that can be used in studies of intra- and trans-structural neural information transfer and of synaptic transmission using multichannel silicon probes.Supported by a scholarship to TT from the German Academic Exchange Service (www.DAAD.de)