We investigate the key factors that shape the dynamic evolution of Day-Ahead spot prices of seven European interconnected electricity markets (Austria, Hungary, Slovenia, Romania, Bulgaria, Greece and Italy), with emphasis on their price surges and discrepancies during the period 2022-2024, that challenge the reliability and efficiency of the European target model. The high differences in the prices of the two groups, has generated political reactions from the countries that ‘suffer’ from these price discrepancies, expressed with different ways (e.g. a noticed reaction is the letter of the Greek Prime Minister sent to European Commission President). To ‘reveal’ the whole path of surging prices (from north to south), we employ combination of Machine Learning (ML) approaches in learning the causal structure of this phenomenon. Local, causal structures learning (LCSL) and Markov Blanket (MB) learning are combined to ‘lift the blanket’ that covers the ‘true structure’ of the path of causalities, responsible for the price disparity. Markov Blanket Learning is useful for identifying key fundamental variables but should be combined with causal structure learning to uncover true causes of price surges. Finally, we compute the correlation curves of rolling volatility of spot prices as well as of cross-border transfer availabilities (CBTA) identified as crucial factors by MB and LCSL, of all markets, to study their volatility spillover (a tool to detect the entire path of volatility propagation from the upstream to downstream SEE countries)…