<p>This study investigates the feature importance of return and volatility connectedness between Islamic sectoral indices and meme assets across quantiles. Using daily data from August 2020 to March 2024, we adopt a multistage approach combining the quantile extended joint connectedness framework, robust OLS regression, and tree-based machine learning of feature importance. The findings reveal that Islamic indices act as net return transmitters in bearish and bullish markets, while meme assets are net receivers; both are disconnected in normal conditions. In volatility spillovers, Islamic indices lead during bearish and normal markets, whereas meme assets dominate during bullish periods. Feature importance analysis shows that the global financial stress index is the most influential driver of return spillovers, while cyber-attacks emerge as the primary determinant of volatility spillovers. These findings highlight that spillover dynamics, and thus portfolio diversification benefits, vary significantly across market states and should be incorporated into risk-sensitive asset allocation strategies.</p>