Abstract
Aperiodic, 1/f-like EEG activity is increasingly used to investigate neural dynamics in cognitive and clinical research, with applications ranging from experimental manipulations to clinical biomarker development. However, these applications assume, without systematic testing, that aperiodic parameters primarily reflect neural activity rather than non-neural sources. To address this critical gap, we systematically quantified how data quality and physiological artifacts influence aperiodic parameter estimation across two independent datasets (N=99 and N=103) using complementary methodological approaches. Poor data quality and ocular artifacts significantly increased aperiodic offsets and exponents, whereas muscular artifacts showed opposite effects. These spatially widespread effects reached magnitudes comparable to neurophysiologically meaningful group differences and were only partially attenuated by state-of-the-art preprocessing. Cardiac artifacts showed modest effects limited to the offset. Critically, regression-based statistical correction effectively mitigated artifact-induced biases. Our findings establish that differential artifact rates between experimental conditions or populations can substantially bias neural interpretations of aperiodic parameters and provide methodological guidelines for ensuring valid inference.