It has recently been suggested that the analytical methods that have been developed to date for the determination of acrylamide (AA) may underestimate the concentration of AA in certain foods, because significantly higher results were obtained upon extraction of the food matrix under alkaline conditions. The present study employs food (potato, rye) and chemical model systems to better understand the tentative release of AA under high pH extraction conditions. The experimental design is based on the generation of AA in an environment containing an AA-isotopomer, and by comparing the ratio of AA, respectively the AA-isotopomer, after extraction at pHs 7 and 12. The results show that the additional AA released is not due to improved extractability of AA from the food matrix, and should therefore be regarded as an extraction artefact. Strongly alkaline conditions seem to induce net formation of AA from water-soluble precursors formed during thermolysis.