Atmospheric pressure chemical ionization mass spectrometry (APCI-MS) or proton transfer reaction mass spectrometry (PTR-MS) has been in use frequently for in vivo aroma release studies. In APCI-MS, protonated water clusters formed from moisture in the expired air are used as reagent ions. Yet the influence of a change in relative water content in the ionization gas on the data collected have been rarely studied and conflicting results were obtained. In this context our first objective was to study the impact of water content of the expired air on the level of aroma release measured by APCI-MS and to compare with the results obtained by PTR-MS. Otherwise, Le Quéré et al. showed in vitro a competition between aroma compounds whatever the ionization method used (APCI or PTR). This competition could be influenced by the chemical characteristics of aroma compounds and their respective amount in the mixture. In this context our second objective was to check in vivo the reality of this competition phenomenon. In a first experiment, 12 subjects with various levels of water in the expired air were selected. Aroma release during consumption of fresh cottage cheese flavoured with ethyl propanoate (EP) at 25 ppm or nonan-2-one (NO) at 8 ppm or a mixture of these two compounds was followed by APCI-MS and PTR-MS. Results showed no significant correlation between detection level and quantity of aroma release and the water content of expired air whatever the MS-method used. In a second experiment, aroma release of twelve subjects was followed by APCI- and PTR-MS during their consumption of eight fresh cottage cheeses flavoured either with EP at 0, 10 or 20 ppm or by NO at 0, 4 or 10 ppm and all possible binary mixtures. Confirming previous studies obtained in vitro, our results showed a competition between the two aroma compounds studied whatever the MS-method used (APCI-MS or PTR-MS).