Background: Amongst the techniques to assess olfactory functions, discrimination and cued identification are those most prone to the influence of odour short-term memory (STM). Discrimination task requires participants to detect the odd one out of three presented odourants. As re-smelling is not permitted, an un-intended STM load may generate, even though the task purports to assess discrimination ability. Analogously, cued identification task requires participants to smell an odour, and then select a label from three or four alternatives. As the interval between smelling and reading each label increases this too imposes a STM load, even though the task aims to measure identification ability. New method: We tested whether modifying task design to reduce STM load improve performance on these tests. We examined five age-groups of participants (Adolescents, Young adults, Middle-aged, Elderly, very Elderly), some of whom should be more prone to the effects of STM load than others, on standard and modified tests of discrimination and identification. Results: We found that using a technique to reduce STM load improved performance, especially for the very Elderly and Adolescent groups. Comparison with existing methods: Sources of error are now prevented. Conclusions: Findings indicate that STM load can adversely affect performance in groups vulnerable from memory impairment (i.e., very Elderly) and in those who may still be acquiring memory-based representations of familiar odours (i.e., Adolescents). It may be that adults in general would be even more sensitive to the effects of olfactory STM load reduction, if the odour-related task was more difficult.