The olfactory working memory capacity (OWMC) paradigm is able to detect cognitive deficits in 5XFAD mice (an animal model of Alzheimer's disease [TG]) as early as 3 months of age, while other behavioral paradigms detect cognitive deficits only at 4–5 months of age. Therefore, we aimed to demonstrate that the OWMC paradigm is more sensitive and consistent in the early detection of declines in cognitive function than other commonly used behavioral paradigms. The prefrontal cortex (PFC), retrosplenial cortex (RSC), subiculum (SUB), and amygdala (AMY) of 5XFAD mice were harvested and subjected to immunostaining to detect the expression of β‐amyloid (Aβ). Additionally, we compared the performance of 3‐month‐old male 5XFAD mice on common behavioral paradigms for assessing cognitive function (i.e., the open field [OF] test, novel object recognition [NOR] test, novel object location [NOL] test, Y‐maze, and Morris water maze [MWM]) with that on the OWMC task. In the testing phase of the OWMC task, we varied the delay periods to evaluate the working memory capacity (WMC) of wild‐type (WT) mice. Significant amyloid plaque deposition was observed in the PFC, RSC, SUB, and AMY of 3‐month‐old male 5XFAD mice. However, aside from the OWMC task, the other behavioral tests failed to detect cognitive deficits in 5XFAD mice. Additionally, to demonstrate the efficacy of the OWMC task in assessing WMC, we varied the retention delay periods; we found that the WMC of WT mice decreased with longer delay periods. The OWMC task is a sensitive and robust behavioral assay for detecting changes in cognitive function.