It is known that anxiety has a lot of comorbidity such as depression and may cause older subjects to be more prone to anxiety. Animal studies have also observed that older rats have lower movement in different situations, somewhat like anxiety symptoms (avoidance and freezing). The anxiety level of rodents is commonly evaluated using the elevated plus maze test, and videos are often used to record and quantify the activities of animals during the test. Furthermore, aging can reduce the frequency and activity of locomotion-induced hippocampal theta rhythm, and these reductions are also reflected in sensory–motor integration and cognition. Thus, directly analyzing the hippocampal theta rhythm and the quantity and quality of physical activity (PA) during the anxiety test can help elucidate the mechanisms underlying the anxiety-like behaviour. We explored alterations in the hippocampal theta rhythm, PA, and behavioural consistency (cross-correlation with 8-s window length) from triaxial acceleration vectors and subsequently examined the effects of age during the elevated plus maze test. We hypothesized that middle-aged rats would exhibit more anxiety, which is correlated with a reduction in behavioural consistency and hippocampal theta frequency during the elevated plus maze test. The middle-aged rats had a lower open-arm frequency and more low-frequency theta oscillations. The middle-aged rats had a lower mean power frequency (Frq) of the hippocampal theta rhythm, PA, and behavioural consistency. The Frq of the middle-aged rats was positively correlated with behavioural consistency. The results indicated that behavioural consistency could distinguish age-related anxiety-like behaviour. Behavioural consistency can serve as an index for the quality of movement and be used to distinguish age-related anxiety.