The same animal can look like a sprinter or a wanderer depending on the timescale
The clearest explanation in the evidence is statistical rather than behavioral. Researchers tracked the marsupial Didelphis aurita at two scales: daily paths recorded with a spool-and-line technique, and distances estimated over weeks using radio-tracking. The daily paths were well described by a "truncated Lévy flight" — a movement pattern with many short steps and a few long ones — while the distances over weeks looked like Brownian motion, the ordinary random diffusion of a particle [3].
When the researchers built a model from the daily data, it converged toward that random-diffusion behavior after roughly one week. Their interpretation is that the two apparent "movement phases" can arise purely from how the data are aggregated, without needing separate ecological drives operating at different scales [3]. In plain terms, intermittent-looking movement at close range may partly be an artifact of the window through which it is measured, not necessarily a distinct strategy the animal switches into.
Cover and open ground change how far each step goes
For Mexican woodrats translocated across burned and unburned patches, movement was not explained directly by how severely the area had burned. Instead, vegetation structure and composition mattered most: step length was shorter where there was more bare ground, and squared displacement was higher in areas with more logs [5]. The authors suggest the selection for logs and avoidance of dense vegetation may relate to auditory concealment and ease of travel, though they present this as a possible explanation rather than a demonstrated cause [5].
That pattern fits the intuition that a small mammal crossing exposed ground makes short, cautious hops between cover, while a log-rich route allows longer, more confident moves. The study measured path tortuosity — how much a path twists — as an indicator of microhabitat selection, foraging, and perceived predation risk, so the stops and starts are being read as responses to what the animal encounters, not just to its own body size [5].
Smaller animals within a species may also be the more active movers
Body size shows up in a different way in urban rats. Using Rhodamine B, a non-toxic biomarker that leaves a fluorescent mark in whiskers, researchers found rats moving up to 90 meters in urban slums, and smaller rats traveled more actively than bigger ones [2]. That is a comparison within one group of small mammals rather than a general law about all small species, and the study was designed to measure distance traveled for rodent control, not to explain intermittent movement [2].
Other small-mammal work in the supplied evidence measures activity and movement without addressing whether it is intermittent. Camera trapping at Sydney's urban fringe found black rats were the most active small mammal in residential yards, with activity higher in yards next to bushland, and only about 6% of rats crossed the urban fringe [4]. A study of rainforest fragmentation and disturbance notes that small mammal movement can be intermittent, but its abstract gives no mechanism for why [1]. Taken together, the evidence supports scale and habitat as the best-grounded explanations, while the question of a built-in reason for stopping and starting remains open.
Sources used in this answer
The effect of rainforest fragmentation and disturbance on small ground-dwelling mammals
The paper examines rainforest fragmentation and disturbance effects on small ground-dwelling mammals and notes that small mammal movement can be intermittent, without detailing a cause.
Using Rhodamine B to assess the movement of small mammals in an urban slum
Using Rhodamine B in urban slums, researchers found rats move up to 90 m and that smaller rats travel more actively than bigger rats.
Exploring the interplay between small and large scales movements in a neotropical small mammal
Daily paths of the marsupial Didelphis aurita fit a truncated Lévy flight, but aggregated over about a week they converge toward Brownian motion, so apparent movement phases can arise from the scale of data aggregation.
Activity and movement of small mammal tick hosts at the urban fringes of Sydney, Australia
At Sydney's urban fringe, black rats were the most active small mammal in residential yards, activity was higher in yards adjacent to bushland, and only about 6% crossed the urban fringe.
Movement response of small mammals to burn severity reveals importance of microhabitat features
Mexican woodrats took shorter steps with more bare ground and showed higher squared displacement where logs were more abundant, with burn severity acting through vegetation structure rather than directly.