Urbanization and the effects of humans on the planet’s flora and fauna have exponentially accelerated since the onset of the industrial revolution, causing native species to decline in numbers or be lost completely at an unprecedented rate. On the other hand, the change in the environment created novel niches for species that are able to adapt, for example, by decreasing the chance of predation, offering new food resources, and often ameliorate climatic conditions (McKinney, 2002; Gross, 2018; Szulkin et al., 2020). Currently, our understanding of why some species are able to adapt to and thrive in urban environments (so-called “urban dwellers”) while others merely cope, is limited. Urban dwellers are hypothesized to possess specific traits which enable them to cope with challenges such as frequent anthropogenic disturbances, altered social environments, light and air pollution or noise and habitat fragmentation that are presented by urban habitats (McKinney, 2002; Sih, 2013).

Behavior is generally the fastest response to cope with environmental challenges and changes (Sih et al., 2010; Lowry et al., 2013). Therefore, behavior has been proposed to be an important component in adapting to human-altered habitats. Indeed, several behaviors have already been shown to differ between urban and rural populations of the same animal species. For example, enhanced problem-solving abilities, reflecting high rates of flexibility and innovation, are being recognized as significant contributors for animals to thrive in human-altered environments (Reader, 2003; Sih, 2013; Vrbanec et al., 2021). In addition, behaviors that allow animals to cope with novel stimuli, such as high levels of neophilia and boldness or low levels of disturbance tendencies, are often altered in urban populations (Tryjanowski et al., 2016; Sol et al., 2018; von Merten et al., 2022). In addition to such behavioral adaptations, several studies indicate that life-history characteristics such as developmental time, longevity, and emphasis on reproduction, are altered in animals living in urban environments. For instance, Calliphora vicina Robineau-Desvoidy, a forensically important blowfly, is common in southern England and can occur in both urban and rural areas as well as in transition areas. C. vicina has altered life histories in urban areas that often have higher temperatures than the surrounding rural areas. Between 16 and 20°C, females grow slower but to a larger final mass than males, while at 28°C, both sexes grow similarly but males reach a larger maximum body mass (Hwang and Turner, 2009). A recent meta-analysis showed that urban birds generally have higher adult survival rates and thus tend to have a longer life (Sepp et al., 2018). In addition, urban birds seem to shift to fewer offspring for which they care more. In general, however, and for taxa other than birds, knowledge on which traits differ between urban and rural populations, is still largely missing.

Traits facilitating urban-dwelling might adapt through different mechanisms. In general, it is assumed that different forms of phenotypic plasticity, such as within-generation (e.g., learning and behavioral flexibility) and intergenerational plasticity (e.g., grandparental and parental effects or epigenetic processes) play an important role in the initial steps of shifting to an urban-dwelling life-style and may prevent individuals from being eliminated by changing environmental conditions (Lowry et al., 2013; Miranda et al., 2013). Therefore, phenotypic plasticity might often precede micro-evolutionary changes in populations adapting to urban habitats (Miranda, 2017). Micro-evolution is a process involving changes in gene frequencies across several generations of a population and thus occurs over prolonged time periods, eventually leading to speciation processes (Boutin and Lane, 2014; Nicoglou, 2015). Given the paucity of scientific studies investigating adaptations to urban environments, actual evidence for micro-evolutionary changes during the urban adaptation process is currently very scarce.

In this study, we aim to investigate whether living closely together with humans has an impact on the microevolution of mouse behavioral and life-history traits such as boldness, stress-coping, growth, longevity, and emphasis on reproduction. Therefore, we conducted a common-garden experiment with descendants of three species of wild mice; Apodemus uralensis, Mus spicilegus, and Mus musculus (house mice). Descendants of all mice have been kept and bred under standard laboratory conditions for several generations (8–16), thus, minimizing the potential of phenotypic plasticity to influence differences in traits. While A. uralensis and M. spicilegus have (on an evolutionary time-scale) recently begun to use human resources such as the abundant food supply on agricultural fields, M. musculus already has a long evolutionary history of close association with humans, representing one of the best studied examples of human-commensalism. Commensals are species that have already adapted to a life with and often off humans and have often already diverged from their ancestral form (Banks and Smith, 2015). Early human settlements led to the evolution of such human commensals and produced urban specialists (Johnson and Munshi-South, 2017). Studying adjustments of such undoubtedly successful exploiters of human habitats provides a natural evolutionary experiment and may also help to understand trait adjustments of non-commensal species to an urban life style.

Apodemus uralensis is a strictly non-commensal species occurring in central and eastern Europe through Russia and the Caucasus into China. Its preferred habitats are forest edges and habitats adjacent to woodlands, foraging both, on the ground and in trees, where it also often builds its nests (Denys et al., 2017). While a few newer studies report the occurrence of this species in agricultural areas, there are no reports that they have been encountered within human settlements (Cichocki et al., 2011). Mus spicilegus also does not use human settlements as a habitat but they have a much closer connection to humans, often occurring on harvested cereal fields and other agricultural areas (Denys et al., 2017). Thus, they are recognized as synanthropomorpic but not commensal (as for example, house mice) species. No official records about the social system of A. uralensis exist but the closest known relative, Apodemus sylvaticus, has a promiscuous mating system (Bartmann and Gerlach, 2001). Mus spicilegus on the other hand is known as the only Mus species practicing social monogamy (Denys et al., 2017).

The house mouse, M. musculus, diverged into several different subspecies. At least three of those subspecies have independently evolved a commensal lifestyle during the last several thousand years (Cucchi et al., 2012). The three subspecies Mus musculus musculus (MMM), Mus musculus domesticus (MMD), and Mus musculus casteneus (MMC) diverged about 350–500 thousand years ago. Records show that MMD already occurred in human settlements as early as 11000–13000 years ago (Bonhomme and Searle, 2012; Cucchi et al., 2012). MMM first occurred in human settlements around 8000 years ago in today’s Balkan area. MMC has spread into India, East Asia, and Southeast Asia and started to live commensally with humans around 7600–3800 years ago (Suzuki et al., 2013). To minimize potential effects of geographical origin rather than a subspecies effect, we used two populations of each subspecies originating from different geographical locations within the subspecies range. The social system of commensally living house mice is flexible, with polygynous males holding territories in which female groups with close kin associations live and raise their offspring communally (Denys et al., 2017).

House mice have long been a classic species to investigate changes due to commensal living. Morphological adaptations such as longer tails as well as changes in behavior and life history have been described previously (Berry, 1981; Pockock et al., 2005). Behaviors associated with flexibility (i.e., innovation propensity and inhibitory control) are heightened in house mouse subspecies that have lived together with humans for longer (Vrbanec et al., 2021). In addition, Frynta et al. (2018) found commensal populations to show increased exploration behavior of elevated places over non-commensal house mouse populations. Based on these previous results, we hypothesize that mice living together with humans more closely (species level) or for longer (subspecies level) show more novelty seeking behavior in general and a more active stress-coping strategy. In addition, we hypothesize these animals have evolved life history traits resembling a slower pace of life in accordance with what has been observed in birds (Sepp et al., 2018). In particular, house mice should have smaller litters and should be weaned at a higher proportional weaning weight than M. spicilegus and A. uralensis. A similar pattern should be observed on the subspecies level within house mice.

In this study, we explored patterns in risk-taking, stress-coping and life history variation of three small rodent species for concordance with micro-evolutionary changes depending on how close their relationship with human habitats are. We found several differences on the species, subspecies, and population level, albeit no pattern that suggests an evolutionary adaptation to living in close proximity to humans.

On the species level, we did not find any conclusive evidence that small, muroid mammals evolved specific behavioral responses in stress-coping or risk-taking behaviors to adjust to a life close to humans. For the distance covered in an Open Field, a situation that is considered mildly stressful for rodents (Archer, 1973), both, A. uralensis and M. spicilegus covered intermediate distances compared to three different subspecies of M. musculus. Regarding the time spent in the central area, which reflects the avoidance of being in the open without cover and possibility for retreat to safety, A. uralensis showed the most avoidance and spent significantly less time in the central zone compared to M. spicilegus and M. musculus while there was no difference between M. spicilegus and M. musculus. This suggests an adaptation to the natural habitat of A. uralensis, which consists of shrublands and forest-edges and agricultural areas close to such vegetation, while M. spicilegus is more often found on open, agricultural areas and M. musculus is usually found within or in close proximity to human settlements. Initially, this may suggest that species that live closer to human settlements may show an altered behavioral pattern, but we would have expected a similar finding then for the Elevated Plus Maze, another mildly stressful situation for rodents. Here, however, both, A. uralensis and M. spicilegus tend to cover more distance compared to M. musculus and do not differ from each other. Also for the time spent in the bright arms of the Elevated Plus Maze, we found both non-commensal species to be intermediate with M. musculus and to not differ from each other. Likewise, A. uralensis and M. spicilegus did not differ in their responses to novelty and always overlapped with at least two of the M. musculus subspecies. Finding the three species to overlap strongly in many of the behavioral measures suggests only (if any) little phylogenetic influence. This would be in agreement with a recently published meta-analysis on behavioral responses toward human-induced rapid environmental change, which only found a phylogenetic signal for birds and for studies conducted in the wild (Gunn et al., 2021). However, some caution should be taken here since the number of studies on mammals was very small and in the present study, we also only used three species.

At the within-species level, we compared two geographically distant populations of the three most common subspecies of M. musculus each. While MMD already started to live with humans more than 10.000 years ago, MMC only started living with humans relatively recently (3.800–7.500 years ago) and MMM is intermediate between the two. While the subspecies significantly differed from each other for several behaviors such as the distance covered in the Open Field and the distance covered in the Elevated Plus Maze, we never found a directional change depending on the time that the subspecies lived with humans. In addition, for many behaviors, the two sampled populations within each subspecies also differed from each other considerably. These findings suggest no directed evolution of behavioral patterns to adjust to an urban-dwelling life-style, which is in contrast to an earlier study on problem-solving and cognitive flexibility in house mice (Vrbanec et al., 2021). In this study, the authors found a general pattern of a better evolved problem-solving ability reflecting cognitive flexibility in house mice populations that lived with humans longer.

Traits in general need to have a genetically determined and thus heritable basis to be able to influence evolutionary processes over long time scales. The behaviors we chose to study here are usually heritable across species (Stirling et al., 2002; Dochtermann et al., 2015) and especially for mice (DeFries et al., 1970; Rinaldi et al., 2001, although information is only available for inbred mouse strains). Repeatability, i.e., the proportion of among-individual variation divided by the total phenotypic variation, is often used to assess the upper boundary of heritability for any given trait (Falconer and Mackay, 1996). At the same time, repeatability reflects the potential for within-individual flexibility in reacting to environmental changes (Royauté and Dochtermann, 2017). Here, we found behaviors shown in the Open Field and Elevated Plus Maze as well as the duration of interaction with an unknown object to be repeatable in all three species with similar estimates and strongly overlapping confidence intervals. This indicates that the measured behaviors represent important and consistent traits in all tested species.

While we did not observe any directed changes in behavioral means in response to a longer association with humans, we interestingly observed a general pattern that the two non-commensal species show less variation in many traits compared to populations of M. musculus. This effect was consistent despite similar age at testing, similar number of animals and identical experimental conditions. This pattern was observed for the Open Field, Elevated Plus Maze, and Novel Object but not for life history traits, thus, indicating a general pattern for behavioral traits. Previous studies reported more variation in urban compared to rural populations measured under field conditions with unknown population history (e.g., Bókony et al., 2012; Dammhahn et al., 2020; von Merten et al., 2022). Variance of flight initiation distance was found to be initially reduced when birds colonize urban habitats (Carrete and Tella, 2011) because only a specific subset of phenotypes would lead the colonization process. During the following population establishment and increase, however, among-individual variation increased considerably and eventually became larger in urban populations (Møller, 2010). The original interpretation of this pattern was that urban populations gained variation in behavioral flexibility through adapting to urban habitats (Møller, 2010). A recent review by Thompson et al. (2022) found increased variance in morphological traits across 13 pairs of rural and urban tit pairs, suggesting that increased variance in urban populations is a consistent phenomenon. Whether such changes in variance are due to within-individual (i.e., flexibility) or between-individual (i.e., a wider spread of phenotypes) variation has not been investigated in many studies. A study conducted on common voles (Microtus arvalis) supports an enormous role of behavioral flexibility in adapting to human-altered habitats (Mazza et al., 2020). The authors found higher levels of risk-taking and exploration behavior in urban compared to rural populations when animals were directly tested after capture. After relocating animals to the laboratory and giving them 9–12 weeks habituation time, initial differences were lost due to a highly flexible adjustment in behavior of the urban animals (Mazza et al., 2020). Another study comparing rural with urban populations in two species of shrews found urban populations to harbor consistently more among-individual variance, i.e., found individuals to differ more from each other in urban habitats (von Merten et al., 2022). Here, we found different patterns for different traits. While, for example, behaviors measured in the Open Field mainly differed for between-individual variance when compared between species, behaviors measured in the Elevated Plus maze showed huge species differences for both variance components. Since our sample size is rather low to thoroughly compare variance components with each other, however, we suggest to interpret them cautiously. To disentangle effects of urbanization on behavioral variation and elucidate their potential ecological and evolutionary potential, future studies should focus on changes in population variance rather than just focusing on the population mean.

Life history traits related to survival and reproductive investment are also sometimes found to differ between rural and urban populations and a recent meta-analysis formally suggested that an urban-dwelling life-style should lead to a generally slower pace-of-life (Sepp et al., 2018). Traits indicative of a slow pace-of-life should include a slower growth, later onset of reproduction and less investment into reproduction at a given point in time. This combination of traits should then increase survival and potentially lead to a higher longevity (Stearns, 1992). In birds, meta-analytic records indeed support higher adult survival rates and smaller clutch sizes while no such information is yet available for other taxa (Sepp et al., 2018). Under natural conditions though, smaller clutches could also be the result of poorer nestling development and survival due to a constraint, non-optimal diet rather than reflecting an adaptation toward an urban-dwelling life-style (Peach et al., 2008; Seress et al., 2012). In the present study, we did not find evidence for a decrease in litter size within M. musculus subspecies, however, body mass at weaning increased in populations of subspecies which have lived with humans for longer. This might suggest that populations that have lived with humans for longer invest more into each single offspring even in the absence of the direct environmental influences and after having been bred under standardized laboratory conditions for several generations. Such a pattern suggests that the increased investment into single offspring might be an adaptive pattern rather than being a constraint due to poorer developmental conditions. On the other hand, we could not find the predicted increased longevity, thus, observed adult survival rates under natural conditions might more likely reflect benefits from better immune responses, less likelihood to fall prey or similar rather than reflecting a physiological adaptation for a longer life span. In fact, both non-commensal species we tested in our study, easily outlive the commensal house mice. One potentially important aspect we could not assess in our study is, however, the effect of the original social and breeding system of the mice. While M. spicilegus is a monogamous species and hence our breeding design reflected the natural situation of the species quite closely, no information on the natural breeding system of A. uralensis is available. It might thus be that the smaller observed litter size in A. uralensis is due to an enforced deviation from their optimal social and breeding system. Likewise, no information on the details of social and breeding systems of M. musculus subspecies is known. The social system of house mice is described as very flexible and adjusted to ecological and habitat characteristics specific populations face. We cannot fully exclude the possibility that aspects of our keeping and breeding design mask differences between species or subspecies because they might deviate from natural conditions more strongly for some species/populations than for others.

Taken together, our findings suggest no or only little micro-evolutionary change in specific behavioral or life history characteristics that is maintained under common-garden conditions across several generations in small rodents. Rather, the absence of an evolutionary adaptation to a life in human-altered habitats argues for an important role of fast and flexible adjustments of behavior and life history induced by different mechanisms of phenotypic plasticity as suggested by the “plasticity first” hypothesis (Levis and Pfennig, 2016; Perry et al., 2018).

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by Ministerium für Energiewende, Landwirtschaftliche Räume und Umwelt, Kiel.

AG designed the study. Data curation and analysis were done by FK. Both authors wrote and edited the manuscript, contributed to the article, and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.