**4. Discussion**

This study investigated different factors that predict hand grip strength in a large, heterogenous adult human sample. While some of the results support the findings of previous studies, our study sheds new light on the variability in grip strength relative to sex, age, hand shape, hand dominance asymmetry (i.e., laterality) and daily hand use. We discuss these results below and their implications for understanding the evolution of the human hand.

Consistent with findings from previous studies [78,79] and our predictions, we found that males were stronger than females. However, our study also investigated relative grip strength and found that males remained significantly stronger than females even when accounting for variation in hand size. This result is consistent with Leyk et al. [62], who showed that untrained males were, on average, stronger than highly trained female athletes. We also found a significant effect of hand dominance asymmetry, in which the dominant hand was significantly stronger than the non-dominant hand in both sexes, with an average difference being slightly higher for males (5.5%) than for females (4.2%) across all age categories. These mean differences were lower than the reported average of 10% higher grip strength for the dominant hand compared with the non-dominant hand for both sexes reported in previous studies [61,74,87]. Interestingly, we found that the difference in hand strength between the hands seemed to vary with age in both sexes (Table S1), but the differences were not significant. This result could be due to the uneven distributions of the sample across the age categories (e.g., more participants in the 20–24 age category than the 60 years and older), but also requires further investigation through a larger study to examine this general premise of 10% difference in grip strength between the dominant and nondominant hands by examining different age categories.

We also found that the grip strength of both hands decreased significantly with age in females but not in males, which partially supported our expectations. Previous studies have shown an effect of age on grip strength for both sexes [77–79]. The non-significant effect of age for both hands in males in our study may reflect the greater variability in grip strength (Figure 3). Middle-aged participants (ages 35–39 for males and 40–44 for females) showed more hand symmetry with limited differences in grip strength between both hands, while younger and older individuals showed greater asymmetry in hand dominance. Although previous studies have shown that laterality decreased with age [88], this was not the case for grip strength in our male sample. The dominant hand was always significantly stronger than the non-dominant hand for both sexes, even for participants in older age categories (Table S1). One reason why this might be the case is that previous studies have typically only assessed differences in the average grip strength across all ages [61,89,90], and thus we demonstrate for the first time, to our knowledge, important grip strength variation at specific life stages for both sexes.

Our results showed that grip strength in our sample was not an indicator of handedness. We did not find a significant effect of self-reported hand preference, although left-handed participants tended to show a larger difference of grip strength between the dominant and non-dominant hand (males = 8.2%, females = 6.4%) compared with righthanded participants (males = 5.2%, females = 4.0%; Table S1). This result did not support our expectation or previous research showing more symmetry in grip strength between the hands in left-handed compared to right-handed people showing more DA [66,75,91]. However, previous work has yielded mixed results, with some studies finding that left-handed individuals had a relatively stronger non-dominant hand [92,93]. The results of our study (and previous research) may be biased by differences in sample size (n = 60 left-handed vs. n = 594 right-handed participants), given the much lower proportion of left-handed individuals across human populations [16–19]. Left-handed individuals may also be expected to show more symmetry in grip strength between both hands (i.e., a relatively stronger right hand), because the world is adapted for right-handed individuals. Thus, our results are somewhat unexpected and require further investigation through a larger study

of left-handed people across their lifespan to better understand the potential differences in grip strength between right- and left-handed individuals.

We found that grip strength of the non-dominant hand was also significantly influenced by hand shape in both males and females, and in both hands for females only. Participants with wider hands were stronger than participants with longer hands, which is consistent with previous studies that reported that people with wider hands tend to have greater muscular strength (when controlling for height) [67–69]. The fact that hand shape did not significantly influence grip strength of the dominant hand in males may also reflect the greater variability in grip strength for males compared with females. Interestingly, we found that the effect of hand shape is stronger for older females (~>50 age) than younger females (~<30 age), with older females with wider hands being stronger than older females with longer hands. This variation by age may reflect younger females being, on average, more active than older females and potentially using both hands more frequently and/or with more muscular force during a variety of daily activities. In contrast, older females are more likely to develop osteoarthritis within the hand [94,95], and patients with this disease show weaker grip strength in the affected hand than healthy individuals [96]. As humans with longer digits appear to have relatively larger articular areas [45], females with longer digits could be less susceptible to osteoarthritis, and thus could show less reduction in hand strength than females with shorter fingers. Additional research is needed to investigate the potential effect of hand shape on grip strength in older females and the potential clinical implications of this.

Variation in hand dominance asymmetry and hand function was observed according to the lifestyle factors (i.e., occupation, practice of sport and music). We found that the type of occupation had a significant effect on grip strength for males but not for females, which is consistent with the findings of Hossain et al. [76]. The female result could be explained by the relatively fewer number of female participants doing, for example, forceful manual labour (22 females against 36 males), which potentially could have affected the analysis. In particular, we found that males engaging in forceful manual work were significantly stronger than those doing 'office work'. This result supports that of previous studies [52,54] (but see [76]). However, we found an effect only for the non-dominant hand. This result likely reflects the fact that manual labour occupations often involve using both hands more forcefully and frequently than office work does, thus increasing muscle strength [54]. Indeed, middle-aged males doing forceful manual work showed greater similarities in grip strength between the dominant and non-dominant hand compared to office workers (Table S1). We found similar results for the practice of manual sports, which significantly affected grip strength in the non-dominant hand for both sexes, while there was no effect from practicing a musical instrument. Together, these results sugges<sup>t</sup> that middle-aged individuals practicing regular manual activities that require the forceful use of both hands have less strength difference between the two hands (i.e., greater symmetry), while office and precision workers, who are doing more fine motor manipulation and using general tools more often with their dominant hand, may have a greater asymmetry in grip strength between the two hands.

Our findings have interesting implications for the study of human evolution. Both hands are important for modern human daily activities; however, experimental studies have demonstrated the importance of having two strong upper limbs, and hands in particular, for prehistoric activities, such as tool production/use behaviours [1,9–11,97–99], carrying [100], hunting, picking fruit, or dismembering an animal carcass [101]. For some hominins, powerful grip strength in both hands would be critical for climbing as well [102–104]. Thus, there would likely be negative selection for having weak hand grip strength throughout hominin evolution [105–107]. We found a significant influence of hand shape on grip strength such that individuals with wider hands were significantly stronger than those with longer hands. If this relationship held true in the past, there may have been increased selection for relatively shorter fingers and proportionally wider hands. Indeed, fossil evidence demonstrates that hand proportions have changed throughout human

evolution (e.g., [103,108]) and that these changes likely improved dexterity [109–111] and potentially grip strength [112,113].

In modern hunter-gatherer populations, greater grip strength in the dominant hand is associated with better hunting outcomes among Hadza males, but not for Yali males [114]. It would be interesting to also measure the strength of the non-dominant hand in huntergatherer populations to test the hypothesis of the importance of high grip strength in both hands for hunting and other manual activities. An increase in sedentism among recent (nonforaging) humans correlates with a decrease in cortical bone strength (reviewed in [115]) and a reduction in trabecular bone density [116–118] throughout the skeleton, including the hands [116,117]. Given bone's ability to reflect variation in loading throughout life via (re-)modelling, this research suggests that recent, more sedentary humans have a reduced level of forceful manual activity compared with that of hunter-gatherers and/or that increased sedentism has resulted in systemic changes to bone structure throughout the skeleton.

It is possible that the population-level hand asymmetry or bias well-documented in modern human populations, inferred in Neandertals [28,36] and potentially earlier *Homo* species [119], is related to advances in technological and cultural innovations [28,120] or, more generally, to task complexity [24,119,121]. In turn, more frequent use of a dominant hand, rather than both hands, for diverse activities could have favoured an increase in hemispheric specialisation and vice versa. However, early hominins (e.g., *Australopithecus*, *Homo habilis*), and particularly those that likely still used their hands for locomotion, may have been under stronger selection for bimanual manipulative ability and grip strength, such as in extant grea<sup>t</sup> apes [119]. Thus, it is important to investigate both hands in studies of grip strength and laterality to provide a broader evolutionary understanding. In their research on lateralisation through prehistorical tools, Steele and Uomini [122] also highlighted the importance of studying the roles of both hands during bimanual activities (i.e., what they call a "Complementary Role Differentiation" model). We also require a greater appreciation of the effect of lateralisation of specific behaviours on upper limb and hand bone morphology [123–127] to better understand the evolution of human dexterity related to strength of laterality.

There are some limitations to this study which should be considered. As these data were collected as part of the larger Me, Human project and within the rules of the Live Science scheme of the London Science Museum, we were limited to a specific amount of time in which we could keep participants at any one experimental station. As a result, we were not able to collect more detailed data on daily hand use (e.g., the number of musical instruments or specific sports played and for how long). Moreover, although participants were not selected and came voluntarily to the Live Science experiment, potential selection bias could be present and may have affected our results. First, we had more young participants and parents accompanying their children compared to participants from 55 years and older. Moreover, the experiments took place during the summer months when the museum was more likely to attract international visitors. We did not record the ethnicity of the participants while previous, more targeted, studies have shown variation in grip strength across different populations [57,74–76]. Future research on large and more diverse groups of humans providing greater detail on ethnicity and specific hand use activities may provide a more nuanced understanding of the links between performance (e.g., grip strength), hand asymmetry, and hand shape or how variation in hand shape and size impacts dexterity.
