3.1. Introduction
Analysis of several genetic markers, together with assessment of archaeological remains, has shown that dogs were domesticated from Eurasian wolves (Canis lupus) some 15,000–35,000 years BP, although there is still considerable debate about the exact timing and location of the domestication event, how many wolf populations were involved, and the selective mechanism by which wolves were turned into dogs (Vilà et al., 1997; Bokyo et al., 2009; Pang et al., 2009; vonHoldt et al., 2010; Ding et al., 2012; Larson et al., 2012; Wang et al., 2013; Druzhkova et al., 2013). Traditionally, it was thought that dogs had been domesticated only through artificial selection, i.e., by capturing wild wolves and by selectively breeding those bearing desirable traits such as tameness (reviewed in Price, 1984; Clutton-Brock, 1995; Coppinger & Schneider, 1995). Partial support for this view is provided by artificial selection experiments on captive foxes (Vulpes vulpes). In this species, selection for a single behavioural trait, i.e., ‘tameness towards humans’, led to the appearance, through pleiotropic effects, of several morphological and physiological characters that are typically observed in dogs, such as piebald coat, floppy ears, earlier sexual maturation, and dioestrus breeding cycle (Trut, 1999). So, it seems conceivable that similar processes also might have operated during the evolution of dogs. However, more recently, researchers have suggested that natural selective forces also probably contributed to the evolution of wolves into domestic dogs (e.g., Morey, 1994; Clutton-Brock, 1995; Coppinger & Coppinger, 2001; Zeder, 2012; Wang et al., 2013). Actually, one problem with the hypotheses of dog domestication based entirely on artificial selection is that, although genetic studies indicate that several hundred wolves were probably involved in the domestication process (Savolainen, 2007; Vilà & Leonard, 2007; Pang et al., 2009; Ding et al., 2012), there is currently no archaeological evidence that Mesolithic humans artificially bred such a large number of wolves (Coppinger & Coppinger, 2001). For these and other reasons, Coppinger and Coppinger (2001) hypothesised that domestication of dogs was initiated through natural selection when some wolves started scavenging on food leftovers around the first permanent human settlements. According to this hypothesis, wolves that had a shorter flight distance from humans (or higher tameness) had a natural selective advantage over shyer individuals because they were more efficient in exploiting the new available food source provided by humans. At a later stage, once wolves had already evolved into tamer primitive dogs, humans began selecting dog breeding types to meet their requirements and to make dogs suitable for performing specific tasks. The hypothesis of dog self-domestication by natural selection (Coppinger & Coppinger, 2001) seems to be partially supported by two recent genetic studies: one showing that domestication of dogs was accompanied by a positive selection on genes involved in starch metabolism (Axelsson et al., 2013), which suggests an adaptation of dogs to a different ecological niche, and a parallel switch from a strictly carnivorous diet to a more generalised, omnivorous diet; the other one suggesting that during domestication dogs underwent a much milder genetic bottleneck if compared to other domesticated animals (Wang et al., 2013), indicating that domestication was most probably a continuous and dynamic process.
Whichever the mechanisms that led to dog domestication, it is clear that dogs have evolved in association with humans for a very long time (Axelsson et al., 2013; Wang et al., 2013), and that in this new ‘domestic environment’ they were subjected to selective pressures (both natural and artificial) that were quite different from those experienced by wolves in their original environment of adaptation. Consequently, it is expected that important behavioural differences between dogs and wolves also evolved in the meantime, some of these probably reflecting adaptations to different ecological niches, and others resulting from artificial selection on dogs. Moreover, during their long association with humans, dogs have very often formed heterospecific social groups with them, and the complexity of the human social system may have provided further selective pressures leading to the evolution of new social skills in these animals (Miklósi et al., 2004; Miklósi & Topál, 2013). Most studies aimed at investigating the effect of domestication on the behaviour of dogs have focussed on the dog–human relationship, and results suggest that dogs may have evolved a higher ability to understand human communicative gestures in comparison to wolves, possibly due to their reduced emotional reactivity and higher capacity for attention towards human beings (Miklósi et al., 2003; Hare & Tomasello, 2005; Gacsi et al., 2009; Topál et al., 2009; Hare et al., 2010). However, other researchers have stressed that at le...