Although Vigil provides a very plausible ultimate account on the evolution of sex differences in emotionality based on male philopatry and female dispersal, some of the premises of the socio-relational framework of expressive behaviours (SRFB) remain less convincing. The two main aspects of the model circle around advertising capacity (relating to masculine dominance behaviours) and trustworthiness (relating to feminine emotional expressions), which Vigil ties in with numerous examples from diverse literature. While dominance cues might provide the observer with information about the importance of the actor as a social partner, the idea that emotive signals such as crying have evolved as a display of trustworthiness is less compelling.

The purpose of adult crying is still very much disputed. Crying is a communicative signal (Zeifman Reference Zeifman2001) functioning differently in multiple contexts (Peter et al. Reference Peter, Vingerhoets and Van Heck2001). However, it is debatable whether adult tears represent an honest signal of cooperative intentions, and whether crying is perceived as such by receivers of the signal. Although some forms of crying do signify vulnerability and a need of help (Frijda Reference Frijda, Vingerhoets, Van Bussel and Boelhouwer1997), it is more likely to induce others to help without the expectation of reciprocity. In fact, research suggests that, although crying amplifies the perceived sadness in the face (Provine et al. Reference Provine, Krosnowski and Brocato2009) and elicits emotional support from others, crying individuals are sometimes perceived negatively (Hendriks et al. Reference Hendriks, Croon and Vingerhoets2008), and even labelled as being manipulative (Buss Reference Buss1992; Frijda Reference Frijda, Vingerhoets, Van Bussel and Boelhouwer1997). Moreover, empirical, cross-cultural evidence on sex differences in crying is sparse, and the relative importance of socialisation and culture versus biological processes is not clear. Rosenblatt et al. (Reference Rosenblatt, Walsh and Jackson1976), for example, analysed sex differences in crying in 60 societies and found that in over half of the cultures, women did not cry more than men did. Ross and Mirowsky (Reference Ross and Mirowsky1984) suggested that, rather than an evolved signal, crying in adults is socially conditioned behaviour, dependent on factors such as socio-economic status and sex-role identity of the individual. It is possible that crying in adults is an extension of attachment-related behaviour (Nelson Reference Nelson2005), but does not function as a signal signifying reciprocity potential.

Rather than crying, smiling is a more likely signal of reciprocity potential. Research has found that smiling is related to altruistic dispositions (Brown & Moore Reference Brown and Moore2000; Brown et al. Reference Brown, Palameta and Moore2003) and is used in cooperative context (Mehu et al. Reference Mehu, Grammer and Dunbar2007). Moreover, unlike crying, smiling induces trust and positive evaluations by the receivers of the signal (Mehu et al. Reference Mehu, Little and Dunbar2008; Scharlemann et al. 2003). Sex differences in smiling and decoding of smiles could relate, ultimately, to the need to form reciprocal relationships with unrelated individuals. Females, both children and adults, are more expressive than males are (LaFrance et al. Reference LaFrance, Hecht and Paluck2003; Provine Reference Provine2000), and perform better in decoding emotional facial expressions (Hall et al. Reference Hall, Carter, Horgan and Fischer2000; Rotter & Rotter Reference Rotter and Rotter1988; Thayer & Johnsen Reference Thayer and Johnsen2000). Overall, females are better in processing and sending out signals that facilitate social interactions. At a neural level, a recent study found that when processing social information, females recruit areas containing mirror neurons more than males do (Schulte-Rüther et al. Reference Schulte-Rüther, Markowitsch, Fink and Piefke2007), which could provide a possible proximate mechanism facilitating sex differences in social cognition. Ultimately, the superiority of women in reading and sending out nonverbal messages is probably a result of evolutionary pressures for heightened social sensitivity needed for competing and forming alliances in non-kin–based social networks.

The female advantage in sending and receiving social signals fits well with Vigil's idea of coevolution between perceptual systems aiding in identifying cooperators, and honest expressive behaviours in advertising trustworthiness. Ultimately, the coevolution could explain the problem of cooperation via the Green Beard effect, which postulates that altruists possess a conspicuous phenotypic behavioural of physical trait, which can easily be identified by others with the same characteristic (Dawkins Reference Dawkins1976/1989). Although there is some evidence that people can recognise potential cooperators (Frank et al. Reference Frank, Gilovich and Regan1993; Lyons & Aitken Reference Lyons and Aitken2008; Pradel et al. Reference Pradel, Euler and Fetchenhauer2008), not much is known about individual differences that might account for this ability. Moreover, it is possible that altruistic individuals are equipped with superior social intelligence, helping them in identifying the altruistic dispositions in others. There is some evidence that social intelligence in the form of Theory of Mind is related to self-rated cooperativeness (Paal & Bereczkei Reference Paal and Bereczkei2007), but it is unclear whether social intelligence is used when assessing the honesty of another person.

It is equally possible that evolution has produced polymorphism in perceptual and signalling systems, resulting in the existence of mixed strategies. Experimental work suggests that human populations consist of different cooperative types, which are under substantial genetic influence (Cesarini et al. Reference Cesarini, Dawes, Fowler, Johannesson, Lichtenstein and Wallace2008; Kurzban & Hauser Reference Kurzban and Hauser2005). Further research is needed in order to demonstrate how these types relate to individual differences in accuracy in assessing altruistic dispositions in others. Some individuals might be good in detecting trustworthiness, but not actually be trustworthy – a good candidate would be Machiavellian individuals (for the Machiavellian intelligence hypothesis, see McIlwain Reference McIlwain, Repacholi and Slaughter2003; cf. Byrne & Whiten Reference Byrne and Whiten1997). Others might, in turn, be trustworthy, but not competent in recognising the same trait in others (e.g., people with William's syndrome). It remains to be demonstrated how individual differences in social perceptiveness (e.g., Theory of Mind) and trustworthiness (e.g., Machiavellianism) relate to individual and sex differences in sending and receiving social cues.