"Personal social networks in humans appear to consist of a series of sub-groupings arranged in a hierarchically inclusive sequence (Kahn and Antonucci 1980, Hill and Dunbar 2003, Zhou et al. 2005, Roberts 2010, Sutcliffe et al. 2012). An individual can be envisaged as sitting in the centre of a series of concentric circles of acquaintanceship, which increase in size with a scaling ratio of ~3 (Hill and Dunbar 2003, Zhou et al. 2005) and differ in relationship quality. These have been labelled the support clique (of ~5 members), sympathy group (~15), affinity group (~50) and active network (~150) (Sutcliffe et al. 2012)."
"the support clique of approximately five individuals, ... can be defined as all those individuals from whom one would seek advice, support, or help in times of severe emotional or financial distress (Dunbar and Spoors 1995). Many different measures have been used to determine the number of individuals in a typical support clique. One of the standard questions, used in the US General Social Survey in 1985 and 2001, is ‘looking back over the last 6 months—who are the people with whom you have discussed matters important to you?’(Marin 2004, Marsden 1987, McPherson et al. 2006, Ruan 1998, Straits 2000). This produces a mean size of support clique of between 2.1 (McPherson et al. 2006, but see Fischer 2009) and 5.6 (Marin 2004). However, limiting the time frame to the last six months may exclude people who are important to the respondent, but whom they have not contacted recently, and asking the same question without this time limit produces a value of 6.9 (Bernard et al. 1990)."
"Other methods of eliciting the support clique have used questions relating to whom the respondent would turn to in the event of a major personal problem (Nettle 2007, Stiller and Dunbar 2007), whom they have relied on for advice and/or help at the personal level (Dunbar and Spoors 1995), whom the respondent feels ‘especially close to’(Fischer 1982) or to whom the respondent is ‘socially close’(Wellman et al. 1988) or feels so close to that it would be hard to imagine life without them (Lang 2000, van Sonderen et al. 1990) or whose ‘opinions of your personal life area you would consider important’(Johnson and Milardo 1984). All these different definitions produce a mean network size of between 4.7 and 7.4, indicating that they are eliciting the same, inner layer of the social network. This support clique usually consists of immediate kin (parents, siblings, adult children), one or two very close, long-term friends and, if the respondent is in a relationship, a partner (Dunbar and Spoors 1995, Kahn and Antonucci 1980, Wellman and Guilia 1999, Wellman and Wortley 1990)."
"The next layer out is the sympathy group. Typically, this layer of relationships has been elicited by asking, following Buys and Larson (1979), ‘whose death tomorrow would you be upset by?’"
...
"During ... repeated interactions in the early stages of friendship, more intimate information is gradually exchanged. This exchange of intimate information can be seen as a sign of trust in the relationship partner—this information could potentially be damaging or embarrassing if it was widely disseminated, so by telling the friend the information you are trusting them not to pass it on. This process is especially pronounced in female–female friendships, whereas male–male friendships tend to be based more around shared physical activities (Benenson and Christakos 2003, Dindia and Allen 1992, Reis et al. 1985)."
...
"At the sympathy group level, maintaining these close, emotionally intense relationships is cognitively extremely demanding. Each relationship is unique, in the sense that ‘the partner is important as a unique individual and is interchangeable with none other’(Ainsworth 1989: 711). It takes a long history of interaction in a variety of contexts, as well as emotional commitment, to build up and maintain these relationships—very close relationships have higher frequencies of both face-to-face and telephone contact than less close (but still important) relationships (Mok et al. 2007, Roberts and Dunbar 2011a)."
...
"there is growing evidence that the size of support and sympathy groups correlates with individual differences in social cognitive competences."
[
Source]