Cognition vs time as constraints in the structuring of human social networks Robin Dunbar British Academy Centenary Project Institute of Cognitive & Evolutionary Anthropology University of Oxford robin.dunbar@anthro.ox.ac.uk
Convergence of Three Projects Convergence of Three Projects British Academy’s “Lucy Project” � http://www.liv.ac.uk/lucy2003/ Liverpool (Archaeology + Psychology), Kent (Social Psychology) – how social bonds work – cognition and brain evolution (Social Brain Hypothesis) – � EPSRC/ESRC DTESS Project http://www.informatics.man.ac.uk/research/groups/isd/projects/dtess Manchester Business School + Sheffield Hallam – Integrating Small-Groups-as-Dynamic-Systems Theory with – Social Brain Hypothesis � EU-FP7 SOCIALNETS Project http://www.social-nets.eu/ Computer Sciences at Cambridge and Cardiff; + EU partners – – How to design better networking technology
The Social Brain Hypothesis The Social Brain Hypothesis Primates have big brains because they live in a complex social world � Predicted group size for humans is ~150 � “Dunbar’s Number”
Human Human “Reverse” Social Networks Social Networks Small World Experiments These all have mean sizes of Killworth et al (1984) 100-200 Neolithic villages 6500 BC 150-200 10000 military units (company) (N=10) 180 * Hutterite communities (N=51] 107 1000 Nebraska Amish parishes (N=8) 113 Hunter-Gatherer business organisation <200 100 Societies ideal church congregations <200 Doomsday Book villages 150 Dunbar (1993) 10 C18th English villages 160 * GoreTex Inc’s structure 150 1 Research sub-disciplines (N=13) 100-200 10 Individual Tribes 9 0 10 20 30 8 7 Small world experiments (N=2) 134 Number of Cases 6 Hunter-Gatherer communities 148 5 4 Xmas card networks 154 Xmas Card 3 2 Networks 1 0 0 2 5 7 1 1 1 1 2 2 2 2 3 3 3 - 5 0 5 0 2 5 7 0 2 5 7 0 2 5 2 - - - 0 5 0 5 0 5 0 5 0 5 0 Hill & Dunbar (2003) 4 4 7 9 - - - - - - - - - - - 9 4 9 1 1 1 1 2 2 2 2 3 3 3 2 4 7 9 2 4 7 9 2 4 7 4 9 4 9 4 9 4 9 4 9 4 Maximum Network Size
What Makes it Work? What Makes it Work? � Personalised relationships � Trust � Expectations of reciprocity � In traditional societies: – kinship – a shared history The Atapuerca “family” [ Homo heidelbergensis ]
Hidden Structure of Social Hidden Structure of Social Networks Networks � Stable points in 160 140 group size at: 120 Cumulative Network Size 5-7 100 ? 80 12-15 60 ~35 40 20 Maximum ~80? 0 Contacted -7.6 -4.7 -4.6 -4.4 -4.1 -3.9 -3.6 -2.7 -1.7 -.7 .1 1.4 3.4 5.0 7.3 7.4 ~150 Residual Contact Frequency Hill & Dunbar (2003)
Horton Order Analysis of The Fractal Periodicity of The Fractal Periodicity of Hunter-Gatherer Group Sizes Human Group Sizes Human Group Sizes Peak at ω =5.4 Social Groupings Hamilton et al (2007) Database [N=60] Peak at ω =5.2 Xmas Card Database Scaling ratio = exp(2 π / ω ) Zhou, Sornette, Hill & Dunbar (2005) = 3.2 and 3.3
Intimacy, Frequency and Trust Intimacy, Frequency and Trust 8 � Relationship between Mean Time Since Last Contact (Months) frequency of contact 6 and intimacy 4 � Trust and obligation 2 seem to be important 0 0 1 2 3 4 5 6 7 8 9 10 Emotional Closeness Hill & Dunbar (2003)
The Circles of Acquaintanceship The Circles of Acquaintanceship • A hierarchically inclusive series of levels of acquaintanceship Intensity • Levels reflect familiarity and emotional closeness 5 • There are at least 15 TWO more layers at 50 ~500 and ~1500 150 [is this where weak 500 “work” ties lie?] 1500
Friends ≠ ≠ Kin Kin Friends 7.00 Unrelated Alters Mean emotional closeness to unrelated alters � Friends and Kin are not 6.50 the same thing 6.00 � Friendship requires 5.50 emotional closeness 5.00 � We have no choice 4.50 about Kin Under 47 (n=85) 47-76 (n=84) Over 76 (n=81) Total network size � Hence: Friendships are 7.00 fragile…. Related Alters Mean emotional closeness to related alters 6.50 ….Kinship is robust 6.00 [We put up with them even 5.50 though we don’t particularly like them] 5.00 4.50 Under 47 (n=85) 47-76 (n=84) Over 76 (n=81) Small Medium Large Total network size Network Size
Strong ties (EC 8- Structure of Networks Structure of Networks Medium ties (EC 5 70 Unrelated Alters Weak ties (EC 1-4 Strong ties (EC 8- Medium ties (EC 5 60 Weak ties (EC 1-4 Median percentage of unrelated network Medium 50 40 � For relationships indexed on a 30 Weak 1-10 scale: 20 10 Strong � Among UNRELATEDs: 0 – medium strength links predominate Under 55 56-82 83 and over <55 55-82 >82 Total network size – large networks exhibit more Strong ties (EC 8 Medium ties (EC STRONG links 70 Weak ties (EC 1- Related Alters Strong ties (EC 8 Medium ties (EC Weak ties (EC 1- 60 Median percentage of related network Weak � Among RELATEDs: 50 – Weak and Medium links 40 predominate 30 Medium – large networks exhibit more WEAK 20 links 10 Strong 0 <55 55-82 >82 Under 55 56-82 Over 82 Total Network Size Total network size
Blood is is Thicker Thicker Blood 80 than water than water 80 close networks 60 Total Non-Kin 40 � Kin are given 20 priority over 0 Friends 0 10 20 30 40 50 60 Total Kin 120 � Kinship may 250 complete � networks 100 reduce the 80 Unrelated network size 60 cognitive load? 40 20 0 0 20 40 60 80 100 120 140 Related network size Total Kin
Estimating the Limit on Network Size Maximum Network N P Size 120 100 6 0.011 150.0 80 Unrelated network size 8 0.002 146.1 60 40 10 0.001 144.5 20 12 0.004 145.3 0 0 20 40 60 80 100 120 140 Related network size 14 0.004 141.8 16 0.001 136.3
Two Unresolved Questions Two Unresolved Questions Are human groupings Is the limit at: • higher level, with the limited by: internal structure a ⇒ frequency of consequence of interaction fragmentation [top down] ? ⇒ capacity for • lower level, with emotional closeness higher levels simply [i.e. cognition] being small-world emergent properties [bottom-up] ?
A Role for the Social Brain A Role for the Social Brain Intentionality as a reflexively hierarchical sequence of belief states 5 Humans Achievable Intentionality Level 4 The Levels of Intentionality 3 …that may be very Apes 2 costly in Monkeys computational terms 1 0 1000 2000 3000 4000 5000 6000 Frontal Lobe Volume (cc)
The Limits to Intentionality... The Limits to Intentionality... % Correct 120 A natural limit at 5 th order 100 intentionality: 80 60 “I intend that you believe that 40 Fred understands that we ToM want him to be willing to [do 20 Physical something]…” [level 5] 0 2 3 4 5 6 7 Intentionality Level Kinderman, Dunbar & Bentall (1998).
The Story- -Teller Teller’ ’s Art s Art The Story Othello - An Everyday Story of Deception � BUT… 6 Shakespeare 5 had to do SIX Cassio Othello Iago • The audience 3 2 1 has to do FIVE orders of intentionality 4 Stories (especially “origins” stories) are an integral part of community-bonding Desdemona
Is Mentalising Mentalising Costly? Costly? Is Two Experiments Two Experiments Reaction Time Experiment N = 8 Mentalising vs Memory (controlling for order) accuracy: p = 0.919 RT: p < 0.05 Functional Imaging Experiment fMRI [BOLD] 5 stories with 20 mentalising and memory questions @ levels 2, 3 and 4 N=17
The Cognitive Demands of The Cognitive Demands of Mentalising? ? Mentalising Areas with significant parametric effects on the contrast [intentionality > memory] at p=0.001 uncorrected After FWE correction [p=0.05]: right TPJ, bilateral TP, right inferior FG, cerebellum Significant effects for parametric contrast [ToM>memory] masked by nonparametric fMRI contrast N=17 [ToM>memory] Temporal- Parietal junction (p<0.005 uncorrected) Lewis, Birch & Dunbar (in prep)
Cognitive Limits to Sociality? Cognitive Limits to Sociality? 20 Frequency of failure � Achievable intentionality level indexed from stories 10 � 5 th order seems to be the limit 0 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Level of intensionality � Intentionality correlates 30 with clique size 20 Clique size � We now have two neuroimaging 10 studies to support this 0 0 2 4 6 8 10 [Stiller & Dunbar 2006] Level of intensionality
A Volumetric Perspective A Volumetric Perspective Optimised VBM VBM Optimised with modulation with modulation [N=29 subjects, aged 18- -50] 50] [N=29 subjects, aged 18 Grey matter volume correlates of network size for ToM > memory contrast Masked analysis for both ToM and network size [corrected p<0.005]: Middle frontal gyrus Orbitofrontal area Dorsolateral PFC ACC Hippocampus Amygdalla among others, most bilaterally Lewis, Browne & Dunbar (in prep) Orbitofrontal
Social Bonding Social Bonding Primate- -Style Style Primate � Primate social bonds seem to involve two distinct components: � An emotionally intense component [=grooming] � A cognitive component [=brain size + cognition]
Recommend
More recommend
