Genomic Imprinting and the Cognitive Architecture Mediating …

   Genomic Imprinting and the
  Cognitive Architecture Mediating
          Human Culture
                      W ILLIAM M ICHAEL B ROWN

                                        ABSTRACT
     Genomic imprinting may be implicated in the origin and maintenance of the cognitive
architecture required for cultural transmission. Relatedness asymmetries are expected to
lead to increases in the receptibility of matrilineally transmitted information. This may help
explain why maternal genes contribute preferentially to the neocortex. That is, maternal
genes could influence biases in the transmission and/or acquisition of information. This
perspective is complementary to gene-culture coevolutionary approaches.




     Genomic imprinting may be a significant influence in the evolution and
development of the cognitive capacities required for social transmission.
Genomic imprinting is the inactivation of a particular inherited allele
dependent upon the sex of the parent (Haig 1999). Genomic conflict
theory (Haig 2000) argues that multiple paternity evolutionarily favours the
differential expression of maternal and paternal alleles so that a) paternal
alleles increase the cost to the offspring's mother; and b) the maternal
alleles reduce these costs. Haig (2000) hypothesized that a genomic conflict
approach applies to all costs imposed on maternal genes that benefit
the offspring. Moreover, all social interactions between kin with different
maternal and paternal coefficients of relatedness can potentially lead to
conflict (Haig 1997). Haig (1997) refers to these differing patterns of
coefficients of relatedness for paternal and maternal genes as "relatedness
asymmetries".


     Department of Psychology, Life Sciences Centre, Dalhousie University, 1355 Oxford

St., Halifax, Nova Scotia, Canada, B3H 4J1, e-mail: wmbrown@is2.dal.ca


c Koninklijke Brill NV, Leiden, 2001                         Journal of Cognition and Culture 1.3
252                       WILLIAM MICHAEL BROWN


    While commonly evoked, traditional inclusive fitness calculations may
not be sufficient for explaining culture due to relatedness asymmetries.
For example, Takahasi (1998) wrote: "Genetic determination of the
preference seems to preclude the emergence of a cultural runaway [of
a maladaptive variant], and only preferences that favour adaptive variants
may eventually evolve" (p. 154). The problem with Takahasi's (1998) logic
in the above passage is that an adaptive variant for paternal genes may
be disadvantageous to maternal genes (and vice versa). More recently,
Henrich & Boyd (2001) assume in a cultural evolutionary model that
humans may have a psychological bias for copying the majority and/or
the most successful individual. However the `most successful' individual in
terms of maternal genes or memes may not be the most successful in terms
of paternal genes or memes. In kin groups, the information transmitted
between siblings could have different levels of reliability and adaptiveness
because of differing maternal and paternal coefficients of relatedness.
    Genomic imprinting influences behaviour via the differential expression
of paternal and maternal genes in the mammalian brain (Isles & Wilkinson
2000). In mice, maternal genes are expressed in cells found in the
neocortex (involved in decision-making) and paternal genes are expressed
in the cells of the hypothalamus, involved in homeostasis, consummatory
and sexual behaviours (Allen et al. 1995; Dhawan, Kumar Govindaraju
& Raghunathan 1998; Isles & Wilkinson 2000; Keverne et al. 1996;
Plata-Salaman 1998; Swaab & Hofman 1995). If maternal genes are
involved in the filtering of socially transmitted information favouring
paternal genes it may follow that they (i.e. mother's genes) would express
themselves in the decision-making areas of the brain. In terms of socially
transmitted information, the decision rules to adopt a meme for cooperative
information sharing among relatives may be costly to paternal genes (or
maternal x-linked genes). Maternal genes could benefit from filtering selfish
paternal inclinations within siblings. Benefits may also be accrued from
cooperative information sharing among maternal kin. It should be pointed
out that there is nothing preventing the hypothalamus from filtering
information in favour of paternal genes. Furthermore the manipulative
use of signals could be a tactic used in genomic conflicts over socially
transmitted information between collateral kin (i.e. Having an ancestor in
common but descended from a different line).
                         GENOMIC IMPRINTING                              253

     Under conditions of multiple paternity, social transmission between
siblings could be detrimental to maternal genes but beneficial to pater-
nal genes and vice-versa. For example, acquiring selfish benefits from the
public good could favour paternally derived alleles and simultaneously be
costly to maternal genes (Haig & Wilkins 2000). The evolutionary pre-
dictions regarding conflicts are simple when siblings have higher maternal
than paternal relatedness. That is, maternal genes may favour sharing in-
formative memes with maternally related kin and/or suppress misleading
memes exchanged between paternally unrelated siblings. However, when
one or several males dominate paternity, the sources of conflict change.
Specifically, high coefficients of relatedness in male sex-linked genes among
daughters could also favour cooperative networks and information sharing
among daughters. This is due to the father's x-linked allele having a high
probability of transmission to daughters and remaining in the group for
one generation (given male-biased dispersal). In contrast, the probability
that daughters share the x-linked allele inherited from mother is 1/2 (the
same as autosomal genes). When homogametic (xx) offspring of a het-
erogametic (xy) parent share an entire chromosome subject to genomic
imprinting there should be a predisposition towards cooperative informa-
tion exchange among daughters with common paternity (Haig 2000). If
genomic conflicts have interacted with cooperative information exchange
among animals it would not be surprising to find examples of gene-meme
linkage or gene-culture coevolution.
     An empirical example of gene-culture coevolution induced by genomic
conflicts may be `cultural hitchhiking' in whales (Whitehead 1998; Rendell
and Whitehead 2001). Whitehead (1998) found that mtDNA and variation
in socially learned dialect correlated between matrilineal groups of whales.
That is groups with more similar dialects also had less within group
variation in mtDNA. Any heritable unit (mtDNA or cultural information)
transmitted with high fidelity from mother to daughter can lead to
tight social networks due to high maternal relatedness (Haig 2000).
When interests of genes and memes coincide it is predicted that genetic
and cultural information should begin to covary over time. Relatedness
asymmetries, genomic imprinting and paternity are factors that favor
matrilineal sociality (Haig 2000). In humans the ties between mothers
and daughters are positive and enduring (Fingerman 2001). Emotions
254                          WILLIAM MICHAEL BROWN


may maintain mother-daughter cooperation. For example, daughters feel
particularly guilty when mothers disapprove of their behaviour (Fingerman
2001). Social transmission should be most effective (i.e. cooperative)
between mother and daughters (and among daughters) due to the high
mtDNA coefficient of relatedness. Is the variability in matrilineal cultural
practices in humans linked to variability in mtDNA? A possible example
of `cultural hitchhiking' in humans is the strong relationship between
patrilineal surname and Y-chromosome haplotype distribution (Sykes &
Irven 2000). Specifically, individuals with the surname "Sykes" are more
likely to have a particular Y-chromosome haplotype. This is likely due to
the historical accumulation of nonpaternity events (Sykes & Irven 2000).
If patrilineal surname descent is a meme designed to reduce cuckoldry, it
appears to be effective in this sample. Does patrilineal descent of surnames
favour maternal and/or paternal genes and memes? Perhaps the cultural
practice induces males to invest more in unrelated offspring to the benefit
of maternal genes.

Who benefits from meme filters?
Signal-receptor systems designed for the acquisition of information are sus-
ceptible to super-stimulation (Ryan 1990). As a result, cultural transmission
provides opportunities for sensory exploitation via cultural transmission
(Rice and Holland 1997). A potential example of a meme-gene conflict
mediated by sensory exploitation in humans could be the tension between
sexual desires and religious celibacy. Celibacy memes (if an indicator of
commitment) may be beneficial to a religious group's cultural longevity
(Sosis 1999) but costly to the genes predisposing receptibility to celibacy.
If these were the only forces at work, religious celibacy should gradu-
ally disappear from the population. However, kin selection for reputation-
enhancement of one's kin group could also be involved in the receptibility
of one's family members to the idea of religious celibacy. If large families
gained inclusive fitness benefits from having an offspring known for their
religious celibacy and commitment, the family may collectively encour-
age such memes within the group. Based on this reasoning it would be
expected that the collective encouragement of celibacy memes would be
directed more often towards later borns. In addition celibates should more
often be found in large families. Interestingly genomic conflicts could ob-
                          GENOMIC IMPRINTING                                 255

scure a relationship between family size and celibacy. In cases of multiple
paternity it is expected that paternal genes will discourage while maternal
genes will encourage celibacy memes regardless of family size. In fami-
lies with one father, the costs of a celibate son may be outweighed by
the reputation-enhancement to other reproducing sons who share the Y-
chromosome inherited from the same father. The same holds for daughters
who share mtDNA from the same mother. When there is monogamy, the
additional offspring added due to reputation-enhancement from celibacy
(adjusted for coefficient of relatedness) outweighs the number of offspring
lost (adjusted for coefficient of relatedness) there may be selection for sus-
ceptibility to celibacy memes. Under multiple paternity relatedness asym-
metries are predicted to cause conflicts over the acceptance and rejection
of celibacy memes.
      Imprinted gene expression predisposing cooperative information ex-
change may be favoured when recipients in the social group have different
probabilities of sharing a donor's maternally and paternally derived alleles
(Haig 2000). When there is multiple paternity it is predicted that mater-
nal genes predispose offspring to trust information learned from mother
(i.e. vertical transmission) rather than siblings (i.e. horizontal transmission).
There can be inclusive fitness costs to maternal genes that treat all in-
formation as equally reliable. For example, conflict over the transmission
and reception of information between- and within-individuals is possible
(Trivers 2000). Non-paternally related siblings could encourage behaviours
costly to the learner and/or discourage behaviours costly to the transmitter.
When there is a premium on the reliability of information, maternal genes
may benefit from suppressing deceptive information transmitted between
sibs.
      Organisms are likely designed to filter information that causes devia-
tions from optimum behaviour (Rice and Holland 1997; Takahasi 1998).
Within- and between-group components of selection may have differing
`views' on what is optimal however (Boehm 1999; Richerson and Boyd
2000). The same unit of information (e.g. foraging strategy, behavioural
strategy to gain access to mates, etc.) could be designed to be costly for
maternal genes and beneficial to paternal genes. Just like good genes for
one sex may be costly to the other (Rice and Holland 1997), so too can
information have differential costs for genes, individuals and groups.
256                            WILLIAM MICHAEL BROWN


    Misinformation has been hypothesized to favour better filtering mech-
anisms in the human brain (Rice and Holland 1997). One way to filter
misinformation is if individuals have cognitive architecture designed to
make fine-grained assessments of asymmetries in paternal and maternal re-
latedness. Genomic conflicts are important for cognitive studies of human
culture, if genomic imprinting influenced the mosaic evolution of mam-
malian brain structure (Barton & Harvey 2000; Dunbar 1992; de Winter
& Oxnard 2001) designed to transmit and acquire information.
    Acknowledgements: Chris Moore, Pete Richerson and Natasha Crewd-
son made helpful comments on this topic. This work was supported by the
Natural Sciences and Engineering Research Council of Canada and an
Isaak Walton Killam Memorial Postgraduate Scholarship.


                                     REFERENCES

A LLEN , N.D., L OGAN , K., L ALLY , G., D RAGE , D.J., N ORRIS , M.L. & K EV -
ERNE ,  E.B.
     1995 Distribution of parthenogenetic cells in the brain development and behavior.
           Proceedings of the National Academy of Sciences 92, 10782-10786.
B ARTON , R.A. & H ARVEY , P.H.
     2000 Mosaic evolution of brain structure in mammals. Nature 405, 1055-1058.
B OEHM , C.
     1999 The natural selection of altruistic traits. Human Nature 10, 205-252.
D HAWAN , J.K., K UMAR , V.M., G OVINDARAJU , V. & R AGHUNATHAN , P.
     1998 Changes in magnetic resonance imaging and sex behavior after 6-OHDA
           injection in the medial preoptic area. Brain Research Bulletin 45, 333-339.
D UNBAR , R.I.M.
     1992 Neocortex size as a constraint on group size in primates. Journal of Human
           Evolution 20, 469-493.
F INGERMAN , K.
     2001 Aging Mothers and Mixed Emotions. Springer: New York.
H AIG , D.
     1997 Parental antagonism, relatedness asymmetries, and genomic imprinting. Proceed-
           ings of the Royal Society of London Series B-Biological Sciences 264, 1657-1662.
     1999 Multiple paternity and genomic imprinting. Genetics 151, 1229-1231.
     2000 Genomic imprinting, sex-biased dispersal, and social behavior. Annals of the New
           York Academy of Sciences 907, 149-163.
H AIG , D. & W ILKINS , J.F.
     2000 Genomic imprinting, sibling solidarity and the logic of collective action.
           Philosophical Transactions of Royal Society London B 355, 1593-1597.
                               GENOMIC IMPRINTING                                          257

H ENRICH , J. & B OYD , R.
     2001 Why people punish defectors: Weak conformist transmission can stabilize costly
            enforcement of norms in cooperative dilemmas. Journal of Theoretical Biology 208,
            79-89.
I SLES , A.R. & W ILKINSON , L.S.
     2000 Imprinted genes, cognition and behaviour. Trends in Cognitive Sciences 4, 309-318.
K EVERNE , E.B., F UNDELE , R., N ARASIMHA , M., B ARTON , S.C. & S URANI , M.A.
     1996 Genomic imprinting and the differential roles of parental genomes in brain
            development. Developmental Brain Research 92, 91-100.
P LATA -S ALAMAN , C.R.
     1998 Hypothalamus and the control of feeding: Fifteen decades of direct association.
            Nutrition 14, 67-70.
R ENDELL , L. & W HITEHEAD , H.
     2001 In Press Culture in whales and dolphins. Behavioral and Brain Sciences.
R ICE , W.R. & H OLLAND , B.
     1997 The enemies within: intergenomic conflict, interlocus contest evolution (ICE),
            and intraspecific Red Queen. Behavioral Ecology and Sociobiology 41, 1-10.
R ICHERSON , P.J. & B OYD , R.
     2000 The evolution of human ultrasociality. In I. Eibl-Eibesfeldt & F.K. Salter (Eds),
            Indoctrinability, ideology and warfare (pp. 71-96). New York: Berghahn Books.
R YAN , M.J.
     1990 Sexual selection, sensory systems, and sensory exploitation. Oxford Surveys in
            Evolutionary Biology 7, 156-195.
S OSIS , R.
     1999 Costly signaling and commitment: A test of a hypothesis using historical data on
            the longevity of 19th century utopian communes. Paper presented at the Annual
            Meeting of the Human Behavior and Evolution Society June 2-6, 1999 Salt Lake City,
            Utah, USA.
S WAAB , D.F. & H OFMAN , M.A.
     1995 Sexual differentiation of the human hypothalamus in relation to gender and
            sexual orientation. Trends in Neurosciences 18, 264-270.
S YKES , B. & I RVEN , C.
     2000 Surnames and the Y Chromosome. American Journal of Human Genetics 66, 1417-
            1419.
T AKAHASI , K.
     1998 Evolution of transmission bias in cultural inheritance. Journal of Theoretical Biology
            190, 147-159.
T RIVERS , R.
     2000 The elements of a scientific theory of self-deception. Annals of the New York Academy
            of Sciences 907, 114-131.
W HITEHEAD , H.
     1998 Cultural selection and genetic diversity in matrilineal whales. Science 282, 1708-
            1711.
258                           WILLIAM MICHAEL BROWN


DE   W INTER , W. & O XNARD , C.E.
     2001 Evolutionary radiations and convergences in the structural organization of
          mammalian brains. Nature 409, 710-714.