altruism towards cousins

here’s a neat, little paper i came across during my r&r: Altruism towards Cousins [pdf]. sounds like it should be right up my alley! (^_^)

the authors — joonghwan jeon and david m. buss — conducted a study of some “w.e.i.r.d.” students at the univ. of texas (n=195) to see if they might treat and feel about their different cousins differently — the different cousins being (hold on!): father’s brother’s (FB) kids, father’s sister’s (FZ) kids, mother’s brother’s (MB) kids, and mother’s sister’s (MZ) kids. they quizzed their subjects on such things like, “would you run into a burning building to save your father’s brother’s son?” — and so on and so forth.

their hypothesis was that, because of paternity uncertainty, people ought to feel closer to, and be more altruistic towards, their mother’s sister’s kids since the connection between those cousins is via two mothers (mother and mother’s sister) and, therefore, the relatedness pretty certain. on the other hand, people ought to show the least altruism towards their father’s brother’s kids since, in that case, the connection is via two fathers (father and father’s brother) thus a LOT of paternity uncertainty there.

they figured that the pattern of altruism ought to go like this: MZ kids first, FB kids last, MB and FZ kids somewhere in between.

so, what did they find?

well, after tweaking the data a bit — i.e. to include things like residential proximity (or not) — they found that, yes, indeed, people are (or say that they would be) the most altruistic towards their MZ kids, then their MB kids, then their FZ kids, and lastly their FB kids. that seems to fit their hypothesis perfectly! however, the authors do note that, all things considered [pg. 1185]:

It is reasonable to conclude, however, that FaSis children [FZ kids] are the least likely to be helped among all the four cousin categories, because (i) the mean rating of FaSis children was invariably the lowest in the willingness-to-help scores, as well as in all the three psychological variables affecting altruism, (ii) the predicted linear trends proved to be highly significant for all the four dependent measures, and (iii) FaSis children were helped more than FaBro children, as predicted, although it just missed conventional significance. Given that MoSis children were always ranked at the top position with a high statistical significance (p less than or equal to 0.001) for all dependent measures, it remains to be revealed why the degree to which FaBro children are the least likely to be helped appears not to be as robust as expected.”

that would make the pattern — MZ kids, MB kids, FB kids, FZ kids — which doesn’t fit their prediction. not perfectly, anyway.

the raw data really don’t fit their prediction (click on table for LARGER view):

here we see three different patterns. although MZ kids always do come first, in two instances, the FB kids came second — and they should come dead last according to the paternity uncertainty theory — AND the MB kids came last twice, which is not right at all:

– willingness-to-help (i.e. burning building) = MZ kids, FZ kids, FB kids, MB kids
– emotional closeness = MZ kids, FB kids, MB kids, FZ kids
– empathic concern = MZ kids, FB kids, FZ kids, MB kids

what i wondered was: could the fact that the different cousins are, on average, related to each other to different degrees be connected to how altruistic cousins are to one another?

the coefficient of relationship between any two first cousins is 0.125 (give or take a few genes here and there) — but this figure doesn’t take into account the differential inheritance of the x- and y-chromosomes in humans, and that that alters the relatedness between different family members depending on whether they are male or female (for some background on this, see this post — also, see this paper).

a long time ago in you know where…, i sat down and calculated the coefficients of relationship for different family members with the differential inheritance of the x- and y-chromosomes in mind. the plan was to do that for ALL the family members — all the various cousins, etc. — but, the best laid plans of mice and female bloggers….

so, unfortunately, i don’t have any new-and-improved coefficients of relationship for all the various cousins (nor do i feel like working on them today), but i DO have them for brothers and sisters, i.e. FB, FZ, MB and MZ. i think we can use them as a proxy for the cousin coefficients. here they are (assuming no inbreeding):

FB (B-B on the chart if you follow the link) = 0.5050
FZ (B-Z) = 0.4951
MB (Z-B) = 0.4872
MZ (Z-Z) = 0.5128

and, rearranged to be in order of relatedness:

MZ = 0.5128
FB = 0.5050
FZ = 0.4951
MB = 0.4872

so, if I were to predict which type of cousin people would be most altruistic towards, i would pick the MZ kids (score! three points to moi) — the least would the MB kids (two points! according to the raw data) — and then the FB kids second (two more points for me!) — and the FZ kids third (one point). that’s eight points to me and … six points to jeon and buss (using their prediction and the raw data). (~_^)

i kinda like my explanation. (^_^) but i could be wrong, of course (the likelihood is that i am, right?).

see also: genealogical terminology

(note: comments do not require an email. cousins?)


  1. My mother’s brother never had children. But if he had, are you saying that they’d like me/be more altruistic toward me more than I would be toward them? On the other hand I definitely prefer my cousins from my mother’s sister to my cousins from my father’s brothers (my father had no sisters). However there is a much larger age gap on my father’s side than my mother’s side, and my father’s side lives all over the US.


  2. @anonymous – “But if he had, are you saying that they’d like me/be more altruistic toward me more than I would be toward them?”

    that is the implication, yes. (^_^)

    if jeon and buss are right and this differential altruism between family members is based upon paternity uncertainty, then there ought to be other differences in altruistic behaviors by family members based upon that — like paternal grandparents being less altruistic towards their grandkids than maternal grandparents. and i think that there is some evidence out there to suggest that (quoted in the jeon and buss paper, i think).

    if the theory i’m talking about is right, then differential altruism within families ought to follow along the lines of genetic relatedness (see my chart) — fox, et al., found some evidence for this. if this is right, then brothers ought to be more altruistic (on average) to their sisters than vice versa. (~_^) and fathers ought to be much more altruistic towards their daughters than to their sons.

    these two explanations don’t have to be mutually exclusive, i suppose.

    @anonymous – “However there is a much larger age gap on my father’s side than my mother’s side, and my father’s side lives all over the US.”

    yes. this is why jeon and buss controlled for those sorts of factors. if you’ve never really had a chance to get to know a certain set of cousins ’cause they live far away, then you’d think it would be pretty unlikely for you to feel very altruistic towards them.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s