A pacific culture among wild baboons: its emergence and transmission

Abstract

Reports exist of transmission of culture in nonhuman primates. We examine this in a troop of savanna baboons studied since 1978. During the mid-1980s, half of the males died from tuberculosis; because of circumstances of the outbreak, it was more aggressive males who died, leaving a cohort of atypically unaggressive survivors. A decade later, these behavioral patterns persisted. Males leave their natal troops at adolescence; by the mid-1990s, no males remained who had resided in the troop a decade before. Thus, critically, the troop's unique culture was being adopted by new males joining the troop. We describe (a) features of this culture in the behavior of males, including high rates of grooming and affiliation with females and a "relaxed" dominance hierarchy; (b) physiological measures suggesting less stress among low-ranking males; (c) models explaining transmission of this culture; and (d) data testing these models, centered around treatment of transfer males by resident females.

Figure 1. Quality of Male–Male Dominance Interactions

(A) Percentage of male approach–avoidance dominance interactions occurring between males more than two ranks apart. (B) Percentage of male approach–avoidance interactions occurring between males of adjacent ranks. (C) Percentage of approach–avoidance interactions representing a reversal of the direction of dominance within a dyad by a male more than two steps lower ranking. Mean ± SEM, ** and *** indicate p < 0.02 and p < 0.01, respectively, by t-test, treating each male/year as a data point. Data were derived from a total of ten different males in F93–96, 31 different males in T93–98, and 19 different males in F79–82. Potentially, the result in (B) could have arisen from different numbers of males in F93–96 versus the other two troops (a smaller group size does not change the number of adjacent animals available to any given subject, but decreases the number of nonadjacent animals available). However, the same results were found if the numbers of males in the three troops were artificially made equal by excluding excess males from either the top or the bottom of the hierarchy (data not shown).

Figure 2. Quality of Affiliative Behaviors

(A) Amount of grooming involving adult males in Forest Troop 1993–1996 and Talek Troop 1993–1996. The first pair of columns represents mean time adult males spent grooming adult females; the second pair, mean time adult males were groomed by adult females. (B) Comparison of average number of neighbors (i.e., within 3 m) of adult males in the two troops. Mean ± SEM. *, **, and *** indicate p < 0.05, 0.02, and 0.01, respectively, by unpaired t-test. Data were derived from a total of ten different males and 17 different females in F93–96, 31 different males and 21 different females in T93–98, and 19 different males and 23 different females in F79–82.

Figure 3. Stress-Related Physiological Profiles

(A) Basal glucocorticoid levels (μg/100 ml). Males were split into higher- and lower-ranking 50%, by approach–avoidance criteria. The primate glucocorticoid, cortisol, was measured by radioimmunoassay. (B) Number of anxiety-related behaviors observed 10–20 min after β-carboline-3-carboxylic acid administration (M-156, Research Biochemicals International, Natick, Massachusetts, United States), after subtracting the number observed 10–20 min after vehicle administration (dextrin in 1 ml saline); 0.5 g of the drug in 1ml saline was delivered intramuscularly by dart syringe (Pneu-Dart, Inc., Williamsport, Pennsylvania, United States) fired from a blowgun at 5 m. Mean ± SEM. * and *** indicate p < 0.05 and p < 0.01, respectively, by unpaired t-test. Data were derived from a total of ten different males in F93–96, 31 different males in T93–98, and 18 different males in F79–82.

Figure 4. Quality of Interactions between Resident Females and Transfer Males

(A) Latency, in days, until a newly transferred male is first groomed by a female (left) or presented to by a female (right). (B) Average number of adult female neighbors per scan (i.e., within 3 m; left) and average number of grooming bouts with females per 100 h of observation (right) for transfer males. Mean ± SEM. * and *** indicate p < 0.05 and p < 0.01, respectively, by unpaired t-test. Latency until first presented to by a female approached significance (p < 0.08). Data were derived from a total of ten different males and 17 different females in F93–96, and 31 different males and 21 different females in T93–98.