Computer vision enables taxon-specific identification of African carnivore tooth marks on bone.

Taphonomic works aim at discovering how paleontological and archaeofaunal assemblages were formed. They also aim at determining how hominin fossils were preserved or destroyed. Hominins and other mammal carnivores have been co-evolving, at least during the past two million years, and their potential interactions determined the evolution of human behavior. In order to understand all this, taxon-specific carnivore agency must be effectively identified in the fossil record. Until now, taphonomists have been able to determine, to some degree, hominin and carnivore inputs in site formation, and their interactions in the modification of part of those assemblages. However, the inability to determine agency more specifically has hampered the development of taphonomic research, whose methods are virtually identical to those used several decades ago (lagged by a high degree of subjectivity). A call for more objective and agent-specific methods would be a major contribution to the advancement of taphonomic research. Here, we present one of these advances. The use of computer vision (CV) on a large data set of images of tooth marks has enabled the objective discrimination of taxon-specific carnivore agency up to 88% of the testing sample. We highlight the significance of this method in an interdisciplinary interplay between traditional taphonomic-paleontological analysis and artificial intelligence-based computer science. The new questions that can be addressed with this will certainly bring important changes to several ideas on important aspects of the human evolutionary process.

Biomechanical and taxonomic diversity in the Early Pleistocene in East Africa: Structural analysis of a recently discovered femur shaft from Olduvai Gorge (bed I).

Recent Plio-Pleistocene hominin findings have revealed the complexity of human evolutionary history and the difficulties involved in its interpretation. Moreover, the study of hominin long bone remains is particularly problematic, since it commonly depends on the analysis of fragmentary skeletal elements that in many cases are merely represented by small diaphyseal portions and appear in an isolated fashion in the fossil record. Nevertheless, the study of the postcranial skeleton is particularly important to ascertain locomotor patterns. Here we report on the discovery of a robust hominin femoral fragment (OH 84) at the site of Amin Mturi Korongo dated to 1.84 Ma (Olduvai Bed I). External anatomy and internal bone structure of OH 84 were analyzed and compared with previously published data for modern humans and chimpanzees, as well as for Australopithecus, Paranthropus and Homo specimens ranging from the Late Pliocene to Late Pleistocene. Biomechanical analyses based on transverse cross-sections and the comparison of OH 84 with another robust Olduvai specimen (OH 80) suggest that OH 84 might be tentatively allocated to Paranthropus boisei. More importantly, the identification of a unique combination of traits in OH 84 could indicate both terrestrial bipedalism and an arboreal component in the locomotor repertoire of this individual. If interpreted correctly, OH 84 could thus add to the already mounting evidence of substantial locomotor diversity among Early Pleistocene hominins. Likewise, our results also highlight the difficulties in accurately interpreting the link between form and function in the human fossil record based on fragmentary remains, and ultimately in distinguishing between coeval hominin groups due to the heterogeneous pattern of inter- and intraspecific morphological variability detected among fossil femora.

African bovid tribe classification using transfer learning and computer vision.

Objective analytical identification methods are still a minority in the praxis of paleobiological sciences. Subjective interpretation of fossils and their modifications remains a nonreplicable expert endeavor. Identification of African bovids is a crucial element in the reconstruction of paleo-landscapes, ungulate paleoecology, and, eventually, hominin adaptation and ecosystemic reconstruction. Recent analytical efforts drawing on Fourier functional analysis and discrimination methods applied to occlusal surfaces of teeth have provided a highly accurate framework to correctly classify African bovid tribes and taxa. Artificial intelligence tools, like computer vision, have also shown their potential to be objectively more accurate in the identification of taphonomic agency than human experts. For this reason, here we implement some of the most successful computer vision methods, using transfer learning and ensemble analysis, to classify bidimensional images of African bovid teeth and show that 92% of the large testing set of images of African bovid tribes analyzed could be correctly classified. This brings an objective tool to paleoecological interpretation, where bovid identification and paleoecological interpretation can be more confidently carried out.

Computer vision supports primary access to meat by early Homo 1.84 million years ago.

Human carnivory is atypical among primates. Unlike chimpanzees and bonobos, who are known to hunt smaller monkeys and eat them immediately, human foragers often cooperate to kill large animals and transport them to a safe location to be shared. While it is known that meat became an important part of the hominin diet around 2.6-2 Mya, whether intense cooperation and food sharing developed in conjunction with the regular intake of meat remains unresolved. A widespread assumption is that early hominins acquired animal protein through klepto-parasitism at felid kills. This should be testable by detecting felid-specific bone modifications and tooth marks on carcasses consumed by hominins. Here, deep learning (DL) computer vision was used to identify agency through the analysis of tooth pits and scores on bones recovered from the Early Pleistocene site of DS (Bed I, Olduvai Gorge). We present the first objective evidence of primary access to meat by hominins 1.8 Mya by showing that the most common securely detectable bone-modifying fissipeds at the site were hyenas. The absence of felid modifications in most of the carcasses analyzed indicates that hominins were the primary consumers of most animals accumulated at the site, with hyenas intervening at the post-depositional stage. This underscores the role of hominins as a prominent part of the early Pleistocene African carnivore guild. It also stresses the major (and potentially regular) role that meat played in the diet that configured the emergence of early Homo.

Neo-taphonomic analysis of the Misiam leopard lair from Olduvai Gorge (Tanzania): understanding leopard-hyena interactions in open settings.

Misiam is a modern wildebeest-dominated accumulation situated in a steep ravine covered with dense vegetation at Olduvai Gorge (Tanzania). It is interpreted here as a leopard lair to which carcasses have been transported for several years. Felid-specific bone damage patterns, felid-typical skeletal part profiles, taxonomic specialization and the physical presence of leopards observed by the authors show that leopards at Misiam can be specialized medium-sized carcass accumulators. Hyenas also intervened at intervals in the modification of the retrieved faunal assemblage. This makes Misiam a carnivore palimpsest. Here, we additionally show that leopards only transport and accumulate carcasses on occasions, that they can seem highly specialized despite being dietary generalists, and that such a behaviour may be prompted by seasonal competition or during the breeding season or both. Misiam is the first open-air leopard lair with a dense bone accumulation reported. There, leopards engaged in intensive accumulation of carcasses during the wet season, when the southern Serengeti short-grass plains undergo the effect of the famous wildebeest migration and this migratory species reaches the gorge. The ecological importance of this behaviour and its relevance as a proxy for reconstructing prehistoric carnivore behaviours are discussed.

New site at Olduvai Gorge (AGS, Bed I, 1.84 Mya) widens the range of locations where hominins engaged in butchery.

Outstanding questions about human evolution include systematic connections between critical landscape resources-such as water and food-and how these shaped the competitive and biodiverse environment(s) that our ancestors inhabited. Here, we report fossil n-alkyl lipid biomarkers and their associated δ13C values across a newly discovered Olduvai Gorge site (AGS) dated to 1.84 million years ago, enabling a multiproxy analysis of the distributions of critical local landscape resources across an explicit locus of hominin activity. Our results reveal that AGS was a seasonally waterlogged, largely unvegetated lakeside site situated near an ephemeral freshwater river surrounded by arid-adapted C4 grasses. The sparse vegetation at AGS contrasts with reconstructed (micro)habitats at the other anthropogenic sites at Olduvai Gorge, suggesting that central-provisioning places depended more heavily on water access than vegetation viz. woody plants as is often observed for modern hunter-gatherers. As hominins at AGS performed similar butchering activities as at other Bed I sites, our results suggest they did not need the shelter of trees and thus occupied a competitive position within the predatory guild.

Sabertooth carcass consumption behavior and the dynamics of Pleistocene large carnivoran guilds.

Apex predators play an important role in the top-down regulation of ecological communities. Their hunting and feeding behaviors influence, respectively, prey demography and the availability of resources to other consumers. Among the most iconic-and enigmatic-terrestrial predators of the late Cenozoic are the Machairodontinae, a diverse group of big cats whose hypertrophied upper canines have earned them the moniker "sabertooths." Many aspects of these animals' paleobiology, especially their prey preferences and carcass consumption behavior, remain unsettled. While skeletal anatomy, dental morphology and wear, and isotopic profiles provide important insights, the most direct way to resolve these issues is through the fossil remains of sabertooth prey. Here, we report on a taphonomic analysis of an early Pleistocene faunal assemblage from Haile 21A (Florida, USA) that preserves feeding damage from the lion-sized sabertooth Xenosmilus hodsonae. Patterns of tooth-marking and bone damage indicate that Xenosmilus fully defleshed the carcasses of their prey and even engaged in some minor bone consumption. This has important implications for Pleistocene carnivoran guild dynamics, including the carcass foraging behavior of the first stone-tool-using hominins.

A case of hominin scavenging 1.84 million years ago from Olduvai Gorge (Tanzania).

Meat eating is one of the hallmarks of human evolution. It has been linked to the beginning of stone tool use, to physiological changes leading to crucial anatomical transformations defining our genus, and to new socioreproductive and cognitive behaviors. Uncontroversial evidence of meat eating goes back to 2.6 million years ago; however, little is known about the frequency and timing with which early hominins acquired animal resources. Here, we show that the combination of hunting and scavenging documented in some modern human foragers may have a long evolutionary trajectory. Using a new set of artificial intelligence methods for objective identification, we present direct evidence of an episode of hominins scavenging from large felids-probably lions-discovered at Olduvai Gorge (DS site, Bed I). This casts a new perspective on the diversity of hominin carcass acquisition behaviors and survival strategies, and places some early Pleistocene hominins in ecological proximity to African large carnivore guilds.

Early Pleistocene faunivorous hominins were not kleptoparasitic, and this impacted the evolution of human anatomy and socio-ecology.

Humans are unique in their diet, physiology and socio-reproductive behavior compared to other primates. They are also unique in the ubiquitous adaptation to all biomes and habitats. From an evolutionary perspective, these trends seem to have started about two million years ago, coinciding with the emergence of encephalization, the reduction of the dental apparatus, the adoption of a fully terrestrial lifestyle, resulting in the emergence of the modern anatomical bauplan, the focalization of certain activities in the landscape, the use of stone tools, and the exit from Africa. It is in this period that clear taphonomic evidence of a switch in diet with respect to Pliocene hominins occurred, with the adoption of carnivory. Until now, the degree of carnivorism in early humans remained controversial. A persistent hypothesis is that hominins acquired meat irregularly (potentially as fallback food) and opportunistically through klepto-foraging. Here, we test this hypothesis and show, in contrast, that the butchery practices of early Pleistocene hominins (unveiled through systematic study of the patterning and intensity of cut marks on their prey) could not have resulted from having frequent secondary access to carcasses. We provide evidence of hominin primary access to animal resources and emphasize the role that meat played in their diets, their ecology and their anatomical evolution, ultimately resulting in the ecologically unrestricted terrestrial adaptation of our species. This has major implications to the evolution of human physiology and potentially for the evolution of the human brain.

Tracing the spatial imprint of Oldowan technological behaviors: A view from DS (Bed I, Olduvai Gorge, Tanzania).

DS (David's site) is one of the new archaeological sites documented in the same paleolandscape in which FLK 22 was deposited at about 1.85 Ma in Olduvai Gorge. Fieldwork in DS has unearthed the largest vertically-discrete archaeological horizon in the African Pleistocene, where a multi-cluster anthropogenic accumulation of fossil bones and stone tools has been identified. In this work we present the results of the techno-economic study of the lithic assemblage recovered from DS. We also explore the spatial magnitude of the technological behaviors documented at this spot using powerful spatial statistical tools to unravel correlations between the spatial distributional patterns of lithic categories. At DS, lavas and quartzite were involved in different technological processes. Volcanic materials, probably transported to this spot from a close source, were introduced in large numbers, including unmodified materials, and used in percussion activities and in a wide variety of reduction strategies. A number of volcanic products were subject to outward fluxes to other parts of the paleolandscape. In contrast, quartzite rocks were introduced in smaller numbers and might have been subject to a significantly more intense exploitation. The intra-site spatial analysis has shown that specialized areas cannot be identified, unmodified materials are not randomly distributed, percussion and knapping categories do not spatially overlap, while bipolar specimens show some sort of spatial correlation with percussion activities.

Deep classification of cut-marks on bones from Arroyo del Vizcaíno (Uruguay).

The earliest widely accepted presence of humans in America dates to approximately 17.5 cal kyr BP, at the end of the Last Glacial Maximum (LGM). Among other evidence, this presence is attested by stone tools and associated cut-marks and other bone surface modifications (BSM), interpreted as the result of the consumption of animals by humans. Claims of an older human presence in the continent have been made based on the proposed anthropogenic modification of faunal remains; however, these have been controversial due to the highly subjective nature of the interpretations. Here, we employ advanced deep learning algorithms to objectively increase the accuracy of BSM identification on bones. With several models that exhibit BSM classification accuracies greater than 94%, we use ensemble learning techniques to robustly classify a selected sample of BSM from the approximately 30 kyr BP site of Arroyo del Vizcaíno, Uruguay. Our results confidently show the presence of cut-marks imparted by stone tools on bones at the site. This result supports an earlier presence of humans in the American continent, expanding additional genetic and archaeological evidence of a human LGM and pre-LGM presence in the continent.

Elevated rates of horizontal gene transfer in the industrialized human microbiome.

Industrialization has impacted the human gut ecosystem, resulting in altered microbiome composition and diversity. Whether bacterial genomes may also adapt to the industrialization of their host populations remains largely unexplored. Here, we investigate the extent to which the rates and targets of horizontal gene transfer (HGT) vary across thousands of bacterial strains from 15 human populations spanning a range of industrialization. We show that HGTs have accumulated in the microbiome over recent host generations and that HGT occurs at high frequency within individuals. Comparison across human populations reveals that industrialized lifestyles are associated with higher HGT rates and that the functions of HGTs are related to the level of host industrialization. Our results suggest that gut bacteria continuously acquire new functionality based on host lifestyle and that high rates of HGT may be a recent development in human history linked to industrialization.

A 3D taphonomic model of long bone modification by lions in medium-sized ungulate carcasses.

Here, we present the first three-dimensional taphonomic analysis of a carnivore-modified assemblage at the anatomical scale of the appendicular skeleton. A sample of ten carcasses composed of two taxa (zebra and wildebeest) consumed by wild lions in the Tarangire National Park (Tanzania) has been used to determine element-specific lion damage patterns. This study presents a novel software for the 3D spatial documentation of bone surface modifications at the anatomical level. Combined with spatial statistics, the present analysis has been able to conclude that despite variable degrees of competition during carcass consumption, lions generate bilateral patterning consisting of substantial damage of proximal ends of stylopodials and zeugopodials, moderate damage of the distal ends of femora and marginal damage of distal ends of humeri and zeugopodials. Of special interest is, specifically, the patterning of tooth marks on shafts according to element, since these are crucial to determine not only the type of carnivore involved in any given bone assemblage, but also the interaction with other agents (namely, hominins, in the past). Lions leave few tooth marks on mid-shaft sections, mostly concentrated on certain sections and orientations of stylopodials and, to a lesser extent, of the proximal tibia. Redundant occurrence of tooth marks on certain bone sections renders them as crucial to attest lion agency in carcass initial consumption. Indirectly, this can also be used to determine whether hominins ever acquired carcasses at lion kills.

Do human butchery patterns exist? A study of the interaction of randomness and channelling in the distribution of cut marks on long bones.

Bone surface modifications (BSMs) in faunal assemblages are frequently used to infer past agency and actions of hominins and carnivores, with implications for the emergence of key human behaviours. Patterning of BSMs has mostly been defined as a combination of the intensity of marks per bone portion and sometimes per element. Numerous variables involved in butchery can condition cut mark anatomical distribution, so much so that these variables are widely assumed to be stochastic. Here, we present a new methodological approach using a novel geospatial tool (Ikhnos) which combines the three-dimensional spatial documentation of cut mark patterns with spatial statistics based on wavelets, applied to three experimental and ethnoarchaeological faunal assemblages. We use wavelets to identify patterning of multiple longitudinal series of cut mark distributions on bones, and to establish similarities or differences in patterning within and across different assemblages. This method demonstrates the existence of general and behaviour-specific butchery patterns. It can also be used to effectively assess the proportion of mark clustering that is due to randomness, versus that which is conditioned by the butchery process.

Distinguishing Discoid and Centripetal Levallois methods through machine learning.

In this paper, we apply Machine Learning (ML) algorithms to study the differences between Discoid and Centripetal Levallois methods. For this purpose, we have used experimentally knapped flint flakes, measuring several parameters that have been analyzed by seven ML algorithms. From these analyses, it has been possible to demonstrate the existence of statistically significant differences between Discoid products and Centripetal Levallois products, thus contributing with new data and a new method to this traditional debate. The new approach enabled differentiating the blanks created by both knapping methods with an accuracy >80% using only ten typometric variables. The most relevant variables were maximum length, width to the 25%, 50% and 75% of the flake length, external and internal platform angles, maximum width and number of dorsal scars. This study also demonstrates the advantages of the application of multivariate ML methods to lithic studies.

Artificial intelligence provides greater accuracy in the classification of modern and ancient bone surface modifications.

Bone surface modifications are foundational to the correct identification of hominin butchery traces in the archaeological record. Until present, no analytical technique existed that could provide objectivity, high accuracy, and an estimate of probability in the identification of multiple structurally-similar and dissimilar marks. Here, we present a major methodological breakthrough that incorporates these three elements using Artificial Intelligence (AI) through computer vision techniques, based on convolutional neural networks. This method, when applied to controlled experimental marks on bones, yielded the highest rate documented to date of accurate classification (92%) of cut, tooth and trampling marks. After testing this method experimentally, it was applied to published images of some important traces purportedly indicating a very ancient hominin presence in Africa, America and Europe. The preliminary results are supportive of interpretations of ancient butchery in some places, but not in others, and suggest that new analyses of these controversial marks should be done following the protocol described here to confirm or disprove these archaeological interpretations.

Microbial biomarkers reveal a hydrothermally active landscape at Olduvai Gorge at the dawn of the Acheulean, 1.7 Ma.

Landscape-scale reconstructions of ancient environments within the cradle of humanity may reveal insights into the relationship between early hominins and the changing resources around them. Many studies of Olduvai Gorge during Pliocene-Pleistocene times have revealed the presence of precession-driven wet-dry cycles atop a general aridification trend, though may underestimate the impact of local-scale conditions on early hominins, who likely experienced a varied and more dynamic landscape. Fossil lipid biomarkers from ancient plants and microbes encode information about their surroundings via their molecular structures and composition, and thus can shed light on past environments. Here, we employ fossil lipid biomarkers to study the paleolandscape at Olduvai Gorge at the emergence of the Acheulean technology, 1.7 Ma, through the Lower Augitic Sandstones layer. In the context of the expansion of savanna grasslands, our results represent a resource-rich mosaic ecosystem populated by groundwater-fed rivers, aquatic plants, angiosperm shrublands, and edible plants. Evidence of a geothermally active landscape is reported via an unusual biomarker distribution consistent with the presence of hydrothermal features seen today at Yellowstone National Park. The study of hydrothermalism in ancient settings and its impact on hominin evolution has not been addressed before, although the association of thermal springs in the proximity of archaeological sites documented here can also be found at other localities. The hydrothermal features and resources present at Olduvai Gorge may have allowed early hominins to thermally process edible plants and meat, supporting the possibility of a prefire stage of human evolution.

Deep learning and taphonomy: high accuracy in the classification of cut marks made on fleshed and defleshed bones using convolutional neural networks.

Accurate identification of bone surface modifications (BSM) is crucial for the taphonomic understanding of archaeological and paleontological sites. Critical interpretations of when humans started eating meat and animal fat or when they started using stone tools, or when they occupied new continents or interacted with predatory guilds impinge on accurate identifications of BSM. Until now, interpretations of Plio-Pleistocene BSM have been contentious because of the high uncertainty in discriminating among taphonomic agents. Recently, the use of machine learning algorithms has yielded high accuracy in the identification of BSM. A branch of machine learning methods based on imaging, computer vision (CV), has opened the door to a more objective and accurate method of BSM identification. The present work has selected two extremely similar types of BSM (cut marks made on fleshed an defleshed bones) to test the immense potential of artificial intelligence methods. This CV approach not only produced the highest accuracy in the classification of these types of BSM until present (95% on complete images of BSM and 88.89% of images of only internal mark features), but it also has enabled a method for determining which inconspicuous microscopic features determine successful BSM discrimination. The potential of this method in other areas of taphonomy and paleobiology is enormous.

The meta-group social network of early humans: A temporal-spatial assessment of group size at FLK Zinj (Olduvai Gorge, Tanzania).

Humans are the only primates that maintain regular inter-group relationships and meta-group social networks that enable the inter-group flow of individuals. This is the basis of the band/tribe concept in the anthropology of modern foragers. The present work is a theoretical approach to the development of analytical tools to understand group size and the temporal scale of site occupation in the archaeological record. We selected FLK Zinj as one of the oldest examples of a taphonomically-supported anthropogenic site in which both variables (group size and time) could be modelled using a combination of modern forager regression estimates from their camp sizes and estimates derived from the combined use of modern African foragers' meat consumption rates per day per capita during the dry season and minimum estimates of flesh yields provided by the carcass parts preserved at FLK Zinj. This approach provides the basis for a testable hypothesis which should be further tested in other Oldowan sites. An estimate of 18-28 individuals occupying FLK Zinj was made, which is similar to the estimated 16 individuals of one of the 1.5 Ma Ileret Homo erectus footprint trails. It also shows a similar proportional distribution to Dunbar's equations (group size to neocortex ratio) as documented in modern foragers, which suggest that most of the social network of H. erectus was in the meta-group level as is the case of modern foragers. Irrespective of the range of variation discussed for both variables (group size and length of time represented), it is argued that neither small estimates of time nor small group sizes can account for the formation of FLK Zinj.