Periodic Reporting for period 1 - CRAHL (Comparative Reconstruction of Ape and Hominin Locomotor ontogeny)
Reporting period: 2023-01-09 to 2025-01-08
Summary of the context and overall objectives of the project
Human locomotion is unique among apes. Unlike our closest relatives, humans are fully bipedal, exclusively terrestrial, and develop slowly (altricial). In contrast, other apes are at least partially arboreal and mature more quickly (precocial). Fossil evidence suggests that our early ancestors (hominins) followed a more ape-like developmental trajectory, with a combination of arboreal and terrestrial movement and a precocial pattern of locomotor development.
To study the evolution of hominin locomotion and its developmental changes, we need reliable skeletal traits that respond to locomotor stress and can be identified in the fossil record. Great apes provide a natural comparative framework for this research because they exhibit diverse locomotor behaviors throughout their development. For example:
Humans are fully terrestrial and bipedal from an early age.
Chimpanzees and gorillas start life as highly arboreal but transition to terrestrial knuckle-walking as they mature, with chimpanzees remaining arboreal for a longer period.
Orangutans and gibbons remain predominantly arboreal throughout their lives.
Objectives of the Project
This project aimed to improve our understanding of the evolution and development of locomotion in human ancestors. Specifically, it sought to answer how humans transitioned from an arboreal and rapidly developing ancestor to the fully terrestrial and slowly developing species we are today.
The research focused on two main objectives:
Examining how locomotion develops in humans and non-human primates and its impact on skeletal structure.
Like muscles, bones adapt to the stresses they experience during life. This study focused on trabecular bone—the spongy inner bone structure—which responds dynamically to locomotor forces by adjusting its density and organization.
By analyzing how trabecular bone changes during development in primates with different locomotor strategies, we can determine whether these skeletal adaptations leave a detectable signature of movement patterns.
Applying these findings to the hominin fossil record to reconstruct locomotor development in our ancestors.
The study sought to understand how quickly hominins could walk and climb independently and whether they, like modern apes, exhibited a strongly arboreal phase in early development.
By integrating comparative primate data with fossil evidence, this research provides new insights into the evolutionary shifts that shaped human locomotion and developmental patterns.
To study the evolution of hominin locomotion and its developmental changes, we need reliable skeletal traits that respond to locomotor stress and can be identified in the fossil record. Great apes provide a natural comparative framework for this research because they exhibit diverse locomotor behaviors throughout their development. For example:
Humans are fully terrestrial and bipedal from an early age.
Chimpanzees and gorillas start life as highly arboreal but transition to terrestrial knuckle-walking as they mature, with chimpanzees remaining arboreal for a longer period.
Orangutans and gibbons remain predominantly arboreal throughout their lives.
Objectives of the Project
This project aimed to improve our understanding of the evolution and development of locomotion in human ancestors. Specifically, it sought to answer how humans transitioned from an arboreal and rapidly developing ancestor to the fully terrestrial and slowly developing species we are today.
The research focused on two main objectives:
Examining how locomotion develops in humans and non-human primates and its impact on skeletal structure.
Like muscles, bones adapt to the stresses they experience during life. This study focused on trabecular bone—the spongy inner bone structure—which responds dynamically to locomotor forces by adjusting its density and organization.
By analyzing how trabecular bone changes during development in primates with different locomotor strategies, we can determine whether these skeletal adaptations leave a detectable signature of movement patterns.
Applying these findings to the hominin fossil record to reconstruct locomotor development in our ancestors.
The study sought to understand how quickly hominins could walk and climb independently and whether they, like modern apes, exhibited a strongly arboreal phase in early development.
By integrating comparative primate data with fossil evidence, this research provides new insights into the evolutionary shifts that shaped human locomotion and developmental patterns.
Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far
Work performed for objective 1:
My first task was to collect extant data on the skeletal growth and development of modern primates. I MicroCT-scanned primate skeletons from the collections of Naturalis – Leiden, and the Schulz Anthropological collection – University of Zurich. Data was processed in parallel and continued throughout the project, finishing on time. Results were published in an open access paper in Frontiers in Evolution and Ecology. Analyses for a second paper are completed and the paper is in the final stages of drafting. Submission is expected in March 2025.
I carried out methodological innovations that that resulted in the publication of a new R package (see below).
Achievements:
1. I assembled the largest archive of microCT scans of non-adult primate postcranial bones that will form the basis of numerous future projects.
2. To process and analyse the CT data I wrote an R package called trabmap. This package processes CT scans of bone to quantify, visualize, and statistically analyse 3D variation in the distribution of trabecular bone. Writing this package was not initially planned as part of CRAHL but nonetheless represents a major output of the project and a considerable advancement to the field. The only software that could do this was locked behind an expensive paywall. Now, researchers can process and analyse trabecular structure in 3D for free using trabmap without having to pay thousands of euros per year. Github link: https://github.com/Jaapro/trabmap(opens in new window)
3. I published findings in an open access paper in a special issue of Frontiers in Ecology and Evolution on the evolution of hominin upright walking. In this paper I showed how developmental trajectories of trabecular bone in primates corresponds to developmental changes in locomotor behaviour across primates. I then proposed how these results may be used to reconstruct the evolution of both locomotor development and aspects of life history in fossil hominins.
4. A paper on the interactions between life history, locomotor development, and bone structure across humans, other primates, and fossil hominins is in the final stages of drafting. This work provides unique new insights into modern human skeletal health. Humans possess relatively fragile bones for a primate of our body size, even when accounting for modern lifestyle factors. This skeletal fragility, combined with increasing longevity and reduced physical activity, contributes to osteoporosis risk. In this paper I demonstrate that this fragility originates from our slow, altricial life history—a consequence of our exceptionally large brains. I also demonstrate how this reduction in bone quality evolved over the last 3 million years because of increases in hominin brain size. As clinicians are well aware, bed rest is detrimental to bone health. Because of prolonged motor development due to our large brains, human infants experience a form of “bed rest” in early life, leading to significant early-life bone loss that predisposes individuals to osteoporosis later in life. Understanding these developmental constraints offers crucial perspectives on the evolutionary origins of human skeletal health and potential strategies for mitigating bone loss in aging populations. Submission of the paper is expected in March 2025 and will be submitted to Science.
Work performed for objective 2L
I travelled to Johannesburg in South Africa to spend two weeks at the fossil hominin vault at the University of the Witwatersrand in March 2023. There, I was able to investigate the fossil collection and select specimens to be CT scanned. CT scanning was delayed since the machine was broken during my visit, and a new technician was recruited. I received all fossil CT scans that I requested by the end of the project. Therefore, my time plan was changed dramatically and caused a delay of publications. Data processing was carried out in parallel to data gathering and continued throughout the project as scans became gradually accessible. Since the machine learning model that I intended to use to segment the fossils did not work adequately, I chose to develop a new one trained with fossil data obtained during the project. This required manually segmenting CT scans of fossils and using this data to train a new Deep Learning model. For this task I recruited seven students from the Faculties of Archaeology and Biology at the University of Leiden, they created the necessary training data. This approach prooved to be efficient and safed a lot of my time. The model represents a major step forward for the field, as it can now automatically segment fossil CT scans that prior to its development would have taken weeks or months to do what the model can do in minutes.
I supervised two MSc students that carried out projects on the fossil materials. Both projects are nearly ready to submit as papers to peer-reviewed journals. In addition, I have finished analyses for two more papers on hominin locomotor development that are in the final stages of drafting.
achievements:
1. Collected and processed a large database of Micro CT scans of South African hominin fossils including postcranial elements from Homo naledi, Paranthropus robustus, Australopithecus sediba, and Australopithecus africanus. These CT scans will be used for numerous future publications and projects.
2. I developed a deep learning model to process CT scans of fossil hominins. The preservation of fossils with often damaged specimens and infill with matrix, makes it incredibly difficult to separate bone from other inclusions and diagenetic alterations. The model can segment these messy scans neatly into trabecular bone, cortical bone, and other materials. This model does in minutes what previously took weeks or months of manual segmentation work. Developing this model was not initially planned as part of CRAHL but nevertheless represents a major output of the action and a considerable advancement to the field. The model is available open access from my GitHub page. I provided training to other researchers at Naturalis who continued to use this approach in their own research lines. Seven students from the University of Leiden (Archaeology and Biology) were supervised by me to generate the training data for the model, providing them with hands on experience working with CT scans of fossil hominins and Deep Learning methods. GitHub link: https://github.com/Jaapro/Deep_learning_segmentation(opens in new window)
3. One paper on locomotor development in Homo naledi is in the final stages of drafting with analyses completed. Results show that Homo naledi infants, like modern chimpanzees and gorillas, were more arboreal than adults of the species. This likely represents the ancestral condition in the hominin lineage before the evolution of Homo erectus. The paper will be submitted to the Journal of Human Evolution.
4. Another paper on the relationship between hominin locomotor development, life history evolution, and their effects on modern human bone health is in the final stage of drafting and will be submitted to Science. This is the same paper as D1.2
5. One paper on reconstructing the locomotion of adult Homo naledi is in final stages of drafting. The paper uses a novel approach, using canalized and plastic traits to reconstruct locomotion in this fascinating species of hominin that lived alongside the earliest members of Homo sapiens. We found that Homo naledi is more arboreal than modern humans, which, combined with its small body size and small brain suggests that this hominin inhabited a niche not seen in the human lineage since almost two million years ago. I plan to submit the manuscript to Nature Ecology and Evolution, co-authored with the MSc student I supervised as part of CRAHL.
6. One paper on the role of climbing in South African hominin locomotion is in the final stages of drafting. I compared the trabecular structure in the upper and lower limbs of Australopithecus sediba, Australopithecus africanus, Homo naledi, and Paranthropus robustus. Results show that all of these hominins still relied on arboreal locomotion to a significant degree, despite their derived lower limb morphology. The paper will be submitted to the Journal of Human Evolution, co-authored with the MSc student that was supervised as part of CRAHL.
My first task was to collect extant data on the skeletal growth and development of modern primates. I MicroCT-scanned primate skeletons from the collections of Naturalis – Leiden, and the Schulz Anthropological collection – University of Zurich. Data was processed in parallel and continued throughout the project, finishing on time. Results were published in an open access paper in Frontiers in Evolution and Ecology. Analyses for a second paper are completed and the paper is in the final stages of drafting. Submission is expected in March 2025.
I carried out methodological innovations that that resulted in the publication of a new R package (see below).
Achievements:
1. I assembled the largest archive of microCT scans of non-adult primate postcranial bones that will form the basis of numerous future projects.
2. To process and analyse the CT data I wrote an R package called trabmap. This package processes CT scans of bone to quantify, visualize, and statistically analyse 3D variation in the distribution of trabecular bone. Writing this package was not initially planned as part of CRAHL but nonetheless represents a major output of the project and a considerable advancement to the field. The only software that could do this was locked behind an expensive paywall. Now, researchers can process and analyse trabecular structure in 3D for free using trabmap without having to pay thousands of euros per year. Github link: https://github.com/Jaapro/trabmap(opens in new window)
3. I published findings in an open access paper in a special issue of Frontiers in Ecology and Evolution on the evolution of hominin upright walking. In this paper I showed how developmental trajectories of trabecular bone in primates corresponds to developmental changes in locomotor behaviour across primates. I then proposed how these results may be used to reconstruct the evolution of both locomotor development and aspects of life history in fossil hominins.
4. A paper on the interactions between life history, locomotor development, and bone structure across humans, other primates, and fossil hominins is in the final stages of drafting. This work provides unique new insights into modern human skeletal health. Humans possess relatively fragile bones for a primate of our body size, even when accounting for modern lifestyle factors. This skeletal fragility, combined with increasing longevity and reduced physical activity, contributes to osteoporosis risk. In this paper I demonstrate that this fragility originates from our slow, altricial life history—a consequence of our exceptionally large brains. I also demonstrate how this reduction in bone quality evolved over the last 3 million years because of increases in hominin brain size. As clinicians are well aware, bed rest is detrimental to bone health. Because of prolonged motor development due to our large brains, human infants experience a form of “bed rest” in early life, leading to significant early-life bone loss that predisposes individuals to osteoporosis later in life. Understanding these developmental constraints offers crucial perspectives on the evolutionary origins of human skeletal health and potential strategies for mitigating bone loss in aging populations. Submission of the paper is expected in March 2025 and will be submitted to Science.
Work performed for objective 2L
I travelled to Johannesburg in South Africa to spend two weeks at the fossil hominin vault at the University of the Witwatersrand in March 2023. There, I was able to investigate the fossil collection and select specimens to be CT scanned. CT scanning was delayed since the machine was broken during my visit, and a new technician was recruited. I received all fossil CT scans that I requested by the end of the project. Therefore, my time plan was changed dramatically and caused a delay of publications. Data processing was carried out in parallel to data gathering and continued throughout the project as scans became gradually accessible. Since the machine learning model that I intended to use to segment the fossils did not work adequately, I chose to develop a new one trained with fossil data obtained during the project. This required manually segmenting CT scans of fossils and using this data to train a new Deep Learning model. For this task I recruited seven students from the Faculties of Archaeology and Biology at the University of Leiden, they created the necessary training data. This approach prooved to be efficient and safed a lot of my time. The model represents a major step forward for the field, as it can now automatically segment fossil CT scans that prior to its development would have taken weeks or months to do what the model can do in minutes.
I supervised two MSc students that carried out projects on the fossil materials. Both projects are nearly ready to submit as papers to peer-reviewed journals. In addition, I have finished analyses for two more papers on hominin locomotor development that are in the final stages of drafting.
achievements:
1. Collected and processed a large database of Micro CT scans of South African hominin fossils including postcranial elements from Homo naledi, Paranthropus robustus, Australopithecus sediba, and Australopithecus africanus. These CT scans will be used for numerous future publications and projects.
2. I developed a deep learning model to process CT scans of fossil hominins. The preservation of fossils with often damaged specimens and infill with matrix, makes it incredibly difficult to separate bone from other inclusions and diagenetic alterations. The model can segment these messy scans neatly into trabecular bone, cortical bone, and other materials. This model does in minutes what previously took weeks or months of manual segmentation work. Developing this model was not initially planned as part of CRAHL but nevertheless represents a major output of the action and a considerable advancement to the field. The model is available open access from my GitHub page. I provided training to other researchers at Naturalis who continued to use this approach in their own research lines. Seven students from the University of Leiden (Archaeology and Biology) were supervised by me to generate the training data for the model, providing them with hands on experience working with CT scans of fossil hominins and Deep Learning methods. GitHub link: https://github.com/Jaapro/Deep_learning_segmentation(opens in new window)
3. One paper on locomotor development in Homo naledi is in the final stages of drafting with analyses completed. Results show that Homo naledi infants, like modern chimpanzees and gorillas, were more arboreal than adults of the species. This likely represents the ancestral condition in the hominin lineage before the evolution of Homo erectus. The paper will be submitted to the Journal of Human Evolution.
4. Another paper on the relationship between hominin locomotor development, life history evolution, and their effects on modern human bone health is in the final stage of drafting and will be submitted to Science. This is the same paper as D1.2
5. One paper on reconstructing the locomotion of adult Homo naledi is in final stages of drafting. The paper uses a novel approach, using canalized and plastic traits to reconstruct locomotion in this fascinating species of hominin that lived alongside the earliest members of Homo sapiens. We found that Homo naledi is more arboreal than modern humans, which, combined with its small body size and small brain suggests that this hominin inhabited a niche not seen in the human lineage since almost two million years ago. I plan to submit the manuscript to Nature Ecology and Evolution, co-authored with the MSc student I supervised as part of CRAHL.
6. One paper on the role of climbing in South African hominin locomotion is in the final stages of drafting. I compared the trabecular structure in the upper and lower limbs of Australopithecus sediba, Australopithecus africanus, Homo naledi, and Paranthropus robustus. Results show that all of these hominins still relied on arboreal locomotion to a significant degree, despite their derived lower limb morphology. The paper will be submitted to the Journal of Human Evolution, co-authored with the MSc student that was supervised as part of CRAHL.
Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)
The project led to a novel understanding of the relationship between brain size, life history, and skeletal development that has significant impacts on modern human health. I found strong interactions between life history, locomotor development, and bone structure across humans, other primates, and fossil hominins. This research provides novel insights into modern human skeletal health, particularly our susceptibility to bone fragility and osteoporosis. Humans have relatively fragile bones compared to other primates of similar body size, even when accounting for modern lifestyle factors. This skeletal weakness, combined with increased longevity and reduced physical activity, significantly contributes to osteoporosis risk. I demonstrated that human bone fragility originates from our slow, altricial life history—a direct consequence of our exceptionally large brains. Furthermore, I show that the decline in bone quality evolved over the past three million years in parallel with increases in hominin brain size. Clinicians recognize that prolonged bed rest negatively impacts bone health. Similarly, due to our extended motor development—a result of our large brains—human infants experience a form of "bed rest" early in life. This leads to significant early-life bone loss, predisposing individuals to osteoporosis later in adulthood. Understanding these developmental constraints offers crucial evolutionary insights into human skeletal health and may inform strategies to mitigate bone loss in aging populations.