How woodpecker protects its brain from concussion during pecking compared with chicken and pigeon

A human head often suffers various injuries when it is subjected to an external force. Head injury not only involves damage to the skull and scalp but also often results in traumatic brain injury (TBI).^1,2 Head injury is a major cause of death and disability worldwide, especially in children and young adults. During 2002–2006, the incidence of head injury is about 1.7 × 10⁶ people in the United States each year. About 3% of these incidents leads to death.^3,4 In nature, however, many birds drum with their beak on hard and solid objects to feed themselves without head injury. The woodpecker is a typical and remarkable bird to have a wonderful protecting ability from TBI during its high-speed pecking. Moreover, other birds also evolve their own protective strategy to avoid TBI during feed related pecking.

Studies on woodpeckers focus on the shock resistant ability of the woodpecker’s head. Several studies have reported the pecking data of woodpeckers, including pecking velocity and pecking force, which are measured through experimental observations of the pecking behavior of live woodpeckers.^5–7 The unique anatomical structure of the woodpecker’s head, including stout sharply pointed beaks and the hyoid bone, is the most essential factor in avoiding concussion.^8–15 In particular, the hyoid bone of woodpeckers is usually regarded as a very special structure consisting of a stiff core and a compliant shell, while the posterior section of the hyoid bone has a low bending resistance, which indicates a high degree of flexibility in this region.¹⁶ The periodical changing of Young’s modulus around the woodpecker’s skull could also decrease the stress wave propagating to its brain.^17–19 In addition, Lee et al. reported that the microstructure of the woodpecker’s beak has more elongated keratin scales with lower porosity and the nanoscale wavy structure to admit local shear deformation.²⁰ Meanwhile, the pecking trajectory of the woodpecker, a straight-line in the sagittal plane, can prevent the rotational acceleration impact to the brain.^5,21,22

Several finite element (FE) models for the woodpecker’s head have been built using the reverse engineering method to investigate the impact response during the pecking process.^{6,7,17,23–26} In particular, Zhu et al. established a complete FE model of the woodpecker and performed the entire drumming process.²⁶ They analyzed the propagation process of the stress wave and energy conversion and commented that the special skull structure and the tissue viscoelasticity can effectively spread and decrease the stress wave. It is shown that 99.7% of impact energy is converted into the strain energy (SE) finally stored in the body. There may be no doubt that the woodpecker has the most extraordinary shock resistant ability among all kinds of birds so far. However, limited attention has been paid to the comparison of the protective strategy of other birds. We still wonder whether other birds have the same protective strategy as the woodpecker or they have a different method to resist the concussion.

In this paper, the woodpecker, chicken, and pigeon were selected as the study objects due to their similar feeding habits. Pecking force, pecking frequency, maximal pecking velocity, etc. were tested by performing the pecking experiments of the chicken and the pigeon in vivo. Additionally, we built the FE head models of the woodpecker, chicken, and pigeon including all the essential features based on the micro-computed tomography (CT) scanning technology. Numerical analyses for stress distribution/wave propagation in the three birds’ heads were performed to clarify the mechanical response of their head structure during impacting. The purpose of this study is to investigate the shock resistance of the three birds and analyze their protective strategy during their pecking processes.

The chicken and the pigeon studied in this paper are big-bone chicken and domestic pigeon, respectively. They are all adult males and widely living in Liaoning Province, China. The chicken and the pigeon were selected because they all drum on a solid objective during their pecking. We used one chicken and pigeon to conduct the pecking experiments. They were fed with cereal and monitored to track their health condition for 5 days. Before the experiments, they were stopped feeding for 24 h to ensure their desire for food.

A rigid plate connected to a force sensor (BAB-XS-5M, Transcell Technology, Inc., USA) and placed with 10 g cereal on it was set as the pecking object to obtain their pecking force. Their pecking processes were recorded by using a high-speed camera (I-speed TR, Olympus, Japan) of 100 fps, as shown in Fig. 1.

The pecking frequency and velocity were obtained by analyzing the video. The typical anatomical features of the chicken and the pigeon, such as the beak tip and eyelid, were regarded as the tracing point to ensure that the pecking data are obtained from the translational based motion. The pecking force was collected from the force sensor, and the time is synchronized with the pecking video to correspond to the pecking frequency and velocity for each pecking.

After the pecking experiments, the chicken and the pigeon were subjected to euthanasia to obtain their heads for micro-CT scanning. The woodpecker’s head was harvested from a natural dead gray-headed woodpecker. We used the micro-CT scanner (Explore CT-120 micro-CT, GE, USA) to obtain the morphological information of the three birds’ heads to develop the FE model of the bird head.

Since the pecking experiment cannot directly provide the endo-response in their heads, we conducted the numerical simulations of the bird pecking to investigate the propagation of the stress wave under a single impact. In this paper, the head models of the three birds were divided into several essential parts including soft-tissue, skull, dura, cerebrospinal fluid (CSF), brain, and the upper/lower beak with an additional corneum layer. In particular, the woodpecker has the hyoid bone in its head, which is about four times the length of their beak and bypasses the skull to reinforce the skull.⁸ 

The micro-CT images of the three birds, as mentioned in Sec. II A, were used to build the FE models. The images were then imported into the MIMICS software to generate the scatter-point diagrams [Fig. 2(I)] in order to build a 3D geometry model [Fig. 2(II)]. Then, the Hypermesh software was used to repair and mesh the geometry model to the finite element model with different components [Figs. 2(III) and 2(IV)]. Considering the fact that their heads are almost bilaterally symmetric, the FE models were built as 3D symmetrical models. The FE models of the woodpecker, chicken, and pigeon heads contain more than 380 000 nodes and 2 140 000 elements, 198 000 nodes and 1 030 000 elements, and 20 000 nodes and 1 010 000 elements, respectively.

As mentioned in Sec. II B 1, we divided the woodpecker head model into several essential components as follows: the skull, brain, upper/lower beak with a corneum layer, hyoid bone, and soft-tissue. Previous studies^{6,8,16,17,24,25} have widely reported the mechanical properties of different bone structures in the woodpecker’s head. The elastic moduli of the woodpecker’s skull and beak have been measured by compression and nano-indentation experiments.^6,17,20,25 The mechanical property of the hyoid bone has also been reported by Zhou et al.⁸ Furthermore, Zhu et al. utilized the mechanical property of the human brain in their woodpecker head model considering the biological similarity.²⁴ Here, we also assumed that the material property of the soft-tissue and brain of birds’ head is the same as that of the human being.^27,28

The material properties of different components in the chicken and pigeon heads, however, have been rarely reported before. Since the emphasis of our discussion was the shock resistance of the structure of different birds’ heads, we assumed that the material properties of the chicken and the pigeon were the same as those of the woodpecker due to the biological similarity. Each tissue in the FE model was considered as the linear-viscoelastic material. The material properties employed in FE models were listed in Table I.^6,8,27–30

The impact analyses were performed with the ABAQUS software. The woodpecker utilizes its sharp claws and tail to keep its body firmly stable during impact.^31,32 Moreover, the woodpecker beak is perpendicular to the tree trunk when the woodpecker is drumming. Therefore, the FE head model of the woodpecker was also placed perpendicular to the objective plane. The woodpecker can adjust the pecking force according to the different pecking objects.⁶ The woodpecker drums on the trees in the natural environment. In the pecking experiment for the chicken and the pigeon, the pecking object was also a wood block. Therefore, the impact object was modeled as a wooden block and fixed during the simulation. The position of the chicken and the pigeon FE model was adjusted to the same position as the moment of contact in the video. We only took one single pecking process to study the mechanical response in their head. The birds’ heads were set near to the impacted object with an initial velocity as shown in Fig. 3. The initial impact speed was 7 m/s for the woodpecker, 0.56 m/s for the chicken, and 0.5 m/s for the pigeon based on the pecking experimental data. The head models were not assigned further restriction. We also set several monitor paths and points to record the stress wave propagation, as demonstrated in Fig. 3.

To compare the stress wave propagation paths in different birds’ beaks, we separated the woodpecker beak into three propagation paths, which have also been shown in Fig. 3 as the upper-beak (UB), lower-beak (LB), and hyoid bone. For the chicken and the pigeon, the hollow structure divided their upper-beaks as two similar bones, which could provide two different propagation paths for the impact stress wave. Therefore, we named the two bones as the upper-up beak (UUB) and the upper-down beak (UDB), respectively. The lower-beak of the chicken and the pigeon was named as LB.

The relationship between the pecking force and frequency of the chicken and the pigeon is demonstrated in Fig. 4. The average pecking force/frequency and the peak pecking velocity/acceleration of the chicken and the pigeon are listed in Table II and compared with the experimental data of woodpeckers.^6,23 The pecking frequency of the chicken is mainly concentrated at 2–4 Hz, unlike the pigeon, which is widely distributed below 5 Hz, even though their average pecking frequency is similar. The peak pecking force of the chicken is 0.78 N, which is over 40 times higher than that of the pigeon (0.018 N). The average pecking frequency and force of the woodpecker are significantly higher than those of the chicken and the pigeon. The peak pecking velocity and acceleration of the chicken are similar to those of the pigeon but extremely lower than those of the woodpecker.

Table II also shows the morphological comparison of three birds based on the micro-CT images. The head volume of the woodpecker (18.70 cm³) is nearly the same as that of the pigeon (17.02 cm³) and less than half of the head volume of the chicken (51.98 cm³). However, the woodpecker has the highest skull volume fraction (37.94%) and the beak length fraction (52.81%) among the three birds.

Figure 5 compares the kinetic energy (KE) and KE density of the pecking behavior of the three birds. The complete pecking process can be divided into the following three parts:²⁶ (1) the departure process in which the bird head moves to the object in an increasing acceleration; (2) the collision process in which the beak contacts with the object; and (3) the return process in which the bird head moves away from the object in a decreasing deceleration. During the entire pecking process, the KE reaches the highest at the collision moment, and then, it is converted into the strain energy (SE) and the dissipated energy (DE), reflecting the degree of deformation and heat generation, respectively.³³ Thus, the KE can be regarded as an indicator of the collision intensity. For normal pecking behavior, the woodpecker shows the highest KE (0.451 J) among the three birds. The KE of the chicken and the pigeon is 8.81 × 10⁻³ J and 2.79 × 10⁻³ J, respectively, which are in the same order of magnitude but two orders of magnitude lower than that of the woodpecker. To clarify the size effect on the KE tolerance of each bird, we divide the KE by the corresponding volume to obtain their KE density, as shown in Fig. 5. Although the volume of the chicken head is more than twice that of the woodpecker and pigeon head, the KE density comparison indicates the same trend as the KE comparison.

According to the woodpecker pecking experiments^6,23 and our pecking experiments for the chicken and the pigeon, the computational time is taken as 1 ms, 2 ms, and 2 ms for the woodpecker, chicken, and pigeon, respectively.

Figure 6 shows the von Mises stress distribution of the whole head and bone structure of the three birds over the duration of the impact. The first and last moments correspond to the moments of contact and separation between the beak tip and the impact object, and the three moments of the impact process are selected by a uniform time interval.

The von Mises stress increases significantly in the bones of the three birds due to their higher elastic modulus compared with that of the soft-tissue. Moreover, the relative high-value von Mises stress mainly appears at the birds’ beak tips and does not approach the back of their skulls. This means that their brains avoid the strongest stress wave during the impact. The high-value stress region of the woodpecker appears in both the UB and LB rather than being mainly concentrated in the LB as in the chicken and the pigeon. The maximum stress shows up earlier in the woodpecker beak (0.13 ms) than in the beak of the chicken (0.52 ms) or the pigeon (0.37 ms). The head models of each bird are assigned with the same material properties, which means that the stress wave should propagate with the same speed in the same component of the different models. However, the beak structure before the monitor points (from the beak tip to the monitor points) is different in the three models. As shown in Fig. 2, the upper and lower beaks are covered by the corneum, which is also surrounded by the soft tissue. The different structure may affect the stress wave propagation velocity in each bird head model. Therefore, although we use the same material properties, the time to reach the maximum stress in each model is still different.

The beak tips of each bird are considered to be the zero point of the Z-axis. Along the Z-axis, the peak von Mises stress is selected every 0.1 mm in each propagation path. Figure 7 shows the peak von Mises stress on each propagation path at the corresponding impact moment in Fig. 6. The von Mises stress on each propagation path of the three birds is nearly zero at the first and last moments and peaks at the third moment. The von Mises stress along the LB of the woodpecker is ∼10 MPa, which is lower than the stress along the UB, but both are higher than the stress on the hyoid-bone. The hyoid bone endures a relatively high stress wave mostly at two endings that contact with the beak tip and UB of the woodpecker. The stress along the hyoid bone keeps lower from 40 mm to 110 mm of the Z-axis, where it corresponds to parts of the cerato-branchial bone and epibranchial bone.¹⁶ 

For the chicken [Fig. 7(b)], however, the average von Mises stress on the LB is 75.35% and 29.47% higher than the stress on the UUB and UDB, respectively. The same phenomenon can be observed in the pigeon head [Fig. 7(c)]. The average von Mises stress on the LB of the pigeon is 73.53% and 42.63% higher than the stress on the UUB and UDB, respectively. Furthermore, the peak von Mises stress on the LB of the chicken and the pigeon appears at the interface of the beak corneum and the soft-tissue since the stress is strengthened by the stress wave reflected from the interface of the beak corneum and the soft-tissue.

Several monitored points are set on the beaks and the front/back skulls of the three birds to obtain the stress wave propagation at the different locations of the skull. Figure 8 shows the von Mises stress at each point of the birds and also marks the duration of the impact. The highest stress on the woodpecker’s skull is observed on point A, which is 13.81 MPa. Point B on the LB of the woodpecker endures the second highest stress wave. The peak stress on point B reaches 8.28 MPa. During the impact duration, the von Mises stress of point D on the front skull and point E on the back skull is much lower than that of the points on the UB and LB. Furthermore, the von Mises stress at the four points decreases significantly and periodically after the duration of the impact.

However, the highest von Mises stress for the chicken and the pigeon does not occur at their upper-beaks (A_UB for the UUB and A_DB for the UDB). Point B on the LB suffers the highest stress wave during the impact indicating that the high stress wave mainly propagates through the LB of the chicken and the pigeon. Moreover, the von Mises stress of the skull is much lower than that of the beak. After the impact, the stress on the chicken and the pigeon also drops dramatically.

Furthermore, five monitor points (see Fig. 3) are set at different locations of the three birds’ brains to record the stress during the impact, as shown in Fig. 9. The von Mises stress rapidly increases at the anterior (point a) and superior (point b) of the woodpecker’s brain [Fig. 9(a)]. The peak stress at the posterior (point c) of the brain is almost the same as the stress at the superior (point b) except for a delay of nearly 0.3 ms. The stress at the inferior (point d) and center (point e) of the brain is relatively lower during the simulation. The stress in the woodpecker’s brain is nearly one order of magnitude higher than that in the chicken and pigeon brains.

For the chicken [Fig. 9(b)] and pigeon [Fig. 9(c)], the stress at the anterior (point a) and posterior (point c) increases much higher than that at other positions. The superior (point b) of the brain does not suffer a strong stress wave. The average stress in the center (point e) of both their brains appears as the lowest stress and the least changes during the simulations.

The woodpecker can peck at a much higher frequency, speed, and acceleration than the chicken and the pigeon. Furthermore, the KE and KE density of the woodpecker are the highest among the three birds, which indicates the superior shock-resistance of the woodpecker. Unlike the chicken and the pigeon, the stress wave in the woodpecker pecking simulation propagates mainly through the UB and also transmits in large quantities through the LB. The length of the UB and LB could impact the propagation path, which have been reported in previous studies.^6,24 In the morphological observation of Wang et al., with the corneum, the UB of the woodpecker is 1.6 mm longer than its LB.⁶ On the contrary, without the outer layer, the woodpecker’s UB is 1.2 mm shorter than its LB. They reported that the impact force and the brain strain are minimal when the woodpecker model impact with the LB in the simulation. However, when the UB is longer in their simulations, the pecking force does not have a significant difference with the force data from their woodpecker pecking test. Zhu et al. have a contrary result that the stress wave mainly propagates through the UB when the woodpecker model has a longer UB.²⁴ The FE model of the woodpecker used in our study is built based on the micro-CT images, which also shows that the UB is longer. We introduce the outer layer in our woodpecker model and obtain the similar results as Zhu et al.²⁴ 

In the morphological comparison of the three birds, the woodpecker has the highest skull volume fraction and beak length fraction. The bone structure has the highest elastic modulus among all the main components in the woodpecker’s head. The higher skull volume fraction means fewer other tissues with lower elastic modulus in the head. Furthermore, previous studies^6,25 reported that the non-uniform distribution of Young’s modulus around the ellipsoid-like smooth skull can hinder the stress propagation. Therefore, the higher skull volume fraction could contribute to the higher structural strength for the whole head.

In addition to the small nostrils on the UB, the complete and stouter beaks could also increase the structural strength for the woodpecker’s head to conduct high speed and frequency pecking. Lee et al. reported that woodpeckers have much lower porosity bony layer (9.9%) compared with that of chickens (42.3%), which leads to higher bulk modulus and density of the bony layer to enhance the structural strength of the woodpecker’s head.²⁰ Furthermore, more elongated keratin scales and high aspect ratio wavy structures are also found on woodpeckers’ rhamphotheca along the longitudinal direction. The length of the woodpecker beak accounts for more than half the length of the woodpecker skull. Considering the beak length fraction of the chicken (31.58%) and pigeon (38.33%), we hypothesize that the woodpecker’s higher beak length fraction (52.81%) could increase the length of the stress wave propagation path, provide more space for the long keratin scales to dissipate local shear strain, and also retard more impact stress due to the viscoelastic property of the beak.²⁴ 

Additionally, the von Mises stress on the hyoid bone is lower than that on the UB and LB of the woodpecker. Zhu et al. reported that the hyoid bone shows small deformation and the lowest dissipate energy density during continuous woodpeckers’ pecking.²⁶ They suggested that the majority function of the hyoid bone is to limit the deformation of the skull and enhance the stability of the woodpecker’s head. Liu et al. also came to a similar conclusion that the hyoid bone and muscle together can enhance the structural strength of the woodpecker’s head after comparing the pecking simulation of the woodpecker’s head model with and without the hyoid bone.¹³ In our simulations, the relatively high stress only occurred at two endings of the hyoid bone. Therefore, the hyoid bone may provide a path for the stress wave, but it can hardly be regarded as the main path due to its small volume and being surrounded by the soft-tissue.

In the KE and KE density comparison, the pecking intensity of the chicken and the pigeon is at the same level. The stress wave mainly propagates through the LB in the chicken and the pigeon heads. Comparing with their stouter LBs, UBs of the chicken and the pigeon are separated into two narrow paths and have more soft-tissue around their UB. However, the pecking force and frequency of the chicken and the pigeon are much lower than those of the woodpecker, as shown in Table II, which means that for the chicken and the pigeon, it is not necessary to keep the high structural strength of their skulls as the woodpecker. A complete and stouter upper-beak, such as that of the woodpecker, can not only increase the structural strength of the skull but also lead to high stress in the superior of the brain. As shown in Fig. 9(a), high stress usually occurs at the anterior (point a) and posterior (point c) of three birds’ brain due to the horizontal impact. Because the high stress wave propagates mainly through the woodpecker’s UB, the peak stress at the superior (point b) of the brain increases to 3.94 kPa. Therefore, the stress wave propagating mainly through the LB of the chicken and the pigeon could potentially avoid the higher impact on the superior of the chicken’s and the pigeon’s brain. Furthermore, another advantage of the hollow bone structure is that the pecking kinetic energy of the chicken and the pigeon can be decreased since the density of the soft-tissue is lower than that of the cortical bone.

The primary aim of the shock-resistance evolution for the three birds’ heads is to keep the structural strength on the safety level in order to avoid any injury to the brain, skull, and other tissues. Moreover, it is essential to set a reasonable stress wave propagation path with appropriate length and a different damping material. Woodpeckers have the higher skull fraction and beak length fraction to retard the mechanical loading compared with chickens and pigeons. The unique hyoid bone can also provide a “seat belt” effect to enhance the stability of the woodpecker’s head. In addition, two stout beaks of woodpeckers provide two stress wave propagation paths to overcome the fierce impact. Since chickens and pigeons need not peck as fierce as woodpeckers, their LB is regarded as the primary propagation path to avoid direct impact on their brain. Finally, optimizing the structure could also decrease the structure mass as well as the impact kinetic energy.

In this paper, we conducted the pecking experiments of the chicken and the pigeon to observe their pecking motion and test their pecking force, velocity, and frequency and then compared the results with the pecking data of the woodpecker based on previous studies. Tracing this phenomenon for its source, FE models of the three birds were built based on micro-CT images, material properties, and impact analysis. The stress wave distribution and propagation way in the bird’s heads were demonstrated according to the numerical analyses. The following conclusions can be made from our study:

1. Higher skull volume fraction (37.94%) provides the higher structural strength of the woodpecker’s head. Moreover, the higher beak length fraction (52.81%) of the woodpecker could also provide more space for elongated keratin scales to dissipate mechanical impact and retard more shock waves due to the viscoelastic property of the beak.

2. The KE of woodpeckers is much higher than that of chickens and pigeons. The woodpecker pecking simulation shows that the impact stress wave propagates through both the UB and LB. The hyoid bone may provide a path for the stress wave, but it can hardly be regarded as the main path.

3. Since chickens and pigeons do not need to peck in high speed and frequency as woodpeckers, their stouter LBs are regarded as the main propagation path to avoid direct impact on their brains from their UB.

4. The biomechanical design, ensuring the structural strength and setting a stress wave propagation path to disperse the impact stress, may inspire new approaches to improve and design anti-shock equipment.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

This work was supported by the National Natural Science Foundation of China (Grant Nos. 51775541, U1908233, and 11772086), National Key R&D Project of China (Grant Nos. 2018YFA0704104 and 2018YFA0704103), and NSFC of Liaoning Province (Grant No. 2019-KF-02-01).

The authors declare that they have no conflict of interest.