We initially examined whether this sample fo chimpanzees showed population-level biases in hand performance in the grasping task using a one-sample t-tests. A significant population-level right hand bias was found t(69) = 3.07, p<.001 (Mean HI  = −.137). Independent samples t-tests failed to significant sex differences in either the HI measure or in the percentage of grasping error made by the chimpanzees. Based on the sign of the HI values, 40 chimpanzees were judged to perform better with the right hand compared to 28 with their left hand.

For this analysis, we compared the asymmetry quotients (AQ) in WM as a function of threshold, sex and handedness. Sex and handedness were the between group factors while threshold (30%, 50%, 70%) was the repeated measures. The AQ were computed following the formula [AQ  =  (R – L)/((R + L) *5)] where L and R reflect the percentage white matter for the left and right hemispheres. Thus, positive AQ values indicated a rightward bias while negative values indicated a leftward bias. A significant two-way interaction was found between handedness and threshold F(2, 132)  = 4.10 p<.02. Shown in Figure 3 are the mean AQ values as a function of hand performance. Overall, chimpanzees that make fewer errors in grasping with the right (referred to as right-handed) compared to left hand (referred to as left-handed) show greater leftward asymmetries in WM within the KNOB, particularly at the 70% threshold level.

We next considered the relation between the functional PET activation patterns within the motor-hand area and the proportion of grey matter for the different probabilistic maps derived from the anatomical data using essentially a voxel-of-interest approach to the analysis. Recall from the PET analysis that there was a significant cluster of activation found in the region corresponding to the KNOB (see Table 2) as well as in regions adjacent to the KNOB region, such as the ventral premotor cortex. We isolated and created tracings of these two clusters. We then flipped the resulting map in the left-right plane, and fused it with the normally oriented maps to create bilateral tracings corresponding to the KNOB and ventral premotor cortex PET activation clusters.

For each subject, we then placed these bilateral tracings (corresponding to the KNOB and ventral premotor clusters) on each subject's segmented GM volumes. The proportion of GM found within these tracings was then computed, as was done with the anatomical probabilistic maps. The percentages of GM for the KNOB and ventral premotor cortex clusters were then correlated with the GM percentages derived from the anatomical probabilistic maps of the KNOB. Percentage of GM at the 30% (r = .385, p<.001), 50% (r = .322, p<.008) and 70% (r = .257, p<.04) thresholds were significantly positively correlated with the percentage of GM found in the KNOB cluster identified from the PET analysis. In contrast, the percentage of GM at the 30% (r = −.096, p>.05), 50% (r = −.110, p>.05) and 70% (r = −.110, p>.05) thresholds were not significantly correlated with the percentage of GM found in ventral premotor region cluster. Thus, variation in the proportion of grey matter found in the PET cluster corresponding to the KNOB region were more strongly correlated with the proportion of grey matter of the KNOB region estimated from the probabilistic map.

It could be argued that the absence of a comparison scan in our subjects, which could have been subtracted from the grasping scan as a means of removing general neuronal metabolic activity, limits the interpretation of the PET results in our study. We would argue against this position because the ipsilateral hemisphere was effectively the “comparison” in this study as we intentionally tried to limit the activation of this hemisphere. With that said, in no way are we suggesting that ipsilateral brain regions are not involved in contralateral motor movements [58], [59], but this direct issue was beyond the scope and aims of this experiment. Furthermore, previous “resting state” results reported in chimpanzees, macaques and baboons [60], [61] have failed to report any evidence of asymmetrical activation in the precentral gyrus, specifically within the region corresponding to the motor-hand area. Thus, it is unlikely that the main PET results reported here can be attributed to individual variation in resting state rather than activation associated with the reach-and-grasping actions made by the chimpanzees.

In conclusion, reach-and-grasp responses in chimpanzees are associated with contralateral activation of the primary motor and premotor cortices, particularly in regions corresponding to the motor-hand area. PET activation of the motor-hand area correlated with the anatomical location of the motor-hand area in the chimpanzee based on a probabilistic map of this region. Finally, performance asymmetries in grasping in chimpanzees were associated with anatomical lateralization in the motor-hand area. Collectively the results reinforce the view that individual differences in hand preference and performance are linked to variation in the motor-hand area, not just in humans but also in chimpanzees. Collectively, the results suggest functional, structural and behavioral homology of the motor-hand regions in humans and chimpanzees.

During primate evolution, the brain has become increasingly large relative to body size and consequently resulted in increasing gyrification or cortical folding of the cortex [62], [63]. The KNOB represents one of these cortical folds and, in light of the observation that it can only be visualized in great apes and human brains, suggests that increasing selection for complex manual actions resulted in, in essence, a neuroanatomical foot print of that evolutionary process. Furthermore, the fact that the KNOB appears to be experience dependent [37] and may reflect the development of connections between the primary motor cortex and sub-cortical structures known to be involved in prehension, suggests that it also represents a potentially important marker of brain and behavioral development. Additional studies that focus on the role that different motor experiences have on the development of the KNOB [64], [65] would be instrumental in understanding the foundations of handedness in human and nonhuman primates.

All subjects in the PET and MRI studies were captive born chimpanzees from the Yerkes National Primate Research Center (YNPRC). For the PET study, the subjects were four adolescent chimpanzees including one male (Jolson) and 3 females (Rowena, Edwina, Dara). The subjects ranged in age from 14 to 18 years of age. For the MRI studies, the sample consisted of 70 captive chimpanzees including 22 males and 48 females. Within this cohort, the subjects ranged in age from 6 to 51 years (Mean  = 21.52, s.d.  = 11. 59). All of the research conducted in this study were conducted according to the American Psychological Association guidelines for the ethical treatment of animals and were approved by the Institutional Animal care and Use Committee of Emory University. The YNPRC is fully accredited by the American Association for Accreditation of Laboratory Animal Care.

Subjects were scanned using ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) at a dose of 20 mCi. FDG was selected as the ligand because of its relatively long uptake period (∼80 minutes) and long half-life (approximately 110 minutes). Thus, just as we and other investigators have previously done, we capitalized on these features of ¹⁸F-FDG because they allowed for prolonged behavioral testing during the uptake period and a relatively long time frame to capture neural activity trapped in the cells between the termination of uptake and the interval of time needed to transport and scan the chimpanzees. Previous studies have used identical procedures to scan chimpanzees and have revealed significant and consistent patterns of PET activation (Taglialatela et al., 2008; Taglialatela et al., 2009). At the onset of behavioral testing, subjects were orally administered 20 mCi of ¹⁸F-fluorodeoxyglucose (FDG) that was diluted in approximately 100 ml of a sugar free flavored drink mixture. The subjects then participated in the behavioral task for 40 minutes (see below for description). Following the uptake period, chimpanzees were asked to voluntarily present for an intramuscular injection of Telazol (4 mg/kg). Training for voluntary intramuscular administration of the immobilization agent was done prior to testing, so that the chimpanzees were highly familiar with the procedure. Once anesthetized, the chimpanzees were transported to the PET imaging facility (total time from initial anesthesia induction to scan onset was ∼45 minutes). For the duration of the PET scan, chimpanzees remained anesthetized with propofol administered intravenously and diluted in lactated ringers at a dose of ∼10 mg/kg/hr. After completing PET scanning procedure, the subjects were returned to the YNPRC and temporarily housed in a single cage for approximately 18 h to allow the effects of the anesthesia to wear off and radioactivity to decay. Subjects were then returned to their home cages with their social group.

The PET images were acquired on a High Resolution Research Tomograph (CPS HRRT; CTI/Siemens, Inc.) approximately 1 hour and 45 minutes following ingestion of the ¹⁸F-FDG. Scan procedures were identical for all subjects. Chimpanzees fasted for approximately 5 hours prior to ¹⁸F-FDG administration, and were rewarded with only minimal amounts of frozen sugar free flavored drink cubes during the uptake period. Subjects were placed in the supine position inside the scanner. Six minute transmission scans were followed by 20 minute emission scans. Scan parameters were identical for all subjects: Axial FOV  = 24 cm; Transverse FOV  = 31.2 cm; Slice thickness  = 1.21875 mm. Transaxial Spatial Resolution FWHM is 2.4 mm at the center and 2.8 mm 10 cm from the center. Following scanning, a post reconstruction 2 mm smooth was applied to the images.

During the FDG uptake period, subjects were required to unilaterally grasp small rocks placed in their enclosure. Prior to testing, the subjects were separated from their social groups, but remained in their home enclosure. Three subjects (Rowena, Jolson, Edwina) used their left hand to grasp, whereas the remaining ape (Dara) used her right hand. During each test block, the experimenter would place 20 rocks in the enclosure and the subjects were required to grasp each rock, one by one, and hand them back to the experimenter. After each test block, a one-minute inter-block-interval (IBI) occurred during which the subjects remained seated quietly. After the one-minute IBI, another 20-rock test block was administered. The total number of test blocks varied across subjects but for all subjects, unilateral grasping was nearly uniformly accomplished across the entire uptake period (see Table 1). In order to maintain subject motivation during the 40-minute uptake period, the chimpanzees were given a small, frozen sugar free kool-aid ice-cube and squirt of juice at the end of each block. Subjects were also verbally praised during testing by the experimenter when they retrieved all the rocks. During the IBI, the experimenter remained seated in front of the subject but interacted as little as possible with the chimpanzee.

The individual PET images were spatially aligned to their respective MRI images using 3D voxel registration with a linear transformation (Analyze 8.1, Mayo Clinic). Once aligned, each subject's MRI was used to outline the brain on the PET image in each and every slice in the axial plane. An average PET activation was then calculated based on the registered activity within these slices. Once the mean activation for the whole brain had been computed, each voxel within that entire volume was divided by the mean activation in order to obtain a standardized PET image. Next, images were smoothed with a low pass filter with an isotropic 6 mm kernel. To assess asymmetries in PET activation, the individually registered PET volumes were flipped on the left-right axis. The flipped PET volumes were registered back to the chimpanzee template and the flipped PET image was subtracted from the normal PET scan to create a difference volume. For the chimpanzee that used it right hand to grasp the object, the difference volume was reversed so that its left hemisphere difference values were aligned with the right hemisphere values for the remaining three subjects. This was done so that the hemisphere could be compared along the contralateral and ipsilateral dimension within the same sets of images. The four difference volumes were then spatially registered to one another, and a single average PET volume calculated. Whole brain analysis was conducted for the 4 subjects collectively using Analyze 8.1 (Mayo Clinic, Mayo Foundation, Rochester, Minnesota, USA). Significant areas of activation were identified by calculating a single t-map volume and using a threshold value of t  = 4.54 and a minimum cluster size corresponding to 72 mm³.

Scans were obtained at the time the chimpanzees were being surveyed for their annual physical examinations. For all scans, subjects were first immobilized by ketamine (10 mg/kg) or telazol (3–5 mg/kg) and subsequently anaesthetized with propofol (40–60 mg/(kg/h)) following standard procedures at the YNPRC. Subjects were then transported to the MRI facility. The subjects remained anaesthetized for the duration of the scans as well as the time needed to transport them between their home cage and the imaging facility (total time ∼1.5 h). Subjects were placed in the scanner chamber in a supine position with their head fitted inside the human-head coil. Scan duration ranged between 40 and 60 min as a function of brain size.

The chimpanzees were scanned using a 3.0 T scanner (Siemens Trio, Siemens Medical Solutions USA, Inc., Malvern, Pennsylvania, USA) at the YNPRC. T1-weighted images were collected using a three-dimensional gradient echo sequence (pulse repetition = 2300 ms, echo time = 4.4 ms, number of signals averaged = 3, matrix size  = 320×320, with .6×.6×.6 resolution). After completing MRI procedures, the subjects were returned to the YNPRC and temporarily housed in a single cage for 6–12 h to allow the effects of the anesthesia to wear off, after which they were returned to their home cage. The archived MRI data were transferred to a PC running Analyze 8.1 software for post-image processing.

Individuals tracing the ROIs were blind to the sex and handedness of the chimpanzees. Prior to the measurement of the KNOB, inter-rater reliability was established between the two observers. To assess inter-rater reliability, two individuals measured the KNOB for 10 individual chimpanzees following the criteria and landmarks outlined for this study (see below). The volume measures of the left and right hemispheres were correlated within each individual chimpanzee between the two raters. For the KNOB region, inter-rater reliabilities for the left (r = .96, df = 8, p<.01) and right hemispheres (r = .85, df = 8, p<.01) were positive and significant.

Prior to tracing the KNOB on each individual brain, the skulls were removed from the raw MRI scans and the brains were re-aligned in the axial planes and subsequently virtually cut into consecutive 1 mm slices [66]. Each subject's brain volume was then co-registered to a template of a chimpanzee brain using three-dimensional voxel registration with a linear transformation (Analyze 8.1). The chimpanzee brain template was created using a two-tiered procedure [61]. Initially, the MRI scans of 8 chimpanzees (subjects included in this study) were placed in stereotaxic orientation using AFNI software and then averaged together into a single image. Next, each individual MRI scan was spatially normalized to this template using an affine transformation. Subsequently, all the spatially normalized MRI scans were averaged to create the template.

In creating the probabilistic map of the KNOB, tracing from 62 chimpanzees were used including 21 males and 41 females. These 62 apes were selected because we could reliably trace the KNOB in both hemispheres whereas the KNOB was not clearly visible in one or both of the hemispheres in the remaining 8 apes. The KNOB was quantified separately for the left and right hemispheres, in the axial (transverse) plane on each spatially aligned brain (see Figure 4) following procedures previously used in human and chimpanzees brain specimens ([11], [24]. The dorsal and ventral edges of the knob served as markers for defining the boundaries of the area. The area of the entire knob was traced on each slice and hemisphere using a mouse-driven pointer (see Figure 4). The areas within the traced regions were subsequently summed to derive volumes of the KNOB for each hemisphere. These KNOB tracings for the right and left hemispheres were saved for each individual subject.

The KNOB tracings were then summed across all 62 subjects and color-coded for consistency in the presence or absence of voxels (see Figure 5). The voxel values ranged from 1 to 62; a value of 62 represented voxels that were present in all subjects, whereas a value of 1 represented voxels which were present in only a single subject. The summed KNOB tracings were subsequently thresholded at three different criterion levels including 30%, 50% and 70% voxel overlap, respectively. The different thresholded maps represented the voxels within the individual KNOB tracings that were present in 30%, 50%, and 70% of the chimpanzee sample (i.e., common voxels in at least 19, 31 and 43 individuals, respectively). For each of the thresholded volumes, the resulting probabilistic map within the left and right hemisphere was outlined and saved for subsequent use with the segmented grey and white matter volumes.

For the grey matter (GM) and white matter (WM) analysis, each individual MRI scan was co-registered to the chimpanzee template. The MRI scans were then segmented into grey, white and CSF tissue using FSL (Analysis Group, FMRIB, Oxford, UK) [67], [68]. The KNOB volumetric maps for the left and right hemispheres that were derived for the 30%, 50% and 70% probabilistic maps were then applied to each of the co-registered GM and WM volumes. The volume of GM and WM found within the left and right hemispheres for the three different probabilistic maps were quantified across subjects.

Because the focus of the study was on the neural correlates of grasping, we also compared the KNOB asymmetries in relation to behavioral asymmetries in grasping performance when picking up small food items. We have previously reported that chimpanzees show population-level right hand biases in performance asymmetries when grasping small food items [35], [36]. Following the procedures used by Hopkins et al., [35], small food items were presented to the left and right hands of the chimpanzees and we recorded the number of errors they made when grasping the food items. The food items included small, quartered peanuts, mini m&m's, tiny-tarts, and small pretzel sticks. Each chimpanzee received 20 trials for the left and right hand for each food item with the exception of the quartered peanuts in which the apes received 40 trials for each hand. Thus, all the chimpanzees were required to grasp 60 food items with the left and right hands (total  = 120 trials) when the data were summed across the two studies. Based on these combined data, we computed a handedness index (HI) following the formula [HI  =  (#R - #L)/(#L + #R)] where #R and #L indicate the number of errors made by the left and right hand, respectively. For simplicity, we classified subjects as performing better with the left or right hand based on the sign of the HI value. Subjects with negative HI values were classified as performing better with the right hand while apes with positive HI values were judged to perform better with the left hand. Two chimpanzees performed equally well with both hands or could not complete testing and they were omitted from the analyses.

For the probabilistic maps of the KNOB region, we calculated the percentage of GM and WM within the left and right hemisphere by calculating the GM and WM volume and dividing by the total volume of the probabilistic map for each threshold and hemisphere and multiplying by 100. We chose to use percentages of GM and WM rather than the raw volumes for each threshold because the size of the probabilistic object maps varied between hemispheres and we sought to adjust for those differences in estimating the amount of grey and white matter. All inferential statistics (ANOVA, t-tests) adopted an alpha of p<.05 (unless otherwise stated) and post-hoc tests, when necessary, were conducted using Tukey's Honestly Significant Difference (HSD) test.