M. Melissa Gross

A comparison of three-dimensional and two-dimensional analyses of facial motion

The purpose of this study was to compare the amplitude of facial motion obtained using three-dimensional (3-D) and two-dimensional (2-D) methods. The amplitude of motion of fifteen facial landmarks during five maximal animations (smile, lip-purse, grimace, eye closure, and cheek-puff) was quantified in 3-D and 2-D using a video-based system. Results showed that the 3-D amplitudes were significantly larger than the 2-D amplitudes, especially for landmarks on the lower face during the smile animation. In the latter instance, the 2-D amplitudes underestimated the 3-D amplitudes by as much as 43%. The difference between 3-D and 2-D amplitudes was greater for 2-D amplitudes obtained from one camera rather than from multiple cameras. The results suggest that a 2-D analysis may not be adequate to assess facial motion during maximal animations, and that a 3-D analysis may be more appropriate for detecting clinical differences in facial function.

Effort-Shape and kinematic assessment of bodily expression of emotion during gait

The purpose of this study was to identify the movement characteristics associated with positive and negative emotions experienced during walking. Joy, contentment, anger, sadness, and neutral were elicited in 16 individuals, and motion capture data were collected as they walked while experiencing the emotions. Observers decoded the target emotions from side and front view videos of the walking trials; other observers viewed the same videos to rate the qualitative movement features using an Effort-Shape analysis. Kinematic analysis was used to quantify body posture and limb movements during walking with the different emotions. View did not affect decoding accuracy except for contentment, which was slightly enhanced with the front view. Walking speed was fastest for joy and anger, and slowest for sadness. Although walking speed may have accounted for increased amplitude of hip, shoulder, elbow, pelvis and trunk motion for anger and joy compared to sadness, neck and thoracic flexion with sadness, and trunk extension and shoulder depression with joy were independent of gait speed. More differences among emotions occurred with the Effort-Shape rather than the kinematic analysis, suggesting that observer judgments of Effort-Shape characteristics were more sensitive than the kinematic outcomes to differences among emotions.

Reliability of a three-dimensional method for measuring facial animation: a case report.

Reliable methods of quantifying functional impairment of the craniofacial region are sorely lacking. The purpose of this study was to test the reliability of a three-dimensional method for assessing the functional repertoire of the face. Subjects were instructed to perform repeated sequences of five maximal facial animations. Facial motions were captured by three 60-Hz video cameras, and three-dimensional maximum motion amplitudes were calculated. Students t-test and Pearson product-moment correlation coefficients were used to test for significant differences between repetitions. The results show moderate to excellent reliability of the amplitude of motion for the landmarks over all animations. For each specific animation, certain landmarks demonstrated excellent reliability of motion.

A Comparison of Accelerometers for Predicting Energy Expenditure and Vertical Ground Reaction Force in School-Age Children

In this pilot study of 16 children, we evaluated the reliability and validity of three accelerometers (Mini-Motionlogger [MML], Computer Science Applications, Inc. Actigraph [CSA], and BioTrainer) as indicators of energy expenditure and vertical ground reaction force. The children wore 2 of each type of monitor while they walked, ran, and performed 2 jumping tasks on a force plate and walked, jogged, and ran on a treadmill. Intrainstrument reliability of the monitors ranged from .64 to .98 across the treadmill tasks and from .69 to .98 across the force plate tasks, with the MML and CSA appearing more consistent than the BioTrainer. Analyses of variance were conducted to compare activity counts with criterion measures (oxygen utilization and force plate scores). All of the monitors generally differentiated among the treadmill tasks, mirroring the change in oxygen utilization. The CSA monitors corresponded more closely to the changes in force plate scores than the BioTrainer or the MML. Simple regression analyses indicated that count scores from all of the monitors were associated with oxygen utilization, with the MML and CSA exhibiting stronger relations (R values = .81 and .83, respectively) than the BioTrainer (R= .60). Similar analyses between the activity monitors and the force plate scores were also significant but the relations were not as strong as for oxygen utilization (R values = .46, .51, and .52, respectively). Based on backward elimination regression analyses, caloric expenditure on the treadmill tasks was predicted significantly with each of the MML (37 and 38% of variance) and CSA (39 and 42% of variance) units when body mass was included in the model. For the BioTrainer counts to provide the best prediction of caloric expenditure, both body mass and height were retained in the model, resulting in 20 and 25% explained variance. Future research to evaluate the utility of accelerometers should employ tasks that prevent the confounding of force and caloric expenditure.

Locomotor advantages of Neandertal skeletal morphology at the knee and ankle

We quantified Neandertal knee extensor and ankle plantarflexor moments to determine whether differences between Neandertal and recent human skeletal morphology represent important functional differences. Neandertal skeletal differences in the patella, tibial tuberosity, and calcaneus were used to modify a computer model of recent humans to calculate the moment arms and moments of Neandertal knee extensor and ankle plantarflexor muscles. We also conducted sensitivity studies on the effect of musculotendon parameters on the Neandertal moments. As expected, we found that Neandertal moment arms were greater than recent humans at the ankle (122-141%); however, the magnitude of the increase was not well-predicted from measurements of size differences between Neandertal and recent human skeletons. At the knee, Neandertal moment arms were greater than those of recent humans in the locomotor range (108%) but less so at more flexed knee angles (102%). Not all Neandertal skeletal adaptations at the knee contributed to increased moment arm. Knee extensor moments were enhanced in the Neandertal models in the locomotor range (111%), regardless of musculotendon parameters. At the ankle, however, Neandertal plantarflexor moment was greater than that of recent humans (149-200%) at all joint angles only if muscle fiber length increased proportionately with moment arm. Our results demonstrate that Neandertal skeletal morphology, compared to that of recent humans, generated greater moments at both the knee and ankle in the locomotor range but not at higher angles of knee flexion or ankle plantarflexion.

Emotional influences on sit-to-walk in healthy young adults.

The purpose of this study was to investigate influences of emotional feelings on sit-to-walk (STW). Eighteen healthy young adults performed STW while feeling sadness, anger, joy and neutral emotion. Emotions were elicited using an autobiographical memories task. We used an optoelectronic motion capture system to collect motion data and assessed kinematics of STW. Emotion-related differences in STW kinematics were consistent with differences in movement speed. Compared to neutral emotion, sadness was associated with increased STW duration and phase durations, decreased peak forward and vertical center-of-mass (COM) velocity, increased drop in forward COM velocity, and increased forward and vertical normalized jerk score (NJS). Anger and joy were associated with decreased STW duration and phase durations, increased peak forward and vertical COM velocity, decreased drop in forward COM velocity, and decreased forward and vertical NJS compared to neutral emotion. Findings suggest that emotional feelings affect movement speed, hesitation, and movement smoothness during STW.

The effect of emotion on movement smoothness during gait in healthy young adults

This study aimed to investigate the effect of emotion on movement smoothness during gait. We followed an autobiographical memories paradigm to induce four target emotions, neutral emotion, sadness, anger and joy, in eighteen healthy young adults. Participants performed gait trials while feeling the target emotions. We collected gait data using an eight-camera optoelectronic motion capture system. We measured spatiotemporal gait parameters, smoothness of linear movements for the whole body center-of-mass (COM), head, thorax and pelvis in the anterior-posterior (AP), vertical (VT) and mediolateral (ML) directions, and smoothness of angular movements in the sagittal plane for the shoulder, elbow, wrist, hip, knee and ankle. Movement smoothness was measured as jerk, the first time derivative to acceleration, normalized to movement distance and stride time. Compared to sadness, gait speed increased with anger and joy, and spatiotemporal parameters associated with increased gait speed changed accordingly. In the VT direction, movement smoothness in the whole body COM, head, thorax and pelvis increased for anger and joy compared to sadness. In the AP direction, movement smoothness increased only for the head for neutral emotion, anger and joy compared to sadness. In the ML direction, emotion did not affect movement smoothness. In angular movements, smoothness in the hip and ankle increased for anger compared to sadness. Smoothness in the shoulder increased for anger and joy compared to sadness. The present findings suggest that emotion affects movement smoothness during gait, and that anger and joy are associated with increased movement smoothness.

Methods for Quantifying Emotion-Related Gait Kinematics

Quantitative models of whole body expressive movement can be developed by combining methods form biomechanics, psychology, and statistics. The purpose of this paper was to use motion capture data to assess emotion-related gait kinematics of hip and shoulder sagittal plane movement to evaluate the feasibility of using functional data analysis (FDA) for developing quantitative models. Overall, FDA was an effective method for comparing gait waveforms and emotion-related kinematics were associated with emotion arousal level.

Effects of image-based and text-based active learning exercises on student examination performance in a musculoskeletal anatomy course

Research on the benefits of visual learning has relied primarily on lecture‐based pedagogy, but the potential benefits of combining active learning strategies with visual and verbal materials on learning anatomy has not yet been explored. In this study, the differential effects of text‐based and image‐based active learning exercises on examination performance were investigated in a functional anatomy course. Each class session was punctuated with an average of 12 text‐based and image‐based active learning exercises. Participation data from 231 students were compared with their examination performance on 262 questions associated with the in‐class exercises. Students also rated the helpfulness and difficulty of the in‐class exercises on a survey. Participation in the active learning exercises was positively correlated with examination performance (r = 0.63, P < 0.001). When controlling for other key demographics (gender, underrepresented minority status) and prior grade point average, participation in the image‐based exercises was significantly correlated with performance on examination questions associated with image‐based exercises (P < 0.001) and text‐based exercises (P < 0.01), while participation in text‐based exercises was not. Additionally, students reported that the active learning exercises were helpful for seeing images of key ideas (94%) and clarifying key course concepts (80%), and that the image‐based exercises were significantly less demanding, less hard and required less effort than text‐based exercises (P < 0.05). The findings confirm the positive effect of using images and active learning strategies on student learning, and suggest that integrating them may be especially beneficial for learning anatomy. Anat Sci Educ 10: 444–455.

Effects of image-based and text-based activities on student learning outcomes

Research on benefits of visual learning has relied primarily on lecture-based pedagogy, not accounting for the processing time students need to make sense of both visual and verbal material[8]. In this study, we investigate the potential differential effects of text-based and image-based student learning activities on student learning outcomes in a functional anatomy course. When controlling for demographics and prior GPA, participation in in-class image-based activities is significantly correlated with performance on associated exam questions, while text-based engagement is not. Additionally, students rated activities as helpful for seeing images of key ideas and as being significantly less mentally taxing than text-based activities.