Measurement of human rotation behavior for psychological and neuropsychological investigations

Behav Res (2015) 47:1425–1435 DOI 10.3758/s13428-014-0554-z Measurement of human rotation behavior for psychological and neuropsychological investigations Kaspar Leuenberger · Reto Hofmann · Peter Brugger · Roger Gassert Published online: 15 January 2015 © Psychonomic Society, Inc. 2015 Abstract The investigation of rotation behavior in human beings enjoys a longstanding and enduring interest in laterality research. While in animal studies the issue of accurately measuring the number of rotations has been solved and is widely applied in practice, it is still challenging to assess the rotation behavior of humans in daily life. We propose a robust method to assess human rotation behavior based on recordings from a miniature inertial measurement unit that can be worn unobtrusively on a belt. We investigate the effect of different combinations of lowcost sensors—including accelerometers, gyroscopes, and magnetometers—on rotation measurement accuracy, propose a simple calibration procedure, and validate the method on data from a predefined path through and around buildings. Results suggest that a rotation estimation based on the fusion of accelerometer, gyroscope, and magnetometer measurements outperforms methods based solely on earth magnetic field measurements, as proposed in previous studies, by a drop in error rate of up to 32 %. We further show that magnetometer signals do not significantly contribute K. Leuenberger () · R. Gassert Department of Health Sciences and Technology, ETH Zurich, Leonhardstrasse 27, 8092 Zurich, Switzerland e-mail: R. Gassert e-mail: R. Hofmann Department of Mechanical and Process Engineering, ETH Zurich, Zurich Switzerland P. Brugger Neuropsychology Unit, Department of Neurology, University Hospital Zurich, Zurich Switzerland to measurement accuracy in short-term measurements, and could thus be omitted for improved robustness in environments with magnetic field disturbances. Results also suggest that our simple calibration procedure can compete with more complex approaches and reduce the error rate of the proposed algorithm by up to 38 %. Keywords Locomotion · Circling behavior · Turning behavior · Lateral bias · Motor system · Neuropsychiatry · Dopamine · Inertial measurement unit · Accelerometer · Gyroscope · Magnetometer · Long-term activity monitoring Introduction In the dawn of “embodied cognition” (Barsalou 2008), the measurement of human whole-body movements has gained increased attention. One important movement characteristic, associated with a vast number of cognitive functions arguably mediated by one or the other cerebral hemisphere, is body rotation, or turning bias. Turning bias has been extensively studied in animal species from amphibians (Rogers 2002) to fish (Vallortigara and Bisazza 2002), and especially in rodents (Glick et al. 1976). The model of the “circling rat” (Glick and Ross 1981) has helped to establish dopaminergic imbalances underlying asymmetric manifestations of Parkinson’s disease. It is a robust finding that animals with unihemispheric lesions rotate in the direction of the hemisphere with a lesion-induced dopamine deficiency (Dunnett and Torres 2012). In healthy human beings, rotation behavior enjoys a longstanding and enduring interest in laterality research. Comprehensive field studies (Schaeffer 1928) 1426 of “spiral movements in man” documented a left-sided (counterclockwise) preference during long-term locomotion, ruling out a possible role of peripheral asymmetries, such as leg length (Souman et al. 2009). Rather, the preferred direction of rotation was recognized as a marker of (neuropharmacological and cognitive) asymmetries between the two cerebral hemispheres. More recent research has confirmed the association between psychiatric disease and increased counterclockwise rotations (Bracha et al. 1993). On the basis of sound neuropharmacological evidence, it is assumed that both schizophrenic delusions and left-sided body turns are a direct consequence of a hemi-hyperdopaminergia of the right cerebral hemisphere (Bracha 1989). Similarly, Parkinsonian patients with an asymmetric hemispheric dopamine depletion were shown to preferably rotate towards the more affected hemisphere during unconstrained long-term locomotion (Bracha et al. 1987; Patino et al. 1995). Asymmetric locomotion is also observed after unihemispheric stroke. In rodents, ipsilesional rotation occurs after ischemia-induced focal infarction (Ishibashi et al. 2004), whereas in human patients, turning or veering tendencies reportedly depend on the way of ambulation. While walking shows an ipsilesional bias, driving a powered wheelchair led to a bias towards the opposite, contralesional side (Turton et al. 2009). Standardized assessment of extrapyramidal symptoms like veering or rotation behavior would appear desirable, especially in view of providing individually tailored pharmacological treatment (Ishibashi et al. 2004). A major challenge in such studies is obtaining a reliable measure of rotation behavior over extended periods of time. In animals, methods relying on a human observer (Robinson et al. 1980), automated procedures based on video recordings (Schwarting et al. 1993; Bonatz et al. 1987), as well as mechanical or electrical sensors (Ungerstedt and Arbuthnott 1970; Greenstein and Glick 1975; Heredia-Lopez et al. 2002), have been proposed. The latter solution, often referred to as rotometer , is typically based on a string connected to the animal, which transmits the rotation to a mechanical or electrical counter. The counter records single turns in both directions with a resolution of up to a quarter turn. Field studies investigating rotation behavior in humans during long-term locomotion have made use of the earth’s magnetic field captured by sensors integrated into vests that had to be continuously worn by the subjects (Bracha et al. 1987; Bracha et al. 1993; Mohr, Bracha and Brugger 2003; Mohr, Landis, Bracha, Brugger, et al. 2003; Mohr and Lievesley 2007). These devices capture magnetic north relative to the user’s orientation using a compass transducer. A microcontroller extracts quarter turns and counts a full turn when four consecutive quarter turns in the same Behav Res (2015) 47:1425–1435 direction are registered. This output logic corresponds to the output generated by tethered rotometers that have been used for rodents (Bracha et al. 1987). Whereas the methods proposed for animal studies have been validated and found to be robust, magnetic field sensors used in human studies are known to be heavily disturbed inside and near buildings due to large ferrous structures. It is therefore of interest to characterize the influence of magnetic disturbances on the counts in human rotation behavior studies. Furthermore, previous studies did not document the hardware and algorithms that were used for rotation counting, thus limiting a reproduction and reuse of these systems and methods in future studies. Here, we pres (...truncated)