George P. Moore

Synchronizing human movement with an external clock source

Temporal information processing was studied in humans attempting to tap a key in synchrony with a metronome whose base period was subjected to subliminal random changes. Statistical measures of the sequential timing of metronome and key-tap events were compared with similar time series generated by computer-simulated models of synchronization strategies. From this analysis, synchronization appears to be a sensory-dependent second-order recursive process indexed and stabilized by a combination of internal and external resetting events, with at least two independent sources of timing error.

A Neuromuscular Actuation System Model

Recently, both high quality physiological data and human-operator-describing function data of low variability and large dynamic range have become available. These data lead to control engineering descriptions for neuromuscular actuation systems that are compatible with the available data and that provide insight into the overall human control structure (e.g., the types of feedback systems used for various inputs). In this paper, some of these physiological and human-operator data are briefly reviewed, and a simple neuromuscular actuation system model is presented. The physiological data of interest include recent anatomical and physiological data for the muscle spindle and input-output studies of the muscle. These data indicate that simple linear models can describe the basic behavior of these two elements in tracking tasks. This paper contains two key developments: 1) the variation in system parameters as a function of average muscle tension or operating point; and 2) the role of the muscle spindle both as an equalization element and in its effects on muscle tone or average tension. The simplest neuromuscular model suggested by and compatible with the data is one in which muscle spindles provide four functions in one entity: 1) the feedback of limb position; 2) lead/lag series equalization; 3) the source of at least one command signal to the system; and 4) a signal for adjustment of the spindle gain, equalization, and steady-state spindle output which produces the average muscle tension.

Timings and interactions of skilled musicians

We report here a preliminary study of interactive behavior between two members of a skilled string quartet performing a selected musical passage that required both performers to play several hundred notes in rapid succession at a steady tempo and in synchrony. Bowing movements were recorded using angular velocity sensors attached to their right forearms. The results show a high degree of temporal precision in both players. In addition, both players exhibited embedded rhythmic components in their timekeeping pattern, which arose from the grouping of notes in the musical score: four 16th notes to a beat. Within each group of four notes, we found a consistent timing microstructure: alternate upbows and alternate downbows had different mean durations. Both players’ bowings could be modeled as alternating renewal processes. In addition, we report evidence of interactive coupling between the players as an essential component of their joint performance. The alternating renewal model enables us to propose a note-generation process that has implications for the central generators underlying the observed behavior and their hierarchical organization. We discuss the implications of this model for the organization and execution of more complex motor sequences.

Applications of the Theory of Stochastic Point Processes in the Detection and Analysis of Neuronal Interaction

Abstract The sequence of pulses generated by a neuron and transmitted to other neurons is the most easily measured and quantifiable signal by which neurons communicate and interact with each other. These can be recorded experimentally from the brain of an animal by special microelectrode techniques, and in our laboratory it is a routine matter to record such “spike trains” simultaneously from several neurons Each spike train is then characterized quantitatively by a number of statistical measures which are obtained by a computer analysis of the times of occurrence of each pulse generated by the neuron. We presently find the interspike-interval histogram and autocorrelation histogram the most useful models for representing the long-term signalling properties of a single neuron (8, 10)

Stochastic aspects of motor behavior and their dependence on auditory feedback in experienced cellists

This study aimed to investigate movement accuracy of experienced cellists, the statistical properties of their note sequences during a reciprocal task, and the degree to which these movement characteristics depend on auditory feedback. Nine experienced cellists were asked to shift alternately between two notes using only their index finger to make contact with the string and fingerboard. Shifting sequences continued for two minutes at a rate of one note per second. The task was performed under two conditions: with auditory feedback (provided by the bow) or without auditory feedback (i.e., without the use of bow). When the bow was used, subjects had no difficulty in shifting between target notes with precision and stability. Some variability was present, but notes in these sequences were generally uncorrelated. The contact data and correlations in most bowed trials resembled those expected of a renewal process, a process in which successive values are statistically independent and identically distributed. Without the bow, subjects lost their ability to reach the same target positions accurately; contact locations tended to drift and had a random quality, indicating that without the bow subjects were uncertain of the target location in relation to the spatial location of their fingertips. Within these unbowed sequences, finger positions were highly correlated—within and between note sequences. In some trials without the bow, the statistical correlation patterns of the sequence were consistent with the expectations of a discrete Wiener process. Throughout our study, computer simulations of renewal and Wiener processes enabled us to determine the types of correlations to be expected from these theoretical models. The implications of the statistical results in terms of subject behavior are discussed.

Afternoons with Segundo

G. P. Moore (B) School of Engineering, University of Southern California, Los Angeles, CA 90089, USA e-mail: gmoore@usc.edu Jose (Pepe) Segundo flanked by Curtis Bell, his first UCLA PhD student (1964), on the left and Michael Stiber, his most recent UCLA PhD student (1992), on the right. Photo taken in Montevideo, Uruguay, during the Neural Coding Workshop, NC2007, as mentioned on p. 410 in the Editorial

A Computer‐Based Portable Keyboard Monitor for Studying Timing Performance in Pianists

As a preliminary step in the study of timing in musical performance, we have developed a low-cost, portable computer-based data acquisition system which can be quickly and unobtrusively connected to pianos with standard keyboard dimensions. The system is based on several components linked in a configuration that allows the identity and time of activation and release of each keyboard note to be coded and stored in a memory buffer and sequentially transferred to a host computer for decoding, processing, and analysis. Sixty-four notes of the piano are monitored by two 32-note keyswitch boards (Marantz Piano Co., Inc., Morgantown, NC) which are inserted under the keyboard. Each key, when depressed, closes a spring switch; release opens the switch. At preselected time intervals (typically every millisecond or every 10 milliseconds), a 6502 microprocessor chip sends an interrogatory command to the switch register, which responds with a coded number sent to the microprocessor summarizing the status (up or down) of each of the 64 notes being monitored. This code is compared to the previous coded signal to determine whether the status of any note has changed. If a change of