Presentation Abstracts

CoRR faculty and graduate students delivered the following presentations at the 14th Annual International Conference on Perception and Action (July 2-6, 2007) in Yokohama, Japan.

Perceiving Attachments to Hand and Foot

Claudia Carello, Alen Hajnal, Sergio Fonseca, and Jeffrey Kinsella-Shaw

Using tools effectively entails haptic awareness of their spatial properties. Such awareness is accomplished through muscle deformation registered by mechanoreceptors embedded in the muscles, tendons, and ligaments. In principle, movement by any part of musculo-skeletal system can engender activity in muscle-related mechanoreceptors, thereby supporting the capabilities of dynamic touch. But morphological and physiological characteristics of the hand and arm - the primary tool-using configuration - differ from those of the foot and leg (Kennedy & Inglis, 2002). They differ in their sensory accuity (Kets, van Leerdam, van Brakel, Deville & Bertelsmann, 1996) and in the spatial scale of their exploratory behaviours (Lederman, Klatzky, Collins & Wardell, 1987). Moreover whereas the hand actively attaches itself to the objects it uses, the foot typically does not. Skis, snowshoes, skates, swim fins, spurs, and so on must be attached to the foot so that they can be used. Do these differences have consequences for the perceptual capabilities of hand and foot? Experimental results on length perception reveal that performance by the two limbs is largely comparable, as indexed by accuracy, consistency, and sensitivity to manipulations of the moments of the mass distribution (Hajnal, Fonseca, Harrison, Kinsella-Shaw & Carello, 2007). In a selective attention setting, foot and hand distinguish perceived partial length from perceived whole length in similar fashion (Hajnal, Fonseca, Kinsella-Shaw, Silva, Carello & Turvey, under review). Equivalent performance by the two limbs is perhaps surprising in light of their neuromuscular differences. Certainly the different functional roles typically assumed by foot and hand in everyday life might suggest that they ought to differ. However, data of this kind support the characterization of muscle organization as a softly assembled smart perceptual device. Haptic perceptual instruments are assembled from general dynamic properties, not from specific anatomical components.

Influence of Visual Environment on the Contribution
of the Individual Lower Limbs to the Control of Posture

Jeffrey M. Kinsella-Shaw, Steven J. Harrison, and M.T. Turvey

Center of pressure (COP) excursion generated while standing under a wide variety of task constraints constitutes the foundational data for a broad array of studies of postural control. (See Newell, 1998, Stoffregen et al, 1999, and Winter et al, 1996 for relevant reviews.) Most typically, measures are collected from a single force plate. Analyses have of necessity focused on the behavior of the net COP defined over the total contact area of both feet. Consequently, the literature devoted to the analysis of each lower limb's attendant COP, in isolation and as they contribute to the global COP, is relatively sparse. There is a similar paucity of experiments examining the interactive effects of organismic (e.g. age) and environmental (e.g. available illumination) constraints on COP excursion during standing (see Brooke-Wavell et al., 2002, and Kinsella-Shaw et al., 2005). Therefore, in this study we have employed dual force-plates. This arrangement enabled us to record the COP excursion generated independently by each limb during stnding and to subject it to the nonlinear technique of cross recurrence quantitative analysis (CRQA). CRQA provides measures (% recurrence and maximum line length) that index the noise and attractor strength of each limb's dynamics (Richardson, Schmidt & Kay, 2006). In the present paper we investigate the influences of ambient illumination and optical structure on each lower limb's COP behavior in standing and inquire whether those influences are age related. Results are discussed in the context of age and pathology related reorganization of postural control dynamics.

Reproducing Distance Traversed: Role of the Haptic Informatin and Gait Symmetry

C.A. Romaniak-Gross, R.W. Isenhower, R. Arzamarski, S.J. Harrison, and M.T. Turvey

We investigated whether the haptic information generated during legged locomotion provides a measure of distance. Specifically we asked: In the absence of vision and hearing, is a given distance identical when measured by various gait styles? In Experiment 1 participants wore a blindfold and noise-cancelling headphones and either walked or jogged from a starting point A to an end point B, signalled during locomotion by the experimenter. The task of the participants on each trial was to replicate, by walking from B, the previously travelled distance A-B. In concert with the findings of Schwartz (1999), the results of Experiment 1 suggested that successful distance replication was not based on either the number of steps from A to B, the duration of travel from A to B, or auditory and tactile landmarks. Experiment 2 replicated the design of Experiment 1 but with so-called "gallop-walking" rather than jogging. The gallop-walk is characterized by repetition of the rule "step forward with the right foot, then bring the left foot into alignment with the right foot and pause." Results showed that particpants accurately replicated the outbound distance in the walk condition but signifiantly underestimated the outbound distance in the gallop-walk condition. Given the demonstrated indifference of distance replication to the type and speed of gait used from A to B (Experiment 1; Harrison, 2007; Schwartz, 1999) the contrast between gallop-walk and walk is anomalous. All non-walk gaits used in A to B travel to date have been of like symmetry; both legs do the same thing, 1/2 period out of phase with each other, continuously. The gallop-walk has a different and lower symmetry (Pinto & Golubitsky, 2006). The next round of experiments will investigate whether the symmetry of locomotion as a distance measuring system is a determinant of distance measure.

Hypotheses on the Relation Between Diabetes Mellitus and Postural Control

Cedrick T. Bonnet and Michael T. Turvey

Diabetes mellitus is a metabolic disorder caused by hyperglycaemia. It can lead to neuropathy (damage to sensory and motor nerves), primarily at the peripheral level, with complications for perception and action. A notable complication, mentioned in the literature since 1964, is postural instability. Although the first experimental investigation did not occur until 1982 (Mimoir et al., 1982), there are now more than 40 research publications directed at the relation between postural control and diabetes. Experiments have been directed at (a) understanding the role of the somatosensory system in the control of stance, and (b) finding means to evaluate and improve diabetic patients' lives.

The primary fact is that diabetic patients with peripheral neuropathy sway more than controls and more than diabetic patients without neuropathy. The question posed therefore is one of whether neuropathy (e.g., dysfunctioning at the nerve conduction level) is the direct cause of instability or the cause of body disorders that lead to instability (e.g., weakness of postural muscles, reduction of the base of support). Electrophysiological, perceptual, biomechanical, and motor hypotheses will be discussed along with methodological problems and confounding variables that might explain why amplification of postural sway is symptomatic of only diabetic patients with neuropathy. The ecological approach to perception-action has not yet been brought to bear, experimentally or theoretically, on this particular instance of impaired postural control. Suggestions for how the approach might be applied, in respect to hypotheses, classes of experiments, and methods of amelioration, will be given and discussed.

Bidirectional Coupling Between a Metronome and a Coordinating Actor

Bruce A. Kay and Paula da Silva

A new coordination paradigm is introduced in which an oscillator producing a metronomic signal is coupled to the actions of the actor attempting to synchronize with it. This has serveral implications for the study of coordination from the perspective of dynamical systems theory. First, this generates a system in which the dynamics are fully coupled, unlike the usual metronome paradigm. Second, the detailed dynamical properties of one-half of the coordinating system are known, namely, the dynamics of the oscillator and the coupling it receives from the actor, unlike the situation in other coordination paradigms such as interlimb coordination. Finally, by recording and analyzing the oscillator's time series behavior we can more easily reconstruct the phase space of the full system, as compared with less completely defined systems. The promise of paradigm is indicated in two experiments, one involving discrete pulse coupling, which connects to the literature on synchronization with a metronome via discrete taps, the other involving continuous coupling, which connects to the literature on interlimb coordination.

Remote Effects of Locally Produced Forces on Rhythmic Movement Dynamics

Paula L. Silva, Marisa C. Mancini, Sergio T. Fonseca, Miguel Moreno, and Michael T. Turvey

During the performance of motor activities, such as walking or running, muscle forces generated in one body segment can affect the motion and position of segments far removed from it (Neptune et al., 2001). As a result, the musculoskeletal system is required to deal with the resulting stresses caused by the flow of forces through the kenetic chain. Counteracting these forces locally during the performance of motor tasks is not only unlikely to succeed in a highly integrated system, but is also ineffective (Schenau & Soest, 1996). Bernstein (1967) proposed that movment is organized so as to take advantage of the context of forces in the whole system when a motor task is performed. Under this perspective, advancements in our understanding of movement organization rely on a better appreciation of the remote effects of locally produced forces. The objective of the study was to describe the effects of a tonic force produced at one wrist on the observed kinematic pattern and oscillatory dynamics of the rhythmic movements generated by the opposite wrist. The oscillatory dynamics were evaluated by a graphical and statistical method developed by Beek & Beek (1988). This analysis revealed a change in non-linear stiffness and friction functions underlying the rhythmic movements produced by one hand as the level of force generated by the other had increased. These effects were observed along with increases in the amplitude of oscillation. The results suggest a reorganizaiton of the oscillaroty dynamics as a function of remote force production. It is possible that this reorganizaiton allowed the system to take advantage of the changing context of forces during the performance of the rhythmic movements. Different hypotheses will be discussed regarding the nature of effects observed.

Perceptual Learning and Information Space: Local Effect of Feedback, Nonlocal Effect of Intention

Ryan Arzamarski, Robert W. Isenhower, M.T. Turvey and Claire Michaels

Previous reserach has shown that the dynamic touch paradigm can be used to study perceptual learning when participants are given feedback (Arzamarski, Isenhower, Jacobs & Michaels, 2007). This learning process can be depicted as a movement through an information space that contains all candidate variables for the given task. The current study aims to distinguish two types of shifts in attunement on such a space; that which occurs with the education of attention and that which occurs with a change of intention. Our to-be-wielded objects consisted of a set of variously-sized square parallelepipeds with a uniform handle. In a between-subjects, pretest-posttest-noveltest design, particpants were asked to report either the length or width of each stimulus. In the practice blocks, participants were given veridical feedback about their judgments. The pretest-posttest change reflected learning. For the noveltest, particpants were instructed to report a novel property: If they had initially reported and received feedback about length, they were asked to report width, or vice versa. Results show that participants made gradual and continuous movements across the information space toward the specifying variable. Conversely, when participants were asked to report the novel property, they made a discrete jump across the informatoin space. This jump was to a variable in the "ballpark" of that specific to the novel property. these findings are consistent with the direct learning theory (Jacobs & Michaels, in press) in that switches of intention can be differentiated from the continuous shifts in attunement that occur during perceptual learning. They are also consistent with direct intention theory (Turvey, 1990); Intention to perceive a given property specifies the relevant region in information space.

Weak Anticipation, Strong Anticipation, and Ecological Theory

Nigel Stepp and Michael Turvey

The realm of anticipatory systems is still somewhat loosely defined. By reviewing several descriptions of anticipatory systems from the literature it is possible to develop a general taxonomy. Anticipatory systems appear to vary along two dichotomous dimensions. An instance of anticipation may be identified as being endo- or exoanticipation (Dubois, 2003), depending on whether the self or the environment is the object of anticipation. It is also possible for a particpular instance to exhibit instantaneous or time-shifted anticpation. Instantaneous anticipation is primarily concerned with ideas such as current heading, while time-shifted is meant to evoke a separation between the current event context and one at some finite distance into the future.

Above the taxonomy, however, is the distinction between so-called weak and strong anticipatory systems (Dubois, 2003). Both anticipate using the future values of a given system. If these future values are estimated using a model of the system, the anticipation is labeled weak. Strong anticipation is not model-based, but relies on a lawful connection between the current state of affairs and the relevant future. Weak anticipation is the method which has traditionally been employed to solve problems of prospective control, however, this is not to mean that strong anticipation cannot be taken seriously as well (Michaels, 1992).

Strong anticipation finds a comfortable home in ecological theory, which has the tendency to eschew reliance on models, instead placing importance on ontological structure and lawfulness. In ecological parlance, the information used for prospective control is already informatoin about the future. It is argued that strong anticipation is an expression of the direct perception and action cycle, and that given ecological theory, strong anticipation is the primary choice.

On the Impredicative Nature of Ecological Laws

Stephanie Petrusz and Michael T. Turvey

This talk will, as its title suggests, focus on the impredicative nature of ecological laws. The philosophical background of the issue of lawfulness in the construction of both philosophical and scientific theories will be discussed. The current state of the art of philosophical literature finds no consensus as to the character of laws of nature. The appropriate point of entry into this debate for the ecological scientist is in the consideration of the causal character of laws. Where the philosopher may debate theories of causation in constructing philosphy of laws, the ecological scientist is committed to a certain view of causation in virtue of studying impredicative systems. Therefore, the hallmarks of impredicative systems (self-cause, closure to efficient cause) will be discussed as preconditions for an appropriately ecological account of lawfulness. A distinction should be drawn between the philosophical laws of nature and the role of scientific laws in scientific theories. This distinction is made for the purpose of highlighting that, although ecological laws are properly part of a scientific theory, their causal commitments make them implicitly part of a lager philosophical point of view. This dual role will be considered.

 

Tensegrity Perceptual Hypothesis

Sergio T. Fonseca and Michael T. Turvey

Auditory perception and haptic perception are grounded in contact mechanics and mechanoreceptors. The auditory perceptual system is embedded in a homogeneous, compressible medium (air or water) that when compressed by mechanical disturbances acts to eliminate the compression. The consequence is a vibration field of wave fronts and wave trains with the geometry of the wave fronts and the invariants of the wave trains providing information about distal mechanical events. We hypothesize that the haptic perceptual system is similarly embedded in a compressible medium, one that is constituted by a nesting of tensegrity structures. These are structures whose shape and stability derive from continuous tension and discontinuous compression. For the body as a whole, the muscles, tendons, ligaments, and fascia (infused with mechanoreceptors) are the tension elements and the bones (suspended within the the tension network) are the compression elements. Mechanical disturbances, self- or other-induced, are realized as changes in the force balance in the tension distribution. In our hypothesis, the array of tensions in the body's tensegrity structure, its transformations, and invariants under those transformations, constitute the basis of information about the body and about adjacent surfaces or objects in contact with, or attached to, the body.

Previous Presentations

CoRR faculty and graduate students delivered the following presentations at the
13th Annual International Conference on Perception and Action Conference (July 5-10, 2005) in Monterey, California.

Hypersets, Complexity and Affordances
Open Topics - Session E

Tony Chemero
Scientific and Philosophical Studies of Mind Program
Franklin & Marshall College

Michael T. Turvey
Center for the Ecological Study of Perception and Action
University of Connecticut

Ecological psychology explains behavior by treating animals-in-their-environments as complex systems. Because of this complexity, the mathematical and computational tools of the mainstream cognitive sciences, derived from computability theory and designed for simple systems, are inappropriate for ecological psychology. The purpose of this talk is to introduce non-well founded set theory or hyperset theory (Aczel 1988, Barwise and Etchemendy 1987, Barwise and Moss 1996, Kercel 2003), and show that it is a useful tool to make sense of the complexity inanimal-environment systems. To that end, we will begin by describing hyperset theory and the intuitive, easy-to-understand system of graphs that comes with it. Then, using Rosen's model of living systems (1991) as an example, we will show that hyperset graphs can be used as “complexity detectors”, depicting the nature and source of the complexity in models. We will then use hyperset graphs to analyze the sources of complexity inherent in affordances. Doing so will allow us to address, and perhaps dispel, the controversy in ecological psychology over whether affordances are to be viewed as properties of the environment (Michaels and Carello 1981, Turvey 1992, Reed 1996, Michaels 2003) or viewed as relations between animal and environment (Sanders 1997, Chemero 2003, Stoffregen, 2003).

Antiphase Coordination of the Lower Legs: Lessons Applied to Trans-Tibial Amputees
Symposium 5: Ecological Applications in Rehabilitation

David Tiberio
Center for the Ecological Study of Perception and Action
Department of Physical Therapy
University of Connecticut

Cyclic movements, which are the foundation of locomotion, require the coordination of limbs moving in opposite directions while maintaining a common tempo (1:1frequency locking). Swinging limbs are not unlike pendulums. Pendulums of differentlengths and masses have different natural frequencies; they have different preferred periods of oscillation. The challenge to the neuromuscular machinery of coordinating two limbs with different preferred periods is a real-world problem for amputees. This problem can be investigated by weighting an intact limb and measuring the relative phase between freely swinging limbs.
In normal subjects, the addition of mass to the ankle of one leg has produced the expected impact on relative phase and the variability of relative phase. The limb with the added mass has a longer period of oscillation, thereby creating a shift in relative phase and an increase in the variability when the limbs are coordinated. Applying these manipulations to prosthetic limbs is a logical extension of this line of research.
A device was manufactured to allow mass to be added to a prosthesis while permitting the location of the mass to be changed. The preferred periods of the intact and prosthetic limbs were calculated. Manipulations of the moment of inertia were performed on the prosthesis in order to produce a preferred period similar to the intact limb. Pilot data on trans-tibial amputees suggests that designing prostheses that match the intact limb in terms of moment of inertia, and ultimately the preferred period, may improve the coordination of limbs and reduce energy expenditure.

Lateral Ball Interception: Hand Movements During Planar Ball Trajectories
Open Topics - Session B

Ryan Arzamarski, Steven J. Harrison, Claire Michaels
University of Connecticut

To intercept an approaching object, such as a baseball or football, one must be attuned to the information that will appropriately guide movements. In one such task, catching a ball within arm's reach, disparities have arisen concerning hand-movement reversals. Jacobs and Michaels (2004) found that the unilateral movement reversals observed in the interception of swinging balls were predicted by an optical variable, the
ratio of lateral velocity to expansion velocity. The time course of this variable depends on the ball's trajectory type; so different trajectories ought to yield different hand movements (and movement-reversals). With planar, constant-velocity approaches, the optical variable predicts no movement reversals. However, Montagne, Laurent, Durey, and Bootsma (1999) reported bi-directional reversals in an experiment in which the ball trajectory was planar; these results conflict with predictions generated by Jacobs and Michaels's optical quantity. In the present contribution, we report research that again evaluates movement reversals in the planar-trajectory paradigm. Participants caught an oncoming luminous ball rolling across a flat surface in an otherwise dark room. Using Montagne et al.'s criteria, we observed no movement reversals for any condition. Observed trajectories were thereby consistent with the predictions based on the optical ratio.

ECOLOGICAL AND DYNAMICAL SYSTEMS APPLICATIONS IN REHABILITATION
Symposium 5: Ecological Applications in Rehabilitation

Symposium Organizer: Claire F. Michaels
Collaboratory for Rehabilitation Research
Center for the Ecological Study of Perception and Action
University of Connecticut

In recent years, leaders in the field of physical therapy have called for the clinical practice of physical therapy to become evidence-based. The transition to evidence-based practice has been limited by a lack of research on existing treatment interventions and theabsence of a broad theoretical basis for therapeutic interventions. This symposium seeks to bridge gaps among theory, research, and clinical practice. The premise is that there presently exists a substantial body of knowledge about human movement, including perceptual and environmental factors, that can be useful in solving the practical problems of rehabilitating people with movement disorders. Researchers with commitments to the ecological and/or dynamical-system approaches have been major contributors to the development of theory and to the generation of empirical findings on perception and action. By seeking applications of this work in rehabilitation settings, the benefit of these discoveries can be reaped. Additionally, problems arising in clinical settings can raise novel theoretical issues for the fundamental scientist. Thus, a close connection between clinicians and scientists may help us enjoy the cross-fertilization of theory and practice. The symposium speakers are all practicing clinicians who have been trained in ecological or dynamical-systems approaches to human movement.

Roads to Success in Action: Learning, Calibration, and Perception
Symposium 11: Perceptual Learning and the Control of Action

Claire Michaels
Center for the Ecological Study of Perception and Action
University of Connecticut

Rob Withagen
Center for Human Movement Sciences
University of Groningen

David M. Jacobs
University of Granada

We distinguish three processes by which actions are made to fit with the demands of the environment. We define learning as the process by which an information variable that is more useful comes to be relied upon, and see the process as navigation through a continuous information space. We define calibration as the process of adjusting the function that relates an information variable to an action variable, and also see it as navigation through a continuous space. Finally, perception is the process of letting information guide action; that is, letting the navigated-to information guide the action according to the calibrated function. While ecological psychologists regularly argue that information guides action; our more novel claim is that the processes of learning and calibration are themselves guided by information. We provide evidence that learning and calibration are indeed information-guided. Second, we present results from a dynamic touch paradigm that dissociate learning and calibration; in particular, we show that learning can occur with or without calibration. Attention is also given to how the term calibration is sometimes used by others, and how calibration relates to other concepts such as scaling.

 

The Availability of Optical Structure Influences Postural Control Differently for Young and Older Adults
Symposium 5: Ecological Applications in Rehabilitation

J.M.Kinsella-Shaw, Steven J. Harrison, and C. Colon-Semenza
Collaboratory for Rehabilitation Research
Department of Physical Therapy
University of Connecticut

Changes in both visual sensitivity and control of upright posture occur with aging. The potential impact on function of these changes, both independently and interactively, is largely unknown. This study assessed the influence of the availability of optic structure under different levels of ambient illumination on the dynamics of upright posture in young adults (22-24 years old) and older adults (66-83 years old). Two force plates, one under each foot, measured each participant's center of pressure under 5 conditions. Four conditions were the combination of 2 visual arrays and 2 levels of illumination. The visual arrays consisted of a flat, white screen or a vertical array of 3 rows of 9 dowels. Illumination was either high enough to support photopic vision (440 LUX) or low enough (3 LUX) to support mesopic vision. In the fifth condition, participants stood with eyes closed. Contrast sensitivity and risk for falls were also measured. Analyses of variance revealed effects of age, array type, and illumination on several measures of center of pressure behavior. A 3-way interaction of age, illumination, and array type on center of pressure path-length revealed that the availability of optic structure influences postural control differently in older and younger adults. Recurrence quantification analysis of the center of pressure time-series revealed systematic variations in center of pressure dynamics across conditions. Regression analysis identified measures of contrast sensitivity at different spatial frequencies that predict center of pressure variability, and thus could serve as a diagnostic tool for older adults.