summary of four psychology article

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Smart, L. James] On: 27 April 2010 Access details: Access Details: [subscription number 921691585] Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37- 41 Mortimer Street, London W1T 3JH, UK

Ecological Psychology Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t775653640

Consequence of Self Versus Externally Generated Visual Motion on Postural Regulation Eric M. Littman a; Edward W. Otten a;L. James Smart Jr. a a Department of Psychology, Miami University,

Online publication date: 27 April 2010

To cite this Article Littman, Eric M. , Otten, Edward W. andSmart Jr., L. James(2010) 'Consequence of Self Versus Externally Generated Visual Motion on Postural Regulation', Ecological Psychology, 22: 2, 150 — 167 To link to this Article: DOI: 10.1080/10407411003720106 URL: http://dx.doi.org/10.1080/10407411003720106

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Ecological Psychology, 22:150–167, 2010

Copyright © Taylor & Francis Group, LLC

ISSN: 1040-7413 print/1532-6969 online

DOI: 10.1080/10407411003720106

Consequence of Self Versus Externally Generated Visual Motion on

Postural Regulation

Eric M. Littman, Edward W. Otten, and L. James Smart, Jr.

Department of Psychology

Miami University

A limiting factor in the use of virtual environments to examine perception and

action is that it is dependent on the quality of the interaction. A virtual reality

game (first-person viewpoint with moderate difficulty) was used to investigate the

link between control and postural regulation. Postural regulation was examined

using a motion capture system, and the differences that emerged as a result

of the participant being a passive observer versus an active participant were

evaluated using a fractal procedure: dispersion analysis. A significant interaction

was found between control and health, namely, participants’ postural behavior

tended to be less uniform when they were sick and when they were in control

of the environment. These results lead the authors to suggest that successful

interaction with novel environments depends on the ability to find and consistently

use appropriate control strategies.

In order to achieve most goals, one must be able to successfully assess and

interact with one’s surroundings. These behaviors in turn rely on the ability to

recognize the possibilities for action (affordances) that exist in a given setting.

Successful coordination of these activities allows for people to behave in a

manner that is functional and reduces the distance between the current and goal

states. These activities can be described as being future oriented or prospective

in nature (literally, forward looking; E. J. Gibson, 1969; E. J. Gibson & Pick,

2000; Reed, 1996; Turvey, 1992). Prospective regulation of behavior entails the

use of perceptual information to guide and adjust subsequent actions (Adolf,

Correspondence should be addressed to L. James Smart, Jr., 318 Psychology Building, Miami

University, Oxford, OH 45056. E-mail: [email protected]

150

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 151

Eppler, Marin, Weise, & Clearfield, 2000; J. J. Gibson, 1979/1986). In other

words, prospective control involves the ability to anticipate future events and

modulate behavior based on this to-be-achieved state. In principle prospective

regulation of behavior is a pervasive aspect of our experience and is thought to

be one of the hallmarks of our ability to interact with our environment (Turvey,

1992).

In our everyday experience there exists a relationship between what we per-

ceive and the actions that we produce. This type of relation can be characterized

as closed loop in nature. This relation exists when the person’s actions influence

and are influenced by information from the world. However, there are a number

of situations where the relation between perception and action is open looped. In

this relation, the person’s actions have little or no consequence on the information

generated by the environment but are still influenced by it (e.g., watching a

movie). In both cases the person’s goal is to interact successfully (i.e., in a

manner that reduces the distance between the current and goal states) with the

environment around him or her; it is the manner in which these actions are

accomplished that differ as a function of the type of control available. Problems

can arise when a person implements an inappropriate strategy or is unable to

determine what behaviors will bring him or her closer to a goal state. It has been

suggested that one possible outcome of this deficiency can be the occurrence of

motion sickness (Stoffregen & Smart, 1998).

Motion sickness has been recognized as a problem for travelers since the

ancient Greeks (Reason & Brand, 1975). In recent times, through research

and innovation, engineers have been able to make generally better vehicles

that suppress many of the motion characteristics that make people motion sick

(although this was not the explicit goal for engineers; Crowley, 1987). The boats

that made people sick are now constructed in a manner that dissipates certain

frequencies of motion so people will not feel the shift from the waves. Similarly,

planes are constructed to fly smoothly against the wind so the passengers do not

feel the fluctuation of the plane moving through the air, thus reducing the general

incidence of motion sickness. It is important to note that although technological

advances have reduced the incidence of motion sickness, these advances have

not eliminated the occurrence of motion sickness.

These general improvements in technology have not yielded uniform out-

comes. In the expanding field of virtual reality (VR) and virtual environment

(VE) it has been found that general advances in technology (i.e., increased

perceptual fidelity) often result in more motion sickness rather than less (Biocca,

1992; Crowley, 1987; Nichols & Patel, 2002). For example, high definition

displays, LCDs, and next generation video game consoles (e.g., PlayStation®

3 or Microsoft® Xbox 360TM) provide the user with photorealistic perceptual

information but still employ traditional joystick controls that may not allow

for the types of actions suggested by the abundant perceptual information. In

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

152 LITTMAN, OTTEN, SMART

addition, these displays, video game consoles, simulators, and so on, continue to

depict inertial motion, which has been associated with the occurrence of visually

induced motion sickness (Hettinger & Riccio, 1992; Stoffregen & Smart, 1998).

In particular, these next generation gaming platforms have been shown to induce

motion sickness symptoms (Mehri, Faugloire, Flanagan, & Stoffregen, 2007;

Villard, Flanagan, Albanese, & Stoffregen, 2008).

One possible reason for this phenomenon is that in VE a different kind of

perception-action coordination occurs or is required. That is, although perceptual

fidelity has been steadily increasing over time, action fidelity has not progressed

as quickly (Stoffregen, Bardy, Smart, & Pagulayan, 2003). For example, when

people drive a car they feel the force the car produces against their skin/bodies

whereas in a VE there is no force associated with operating the virtual vehicle.

These changes in the mapping between perception and action not only provide

information to the user about his or her current circumstances (i.e., that she or

he is in a VE) but also require a change in the manner that the person behaves

(Stoffregen et al., 2003). VEs often depict situations (provide information) that

could be perceived as affording control or interaction when in actuality the

environment does not support those behaviors (thus a design goal for VEs should

be to reduce these type of discrepancies; see Stoffregen, Bardy, & Mantel,

2006). These depicted relations may cause the user to adopt inappropriate

strategies to compensate for the perceived relation between his or her actions

and that information. In short, people in these situations will attempt to act in

a prospective manner (where their actions have consequences for future stimuli)

when a more reflexive action (where their actions are a response to changes in

the stimulus) is mandated. The failure to implement the appropriate actions for

a given situation can result in observable behavioral changes or disruptions.

In many studies concerning motion sickness in VEs, researchers are aware that

there are behavioral (postural) disruptions that accompany it—the conventional

interpretation of this has been the view of sway as a symptom of the induced

illness rather than a causal factor (see Riccio & Stoffregen, 1991, and Stoffregen

& Smart, 1998, for a more in depth discussion of this interpretation). One of the

impacts of this view is that postural changes have primarily been examined in

a preimmersion/postimmersion manner (see e.g., Cobb, 1999; Nichols, 1999).

In a similar fashion, Murata (2004) was interested in how postural stability is

affected when immersed in a VE (using the same VE as used in this study).

Murata sampled postural stability (using a force plate) at hour intervals during

a 3-hr session in the VE. Although he found changes in postural motion across

samples gathered during the sessions, Murata concluded that the changes were

an outcome of sensory conflict. It should be noted that he did not differentiate

between sick and well participants, nor did he have participants stop at the onset

of any symptoms (all participants completed the sessions). Additionally, when

Murata sampled participants’ postural motion, they were not engaged in the

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 153

task. This makes it difficult to make inferences about how people are behaving

while engaged in the task. In summary, utilizing pre- postmeasurements to assess

changes in postural regulation has not yielded clear results, further suggesting

that the etiology of motion sickness in VEs may have other (nonbehavioral)

origins.

Alternately, the chosen method (pre/post) of studying postural behavior in

VEs may be nonoptimal. Studies that have looked at postural behavior during

immersion have shown that disruptions in sway precede and can be predic-

tive of motion sickness (Bonnet, Faugloire, Riley, Bardy, & Stoffregen, 2006;

Smart, Otten, & Stoffregen, 2007; Smart, Stoffregen, & Bardy, 2002; Stoffregen,

Hettinger, Haas, Roe, & Smart, 2000; Stoffregen & Smart, 1998). Although

promising, these studies reduced the postural data to a relatively small set

of descriptive measures, which only grossly reflected how postural sway had

changed over time (i.e., provided indices of the extent, spread, or velocity of

motion). These types of measures do not allow one to discriminate between

prospective and compensatory (reflexive) behavior (in short, how the person

is moving rather than how much he or she is moving). To make this kind of

discrimination, it becomes necessary to examine the process (read: structure of

postural motion) as it occurs over time. In order to accomplish this task both

effectively and more efficiently, one must reconsider the utilization of discrete

measures (whereas descriptive are not optimal for understanding processes) in

favor of more process-oriented approaches (such as fractal or other nonlinear,

dynamic measures) of examining the data. These approaches could potentially

provide a different level of description that will allow for clearer understanding

of the changes in movement regulation and coordination as they occur over time.

Given this it seems appropriate to develop studies that allow for the examination

of different modes of postural behavior.

RATIONALE

This study was designed to utilize a process-oriented approach in order to

more accurately examine the behavioral component of postural control and

coordination. We used a VR game (moderate difficulty) to identify parameters

of people’s movement that may lead to motion sickness. It has been shown

that this type of stimuli can be nauseogenic (Mehri et al., 2007; Stoffregen,

Faugloire, Flanagan, Yoshida, & Mehri, 2008). The overall goal of this study

was to explicitly examine the process of postural regulation while engaged in a

given task. The specific goal was to determine the role of control (the ability to

modify the stimuli) on the occurrence of motion sickness and the ability to make

appropriate adjustments (actions that serve a future goal). Research has indicated

that these factors can be influenced by perceived or actual control (Otten et al.,

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

154 LITTMAN, OTTEN, SMART

2003; Rolnick & Lubow, 1991). It has also been shown that decoupled motion

can degrade task performance and increase the reports of motion sickness (Muth,

Walker, & Fiorello, 2006). The current study entailed the measurement and

subsequent analysis of seated participants’ postural motion while immersed in a

VE. It is believed that disruptions in postural regulation that can be produced by

VEs may lead to adverse side effects (cf. Riccio & Stoffregen, 1991) and that

these effects will be more pronounced when the person is exposed to externally

generated motion.

METHOD

Participants

Twenty-two undergraduate students ranging in age from 18 to 22 years par-

ticipated in this study—9 participants were male and 13 were female. All

participants had normal or corrected-to-normal vision and were in good health.

Participants had an average height of 1.72 m and weight of 69.21 kg. Thirty-six

percent (36.36%) of the participants said they had experienced motion sick-

ness before, thirty-six percent (36.36%) said they had not experienced motion

sickness before, and twenty-seven percent (27.27%) did not know if they had

experienced motion sickness before. As part of the demographic data collected,

participants were asked to rate their susceptibility to motion sickness using a

10-point scale where 1 indicated not very susceptible and 10 indicated very

susceptible. The average reported susceptibility of the participants to motion

sickness was 3.86 and did not differ significantly for participants who would later

become sick in the current study. Forty-one percent (40.91%) of the participants

said that they had previously used VR prior to participating in the experiment

and fifty-nine percent (59.09%) said they had not previously used VR prior to

participating in the experiment. Participants were recruited from the Introduction

to Psychology course and were given 2 credits toward their required 12 credits

for participating in this study. Participants were treated in accordance with

American Psychological Association (APA) ethical standards at all times (APA,

1992) and the research protocol was approved by the university’s institutional

review board.

Materials

Display. A Sharp PG-C30XU LCD projector was used to display the VE. The approximate physical image dimensions were 1.5 m high � 2.1 m wide and

2.6 m diagonal yielding a visual angle of 41ı diagonal from a distance of 3 m.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 155

FIGURE 1 Experimental setup and screen shot of stimuli (welder’s goggles not shown).

Game console and software. A Nintendo 64 (Nintendo, Inc.) video game

system, with standard controller, was used to generate the stimuli for the study.

A first person (perspective) shooter (Goldeneye; Rare, Inc.) was used in this

study. The train level was used for the control (watching only) and experimental

trials (see Figure 1). This level of the game was chosen due to its relatively

linear progression through the stage (from rear of the virtual train to the engine

car) and basic goal (get to the front of the train and save the hostage). During

the experimental trials, opponent accuracy, reaction time, and ability to injure

the participant’s character were set to minimal levels to ensure that participants

would be able to interact with the game for the 5-min data collection period.

Motion tracker. A magnetic tracking system (Flock of Birds; Ascension,

Inc.) was used to track postural motion. This is a system that detects motion

in six degrees of freedom (three axes of translation and three of rotation). A

centrally located emitter creates a low-intensity magnetic field of known strength,

extent (˙1.23 m), and orientation. Six sensors (“birds”) move within this field.

The system can detect the position of each sensor to an accuracy of 1 mm.

Four sensors were used (1-lower spine (T12), 2-right wrist, 3-upper spine (C7),

4-head). Data from the sensors were sampled at a rate of 30 Hz and stored on

computer for later analysis.

Goggles. A pair of lightweight, plastic, clear-tinted welder’s safety goggles

was used to restrict the participants’ field of view to that of the stimulus.

Additionally, participants read and signed an informed consent form. Health

status, symptomology, and motion sickness history data were collected using the

Simulator Sickness Questionnaire (SSQ, short form; Kennedy, Lane, Berbaum,

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

156 LITTMAN, OTTEN, SMART

& Lilienthal, 1993); responses were recorded and stored on a computer and

analyzed at the completion of the study.

Procedure

Upon entering the lab, participants were told the nature of the study and were

asked to fill out a consent form, demographics sheet, and an SSQ. Then partic-

ipants performed two balance checks, which the participants were required to

successfully complete in order to participate in the study. The balance checks

consisted of a standard field sobriety test (e.g., walking back and forth on

a straight line in heel-toe fashion) and balancing for 30 s on their preferred

leg with eyes closed. These balance checks were repeated postexposure (again,

participants had to successfully complete both checks) to ensure that participants

were stable enough to safely be dismissed.

Sensors were attached to the participants’ head, right wrist, upper spine (C7)

and lower spine (T12) using Velcro belts/bands. Participants sat on a stool (which

did not provide passive back support) holding the game controller in their hands

(see Figure 1). This setup was intended to mimic typical playing conditions while

also requiring some level of active postural regulation. A total of 14 trials were

administered in each session. The first two trials were used to collect baseline

postural data (one with eyes open (looking at a blank white wall) and one with

eyes closed). Both baseline trials lasted 20 s. The next 4 trials (control) were

used to make sure that visual stimuli would in fact influence the participants’

postural sway. Two of these trials were performed with the participants’ eyes

open and 2 were with eyes closed. During the eyes open control trials the

participant watched a 60-s video of someone playing the train level (the level

they would play in the experimental trials); when his or her eyes were closed

the video was played again, but the participant could only hear what was going

on. All control trials lasted 60 s. The next 6 trials were the experimental trials

(in which the variable of control was manipulated), each lasting 5 min. Three

used “active” control in which the participants had control over the game play.

During these trials the participants’ task was to successfully interact with the

game. Three trials used “passive” control in which the participants watched the

experimenters play the game. The participants’ task during these trials was to

observe the stimulus. The order of active/passive control was alternated across

participants. The final 2 trials were 1 baseline (20 s staring at a blank white

laboratory wall) and 1 control (60 s watching the video recording of the train

level). Both trials were conducted while the participants’ eyes were open.

At the outset of the experiment the participants were instructed to immediately

notify the experimenters if symptoms of motion sickness emerged. Between

trials, the experimenter recorded participants’ comments on their experience

and any reports of sickness. If the participants reported symptoms of motion

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 157

sickness, participation in the experiment was stopped, even if motion sickness

was reported during the middle of a trial, and the participants were asked to

complete the SSQ again, indicating their increased level of symptomology. These

participants were required to remain in lab until their symptoms subsided to a

point where they could pass the balance checks again. In the case where the

participants did not experience symptoms of motion sickness and successfully

completed all trials they were asked to fill out the SSQ and required to pass

both balance checks again before leaving.

RESULTS AND DISCUSSION

Data from all 22 participants were analyzed—the 7 participants who became

motion sick and the 15 remaining participants. For participants who became

motion sick, only the data prior to their reports of sickness onset were used.

Postural sway was collected from all six axes of motion (anterior–posterior [AP],

lateral, vertical, yaw, pitch, and roll) from each sensor, but for the purposes of

this study sway was only analyzed in AP and lateral head motion. AP and

lateral motion have been shown to be significant in previous postural research

(cf. Mehri et al., 2007; Stoffregen & Smart, 1998). The angular counterparts

to these measures, pitch and yaw, were not utilized because these data would

have to be transformed to use in the fractal analysis; second, these data were

highly correlated with their linear counterparts, so it is not clear that they would

have provided a significant amount of additional information. A fractal analysis

(dispersion, see later for explanation of this index) was performed on the motion

data gathered in the control and experimental trials. For the purposes of the

type of analysis employed in this study, motion data were not filtered. The

index obtained from this analysis was analyzed using a 2 (control) by 2 (health)

analysis of variance (ANOVA).

Sickness Incidence

Seven out of 22 (31.81%) participants became motion sick during the experi-

ment. The majority of participants who became motion sick during the study

reported symptoms after a passive trial. As with previous studies (Mehri et al.,

2007; Smart et al., 2002; Stoffregen et al., 2000), reports of motion sickness

were unambiguous.

SSQ

A Kruskal-Wallis nonparametric test was used to analyze the SSQ responses

for four main scales—pre- and postnausea, pre- and postocular-motor, pre- and

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

158 LITTMAN, OTTEN, SMART

TABLE 1

Mean (SD) SSQ Subscale and Total Scores as a Function of Health (Well, Sick)

and Time (Preexposure, Postexposure), N D 22

Nausea Ocular-motor Disorientation Total

Well Sick Well Sick Well Sick Well Sick

Pre 7.00 (11.67) 8.18 (13.97) 10.61 (9.42) 19.45 (19.47) 2.78 (7.80) 13.94 (17.99) 8.73 (9.23) 16.50 (17.89)

Post 13.36 (15.21) 46.38 (18.58) 17.69 (9.79) 41.16 (26.50) 8.35 (10.26) 45.75 (40.74) 16.21 (11.88) 50.74 (29.93)

postdisorientation, and the total score. The analysis, conducted on the preexpo-

sure questionnaire, revealed no significant differences between the preexposure

scores on the Nausea, Disorientation, and Oculomotor subscales as well as the

total scores, indicating that the preexposure scores for the sick participants did

not differ from the well participants. However, when the analysis was conducted

on the postexposure questionnaire, it revealed significant differences between the

postexposure scores on the Nausea, Disorientation, and Oculomotor subscales

as well as the total scores, indicating that the postexposure scores for the sick

participants were higher than those of well participants, p < :05 for all subscales

and total (see Table 1).

Postural Motion Plots

In attempting to understand how postural motion is being regulated under these

novel constraints, it is important to be able to see what changes are occurring and

how these changes are supported by the subsequent quantitative analyses. Given

this, the first step was to create plots that illustrate what the participants were

doing while immersed. The researchers chose to use two different representations

of the participants’ motion—a state space and a phase plot. State spaces depict

motion along two axes (lateral and AP position), whereas the phase plots (lateral

motion vs. lateral velocity) depict spatial as well as temporal properties of motion

and therefore provide a more encompassing “view” how the participants were

behaving. Previous studies have indicated a strong correlation between medial-

lateral balance and overall postural instability (Hertel, 2002). In addition, in

studies where instability (and motion sickness) was self-induced, disruptions in

postural control were most prevalent in the medial-lateral axis (Bonnet et al.,

2008; Smart, Pagulayan, & Stoffregen, 1998), particularly when seated (Mehri

et al., 2007; Stoffregen et al., 2000). It is with these considerations in mind that

the researchers chose to concentrate on lateral motion. It should be noted that

both that state space and phase space plots were intended to be representative of

the data generated in this study but were generated from only 2 participants (1

to represent participants who remained well and 1 to represent participants who

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 159

later became motion sick). Both the state space and phase plots were divided

into four quadrants: A (well-active), B (well-passive), C (sick-active), and D

(sick-passive).

State space plots. (See Figure 2.) In comparing well (A, B) and sick

(C, D), we observed increased variability in the participant who would later

become motion sick, which is comparable to the differences observed in pre-

vious studies (Bonnet et al., 2008; Stoffregen & Smart, 1998; Villard et al.,

2008). In addition, it appears that the participant who later became sick never

seemed to hone into a particular area, unlike the well participant who spent the

majority of the time in one area (a difference that seemed to be common across

sick and well participants). This pattern of motion may represent a reduced

ability to appropriately regulate behavior (see Smart & Smith, 2001). When

comparing the active condition (A & C) with the passive condition (B & D), it

appears that the well participant was reducing the amount, but not the quality,

of motion when moving from an active to a passive role. In contrast, the sick

participant’s responses were equally variable but qualitatively different across

these conditions. Again, this is suggestive of the participant’s reduced ability

FIGURE 2 Representative AP vs. ML position plots for well (top) and sick (bottom)

participants during active (left) and passive (right) trials. Plots are drawn to same scale.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

160 LITTMAN, OTTEN, SMART

to find and maintain an appropriate regulatory strategy. It should be noted that

these types of plots are time compressed (i.e., they represent the entire 300-s

trial), so interpretations about behavioral processes based solely on spatial data

should be made cautiously. It is with this mind that we also examined phase

plots derived from these same data.

Phase plots. (See Figure 3.) In comparing well (A, B) and sick (C, D), we

observed increased variability in velocity as well as position in the participant

who would later become motion sick. It appears that this participant had a

difficult time returning to a stable state after being disrupted (exhibited by

the increased variability of position per velocity). As with the state space, we

interpreted this as a deficit in the participant’s ability to adopt a functional

movement strategy. Although disruptions were also present in the participant

who remained well, the disruptions were much shorter in duration (exhibited

by few excursions from “limit cycle” behavior). When comparing the active

condition (A & C) with the passive condition (B & D), there was a general

reduction in velocity and position as the participants moved between active and

passive roles. However, the participant who would later become sick still seemed

FIGURE 3 Representative Phase (ML position vs. ML velocity) plots for well (top) and

sick (bottom) participants during active (left) and passive (right) trials. Plots are drawn to

same scale.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 161

to have trouble finding and maintaining a given movement strategy (exhibited

by “slow” return to limit cycle behavior).

In general, for both the state space and phase space plots, one can note,

qualitatively, that the well participant exhibited similar “types” of motion while

in both the active and passive roles. Arguably, the well participant was able

to “pick up” on an appropriate behavioral strategy across control conditions.

Even though the magnitude of motion differed between the active and passive

roles, the “type” of behavior (i.e., the pattern of variability) the well participant

displayed was consistent across roles. Conversely, the sick participant’s motion

differed in both magnitude and “type” across both the active and passive roles.

This suggests the way in which the participant is moving may be more indicative

of future problems than magnitude differences alone (a possibility suggested by

Riccio & Stoffregen, 1991). Although these depictions suggest the importance of

honing in on an appropriate strategy and the ability to maintain it, the challenge

was to find a quantitative index that could represent this possibility. It was this

search for an appropriate measure that prompted the following analyses.

Dispersion (Fractal) Analysis

Dispersion analysis is a type of fractal analysis that can be used to examine the

similarity of measured data across various timescales. This analysis calculates

the log of variance (relative dispersion) at multiple timescales and compares them

to the log of the timescales. The slope of a linear regression line determined

by the log-log relationship can be used to calculate the fractal D using the

equation D D 1� slope. The value of D can vary between 1 and 1.5 with

values near 1 indicating uniformity over all timescales; values near 1.5 indicate

random uncorrelated noise (i.e., looking at the system at one timescale does not

inform you about the system at other timescales). As the value of D increases,

it indicates that the larger timescales do not describe the data in a manner

similar to the smaller timescales; in short, as the data become less self-similar,

the dispersion value increases (Bassingthwaighte, Liebovitch, & West, 1994). In

the area of postural sway, we suggest that values of D approaching 1.5 would

indicate instability (i.e., lack of an appropriate regulatory strategy), whereas

values near 1 would indicate stability. The dispersion analysis was conducted

on the control (sample size: 1,800; 60 s at 30 Hz) and experimental (sample

size: 9,000; 300 s at 30 Hz) trials only.

AP Motion

Control trials. There were no significant main effects or interactions. The

self-similarity of participants’ motion did not vary between the eyes open (M D

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

162 LITTMAN, OTTEN, SMART

FIGURE 4 Mean (SE) dispersion values for AP motion as a function of condition (active

vs. passive) and health (well, sick), N D 22.

1:03, SD D :04) and the eyes closed (M D 1:04, SD D :05) conditions or

between participants who remained well (M D 1:03, SD D :04) and those who

later became motion sick (M D 1:03, SD D :05).

Experimental trials. There was no main effect of control, no main effect

of the condition of the participant (sick vs. well), and there was no interaction

between the relative control of the participant and the condition of the participant

(see Figure 4).

Lateral Motion

Control trials. There were no significant main effects or interactions. The

self-similarity of participants’ motion did not vary between the eyes open (M D

1:04, SD D :04) and the eyes closed (M D 1:04, SD D :06) conditions or

between participants who remained well (M D 1:03, SD D :05) and those who

later became motion sick (M D 1:04, SD D :04).

Experimental trials. A 2 (active vs. passive) by 2 (sick vs. well) mixed

model ANOVA resulted in a significant main effect of control, F .1; 61/ D 15:90,

p < :05, �2 D :21, indicating that the value of the dispersion exponent was

smaller when participants were not in control of the environment (M D 1:07,

SD D :06) than when they were in control (M D 1:10, SD D :07). Participants

exhibited more self-similar (fractal) behavior as passive observers and displayed

greater variability in their behavior as active participants.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 163

FIGURE 5 Mean (SE) dispersion values for ML motion as a function of condition (active

vs. passive) and health (well, sick), N D 22.

Additionally, the results indicated a significant main effect of health,

F .1; 61/ D 4:95, p < :05, �2 D :08, indicating that the value of the dispersion

exponent was smaller when participants were well (M D 1:08, SD D :01)

than when they were sick (M D 1:11, SD D :01). These data illustrate that

participants who later became sick displayed more chaotic or erratic (i.e., less

self-similar) behavior than those participants who remained well throughout.

Finally, the results indicated a significant interaction between control (active

vs. passive) and health (sick vs. well), F .1; 61/ D 7:00, p < :05, �2 D

:10. In particular the analysis revealed that whereas participants who remained

well throughout the study maintained similar dispersion values across control

conditions (active, passive), participants who later became motion sick ex-

hibited different dispersion values across control conditions. In particular, for

participants who later reported motion sickness, dispersion values were much

higher when they were in control of the game than when they were simply

watching (where their dispersion values were similar to the well participants; see

Figure 5).

CONCLUSIONS

Overall, the results indicated that participants who later became sick exhibited

less self-similar behavior than those participants who remained well. Addi-

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

164 LITTMAN, OTTEN, SMART

tionally, the behavior of participants tended to be less self-similar while they

were actively controlling the VE. Moreover, participants who later became

sick displayed more chaotic behavior while in control of the environment and

displayed behavior similar to well participants as passive observers, contrary to

our original prediction.

Participants who remained well displayed similar patterns of motion while

in both the active and passive roles. Behavior across well participants differed

somewhat in magnitude; however, their behavior did not differ qualitatively,

which seems to be a critical element. In other words, the structural pattern of

variability was consistent across roles. This suggests that high variability in and

of itself isn’t necessarily maladaptive as long as it allows for the achievement

of the goal. Subsequently, it could be considered beneficial and functionally

adaptive.

Conversely, participants who later became sick displayed qualitatively differ-

ent structural patterns of motion across the different roles. As stated earlier, this

qualitative difference may suggest an inability on the part of the participants who

are susceptible to motion sickness to acquire situation-appropriate movement

strategies in a timely fashion. Given the pattern of results obtained, we feel that

our hypothesis regarding the importance of discovering exploiting the relevant

information and motor constraints is supported despite the fact that key changes

occurred in active trials rather than passive trials (where participants primarily

reported being motion sick).

In general, participants who became sick tended to exhibit more random

motion than participants who remained well. These results suggest that those

participants who remained well were able to find an appropriate stable postural

strategy whereas those participants who became sick were searching for such

a strategy but were not able to find one before exhibiting motion sickness

symptoms.

The results of this study suggest that fractal analyses may allow for a different

understanding of the process of postural regulation than traditional descriptive

measures. The index provides a means to talk about differences in the way in

which posture is regulated; in this study the data suggested that when becoming

motion sick or when placed in a closed-loop and novel situation, postural motion

(regulation) exhibits a different fractal nature. In short, the behavior differs

depending on the timescale being examined; this difference was not seen in

the participants who remained well or when in the passive condition. However,

we do have to be careful in how much interpretation we can make with these

analyses as the effects were significant but not strong. Part of the reason for

this is the size of the data samples used in the analysis, which are 1.5 to 3

times greater than those used in other studies of this type (Bonnet et al., 2006;

Riley, Mitra, Stoffregen, & Turvey, 1997; Villard et al., 2008). The consequence

of these larger samples is that the data can appear more self-similar than they

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 165

actually are (i.e., over longer samples, it is more likely that trends in the data will

be repeated) thus reducing the magnitude of change in the measure (dispersion).

The changes in dispersion value obtained in this study are consistent with other

studies that have used this analysis to look at sick/well differences in postural

motion (see Otten & Smart, 2009) and suggest that the change in fractal structure

of the data does not have to be very large to produce the outcomes observed

(motion sickness).

In summary, we suggest that the advantage that the well participants seemed

to possess was their ability to find an appropriate postural strategy. These results

may also explain the variation in susceptibility across people in that it may be

related to the person’s ability to find and adopt a regulatory strategy appropriate

for the situation. This inability to determine the appropriate behavior (or changes

in behavior) seems to be a key factor in the emergence of postural instability

and subsequent motion sickness.

ACKNOWLEDGMENTS

Portions of this material were presented at the 2006 North American Meeting of

the International Society for Ecological Psychology, Cincinnati, OH; the 2006

Sigma Xi annual student conference, Detroit, MI; and the 15th International

Conference on Perception and Action, Minneapolis, MN. This project was

funded by an undergraduate research award granted to Eric M. Littman.

We thank Robert Leitner, Rae Burgard, Meghan Capistrano, Patrick Harmon,

Marissa Luli, Emily LaPlante, Kelly Vitatoe, Jack Shuler, and Bumi Hadaka for

their help in the collection and analysis of the data and Kaleigh Coughlin for

her help in the preparation of this article. We also thank Dr. Leonard Mark for

his insightful comments and suggestions regarding earlier drafts of this article.

REFERENCES

Adolph, K. E., Eppler, M. A., Marin, L., Weise, I. B., & Clearfield, M. W. (2000). Exploration in

the service of prospective control. Infant Behavior and Development, 23, 441–460.

American Psychological Association. (1992). Ethical principles of psychologists and code of con-

duct. American Psychologist, 47, 1597–1611.

Bassingthwaighte, J. B., Liebovitch, L. S., & West, B. J. (1994). Fractal physiology. Oxford, UK:

Oxford University Press.

Biocca, F. (1992). Will simulator sickness slow down the diffusion of virtual environment technol-

ogy? Presence: Teloperators and Virtual Environments, 1, 333–343.

Bonnet, C. T., Faugloire, E., Riley, M. A., Bardy, B. G., & Stoffregen, T. A. (2006). Motion sickness

preceded by unstable displacement of the center of pressure. Human Movement Science, 25, 800–

820.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

166 LITTMAN, OTTEN, SMART

Bonnet, C. T., Faugliore, E., Riley, M. A., Bardy, B. G., & Stoffregen, T. A. (2008). Self-induced

motion sickness and body movement during passive restraint. Ecological Psychology, 20, 121–145.

Cobb, S. V. G. (1999). Measurement of postural stability before and after immersion in a virtual

environment. Applied Ergonomics, 30, 45–57.

Crowley, J. S. (1987). Simulator sickness: A problem for army aviation. Aviation, Space, and

Environmental Medicine, 58, 355–357.

Gibson, E. J. (1969). Principles of perceptual learning and development. New York: Appleton.

Gibson, E. J., & Pick, A. D. (2000). An ecological approach to perceptual learning and development.

New York: Oxford University Press.

Gibson, J. J. (1986). The ecological approach to visual perception. Hillsdale, NJ: Erlbaum. (Original

work published 1979)

Hertel, J. (2002). Functional anatomy, pathomechanics, and pathophysiology of lateral ankle insta-

bility. Journal of Athletic Training, 37, 364–375.

Hettinger, L. J., & Riccio, G. E. (1992). Visually induced motion sickness in virtual environments.

Presence, 1, 306–310.

Kennedy, R. S., Lane, N. E., Berbaum, K. S., & Lilienthal, M. G. (1993). Simulator sickness

questionnaire: An enhanced method for quantifying simulator sickness. International Journal of

Aviation Psychology, 3, 203–220.

Mehri, O., Faugloire, E., Flanagan, M., & Stoffregen, T. A. (2007). Motion sickness, console video

games, and head mounted displays. Human Factors, 49, 920–934.

Murata, A. (2004). Effect of duration of immersion in a virtual reality environment on postural

stability. International Journal of Human-Computer Interaction, 17, 463–477.

Muth, E. R., Walker, A. D., & Fiorello, M. (2006). Effects of uncoupled motion on performance.

Human Factors, 48, 600–607.

Nichols, S. (1999). Physical ergonomics of virtual environment use. Applied Ergonomics, 30, 79–

90.

Nichols, S., & Patel, H. (2002). Health and safety implications of virtual reality: A review of

empirical evidence. Applied Ergonomics, 33, 251–271.

Otten, E. W., & Smart, L. J. (2009). The effect of open vs. closed-loop optic flow on visually

induced motion sickness. In J. Wagman & C. Pagano (Eds.), Studies in perception and action X

(pp. 149–152). Philadelphia: Taylor & Francis.

Otten, E., Smart, L. J., Amin, M., Bidwell, B., Johnson, B., McLaughlin, M., Pecorak, K., & Smith,

D. (2003, July). Effect of predictability of stimulus on visually induced motion sickness. Poster

presented at the 12th International Conference on Perception and Action, Surfers Paradise, Gold

Coast, Australia.

Reason, J. T., and Brand, J. J. (1975). Motion sickness. London, UK: Academic Press.

Reed, E. S. (1996). Encountering the world: Toward an ecological psychology. Oxford, UK: Oxford

University Press.

Riccio, G. E., & Stoffregen, T. A. (1991). An ecological of motion sickness and postural instability.

Ecological Psychology, 3, 195–240.

Riley, M. A., Mitra, S., Stoffregen, T. A., & Turvey, M. T. (1997). Influences of body lean and

vision on unperturbed postural sway. Motor Control, 1, 229–246.

Rolnick, A., & Lubow, R. E. (1991). Why is the driver rarely motion sick? The role of controllability

in motion sickness. Ergonomics, 34, 867–879.

Smart, L. J., Otten, E. W., & Stoffregen, T. A. (2007). It’s turtles all the way down: A comparative

analysis of visually induced motion sickness. Proceedings of the 51st Annual Meeting of the

Human Factors and Ergonomics Society, 1631–1634.

Smart, L. J., Pagulayan, R. J., & Stoffregen, T. A. (1998). Self-induced motion sickness in unper-

turbed stance. Brain Research Bulletin, 47, 449–457.

Smart, L. J., & Smith, D. L. (2001). Postural dynamics: Clinical and empirical implications. Journal

of Manipulative and Physiological Therapeutics, 24, 340–349.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0

SELF MOTION IN VE 167

Smart, L. J., Stoffregen, T. A., & Bardy, B. G. (2002). Visually induced motion sickness predicted

by postural instability. Human Factors, 44, 451–465.

Stoffregen, T., Bardy, B. G., & Mantel, B. (2006). Affordances in the design of enactive systems.

Journal of Virtual Reality Research, 10, 4–10.

Stoffregen, T. A., Bardy, B. G., Smart, L. J., & Pagulayan, R. J. (2003). On the nature and evaluation

of fidelity in virtual environments. In L. J. Hettinger & M. W. Haas (Eds.), Virtual & adaptive

environments: Applications, implications, and human performance issues (pp. 111–128). Mahwah,

NJ: Erlbaum.

Stoffregen, T. A., Faugloire, E., Flanagan, M., Yoshida, K., & Mehri, O. (2008). Motion sickness

and postural sway in console video games. Human Factors, 50, 322–331.

Stoffregen, T. A., Hettinger, L. R., Haas, M. W., Roe, M., & Smart, L. J. (2000). Postural instability

and motion sickness in a fixed-base flight simulator. Human Factors, 42, 458–469.

Stoffregen, T. A., & Smart, L. J. (1998). Postural instability precedes motion sickness. Brain

Research Bulletin, 47, 437–448.

Turvey, M. T. (1992). Affordances and prospective control: An outline of the ontology. Ecological

Psychology, 4, 173–187.

Villard, S. J., Flanagan, M. B., Albanese, G. M., & Stoffregen, T. A. (2008). Postural instability and

motion sickness in a virtual moving room. Human Factors, 50, 332–345.

D o w n l o a d e d B y : [ S m a r t , L . J a m e s ] A t : 1 6 : 5 6 2 7 A p r i l 2 0 1 0