Investigation of activity performance on laptop in vibration environment

Ramasamy Narayanamoorthy
a,
*
, V. Huzur Saran
b
a
Department of Mechanical Engineering, Francis Xavier Engineering College, Tirunelveli 627 003, India
b
Department of Mechanical and Industrial Engineering, Indian Institute of Technology Roorkee, Roorkee 247 667, India
a r t i c l e i n f o
Article history:
Received 18 March 2011
Received in revised form
5 January 2012
Accepted 13 August 2012
Available online 19 September 2012
Keywords:
Pointing activity
Mouse clicking
Whole-body vibration
Fitts law
Task performance
Activity comfort
a b s t r a c t
The present study investigated effects of vibration direction, vibration magnitude, object size, object
distance and angle of approach on movement time taken by eleven participants using a mouse to perform
pointing and clicking activity on a laptop. From the combination of vibration directions, vibration
magnitudes, object sizes, object distances and angles of approach, a total of 416 exposure conditions were
considered. For successful completion of pointing and clicking activity under different exposure condi-
tions, the time required to nish the task was measured. Using the Fitts law, a relation between movement
time and index of difculty was developed for different vibration directions and magnitudes considered in
the study. Higher vibration magnitudes, smaller size of the objects, and diagonal angular movements were
found to affect the movement time and hence showed difculties to perform the activity.
Relevance to industry: This study investigated the performance of a mouse to execute pointing and
clicking activity on a laptop in vibration environment. The results showed that difculties to perform the
activity were due to higher magnitudes, smaller size of the objects, and diagonal cursor movements.
These research achievements can help humanecomputer interaction design in various dynamic envi-
ronments such as in land and sea vehicles.
2012 Elsevier B.V. All rights reserved.
1. Introduction
Many train passengers during travel experience difculty in
manipulating the mouse while working on laptop computers.
Whole-body vibration (WBV) to which they are exposed onboard,
affects humanecomputer interface (HCI) and degrade their
performance (Grifn, 1990; Suzuki, 1998). An understanding of
speed and accuracy of human motor movements associated with
the computer input devices is important for better design of HCI
and implementation of effective interface systems. Computer use is
primarily dominated by pointing and clicking activities and touch
screen, track ball and mouse are the most commonly used devices
for performing these activities. Both pointing and clicking activities
require precise hand movements which can be severely affected by
vibration. Many models based on the Fitts law are available to
explain the relationship between speed and accuracy of human
motor movements associated with computer devices. The key
factors that inuence movement time considered in these models
include size of object, its distance fromorigin, angle of approach for
a mouse to point and click an icon like target/object (Balakrishnan,
2005; Grossman and Balakrishnan, 2005; Whisenand and Emurian,
1995, 1996; 1999). However, all these models have been develo-
ped for static (no vibration) environment, and are inappropriate
for vibration environments as in running trains/vehicles. Later,
Liu et al. (2007) and Lin et al. (2010) analyzed suitability of pointing
devices such as touch screen, mouse, and a track ball for vibration
environment and found that mouse performed well than other
devices. Further, Manseld et al. (2007) demonstrated that mouse
is better than touchpad under vibrating environments for pointing
and clicking. Besides, the present ride comfort standards too,
provide inadequate guidance on the effect of vibration on human
performance during computer work. The effective interface
between portable computer systems and human is necessary to
ensure a comfortable working environment in moving vehicles.
Developing an interface system to perform pointing and clicking
task on moving vehicles requires thorough investigation of the
interactive effects of vibration directions, magnitudes, object size,
object distance and approaching angle to a displayed object/icon on
hand movement. As the Fitts law is used to express the difculty to
perform hand movement in terms of movement time and used to
evaluate input devices in HCI eld (Lin et al., 2010), the investiga-
tion should consider those interactive effects on movement time
taken by a mouse to select icon like objects displayed on a laptop
monitor. Therefore, the present study investigates the effects of
* Corresponding author.
E-mail address: rn_moorthy@yahoo.com (R. Narayanamoorthy).
Contents lists available at SciVerse ScienceDirect
International Journal of Industrial Ergonomics
j ournal homepage: www. el sevi er. com/ l ocat e/ ergon
0169-8141/$ e see front matter 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ergon.2012.08.005
International Journal of Industrial Ergonomics 42 (2012) 513e524
vibration directions and its magnitudes on an indicator of difculty
such as movement time taken by the mouse while performing the
task (pointing and clicking activity by a mouse) for different angles
of approach, object sizes and distances.
1.1. Convention of vibration axes
WBV occurs when a human is supported by a surface that is
shaking and the vibration affects the body parts remote from the
site of exposure (Manseld, 2005). For an example as in case of
railway passengers, their body is supported at oor (by foot) at seat
(by buttocks) at back rest (by back) and at headrest (by head). The
vibration at these points is then transmitted through body from the
supporting part of the body to the head through the nervous
system, the skeleton, including the spine and ultimately the skull,
which might have its own dynamic responses to the transmitted
vibration (Manseld, 2005). Degraded comfort, interference with
activities, impaired health, perception of low magnitude vibration
and the occurrence of motion sickness are the effects of human
response to WBV. For WBV measurements as per ISO 2631-1, 1997,
the X-axis is dened in fore-and-aft direction, Y-axis in lateral
direction and Z-axis in vertical directions as shown in Fig. 1.
1.2. Human activities under vibration
Pointing and clicking on a laptop monitor involves the transfer
of information from the eye to brain for processing. The results of
processed information are executed by transferring the information
from the brain to hands. During the WBV, there is a higher prob-
ability that different parts of the human body will receive different
vibrations which may alter the exposed persons conception of
either the importance of activity (task) or the performance criteria
which should be targeted (Grifn, 1990). Thus chances of modi-
cation in information at eye (input end of information), the hands
(output end) and performance of input devices (monitor/display
unit) and output devices (mouse) used to perform the activity
(task) may occur which leads to impaired visual acuity and manual
control (Lewis and Grifn, 1978). The total effect of alteration in
information affects the human performance.
Vibration causes image movement on the retina of the eye
which degrades vision. Retinal image movement can occur as
a result of either vibration of the eyes (known as observer vibra-
tion), vibration of the visual eld (display vibration) or vibration of
both the eye and the visual eld (simultaneous observer and
display vibration). This retinal image movement depends on the
vibration frequency and viewing distance. The greatest problems
with display vibration occur at frequency range of about 2e20 Hz
(Stott, 1984). For a seated person, the magnitude of eye motion at
its resonance frequencies is dependent on body posture, seating
characteristics and characteristics of back rest and highly variable
between individuals (Grifn, 1990). Degraded visual performance
due to vibration was evidenced when the head made contact with
the headrest (Johnston and Wharf, 1979). It was reported that the
larger size and shape of characters, and changes in types of char-
acter fonts resulted in reduction of reading errors (Lewis and
Grifn, 1980; Meddick, 1977).
In Swedish inter-city trains, Khan and Sundstrm (2004) found
that passengers using computers were disturbed by jerks and
vibration. A similar study conducted in India by Narayanamoorthy
et al. (2008a) revealed that disruption to perform reading and
writing activities by the vibration and noise among executive class
train passengers although the ride was perceived as comfortable.
Nakagawa and Suzuki (2005) conducted a study on a simulator
with 6 degrees of freedom built as a mockup of railway coach to
address the problems of computer users in trains. For improved
performance on computer in vibration environment, it recom-
mended a value of 95 50 mm as optimum distance between the
center of seat and edge of the table and130 75 mm as the
optimum distance between center of the seat and computer. Time
required by participants to complete the pointing and clicking
activity by using mouse was used by Arora (2006) as a parameter to
predict the effect of WBV and a parameter for measuring the
performance. The study revealed that there was negligible differ-
ence in the human performance between low vibration of their
study (0.508 ms
2
r.s.s) and no vibration (control condition) cases,
but at high vibration (0.878 ms
2
r.s.s) the performance reduced
signicantly.
1.3. Humanecomputer interface
Interaction between users and computers occurs at the user
interface, which includes both software and hardware. The user
interface refers to the graphical, textual and auditory information
presented by a program to the user. In addition, the control
sequences employed to control the program such as keystrokes
with the computer keyboard, movements of the computer mouse,
and selections with the touch screen can also be attributed to user
interface. Even though pointing and clicking task can be performed
by many devices, touch screen, mouse and touchpad are popular
among them. However, mouse was widely used by more number of
computer users (Woods et al., 2002). Using Fitts law, Lin et al.
(2010) investigated performance of pointing devices such as
touch screen, mouse, and track ball to nd its suitability under
dynamic environments like in ships and moving land vehicles. Four
target sizes, 10, 15, 20, and 25 mm, four target distances 35, 45, 100,
and 170 mmand four target angles 0

, 45

, 90

, and 135

were used
in the investigation. Three levels of vibration static (no vibration),
r.m.s. acceleration of 0.22 and 0.34 ms
2
were used. From the
results of the investigation, it was found that the effect of vibration
decreased the performance of the pointing devices, the effects were
not at the same strength and the mouse was strongly affected by
vibration. Based on the results of movement time, error rate, slope
in the Fitts law model and throughput, the mouse was recom-
mended for use in the vibration environment to maintain a high
efciency in computer pointing tasks. Recently, Narayanamoorthy
and Huzur Saran (2011) investigated inuences of vibration
Fig. 1. Typical co-ordinate system for vibration on the human body (adapted from ISO
2361-1, 1997).
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 514
directions, vibration magnitudes, object sizes, object distances and
angles of approach on producing errors while performing pointing
and clicking activity on a laptop monitor similar to passengers
working on moving trains. From the results, it was concluded that
higher vibration magnitudes, smaller size of the objects, and
diagonal angular movements created more errors and affected
accuracy of the activity and hence showed difculties to perform
the activity.
Computer users utilize the mouse to control the motion of
a cursor in a graphical user interface (GUI). Clicking or hovering can
select les, programs or actions from a list of names, or in
a graphical interface, through pictures called icons and other
elements. Point-and-click (pointing) is the simple action of
a computer user which consists of moving a cursor to a certain
location on a monitor (point) by a mouse and then clicking the
mouse button, usually the left one (click). The time required for
human hand movement such as to performa point-and-click action
can be quantitatively modeled with Fitts law (Fitts, 1954;
MacKenzie et al., 1991).
1.4. Fitts law
Fitts law is used to model human movement (MacKenzie et al.,
1991) and predicts the time required to rapidly move from a start-
ing position to a nal object area as a function of the distance to the
object and the size of the object. This lawapplies only to movement
in a single dimension and describes simple motor response of, say,
the untrained movements of human hand. In this study, the Fitts
law is used to model the act of pointing (point and click) on laptop
monitor with a mouse. The commonly used Shannon formulation
of Fitts law for movement along a single dimension is given by
T C MID (1)
Where,
T is the average time taken to complete the movement ein this
study, it is time required to move cursor between white button/
object to yellow button/object
C and M are empirical constants and referred as intercept and
slope respectively. They can be determined by tting a straight
line to appraised time
The Index of Difculty (ID) for the object is written in Fitts law
in logarithmic form and has units of bits. It can be written as
ID log
2

D
W
1

(2)
D is the distance from the starting point to the center of the
object, i.e., the tangential distance between the white and
yellow button
Wis the width of the object measured along the axis of motion.
Fromthe equation, it can be understood that smaller objects and
objects placed farther away require more time to acquire. The task
difculty is analogous to information (MacKenzie et al., 1991),
therefore, the rate of task execution can be interpreted as the
human rate of information processing. Thus, the units for slope M
are seconds/bit and its reciprocal called the Index of Performance
(IP) is expressed in bits/sec. Since IP is expressed in bits/sec, it is
often referred to as bandwidth in the literature. Intuitively, the
higher the bandwidth, the higher will be the rate of human
performance since more information is being articulated per unit
time (MacKenzie, 1995). The constant C represents reaction time
and/or the time required to click a button/object. Fitts law has
proven increasingly useful in HCI research by providing a model to
account for the speed of a users successful movement to capture an
object with a direct manipulation device, such as a mouse
(Manseld et al., 2007).
2. Aim
In order to improve the performance of an input task with
a mouse under motion environment, the effects of both the object
(icon, menu and button) layout and the motion direction have to be
considered simultaneously. A better understanding of the suit-
ability of an HCI for motion environment as like in train could be
obtained by investigating the effects of object size, distance and
approach angle in a simulated train environment to execute
pointing and clicking task with a mouse, an activity frequently
performed by computer users.
Hence, the purpose of this study is:
(a) to investigate the effects of vibration magnitude and direction
on the performance of train passengers executing pointing and
clicking task on a laptop monitor using mouse as the input
device,
(b) to model the mouse movement time with Fitts law for
different vibration conditions,
(c) to analyze the effect of vibration on the mouse motion direc-
tions (angles of approach), size and distance of the object.
The greatest difculties to perform the activity may occur at
5 Hz frequency of vertical vibration due to the mechanical prop-
erties of the human body such as resonance at 5 Hz, maximum
driving force per unit acceleration (apparent mass). The resonance
at 5 Hz leads to interference with simple hand activities. The
maximum apparent mass causes greatest discomfort to human
body and head motions (Grifn, 1990). Hence, in order to evaluate
the performance in this study, the vibration frequency was xed as
5 Hz in all the directions. Based on the studies of Krishnakant
(2007) and Narayanamoorthy et al. (2008a,b, 2009a,b), the
unweighted r.m.s. acceleration magnitude of 0.6, 0.9 and 1.2 ms
2
were selected as the exposure levels in X-, Y- and Z-axis and a no
vibration condition was considered as control condition. For multi-
axis vibration (simultaneous vibration in X-, Y- and Z-axis)
considered in the study, the magnitudes were taken as root sum
squared (r.s.s.) values of each axis.
3. Materials and methods
3.1. Experimental setup
In order to study the activity interference under different
vibration conditions as experienced by railway passengers,
a vibration simulator has been indigenously designed and fabri-
cated as a mockup of railway coach. The vibration simulator
consists of (i) vibrating platform supported by four helical coil
springs and (ii) table along with two chairs and a couch, rigidly
xed on the platform as schematically shown in Fig. 2. Multi-axis
vibration was produced by simultaneously exciting all the three
electrodynamics vibration exciters connected to the vibration
platform (Fig. 2) at the centers of its three principal planes through
spherical joints. These vibration exciters are controlled by three
independent controllers placed outside the simulator room. More
details about the design of vibration simulator are explained else-
where (Narayanamoorthy et al., 2009a). The participant was seated
on a chair in the well lit simulator room, with laptop placed on
a table in front of him.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 515
3.2. Task
Pointing and clicking by a standard optical mouse (Intex Little
Wonder, Intex Technologies (India) Ltd, New Delhi, India) on
a laptop monitor (Lenovo Thinkpad R61, Lenovo, China; 14.1 inch
XGA TFT monitor) was the task selected for present investigation.
Inuences of the object size to be pointed by the cursor, distance of
the object from the starting position of the cursor and approaching
angle to the object were studied by considering different object
sizes, object distances and angles of approach. As the task involved
moving the cursor to another location, two objects were used in the
study. One of the objects was white button/icon referring origin of
the cursor and always positioned at the center of the laptop
monitor. The other object considered was a yellow button having
different sizes, placed at different locations and orientations in
a random order. The different sizes of yellow button used in the
experiment are squares of side 25, 50, 75 and 100 pixels (6, 13, 20
and 27 mm respectively). The object distance was measured as the
tangential length from the center of white button to the center of
yellow button. The different distances used in the study were 62,
124, 186, and 248 pixels (17, 33, 50, and 65 mm respectively). The
angle of approach is dened as the angle between the orientation of
the yellow button and origin measured in clockwise direction
(Fig. 3) The present study employed eight angles e 0

, 45

, 90

,
135

, 180

, 225

, 270

and 315

.
A program code developed in the Labview (Labview V7.1,
National Instruments Corporation, USA) generated the required
graphic views on laptop monitor which was used by the partici-
pants to perform the task on the laptop. At the start of the program
initial entries such as participants number and vibration exposure
conditions were given. At the start of task, initially a white square
button of a specied size appeared at the center of the monitor.
Placing the cursor on the white button and clicking it, caused the
white button to disappear and a yellow square button appeared at
some other location on the monitor. The participant was then
required to immediately move the cursor, place it within the
boundary of the object (yellowbutton) and click it. If the cursor was
placed within the boundary of yellow button while clicking, then
the yellow button disappeared and the white button reappeared at
the center. However, if the cursor was clicked outside the boundary
of yellow button, an error was registered by the program and the
yellow button remains undisturbed on the monitor, until the
participant is able to click inside its boundary. For completion of
one condition, this procedure was repeated 32 times but each time
the yellow button reappeared, its size and position vary according
to the program. The cursor movement time is calculated as the
difference of time period between the clicking of white button to
yellow button which is computed by the program and expressed in
seconds. The output of the customized program yielded the
participant number, exposed vibration direction, vibration magni-
tude, size of yellow button, distance of yellow button from the
white button, angle of approach, and cursor movement time in
Fig. 3. Angle of approach and mouse motion directions (one button only displays at a time on the screen).
Fig. 2. Schematic diagram of mockup of railway coach (vibration simulator).
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 516
separate columns as an MS Ofce Excel supporting le (Microsoft
Ofce 2007, Microsoft Corporation, USA) for further analysis. All the
information about participants was represented by a unique
number initially assigned to each one of them.
Background of the laptop monitor remains black in color
throughout the experiment. Selections of color combinations for
the buttons are based onprevious ndings of Luria et al. (1989) who
reported that a revealed color makes no signicant difference in
search time, but a yellow on black combination is preferred by the
users. However, Mallick (2007) recommended to use black back-
ground and white text combination for better performance while
working on vibration environments. This color combination is also
consistent with the object/background colors used in previous
experiments (Manseld et al., 2007; Narayanamoorthy 2007;
Whisenand and Emurian, 1995, 1996, 1999).
3.3. Exposure conditions
As per the aim, different combinations of vibration magnitudes
and directions yielded a total of 13 different exposure conditions
[product of 4 directions and 3 levels of vibration magnitude along
with one control condition]. One exposure condition comprises of
twice occurrence of sixteen combinations, each of which are
composed of four object sizes and four object distances. This
combination is assigned to the participants with any one angle of
approach in a random order. Latin square design methodology is
used to ensure that there is no effect of the order of exposure
conditions on the performance, and each participant gets
a different order of exposure condition.
3.4. Study group
Eleven male students of Indian Institute of Technology Roorkee,
India, volunteered for this study. Their age, weight and stature
ranged from 19 to 35 (mean age 27.53 5.69) years, 49 to 105
(mean weight 67.23 13.9) kg and 1.63 to 1.78 (mean height
1.68 6.23) m respectively. All participants were right handed and
had a working knowledge of computers. None of the participants
reported uncorrected visual problems or physical limitations that
would inhibit their use of the mouse as an input device. Before the
experiment session, participants were given practice trials to
familiarize themselves with the task. The practice lasted until
participants are comfortable with the task, which is usually about
3 min. The time taken to complete activity at one condition under
no vibration environment by a person was about 1e2 min. The
participants complete entire study at about 40e50 min. The Insti-
tutional Human Ethics Committee of Indian Institute of Technology
Roorkee approved the experimental procedures. A written consent
was obtained from all the participants before the starting of
experiment. All the participants are instructed to move the cursor
and capture the object button as quickly and accurately as possible.
3.5. Vibration measurements
A calibrated tri-axial accelerometer (Kistler K-beam 8393A
series, Kistler Instruments (P) Ltd, Singapore) xed on the vibrating
platform oor beneath the participants chair captured the vibra-
tion signals at a sampling frequency of 1000 Hz, and sent it to
a computer for further analysis through an eight channel data
acquisition system (NI USB 6009, National Instruments Corpora-
tion, USA). Software from National Instruments Corporation (Lab-
view Signal Express, version 3) was used to monitor and analyze
the signal data. Online monitoring of acceleration signal ensured
that a constant vibration level was maintained.
4. Analysis
To begin with, the data of all the conditions (13) for all partici-
pants (11) was sorted based on vibration direction, vibration
magnitude, object size, object distance, angle of approach, and
mouse motion direction (foreeaft and lateral). This was followed by
the separation of error data and performing analyses on the error
free data. Next, the index of difculty is calculated based on the
Fitts law. Then, the scatter graphs of movement time against index
of difculty were drawn separately for all conditions which yielded
regression lines. Further, the slope (M) and intercept (C) of the
regression lines for different conditions were obtained in order to
model the movement as per the Fitts law. Finally, the mean
movement time was calculated for different vibration directions,
vibration magnitudes, object size, object distance and angle of
approach and used for analyzes.
4.1. Statistical analysis
As error data are separated, instead of repeated measures
ANOVA, a multi way analysis of variance (ANOVA), with vibration
directions, vibration magnitudes, object sizes, object distances and
angle of approaches as within subjects independent variables and
measurement time as dependent variable was performed to test
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 1 2 3 4
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
S
e
c
)
Index of Difficulty
Control
X dir
Multi-axis
Y dir
Z dir
Fig. 4. Effect of vibration direction on movement time.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 1 2 3 4
M
o
v
e
m
e
n
t

T
i
m
e

(
S
e
c
)
Index of Difficulty
Zero
Low
Medium
High
Fig. 5. Effect of vibration magnitude on index of difculty.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 517
the signicance. A post-hoc Tukey test was performed if signi-
cance is found within the group. A paired t test was also performed
to compare the means. ANOVA tests were performed on the
Statistical Package for Social Sciences (version 15) (SPSS Inc., Chi-
cago, USA), while paired t test was performed on Microsoft Excel
data analysis package (Microsoft Ofce 2007, Microsoft Corpora-
tion, USA). In all the test, signicance was considered at the level of
p < 0.05.
5. Results
For better understanding, the gures presented (Figs. 4e16) are
grouped with respect to vibration direction, magnitude and
combined vibration direction and magnitude. Further, they are
clustered to understand the effect of object size, distance, angle of
approach on index of difculty. The results of multi-way ANOVA is
presented in Tables 1 and 2.
The results of statistical analysis showed that vibration direc-
tion, magnitude, object size, distance and angle of approach have
their individual effect on mouse movement time (p < 0.05). Inter-
action of vibration direction and magnitude, vibration direction
and object size, vibration magnitude and object distance, and
object distance and size produce signicant effect on the mouse
movement time (p < 0.05). All other possible interactions showed
insignicant effect only.
5.1. Effects of vibration direction and magnitude on movement time
Mean mouse movement time for all successful attempts at
different vibration directions magnitudes and are furnished in
Tables 3e5 respectively. The results fromFigs. 4e9, indicated that the
time taken to perform the task at different vibration directions and
magnitudes increase with the level of difculty. The regression lines
drawn for different vibration directions (Fig. 4) revealed that task
executionunder multi-axis vibrationtook more time thanwithother
three principal directions. Next to multi-axis vibration, the vibration
in Y-axis consumed more time to execute the task. However,
the post-hoc (Tukey) statistical analysis showed insignicance
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 1 2 3 4
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
S
e
c
)
Index of Difficulty
Control
Low
Medium
High
Fig. 6. Effects of vibration magnitude in X-axis on movement time.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 1 2 3 4
M
e
a
n
m
o
v
e
m
e
n
t
T
i
m
e
(
S
e
c
)
Index of Difficulty
Control
Low
Medium
High
Fig. 7. Effects of vibration magnitude under multi-axis vibration on movement time.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 1 2 3 4
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
S
e
c
)
Index of Difficulty
Control
Low
Medium
High
Fig. 8. Effects of vibration magnitude in Y-axis on movement time.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 1 2 3 4
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
S
e
c
)
Index of Difficulty
Control
Low
Medium
High
Fig. 9. Effects of vibration magnitude in Z-axis on movement time.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 518
Movement time (Sec)
0.2 0.4 0.6 0.8 1.0
0.2
0.4
0.6
0.8
1.0
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.2
0.4
0.6
0.8
1.0
0
45
90
135
180
225
270
315
Control
X-axis
Y-axis
Z-axis
Multi-axis
0
0
0
0
0
0
0
0
Fig. 10. Variation of movement time with angles of approach for different vibration directions.
Movement time (Sec)
0.4 0.6 0.8 1.0
0.4
0.6
0.8
1.0
0.4 0.6 0.8 1.0
0.4
0.6
0.8
1.0
0
45
90
135
180
225
270
315
Control
0.6 m s
-2
0.9 m s
-2
1.2 m s
-2
o
o
o
o
o
o
o
o
Fig. 11. Variation of movement time with angles of approach for different vibration magnitudes.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 519
(p 0.726) between Y-axis and Z-axis. As expected, the non-
vibrating or control condition demonstrated the least time to
execute the task and hence least difculty. The slope and intercept of
regression lines for different vibration directions representing Fitts
law model for mouse movement are presented in Table 6.
The Fig. 5, drawn for mean movement time against index of
difculty for different vibration magnitudes expressed that higher
magnitudes required more time to execute the task than the
medium, low and zero (control) magnitudes (p < 0.001). The
corresponding slope and intercept values of regression lines for
different vibration magnitudes that representing Fitts law model
for mouse movement are provided in Table 7.
Direction wise effects of magnitude is displayed in Figs. 6e9 and
its corresponding slope and intercept values of regression lines that
representing Fitts law model for mouse movement is tabulated in
Table 8. It is evident fromthese gures that task execution at higher
magnitudes under all directions of vibration consumed more time
and hence more difculty felt. However, at higher index of dif-
culty, a disparity observed that performing an activity under
medium magnitude required more movement time as compared to
that under high magnitude (Fig. 6).
5.2. Effect of angle of approach on movement time
Figs. 10 and 11 present mean value of movement time across the
eight angles of approach for different vibration directions and
magnitudes. Although all angles are discrete, the data points in the
gures are connected to clarify the graphical representation of the
data. The Fig. 10 illustrates that minimum movement time is
required with angles 0

and 315

, whereas maximum movement

time is necessary with 270

, 225

and 135

for completion of task

under different vibration directions.
In general, it could be concluded that the diagonal angles (45

,
135

, 225

and 315

) consumed more time than the non-diagonal

angles (0

, 90

, 180

and 270

). However, contradictory results

were identied with two angles; while the non-diagonal angle 270

utilized more movement time, the diagonal angle 315

consumed
less time inspite of being a diagonal angle. However, paired t test
results showed that the differences among the movement time of
diagonal angles are insignicant [t (1944) 1.199, two tail p 0.23].
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
0 62 124 186 248
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
s
e
c
)
Distance (Pixels)
Control
X dir
Multi-axis
Y dir
Z dir
Fig. 12. Effects of vibration direction on object distance.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
0 62 124 186 248
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
s
e
c
)
Distance (Pixels)
Control
Low
Medium
High
Fig. 13. Effects of vibration magnitude on object distance.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
0 25 50 75 100
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
s
e
c
)
Size (Pixels)
Control
X dir
Multi-axis
Y dir
Z dir
Fig. 14. Effects of vibration direction on object size.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
0 25 50 75 100
M
e
a
n

m
o
v
e
m
e
n
t

T
i
m
e

(
S
e
c
)
Size (Pixels)
Control
Low
Medium
High
Fig. 15. Effects of vibration magnitude on object size.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 520
Further, it was found that mouse movement in the lateral direction
(0

and 180

) used less time than that of foreeaft direction (90

and
270

). This is supported by paired t test results [t (946) 11.01, two

tail p < 0.0001]. It was also observed that, while the multi-axis
vibration required maximum time, the control condition needed
minimum time to complete the task, for all the approach angles, as
expected.
Fig. 11 exhibits the effect of vibration magnitude on angle of
approach. As anticipated, at high magnitude the movement time
was maximum and at zero (control) magnitude, it was minimum.
From Fig. 11, it was noted that the most time consuming angle of
approach was 270

followed by angles 225

and 135

. The general
trend of Fig. 10 is identical with Fig. 11 for movement time in
diagonal and non-diagonal angles. The mean movement time for all
the approach angles of this study are stated in Table 9.
5.3. Effects of object distance on movement time
The effect of distance under different directions (Fig. 12.)
showed that increasing the distance had proportional effect on the
movement time. For object distances between 62 and 124 pixels,
the trend line showed higher slope in all vibration directions.
However, between other distances, the slope of trend line was less
which indicated that after certain distance (in this study it was 124
pixels), the increase in distance was less pronounced on the
movement time. As pointed out earlier, movement time under
multi-axis vibration was the longest and under control condition it
was shortest, is applicable for different object distances also. The
Movement time (Sec)
0.0 0.2 0.4 0.6 0.8 1.0 1.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0.0 0.2 0.4 0.6 0.8 1.0 1.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0
45
90
135
180
225
270
315
25 pixel
50 pixels
75 pixels
100 pixels
0
0
0
0
0
0
0
0
Fig. 16. Variation of movement time with angle of approach for different object sizes.
Table 1
Results of multi-way ANOVA analyzing the effect of vibration direction, magnitude,
object size and distance on mean movement time.
Source df F Sig. Observed power
Direction (Dir) 3 31.029 0.000 1.000
Magnitude (Mag) 2 58.456 0.000 1.000
Distance (Dis) 3 187.743 0.000 1.000
Size (Siz) 3 337.441 0.000 1.000
Dir Mag 6 4.166 0.000 0.980
Dir Dis 9 1.535 0.130 0.735
Mag Dis 6 2.440 0.023 0.832
Dir Mag Dis 18 1.017 0.436 0.747
Dir Siz 9 2.232 0.018 0.904
Mag Siz 6 1.702 0.116 0.654
Dir Mag Siz 18 0.629 0.880 0.480
Dis Siz 9 3.997 0.000 0.996
Dir Dis Siz 27 0.858 0.675 0.789
Mag Dis Siz 18 0.542 0.939 0.410
Dir Mag Dis Siz 54 0.666 0.972 0.869
Error 3749
df, degrees of freedom; Sig.- signicance.
Table 2
Results of multi-way ANOVA analyzing the effect of vibration direction, magnitude,
and angle of approach on mean movement time.
Source df F Sig. Observed power
Direction (Dir) 3 19.958 0.000 1.000
Magnitude (Mag) 2 33.307 0.000 1.000
Angle (Ang) 7 63.750 0.000 1.000
Dir Mag 6 3.062 0.005 0.917
Dir Ang 21 1.011 0.446 0.795
Mag Ang 14 1.411 0.139 0.835
Dir Mag Ang 42 1.176 0.204 0.985
Error 3853
df, degrees of freedom; Sig.- signicance.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 521
mean mouse movement time for different distances is furnished in
Table 10. Analysis of the effect of distance (Fig. 13) under different
magnitudes demonstrates that increasing the distance had
proportional effect on the movement time. Between distance 62
and 124 pixels, higher slope was observed in all magnitudes.
However, between other distances, the slope was less. The state-
ment, the movement time under high magnitude is longest
whereas it was shortest under control condition was applicable for
different object distances also.
5.4. Effects of object size on movement time
Figs. 14e16 demonstrates the effect of size of the object on
mouse movement time for different vibration directions, magni-
tudes and approach angles, respectively. The results of Figs. 14 and
15 show that the size and movement time are inversely propor-
tional under different vibration directions and magnitudes. The
slope was higher between the sizes 25 and 50 pixels, and less
between other sizes, indicating that sizes larger than 50 pixels had
reduced effect on movement time. Like general trend, multi-axis
vibration and movement under high magnitude required longest
time whereas control condition showed shortest time for different
sizes also. From Fig. 16, it is evident that movement time decreased
with increase in the size of object for the different angles of
approach. It also endorses the previous ndings on approach
angles, i.e., movement time along 0

and 315

were the shortest and

along 135

, 225

and 270

the longest; movement time along

diagonal angles were longer than non-diagonal angles, except 270

and 315

. The mean mouse movement times for different object

sizes are stated in Table 11.
6. Discussion
The results of this study demonstrated the inuence of vibration
direction, magnitude, object size, object distance and angle of
approach to move a cursor by a mouse between two objects pre-
sented on the laptop monitor.
6.1. Effects of vibration magnitude and direction
Performing the pointing and clicking task with mouse in multi-
axis vibration, the participants required longest movement time
indicating maximum difculty in executing the task. Since the
multi-axis vibration was composed of simultaneous vibration in
three mutually orthogonal directions along with some phase
differences, it was likely to increase the complexity of vibration,
thus making it difcult to carry out the coordinated hand move-
ment required in moving the mouse to accomplish the given task.
The next higher difculty in executing the task was found with
vibration in Y-axis. This can be explained, since the hand holding
the mouse is usually supported at the elbow on the table and
therefore, permits only lateral swinging motion while allowing
controlled motion in X-axis. Therefore, it can be expected that the
vibration excited in Y-axis would increase the lateral swing
resulting in the experience of difculty in coordinating the mouse
movement. It was reported that vibration in lateral (Y-axis) direc-
tion caused difculties to stabilize upper parts of the body
(Holmund, 1998) and affected the human performance (Agren and
Westberg, 2003; Wollstrom, 2000; Sundstrm and Khan, 2008).
As expected, higher magnitude had longest movement time
than lower magnitudes and control condition. This was in agree-
ment with Manseld et al. (2007), who reported that signicant
difference in movement time occurred between high
(0.878 ms
2
r.s.s) and low (0.508 ms
2
r.s.s) magnitudes and Lin
et al. (2010) who reported increasing movement time with vibra-
tion level. The result was also consistent with other studies which
reported poor performance at higher vibration magnitudes than
lower magnitudes (Sundstrm, 2006; Westberg, 2000). The
disparity observed in medium and high vibration magnitudes, at
low index of difculty, may be due to the elimination of errors
(Figs. 8, 10 and 11). In general, a combination of vibration directions
and magnitudes caused difculties in performance, and lead to
a slower movement.
6.2. Effects of angle of approach
Appreciably faster movements were found at angles 0

, 90

and
315

, whereas angles 135

, 225

and 270

demonstrated slower
movements. These results conrmand extend the ndings of Boritz
et al. (1991) and Whisenand and Emurian (1999), who revealed
more rapid mouse-driven cursor movements (less movement time)
at 0

in comparison to 270

(ccw-counter clockwise) for item

selection in a pie-shaped menu that consisted of icon-like square
objects similar to those used in the present study. However, they
used counter clockwise angle measurement as compared to this
Table 3
Mean mouse movement time for different vibration directions.
Vibration direction Mean movement time (sec) Standard deviation
Control 0.663 0.177
X-axis 0.753 0.237
Y-axis 0.810 0.301
Z-axis 0.797 0.302
Multi-axis 0.842 0.321
Table 4
Mean mouse movement time for different vibration magnitudes.
Vibration magnitude Mean movement time (sec) Standard deviation
High 0.849 0.349
Low 0.761 0.249
Medium 0.795 0.272
Control 0.663 0.177
Table 5
Mean mouse movement time for different vibration directions under different
magnitudes.
Vibration direction Vibration magnitude Mean time Standard deviation
Control Control 0.663 0.177
X-axis High 0.776 0.259
Low 0.725 0.220
Medium 0.761 0.230
Y-axis High 0.862 0.376
Low 0.786 0.272
Medium 0.799 0.248
Z-axis High 0.890 0.391
Low 0.744 0.232
Medium 0.772 0.258
Multi-axis High 0.880 0.358
Low 0.791 0.266
Medium 0.864 0.335
Table 6
Slope (M), intercept (C) and coefcient of correlation (R
2
) values for different
directions.
Regression values Control X-axis Y-axis Z-axis Multi-axis
Slope (M) 0.156 0.194 0.231 0.215 0.223
Intercept (C) 0.373 0.392 0.39 0.402 0.441
R
2
0.965 0.981 0.986 0.975 0.929
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 522
study. Even though different angle measurement direction (clock-
wise) was employed in the present study, the equivalent angles (0

and 90

) showed consistent results with their studies. Adopting the

counter clockwise angle measurement system, MacKenzie and
Buxton (1992) reported longer movement time for objects at 45

(equivalent to 315

) in comparison to objects at 0

and 90

(equivalent to 0

and 270

) under non-vibrating conditions. Lin

et al. (2010) found lowest movement time at 0

, followed by 90

and 45

and then 135

(CCW). But the ANOVA results showed

insignicant effect between them. However, the present study
found objects at 270

took longer movement time to complete the

task than at 315

(p < 0.001). Further, movements at 0

required
long movement time as compared to movements at 315

, but the
results of statistical analysis revealed that it is insignicant
(p > 0.05).
The study exhaustively analyzed the effect of angles of approach
on movement time and perceived a broad agreement with ndings
of MacKenzie and Buxton (1992). The results also compared well
with those reported by Whisenand and Emurian (1996, 1999).
6.3. Effects of object distance
As distance to an object is a component in the equation for the
index of difculty, it is included in studies of Fitts law applications,
especially to model mouse movements. Intuitively, objects far
from a starting point should take longer time to select objects than
near the starting point. In general, it was found that movement
time increased with the object distance. The present results are
consistent with those of other research studies (Card et al., 1978;
Gillan et al., 1992; Grossman and Balakrishnan, 2005; Manseld
et al., 2007; Narayanamoorthy, 2007; Whisenand and Emurian,
1996, 1999; Yau et al., 2007; Lin et al., 2010). The above results
thus validate the ndings of other similar studies.
6.4. Effects of object size
The size of the object is another important variable used for
computing the index of difculty by Fitts law. Logically, smaller
objects require more precise movements and should, therefore,
take longer time to select than larger objects. The ndings of the
present study validate this logic and further extend its implications
to vibration environment also. Moreover, they are also consistent
with the results of many previous studies (Grossman and
Balakrishnan, 2005; Manseld et al., 2007; Narayanamoorthy,
2007; Whisenand and Emurian, 1996, 1999; Yau et al., 2007; Lin
et al., 2010).
7. Limitations of the study
The simultaneous three axial motions (multi-axis) were ob-
tained fromthree different exciters placed in three principal planes.
Their motions are not synchronized, therefore during simultaneous
excitation (multi-axis) there is some phase difference existing
between different directions. Few studies (Grifn and Whitham,
1977; Shoenberger, 1987) concluded that the phase had little
effect on subjects judgments. However, Shoenberger (1988) re-
ported some evidence of a complex effect of phase when combining
X- and Z- axis vibrations in the range 3e8 Hz. But in the present
study the effect of phase difference has not been considered.
8. Conclusions
The present study investigated the effects of vibration direction,
magnitude, size and distance of object as well as all approach
angles throughout 360

in step of 45

and presented comprehen-

sive and consistent effects on movement time while performing
pointing and clicking activity on laptop monitor. Mouse movement
time was signicantly affected by vibration direction, magnitude,
object size, distance and angle of approach to the object. Interaction
effects of some parameters are also signicant for movement time.
The ndings of this study indicate that the effects of angle of
approach are sensitive to categorical groupings (foreeaft or lateral
approach angles and upward or downward cursor movements).
Table 7
Slope (M), intercept (C) and coefcient of correlation (R
2
) values for different
magnitudes.
Regression values Control Low Medium High
Slope (M) 0.156 0.198 0.212 0.237
Intercept (C) 0.373 0.394 0.413 0.414
R
2
0.965 0.978 0.966 0.981
Table 8
Slope (M), intercept (C) and coefcient of correlation (R
2
) values for different
magnitudes under different directions.
Vibration direction Regression values Vibration magnitude
Control Low Medium High
X-axis Slope (M) 0.156 0.18 0.21 0.192
Intercept (C) 0.373 0.388 0.388 0.418
R
2
0.965 0.964 0.973 0.942
Y-axis Slope (M) 0.156 0.207 0.235 0.26
Intercept (C) 0.373 0.406 0.361 0.393
R
2
0.965 0.971 0.985 0.915
Z-axis Slope (M) 0.156 0.198 0.225 0.238
Intercept (C) 0.373 0.394 0.364 0.448
R
2
0.965 0.978 0.919 0.978
Multi-axis Slope (M) 0.156 0.217 0.243 0.279
Intercept (C) 0.373 0.389 0.403 0.375
R
2
0.965 0.945 0.94 0.974
Table 9
Mean movement time for different angles of approach.
Angles (degrees) Mean movement time (sec) Standard deviation
0 0.649 0.217
45 0.788 0.289
90 0.778 0.272
135 0.859 0.279
180 0.824 0.270
225 0.862 0.292
270 0.914 0.324
315 0.630 0.206
Table 10
Mean mouse movement time for different object distances.
Distance (pixels) Mean movement time (sec) Standard deviation
62 0.639 0.212
124 0.783 0.281
186 0.841 0.275
248 0.888 0.309
Table 11
Mean mouse movement time for different object sizes.
Size (pixels) Mean movement time (sec) Standard deviation
25 0.999 0.333
50 0.814 0.270
75 0.709 0.212
100 0.663 0.211
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 523
Further, the ndings fromthe study could be helpful to improve the
design of GUI systems especially for laptop based control systems
by organizing the frequently used elements such as menus, buttons
closer to neutral cursor position to increase performance. As
vibration directions were considered in this study, it is one of very
few studies that considered vibrations in the level of moving train
these data are valuable to researchers and practitioners interested
in applying a Fitts law model to target acquisition activities in
a two-dimensional plane, such as computers display monitor.
Acknowledgment
The research reported in this paper has been supported by the
grants from the Department of Science and Technology, Govern-
ment of India. The authors are grateful to Dr. Neil Manseld,
Loughborough University, Loughborough, UK and his colleague Dr.
Andrew Rimell, and Mr. Sham Rane, IIT Delhi, New Delhi, India for
their contributions in developing the program used in this study.
The authors also thank all the participants of this study for their
contributions.
References
Agren, J., Westberg, J., 2003. Interference of lateral vibrations on train passenger
activities. Vehicle System Dynamics 37, 409e419.
Arora, Y., 2006. Effect of Whole Body Vibrations on Mouse/Touch Pad Used in Trains.
Report Internship Project Report under EC e Asia Link Program, ASIE/2005/
111000. Loughborough University, UK.
Balakrishnan, R., 2005. A probabilistic approach to modeling two-dimensional
pointing. ACM Transactions on Computer-Human Interaction (TOCHI) 12 (3),
435e459.
Boritz, J., Booth, K.S., Cowan, W.B., 1991. Fitts law studies of directional mouse
movement. In: Proceedings of the Graphics Interface 91, Toronto, Canada, pp.
216e223.
Card, S.K., English, W.K., Burr, B.J., 1978. Evaluation of mouse, rate-controlled
isometric joystick, step keys, and text keys for text selection on a CRT. Ergo-
nomics 21, 601e613.
Fitts, P.M., 1954. The information capacity of the human motor system in controlling
the amplitude of movement. Journal of Experimental Psychology 47, 381e391.
Gillan, D.J., Holden, K., Adam, S., Rudisill, M., Magee, L., 1992. How should Fitts law
be applied to humanecomputer interaction? Interacting with Computers 4,
291e313.
Grifn, M.J., 1990. Handbook of Human Vibration. Academic Press, London.
Grifn, M.J., Whitham, E.M., 1977. Assessing the discomfort of dual-axis whole-body
vibration. Journal of Sound and Vibration 54, 107e116.
Grossman, T., Balakrishnan, R., 2005. A probabilistic approach to modeling two-
dimensional pointing. ACM Transactions on Computer-Human Interaction
(TOCHI) 12 (3), 435e459.
Holmund, P., 1998. Absorbed Power and Mechanical Impedance of the Seated
Human Exposed to Whole-Body Vibration in Horizontal and Vertical Directions.
Doctoral dissertation. Umea University, Umea, Sweden.
ISO 2631-1, 1997. Mechanical Vibration and Shockdevaluation of Human Exposure
to Whole Body Vibration e Part 1: General Requirements. International Orga-
nization for Standardization, Geneva, Switzerland.
Johnston, M.E., and Wharf, J.H., 1979. The Effect of 3e25 Hz Vibration on the
Legibility of Numeric Light Emitting Diode Displays. Technical Memorandum
Space 275, FS 296, Royal Aircraft Establishment, Farnborough, UK.
Khan, M.S., Sundstrm, J., 2004. Vibration comfort in Swedish inter-city trains e
a survey on passenger posture and activities. In: Proceedings of 18th Interna-
tional Congress on Acoustics, Kyoto, Japan.
Krishnakant, P.V., 2007. Evaluation of Ride and Activity Comfort for the Passengers
While Travelling by Rail Vehicles. M.Tech thesis. IIT Roorkee, India.
Lewis, C.H., Grifn, M.J., 1978. A review of the effects of vibration on visual acuity
and continuous manual control. II: continuous manual control. Journal of Sound
and Vibration 56 (3), 415e457.
Lewis, C.H., Grifn, M.J., 1980. Predicting the effects of vibration frequency and axis,
and seating conditions on the reading of numeric displays. Ergonomics 23,
485e501.
Lin, C.J., Liu, C.N., Chao, C.J., Chen, H.J., 2010. The performance of computer input
devices in a vibration environment. Ergonomics 53 (4), 478e490.
Liu, C.N., Lin, C.J., Chao, C.J., 2007. Fitts law model and target size of pointing devices
in a vibration environment. Perceptual and Motor Skills 105, 959e967.
Luria, S.M., Neri, D.F., Schlichting, S., 1989. Performance and preference with various
VDT phosphors. Applied Ergonomics 20, 33e38.
MacKenzie, I.S., 1995. Movement time prediction in humanecomputer interfaces. In:
Baecker, R.M., Buxton, W.A.S., Grudin, J., Greenberg, S. (Eds.), Readings in Humane
Computer Interaction, second ed. Kaufmann, Los Altos, CA, pp. 483e493.
MacKenzie, I.S., Buxton, W., 1992. Extending Fitts law to two-dimensional tasks. In:
Proceedings of the CHI 92 Conference on Human Factors in Computing
Systems. ACM, New York, USA, pp. 219e226.
MacKenzie, I.S., Sellen, A., Buxton, W., 1991. A comparison of input devices in
elemental pointing and dragging tasks. In: Proceedings of the CHI 91 Confer-
ence on Human Factors in Computing Systems. ACM, New York, USA, pp. 161e
166.
Mallick, Z., 2007. Investigating data entry task performance o a laptop under the
impact of vibration: the effect of color. International Journal of Occupational
Safety and Ergonomics 13 (3), 291e303.
Manseld, N.J., 2005. Human Response to Vibration. CRC Press, Boca Raton, Florida,
USA, ISBN 0-415-28239-X.
Manseld, N.J., Arora, Y, Rimell, A.N., 2007. Computer use on moving trains: which
pointing device? In: Contemporary Ergonomics e 2007, Proceedings of Inter-
national conference on Contemporary Ergonomics, 17e19 April 2007. Taylor &
Francis LLC, Nottingham, UK, ISBN: 9780415436380.
Meddick, R.D.L., 1977. The Legibility of Two-Axis Vibrated Test Material When
Presented to Stationary Seated Subjects. M.Phil thesis. University of South-
ampton, UK.
Nakagawa, C., Suzuki, H., 2005. Effects of Train Vibrations on Passenger PC Use. In:
Quarterly Report of RTRI, vol. 46 (3). Railway Technical Research Institute,
Japan.
Narayanamoorthy, R., 2007. Modelling the Fitts Law in Terms of Vibration Magni-
tude for Computer Mouse Use. Report under European Commission e Asia Link
Program (CIRCUS), August 2007, IIT Roorkee, India.
Narayanamoorthy, R., Saran, V.H., Goel, V.K., 2008a. Sedentary activities and
comfort studies in Indian passenger trains. In: Proceedings of International
Ergonomics Conference (HWWE e 2008), 22e24 December 2008, VIT, Pune,
India.
Narayanamoorthy, R., Saran, V.H., Goel, V.K., Harsha, S.P., Khan, S., Berg, M., 2008b.
Determination of activity comfort in Swedish passenger trains. In: Proceedings
of 8th World Congress on Railway Research (WCRR 2008), 18e22 May 2008,
COEX, Seoul, Korea.
Narayanamoorthy, R., Goel, V.K., Saran, V.H., 2009a. Design of a simulator to study
human comfort under vibration, noise and temperature and humidity. In:
International Conference on Emerging Research and Advances in Mechanical
Engineering (ERA e 2009), 19e21 March 2009, VEC, Chennai, ISBN: 13 978-81-
908086-1-3.
Narayanamoorthy, R., Goel, V.K., Saran, V.H., 2009b. Vibration measurement e a tool
for evaluating activity comfort in trains. Journal of the Instrument Society of
India 39 (3), 189e193.
Narayanamoorthy, R., Huzur Saran, V., 2011. Error analysis of task performance with
laptop in vibration environment. Computers in Human Behavior 27 (6), 2263e
2270.
Shoenberger, R.W., 1987. Intensity judgments of vibrations in the Y-axis, Z-axis, and
Y-plus-Z axes. Aviation, Space, and Environmental Medicine 58 (8), 783e787.
Shoenberger, R.W., 1988. Intensity judgments of vibrations in the X axis, Z axis, and
X-plus-Z axes. Aviation, Space, and Environmental Medicine 59 (8), 749e753.
Stott, J.R.R., 1984. The vertical vestibulo-ocular reex and ocular resonance. Vision
Research 24, 949e960.
Sundstrm, J., 2006. Difculties to Read and Write Under Lateral Vibration Expo-
sure: Contextual Studies of Train Passengers Ride Comfort. KTH Doctoral thesis.
Stockholm, Sweden.
Sundstrm, J., Khan, S., 2008. Inuence of stationary lateral vibrations on train
passengers ability to read and write. Applied Ergonomics 39 (6), 710e718.
Suzuki, H., 1998. Research trends on riding comfort evaluation in Japan. Proceedings
of the Institution of Mechanical Engineers Part F e Journal of Rail and Rapid
Transit 212 (1), 61e72.
Westberg, J., 2000. Interference Lateral Vibration on Train Passenger Activities: an
Experiment on Human Ability to Perform Reading, Writing and Drinking.
Master thesis TRITA-FKT Report 2000:62. KTH, Sweden.
Whisenand, T., Emurian, H., 1996. Effects of angle of approach on cursor movement
with a mouse: consideration of Fitts law. Computers in Human Behavior 129
(3), 481e495.
Whisenand, T.G., Emurian, H.H., 1995. Some effects of angle of approach on icon
selection. In: Conference Companion on Human Factors in Computing Systems,
07e11 May 1995, Denver, Colorado, USA, pp. 298e299.
Whisenand, T.G., Emurian, H.H., 1999. Analysis of cursor movements with a mouse.
Computers in Human Behavior 15, 85e103.
Wollstrom, M., 2000. Effects of Vibrations on Passenger Activities e Writing and
Reading, a Literature Study. TRITA-FKT Report 2000:64. Railway Technology,
KTH, Stockholm, Sweden.
Woods, V., Hastings, S., Buckle, P., Haslam, R., 2002. Ergonomics of Using a Mouse or
Other Non-Keyboard Input Device. HSE Research Report 045. Health and Safety
Executive, London, UK.
Yau, Y.J., Chao, C.J., Hwang, S.L., Lin, J.T., 2007. A study of control performance in low
frequency motion workstation. In: Proceedings of the 4th International
Conference on Universal Access in HumaneComputer Interaction: Ambient
Interaction, 22e27 July 2007, Beijing, China.
R. Narayanamoorthy, V. Huzur Saran / International Journal of Industrial Ergonomics 42 (2012) 513e524 524