Available via license: CC BY 4.0
Content may be subject to copyright.
Citation: Kim, G.; Okatomo, S.;
Maruyama, H. Weight Illusion
Caused by Sinusoidal Vibration
Correlates with Grip Force
Adjustment. Appl. Sci. 2023,13, 2717.
https://doi.org/10.3390/
app13042717
Academic Editors: Zhihan Lv, Kai Xu
and Zhigeng Pan
Received: 29 January 2023
Revised: 11 February 2023
Accepted: 16 February 2023
Published: 20 February 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
applied
sciences
Article
Weight Illusion Caused by Sinusoidal Vibration Correlates
with Grip Force Adjustment
Giryeon Kim 1,2,*, Shogo Okamoto 1,2 and Hisataka Maruyama 3
1Department of Mechanical Systems Engineering, Nagoya University, Nagoya 464-8601, Japan
2Department of Computer Science, Tokyo Metropolitan University, Tokyo 191-0065, Japan
3Department of Micro-Nano Systems Engineering, Nagoya University, Nagoya 464-8601, Japan
*Correspondence: kim.giryeon.y2@s.mail.nagoya-u.ac.jp
Abstract:
Our research team previously identified a weight illusion in which a lifted object feels heavy
when it continuously presents a sinusoidal vibration to the fingertips. However, the mechanism
underlying this illusion remains unknown. We thus hypothesized that the autonomous grip force
adjustment against a vibrating object would be one of the factors underlying the weight illusion.
The autonomous grip force adjustment increases the motor outputs of a human hand system, subse-
quently raising the sense of effort to keep holding the lifted object. The grip forces and perceived
heaviness were evaluated using vibratory stimuli with five different frequencies (30 Hz, 60 Hz, 100 Hz,
200 Hz, and 300 Hz) and three different amplitudes (156
µ
m, 177
µ
m, and 203
µ
m). The results
showed that the stimuli at lower frequencies or large amplitudes increased the grip forces more and
felt heavier than the stimuli at higher frequencies or small amplitudes. Specifically, the 30 Hz stimuli
felt the heaviest and increased the grip force the most. An increase in the grip force was positively
correlated with the perceived heaviness. These results indicate that vibratory stimuli influence both
the grip force and weight perception. Our findings can contribute to developing haptic displays to
present virtual heaviness.
Keywords:
weight illusion; vibration; grasp force; grip force adjustment; heaviness; mechanoreceptor
1. Introduction
The weight of an object is defined as the product of its mass and gravitational ac-
celeration. However, humans do not always perceive weight according to its physical
definition. Instead, weight is often estimated based on generated motor commands and
sensations during motions [
1
–
4
] and various types of physical quantities, such as density
and inertia [
5
–
7
]. By controlling specific factors that affect weight perception, an object can
feel heavier or lighter than its actual weight. These phenomena are referred to as weight
illusions. Investigating weight illusions leads to an understanding of weight perception
mechanisms.
The size-weight illusion is a representative example of weight illusions, in which an
object with a smaller volume is perceived as heavier than an object with a larger volume,
despite weighing the same [
8
]. The underlying principle behind this illusion however is
still under debate. Nevertheless, a gap between the perceived weight of the lifted object and
its expected weight before lifting is considered to be part of the reason for the illusion [
1
,
9
].
Furthermore, the material, color, and brightness also cause weight illusions, in which an
object that is expected to be heavy from its appearance is perceived as being light [
10
–
12
].
As evident from the above examples, the appearance of an object is one of the factors
underlying the weight illusion.
The presentation of tactile stimuli to the hand also leads to the weight illusion. Deep
sensations primarily influence weight perception, but cutaneous sensations are also influ-
ential, especially for light objects [
13
–
16
]. When lifting an object, the skin deforms in the
Appl. Sci. 2023,13, 2717. https://doi.org/10.3390/app13042717 https://www.mdpi.com/journal/applsci
Appl. Sci. 2023,13, 2717 2 of 13
direction of the gravity or force acting on the held object. Therefore, an additional skin
deformation in the direction of gravity makes objects feel heavier [15,17,18].
Some researchers have devised methods to increase the perceived weight using vibra-
tory stimuli. When asymmetric vibration is presented to the hands, an additional force
is felt in the direction of greater acceleration [
19
–
22
]. Therefore, asymmetric vibration in
the direction of gravity makes objects feel heavier [
20
]. Furthermore, vibratory stimuli
coinciding with the acceleration of moving hands can increase the perceived weight of an
object [23,24].
Our research team previously identified a weight illusion caused by a continuous
sinusoidal vibration to the fingertips [
25
]. When a vibrating object is lifted, it feels heavier
than a still object with the same mass. In our previous study, 13 out of 15 participants
reported this illusory weight perception [
25
]. We call this phenomenon the “vibration-
weight illusion”. Continuous vibrations over a wide range of frequencies from 30 Hz to
300 Hz generated the vibration-weight illusion [
25
]. As mentioned above, earlier studies
used asymmetric vibration [
19
,
20
] or vibration coinciding with moving hands [
23
,
24
] to
present the weight. In contrast, the vibration-weight illusion is caused by symmetric and
continuous vibrations. The characteristics of the stimuli in the vibration-weight illusion
differ from those of other weight presentation methods reported by earlier researchers.
Therefore, investigating the mechanisms underlying this illusion may lead to new insights
into the perceptual effects of vibratory stimuli and the mechanisms of weight perception.
We hypothesized that an increase in the grip force will be the main factor underlying
the vibration-weight illusion. Humans feel their own motions and forces intensely when the
related motor commands are large [26–29]. In the case of the vibration-weight illusion, an
additional grip force may be used to stabilize the vibrating object, and the additional force
may make the object feel heavier than its actual weight. Therefore, this study investigated
the relationship between the grip force adjustment and vibration-weight illusion. Especially,
this study confirmed how the grip force and perceived heaviness are changed by sinusoidal
vibrations of different frequencies (Experiment 1) or amplitudes (Experiment 2).
2. Related Studies and Hypothesis of the Vibration-Weight Illusion
An asymmetric vibration is an example of force presentation by vibratory stimuli [
19
,
20
].
It comprises a faster motion of mass in one direction and a slower motion in the other, and
the illusory force felt in the former direction. When an object held in the hand vibrates
asymmetrically, it generates higher forces in the direction of a greater vibratory amplitude
than in the opposite direction [
21
]. Moreover, the skin deforms more in that direction [
22
].
The nonlinearity between the physical stimuli, that is, the force and skin deformation, and
their subjective magnitudes is a potential cause of the illusory force perception. An asym-
metrically vibrating object feels heavier when the vibration is in the direction of gravity [
20
].
However, the vibration-weight illusion discussed in the present study uses a sinusoidal
vibration, which is symmetric. Moreover, the vibration-weight illusion occurs even when
the vibration is presented in a direction perpendicular to gravity.
Okamoto et al. and Nagano et al. proposed a method to increase the perceived inertia
or mass by controlling the presentation timing of vibratory stimuli [
23
,
24
]. The inertial force
generated by swaying an object causes finger skin deformations that activate mechanore-
ceptive units. They intended to additionally stimulate the mechanoreceptive units of the
fingertips by vibration and cause an illusory sense of a magnified skin deformation. They
thus made a swayed object feel heavier by presenting vibrations [
23
,
24
]. However, the
vibration-weight illusion discussed in the present study is generated by a continuous
vibration while the object is statically grasped. It thus differs from their findings.
The tonic vibration reflex is a muscle contraction evoked by the vibration applied to the
muscle below the skin that stimulates muscle spindles, leading to a muscle contraction re-
flex [
30
,
31
]. The stimulation of agonist muscles makes an object feel lighter, and conversely,
the stimulation of antagonist muscles makes the object feel heavier [
26
]. There are no re-
ports of a tonic vibration at the fingers because fingers lack muscles; however, in the case of
Appl. Sci. 2023,13, 2717 3 of 13
the forearm, a vibration with an amplitude of 0.2–0.3 mm at a frequency of 100–200 Hz was
found to effectively generate a tonic vibration reflex [
32
]. If the tonic vibration reflex causes
the vibration-weight illusion, the vibration delivered to the fingertips should be transferred
to the antagonist muscles of the wrist or forearm. However, vibrations with amplitudes
of less than 0.1 mm to the fingertips generated the vibration-weight illusion [
25
]. This
stimulus is small when compared to that of a previous study [
32
]. Therefore, we speculated
that the tonic vibration reflex does not cause the vibration-weight illusion.
We hypothesized that the vibration applied to an object will cause the grip force to
increase as a result of the automatic grip force adjustment for stable grasping. Motions that
generate more motor commands are perceived as higher intensity [
26
–
29
]. When the motor
commands related to motions are larger, the object’s weight is perceived as heavier [
1
–
4
].
Therefore, even if the weight of the object does not change, the increase in the grip force
may make the grasped object feel heavier.
3. Experiment 1: Frequency Dependence of Vibration-Weight Illusion
Two different tasks were conducted in Experiment 1. The order of the two tasks was
randomized for individual participants.
3.1. Participants
A total of 11 participants (10 males and 1 female) participated in the experiment.
All participants were right-handed, healthy university students, aged 20 years and above.
3.2. Apparatus and Stimuli
A voice coil motor (Vp408, Acouve Laboratory, Inc., Tokyo, Japan, 85 g) shown in
Figure 1A was used to present vibratory stimuli. A load cell (FS2050-0000-1500-G, TE
Connectivity, Schaffhausen, Switzerland, 5 g) was attached to the center of the voice coil
motor to measure the grip force.
Figure 1.
Apparatus. (
A
) Appearance of the apparatus. Load cell is attached to the center of the voice
coil motor. (B) Direction of vibratory stimuli and how to grip it during the experiments.
Five sinusoidal vibratory stimuli with different frequencies and accelerations were
prepared, as shown in Table 1. The accelerations of the vibratory stimuli were set to 3 dB
higher than the thresholds of the vibration-weight illusion reported in a previous study [
25
],
in which the minimum intensities of vibratory accelerations needed to cause the illusion
were determined using the psychophysical method of limits. Acceleration was measured
using an accelerometer (Model-2320B, Showasokki Co., Ltd., Japan) through an amplifier
(Model-4035-50, Showasokki Co., Ltd., Japan). The accelerometer was attached to the
center of the vibrator’s surface, where the participant’s thumbs contacted. Amplitudes
were calculated from their frequencies and accelerations. During the experiment, the
accelerometer was neither used nor fixed on the voice coil motor.
Here, we describe how the aforementioned stimuli were designed. This study aimed
to investigate the relationship between the vibration-weight illusion and grip force adjust-
ment. The types of mechanoreceptive units that mediate grip force adjustment have been
discussed [
33
–
35
]. Different types of mechanoreceptive units exhibit different frequency
responses [
36
]. Hence, we covered the frequency range of 30–300 Hz. The minimum ampli-
tude (i.e., threshold) to evoke the illusion depends on the frequency [25]. Considering the
Appl. Sci. 2023,13, 2717 4 of 13
properties of the actuator, it would be impossible to maintain equal amplitudes and induce
the illusion over such a wide frequency range. Therefore, we determined the amplitudes
of the vibrations based on the threshold for each frequency. Specifically, the amplitudes
were 3 dB above the threshold so that the illusory weight was experienced by almost
all participants.
Table 1.
Acceleration, amplitude, and input voltage at each frequency of the vibratory stimuli used
in Experiment 1. Accelerations and amplitudes of the stimuli were 3 dB higher than the thresholds at
which the vibration-weight illusions were observed at individual frequencies. Voltage indicates the
amplitude of the voltage applied to the voice coil motor (half to peak-to-peak).
Frequency Acceleration Amplitude Voltage
30 Hz 5.5 m/s2156 µm 2.44 V
60 Hz 11.9 m/s284 µm 0.86 V
100 Hz 17.4 m/s244 µm 1.06 V
200 Hz 75.3 m/s248 µm 4.06 V
300 Hz 138.1 m/s239 µm 7.82 V
3.3. Task 1-1: Grip Force at Different Frequencies
3.3.1. Procedures
Participants were exposed to five different vibratory stimuli, as listed in Table 1. They
gripped the voice coil motor with their dominant hands, as shown in Figure 1B, and placed
their thumbs on the load cell while the other four fingers were on the opposite side. They
were instructed to lift the voice coil motor using their entire arm but were not allowed to
place their arms or wrists on the table. They were asked to grip the voice coil motor with a
minimum grip force without letting it slip.
Participants lifted the voice coil motor that was not yet vibrating and continued
holding it. Vibration stimuli were then initiated for 3 s after a random waiting time.
The participants were unaware of the presentation timing. They then laid down the voice
coil motor and lifted 100 g and 200 g weights in randomized orders to neutralize their
sense of weight. The direction of the vibration was perpendicular to gravity. To prevent the
influence of the sound made by the motor, pink noise was played through headphones.
The grip force of each participant was measured five times for each of the five different
vibratory stimuli. In total, 25 (five frequencies
×
five repetitions) trials were tested in a
randomized order for each participant. Between the trials, the participants were asked if
they were fatigued and provided breaks, if necessary.
3.3.2. Analysis
The outputs from the load cell were submitted to a moving-average low-pass filter with
a window size of 0.1 s. The grip forces before and after the presentation of the stimuli were
compared to evaluate the increase in the grip force for each vibratory stimulus. The grip
force before the presentation of the vibration was the mean of the 0.5 s immediately before
the presentation of a stimulus (
−
0.5–0 s), as in Figure 2, where 0 s was the onset of the
vibration. The grip force after the presentation of the stimulus was the mean during the
0.5 s, 1 s after the start of the presentation (1–1.5 s). The grip forces were transient for 0–1 s,
and this period was not used for the analysis. The means and standard errors of the change
in the grip force for each vibratory frequency were calculated, and two-tailed
t
-tests were
conducted to confirm whether the stimuli changed the grip forces significantly.
Appl. Sci. 2023,13, 2717 5 of 13
Figure 2.
Grip force over time during Experiment 1. The red lines and dashed black lines indicate
the means and range of standard errors, respectively. Stimuli started at 0 s and lasted 3 s. For the
analysis of the change in grip force, means of 0.5 s before the stimulus (
−
0.5–0 s) and after the
stimulus (1–1.5 s), which are surrounded by the black and green dashed lines, respectively, were used.
3.4. Task 1-2: Perceived Heaviness at Different Frequencies
3.4.1. Task
In addition to the five types of vibratory stimuli used in the previous task (Table 1), a
condition of no vibratory stimulus was included in Task 1-2. Hence, a total of six stimuli
(30 Hz, 60 Hz, 100 Hz, 200 Hz, 300 Hz, and no vibration) were presented in this task.
The participants lifted the voice coil motor with their dominant hands in the same posture
as in Task 1-1. During the task, participants continuously held the voice coil motor without
releasing it. Similar to Task 1-1, each type of vibration lasted only 3 s to prevent perceptual
adaptation. They could freely compare the six types of stimulus conditions and rank them
in the order of perceived heaviness without tie ranks. They were allowed to experience
the vibratory stimuli repeatedly until they made their decisions. The frequencies of the
stimuli were not disclosed to them. Each participant performed the above tasks twice, with
an interval of several minutes.
3.4.2. Analysis
For each participant, the mean ranks of the two ranking tasks were calculated. These
mean ranks included ties. Ranks among the vibratory stimuli, including the control stimu-
lus with no vibration, were compared using the Wilcoxon signed-rank test with a Bonferroni
correction, with the correction factor being 15 (
6
C
2
). Furthermore, the relationship between
the change in the grip force (Task 1-1) and perceived heaviness (Task 1-2) was examined
using Spearman’s rank correlation.
3.5. Results
3.5.1. Grip Force at Different Frequencies (Task 1-1)
Figure 2indicates the means and standard errors of the time-series grip force among
all the participants. For all types of vibratory stimuli, the grip force began to increase within
0.1 s after the onset of the vibration. After a rapid and large increase in the grip force, the
excessive grip force was adjusted to potentially minimum levels to hold the activated voice
coil motor during 0–0.5 s. Similar changes in the grip force have also been observed in
other studies investigating its adjustment [34,37,38].
Appl. Sci. 2023,13, 2717 6 of 13
The means and standard errors of the increase in the grip force for each vibratory
stimulus are shown in Figure 3A. The lower the frequency, the greater the increase in the
grip force. There was a negative correlation between the frequency and increase in the
grip force (
ρ=−0.30
,
p
= 0.027). The presentation of the 30 Hz (
t(
10
) =
4.03,
p=
0.002,
two-tailed
t
-test) and 100 Hz (
t(
10
) =
2.31,
p=
0.043) vibrations significantly increased
the grip force. For the 60 Hz vibratory stimuli, the grip force tended to increase but not
significantly (
t(
10
) =
2.21,
p=
0.051). The mean grip force before the presentation of the
stimuli was 58.7 g, and the 30–100 Hz vibratory stimulus increased the grip force by 9.8 g
(17%) on average.
Figure 3.
Results of Experiment 1. Stimuli with five different frequencies were used and their
amplitudes were 3 dB higher than the thresholds for the weight illusion at each frequency. (
A
) Increase
in grip force for five vibratory frequencies. (
B
) Ranks of perceived heaviness. A higher rank number
indicates a perception of greater heaviness. * and ** indicate significant differences at the 5% and 1%
levels, respectively. There was a positive correlation between the increase in grip force and perceived
heaviness (ρ= 0.66, p= 1.24 ×10−9).
3.5.2. Perceived Heaviness at Different Frequencies (Task 1-2)
The ranks of the perceived heaviness for different vibration conditions are shown in
Figure 3B. A higher rank indicates that the participants perceived the stimulus as heavier.
The lower the frequency of the vibratory stimulus, the heavier its perception. For instance,
the participants perceived the 30 Hz vibration as the heaviest. As shown in Table 2, the
vibratory stimuli except for 300 Hz were judged as being significantly heavier than the
condition with no vibratory stimulus.
Table 2.
Comparison of heaviness ranks.
p
-values and (
t
-values) between vibration conditions in
Task 1-2 (Wilcoxon signed-rank test with Bonferroni correction).
p-Value (t-Value) 60 Hz 100 Hz 200 Hz 300 Hz None
30 Hz 0.582 (161.0) 0.020 (176.0) 0.086 (171.5) 0.004 (181.5) 0.001 (186.0)
60 Hz - 0.069 (172.5) 0.134 (171.5) 0.006 (182.5) 0.002 (186.5)
100 Hz - - 0.616 (161.0) 0.021 (176.5) 0.001 (187.0)
200 Hz - - - 0.082 (172.5) 0.005 (183.0)
300 Hz - - - - 0.057 (172.0)
3.5.3. Correlation between Grip Force and Perceived Heaviness at Different Frequencies
The changes in the grip force (Figure 3A) and perceived heaviness (Figure 3B) exhibited
a positive rank correlation coefficient of 0.66 (
p
= 1.24
×
10
−9
). This correlation indicates
that the stimulus conditions with the larger increase in the grip force were perceived to
be heavier.
To investigate the influence of the time intervals to the average grip force on the results,
we calculated a rank correlation coefficient with different time interval conditions, namely
Appl. Sci. 2023,13, 2717 7 of 13
when the time intervals before and after the stimulus onset were
−
1 s to 0 s and 1 s to 2 s.
The correlation coefficient was 0.65 (
p
= 3.73
×
10
−9
), which is close to the result (
ρ
= 0.66)
with intervals of −0.5 s to 0 s and 1 s to 1.5 s.
4. Experiment 2: Amplitude Dependence of Vibration-Weight Illusion
4.1. Participants
A total of 11 participants (6 males and 5 females) participated in the experiment.
All participants were university students, aged over 20 years. Two participants also
participated in Experiment 1, but Experiments 1 and 2 were conducted on different days.
4.2. Apparatus
The same apparatus as in Experiment 1 (Figure 1) was used in Experiment 2. The main
components included a voice coil motor (Vp408, Acouve Laboratory, Inc., Japan, 85 g) and
a load cell (FS2050-0000-1500-G, TE Connectivity, Switzerland, 5 g) to present the vibration
stimuli and measure the grip force, respectively.
4.3. Stimuli
Three stimuli with the same frequency but different amplitudes as listed in Table 3
were used. The frequencies of the stimuli were set to 30 Hz, which increased the grip force
and perceived heaviness the most in Experiment 1. The stimulus with the lowest amplitude
among them was identical to the 30 Hz stimulus used in Experiment 1.
Table 3.
Accelerations, amplitudes, and input voltages of the vibratory stimuli used in Experiment 2.
Amplitude Acceleration Frequency Voltage
156 µm 5.5 m/s230 Hz 2.44 V
177 µm 6.4 m/s230 Hz 2.85 V
203 µm 7.2 m/s230 Hz 3.25 V
4.4. Task 2: Grip Force and Perceived Heaviness at Different Amplitudes
In Experiment 2, the changes in grip force and perceived heaviness were investigated,
similar to Experiment 1. Unlike Experiment 1, the number of stimuli was small, which
allowed us to simultaneously investigate the rank of perceived heaviness and grip forces.
Participants gripped and lifted the voice coil motor in the same manner as in Experi-
ment 1 (Figure 1B). After a random waiting time, a vibration stimulus, which was randomly
selected among the three stimuli, was presented for 3 s, and the grip force was recorded
continuously before and after the vibration period. They then laid down the voice coil
motor and lifted 100 g and 200 g weights in randomized orders to neutralize their sense
of weight. Each stimulus was presented only once, and after all the stimuli had been
presented, participants ranked the perceived heaviness of the stimuli. The above task took
approximately 2 min for each participant.
4.5. Analysis
The data were analyzed in the same manner as in Experiment 1. The grip forces
measured by the load cell were submitted to a moving-average low-pass filter with a
window size of 0.1 s to remove the effect of the vibration stimulus. The means of the grip
forces before and after the vibration were calculated from the grip forces during the 0.5 s
before the vibration and during 0.5 s from 1 s after the commencement of the vibration,
respectively. For each stimulus, the mean and standard error of the change in the grip force
among the participants were calculated, and two-tailed t-tests were conducted to confirm
whether the changes were significantly different from zero. The ranks of the perceived
heaviness were compared using the Wilcoxon signed-rank test with a Bonferroni correction,
with the correction factor being 3 (
3
C
2
). In the last step, the relationship between the change
in the grip force and perceived heaviness was examined using Spearman’s rank correlation.
Appl. Sci. 2023,13, 2717 8 of 13
4.6. Results
4.6.1. Grip Force at Different Amplitudes
Figure 4shows the means and standard errors of the time-series grip force among
all the participants for Experiment 2. The means and standard errors of the increases
in the grip force for the three stimuli with different amplitudes are shown in Figure 5A.
The grip force increased more for the stimuli with higher amplitudes. The grip force
increased significantly for all three stimuli (156
µ
m:
t(
10
) =
5.32,
p=
3.38
×
10
−4
, 177
µ
m:
t(
10
) =
9.15,
p=
3.56
×
10
−6
, 203
µ
m:
t(
10
) =
16.50,
p=
1.39
×
10
−8
, two-tailed
t
-test).
The average grip force before the stimuli was 58.8 g, and even the stimulus with the smallest
increase in the grip force, that is, 156
µ
m, increased the grip force by 8.7 g (15%) on average.
Figure 4.
Grip force over time during Experiment 2. The red lines and dashed black lines indicate the
means and range of standard errors, respectively. Stimuli started at 0 s and lasted 3 s. For the analysis
of the change in grip force, means of 0.5 s before the stimulus (
−
0.5–0 s) and after the stimulus
(1–1.5 s), which are surrounded by the black and green dashed lines, respectively, were used.
Figure 5.
Results of Experiment 2. Three different stimuli with different amplitudes but the same
frequency (30 Hz) were used. (
A
) Increase in grip force for three different amplitudes. (
B
) Ranks of per-
ceived heaviness. A higher rank number indicates a perception of greater heaviness.
* and ** indicate
significant differences at the 5% and 1% levels, respectively. There was a positive correlation between
the increase in grip force and perceived heaviness (ρ= 0.77, p= 1.37 ×10−7).
4.6.2. Perceived Heaviness at Different Amplitudes
The ranking of the perceived heaviness against the three stimuli with different am-
plitudes is shown in Figure 3B. The stimuli with higher amplitudes were perceived as
heavier. The stimuli with the amplitude of 203
µ
m were perceived as significantly heavier
than the other stimuli (203
µ
m vs. 156
µ
m:
t(
10
) =
67.5,
p=
0.0018, 203
µ
m vs. 177
µ
m:
t(
10
) =
75.5,
p=
0.0145, Wilcoxon signed-rank test with Bonferroni correction), but there
was no significant difference between the amplitudes of 156 µm and 177 µm.
Appl. Sci. 2023,13, 2717 9 of 13
4.6.3. Correlation between Grip Force and Perceived Heaviness at Different Amplitudes
The increase in the grip force and perceived heaviness were positively correlated
(
ρ=
0.77,
p
= 1.37
×
10
−7
, Spearman’s rank correlation). The grip forces and perceived
heaviness increased with the amplitude of the vibration when the frequency was constant.
5. Discussion
The experimental results revealed a positive correlation between the increase in the grip
force and perceived heaviness at different vibratory frequencies in Experiment 1 (
ρ= 0.66
,
p
= 1.24
×
10
−9
) and at different amplitudes in Experiment 2 (
ρ=
0.77,
p= 1.37 ×10−7
).
Specifically, the vibratory stimuli felt to be heavier introduced significant increases in the
grip force. The results are consistent with our hypothesis, which posited that the grip
force adjustment is the main factor underlying the vibration-weight illusion. The grip
force adjustment might have functioned autonomously to retain the vibrating object. The
increase in the grip forces indicates an increase in the motor commands, also known as the
sense of effort, which in turn leads to an increase in the perceived weight [26–28].
In Experiment 1, though we did not fully equalize the vibratory amplitudes at different
frequencies, those at 100 Hz (44
µ
m), 200 Hz (48
µ
m), and 300 Hz (39
µ
m) were similar.
Among these three frequencies, the lowest frequency, that is, 100 Hz, led to the greatest
and most significant increase in the grip force, whereas the other two frequencies exhibited
smaller increases in the grip force. This suggests that the vibration-weight illusion depends
on the frequency such that low vibration frequencies are effective. In Experiment 2, the
greater the amplitude, the greater the increase in the grip force and perceived weight when
the frequency was 30 Hz. This result suggests that the weight illusion also depends on
the amplitude.
One might query whether the differences in the outputs of the load cell were caused
by the vibratory amplitudes and resultant finger pad restoration forces and not by the grip
force. However, this factor did not have a major impact on the results. In Experiment 1,
although the amplitudes of the 100, 200, and 300 Hz stimuli were nearly equal, 44, 48, and
39
µ
m, respectively, as listed in Table 1, only the 100 Hz vibration significantly increased
the grip force. Furthermore, the increases in the grip force were nearly identical for the
60 Hz and 100 Hz stimuli, despite their amplitudes differing by a factor of almost two.
Therefore, the amplitudes of the stimuli did not notably affect the outputs of the load cell.
Experiment 1 showed that the 30 Hz vibratory stimulus increased the grip force the
most. Additionally, the lower the frequency of the vibratory stimuli, the greater the increase
in the grip force. This may be due to the characteristics of the fast-adapting type 1 (FAI)
mechanoreceptive units that affect the grip force adjustment [
33
,
39
]. The FAI units play an
important role in the grip force adjustment by detecting fast and subtle skin deformations,
such as slippage, with a peak of activation between 30 Hz and 40 Hz [
36
,
40
], whereas
the mediation of other types of receptive units are not negated [
39
]. The presentation
of the 30 Hz local vibrations to the fingertips, which is expected to effectively stimulate
the FAI units, was reported to increase the grip force, even without actual slippage [
34
].
It is possible that the stimuli with lower frequencies in the present study also effectively
activated the FAI units. This might have resulted in an increase in the grip force, even when
there was no actual slippage between the fingertips and surfaces of the voice coil motor.
Here, one question arises, namely whether the vibratory stimuli at frequencies below
30 Hz effectively cause the weight illusion. To investigate this, as a post hoc experiment,
we compared the intensities of the illusion between the 15 Hz and 30 Hz stimuli. The same
voice coil motor as in Experiments 1 and 2 was used, and the accelerations of the stimuli
were 3 dB higher than the thresholds of the illusion (15 Hz: 1.39 m/s
2
, 30 Hz: 5.5 m/s
2
).
The threshold of the 15 Hz stimulus was determined by the psychophysical method of
the limits with five participants, following a previous experiment [
25
]. Ten out of twelve
participants (eight males and four females) reported that the 30 Hz stimulus was heavier
than the 15 Hz stimulus. The intensity of the illusion may peak at 30 Hz, at which frequency
FAI units are sensitive [36,40].
Appl. Sci. 2023,13, 2717 10 of 13
Nonetheless, it is possible that the vibratory stimuli caused actual slippage. During
the experiment, a participant accidentally dropped the voice coil motor from his hand at the
onset of the vibratory stimuli. Such an accidental case was observed only once throughout
the experiment. Further, we did not observe any other evident full slippages; however,
it is known that incipient slippage occurs in the finger pad’s contact area [
41
,
42
] and the
tactile system permits the detection of this partial slippage [
39
,
43
]. The vibration reduces
the sliding friction, and the effect of the vibration on the friction reduction is prominent
with a greater vibratory amplitude and at a greater frequency [
44
]. As shown in Table 1, the
vibrations at lower frequencies exhibited greater amplitudes in our setup. It is possible that
the vibration had reduced the friction between the finger pad and voice coil motor, and
the grip force adjustment might have occurred to readjust to the surface with a lowered
coefficient of friction. From this perspective, the vibration-weight illusion can be compared
with the material-weight illusion, wherein an object with a smooth surface feels heavier
than an object with a rough surface [
10
]. The material-weight illusion was explained as a
large grip force required to hold an object with a slippery smooth surface that magnified
the perceived weight of the object [45,46].
The following are some limitations of the present study. In Experiment 1, to produce
the weight illusion clearly, we presented vibration stimuli with accelerations of 3 dB above
the threshold [
25
], at which amplitude the illusion barely occurred at each vibration fre-
quency. These settings limit the generalizability of the results. Because the present study
investigated the effect of the amplitude only for 30 Hz stimuli, it is not possible to conclude
how the amplitude dependence of the weight illusion changes with other frequencies. In
Experiment 1, there were six different stimuli, so the participants had to experience the
stimuli multiple times and take time to make a decision. Therefore, the changes in the
grip force (Figure 3A) and perceived heaviness (Figure 3B) were measured in different
experiments, and the conditions of these two Experiments were not perfectly the same. This
difference in the experimental conditions might have affected the experimental results. In
Experiment 1, however, the grip force measurement was repeated several times and the par-
ticipants spent time ranking the perceived heaviness. Thus, the reliability of Experiment 1
is not particularly questionable. Further, in Experiments 1 and 2, the intensities of the
perceived heaviness were ranked, and the rank correlations were calculated to investigate
whether the frequency or amplitudes affect the perceived heaviness. Therefore, we cannot
discuss whether the effects of the frequency and amplitude on the illusion are linear or
not. Finally, this study suggests a relationship between the grip force adjustment and the
vibration-weight illusion. However, it does not negate other factors. It is also inconclusive
as to why the vibratory stimuli increase the grip force, that is, whether the grip force is
adjusted due to actual partial slippage.
Some points remain to be studied in the future. In the experiments, the participants
gripped the voice coil motor with their finger pads. However, the vibration-weight illusion
also occurs when other holding methods are adopted. The relationship between the
holding methods and the intensity of the illusion is one of the interesting topics to be
studied. Furthermore, interestingly, the temporal profiles of the grip forces differed among
the frequency conditions. Nonetheless, we cannot explain the reasons behind this difference
at this point.
6. Conclusions
This study is based on the vibration-weight illusion [
25
] wherein a vibrating object
feels heavier than a still object. The continuous and sinusoidal vibration adequately causes
the illusion, and understanding this perceptual phenomenon might help with the design
of haptic displays that virtually and effectively present the sense of heaviness. This study
investigated whether the grip force adjustment affects the vibration-weight illusion and
how the grip force is changed by a sinusoidal vibration. Five types of vibratory stimuli at
different frequencies and three types of stimuli at different amplitudes were presented, and
the changes in the grip force and perceived heaviness were measured. The results showed
Appl. Sci. 2023,13, 2717 11 of 13
positive correlations between the increase in the grip force and perceived heaviness at both
different frequencies (
ρ
= 0.66,
p
= 1.24
×
10
−9
) and amplitudes (
ρ=
0.77,
p
= 1.37
×
10
−7
).
The increases in the grip force and the intensities of the weight illusion were greater for the
low-frequency and large-amplitude vibratory stimuli. The results of this study suggest that
sinusoidal vibrations change the grip force and weight perception, and the increase in the
grip force makes the object feel heavy. Further studies are needed to clarify the reasons as
to why the grip force is increased by the vibratory stimuli. Practically, the vibration-weight
illusion could be implemented in game controllers or virtual reality, for which vibrotactile
stimuli are commonly used [
47
], to present virtual loadings on avatars’ hands. In immersive
virtual environments, other illusory techniques called pseudo-haptics [
48
] can be applied
with the vibration-weight illusion to influence the perceived heaviness. It is difficult to
eliminate the perception of vibration during the illusion; hence, suitable applications may
include fishing games and scenes that involve hugging moving objects or living creatures,
during which vibratory motions and loadings naturally co-occur.
Author Contributions:
Conceptualization, G.K. and S.O.; methodology, G.K. and S.O.; software,
G.K.; validation, G.K., S.O. and H.M.; formal analysis, G.K. and S.O.; resources, G.K.; data curation,
G.K. and S.O.; writing—original draft preparation, G.K.; writing—review and editing, G.K., S.O. and
H.M.; project administration, S.O.; funding acquisition, S.O. All authors have read and agreed to the
published version of the manuscript.
Funding: This research was in part funded by MEXT Kakenhi #21H05819.
Institutional Review Board Statement:
This study was approved by the Institutional Review Board
of Hino Campus, Tokyo Metropolitan University (#21-008).
Informed Consent Statement:
The participants provided written informed consent before the experiment.
Data Availability Statement: The readers can contact the authors for potential data provision.
Acknowledgments:
We thank Yasuhiro Akiyama and Yoji Yamada for their thoughtful suggestions
on the study.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Ross, H.E. When is a weight not illusory? Q. J. Exp. Psychol. 1969,21, 346–355. [CrossRef] [PubMed]
2. Davis, C.M.; Roberts, W. Lifting movements in the size-weight illusion. Percept. Psychophys. 1976,20, 33–36. [CrossRef]
3.
Gordon, A.; Forssberg, H.; Johansson, R.; Westling, G. Visual size cues in the programming of manipulative forces during
precision grip. Exp. Brain Res. 1991,83, 477–482. [CrossRef] [PubMed]
4.
Taylor, J.L.; Todd, G.; Gandevia, S.C. Evidence for a supraspinal contribution to human muscle fatigue. Clin. Exp. Pharmacol.
Physiol. 2006,33, 400–405. [CrossRef] [PubMed]
5.
Stevens, J.C.; Rubin, L.L. Psychophysical scales of apparent heaviness and the size-weight illusion. Percept. Psychophys.
1970
,
8, 225–230. [CrossRef]
6.
Amazeen, E.L.; Turvey, M.T. Weight perception and the haptic size–weight illusion are functions of the inertia tensor. J. Exp.
Psychol. Hum. Percept. Perform. 1996,22, 213. [CrossRef] [PubMed]
7.
Zhu, Q.; Bingham, G.P. Human readiness to throw: The size–weight illusion is not an illusion when picking the best objects to
throw. Evol. Hum. Behav. 2011,32, 288–293. [CrossRef]
8.
Charpentier, A. Analyse experimentale de quelques elements de la sensation de poids. Arch. Physiol. Norm. Pathol.
1891
,
3, 122–135.
9.
Flanagan, J.R.; Bittner, J.P.; Johansson, R.S. Experience Can Change Distinct Size-Weight Priors Engaged in Lifting Objects and
Judging their Weights. Curr. Biol. 2008,18, 1742–1747. [CrossRef]
10. Wolfe, H.K. Some effects of size on judgments of weight. Psychol. Rev. 1898,5, 25. [CrossRef]
11. Warden, C.J.; Flynn, E.L. The effect of color on apparent size and weight. Am. J. Psychol. 1926,37, 398–401. [CrossRef]
12. Walker, P.; Francis, B.J.; Walker, L. The brightness-weight illusion. Exp. Psychol. 2010,57, 462–469. [CrossRef]
13.
Johansson, R.S.; Westling, G. Roles of glabrous skin receptors and sensorimotor memory in automatic control of precision grip
when lifting rougher or more slippery objects. Exp. Brain Res. 1984,56, 550–564. [CrossRef]
14.
Johansson, R.S.; Birznieks, I. First spikes in ensembles of human tactile afferents code complex spatial fingertip events. Nat.
Neurosci. 2004,7, 170–177. [CrossRef]
Appl. Sci. 2023,13, 2717 12 of 13
15.
Minamizawa, K.; Fukamachi, S.; Kawakami, N.; Tachi, S. Interactive representation of virtual object in hand-held box by
finger-worn haptic display. In Proceedings of the IEEE Symposium on Haptic Interfaces for Virtual Environment and Teleoperator
Systems, Reno, NE, USA, 13–14 March 2008; pp. 367–368.
16.
Van Beek, F.E.; King, R.J.; Brown, C.; Luca, M.D.; Keller, S. Static weight perception through skin stretch and kinesthetic
information: Detection thresholds, JNDs, and PSEs. IEEE Trans. Haptics 2020,14, 20–31. [CrossRef]
17.
Guinan, A.L.; Montandon, M.N.; Caswell, N.A.; Provancher, W.R. Skin stretch feedback for gaming environments. In Proceedings
of the IEEE International Workshop on Haptic Audio Visual Environments and Games, Munich, Germany, 8–9 October 2012;
pp. 101–106.
18.
Park, J.; Oh, Y.; Tan, H.Z. Effect of Cutaneous Feedback on the Perceived Hardness of a Virtual Object. IEEE Trans. Haptics
2018
,
11, 518–530. [CrossRef]
19.
Amemiya, T.; Ando, H.; Maeda, T. Lead-me interface for a pulling sensation from hand-held devices. ACM Trans. Appl. Percept.
2008,5, 1–17. [CrossRef]
20.
Amemiya, T.; Maeda, T. Asymmetric oscillation distorts the perceived heaviness of handheld objects. IEEE Trans. Haptics
2008
,
1, 9–18. [CrossRef]
21.
Tanabe, T.; Yano, H.; Iwata, H. Properties of proprioceptive sensation with a vibration speaker-type non-grounded haptic interface.
In Proceedings of the IEEE Haptics Symposium, Philadelphia, PA, USA, 8–11 April 2016; pp. 21–26.
22.
Culbertson, H.; Walker, J.M.; Okamura, A.M. Modeling and design of asymmetric vibrations to induce ungrounded pulling
sensation through asymmetric skin displacement. In Proceedings of the IEEE Haptics Symposium, Philadelphia, PA, USA,
8–11 April 2016; pp. 27–33.
23.
Okamoto, S.; Konyo, M.; Tadokoro, S. Vibrotactile stimuli applied to finger pads as biases for perceived inertial and viscous loads.
IEEE Trans. Haptics 2011,4, 307–315. [CrossRef]
24. Nagano, H.; Okamoto, S.; Yamada, Y. Vibrotactile cueing for biasing perceived inertia of gripped object. Haptic Interact. Percept.
Devices Appl. 2015,277,17–20.
25.
Kim, G.; Okamoto, S.; Akiyama, Y.; Yamada, Y. Weight illusion by presenting vibration to the fingertip. Front. Virtual Real.
2022
,
3, 797993. [CrossRef]
26.
McCloskey, D.; Ebeling, P.; Goodwin, G. Estimation of weights and tensions and apparent involvement of a “Sense of effort”.
Exp. Neurol. 1974,42, 220–232. [CrossRef] [PubMed]
27.
Gandevia, S.; Killian, K.; Campbell, E. The effect of respiratory muscle fatigue on respiratory sensations. Clin. Sci.
1981
,60, 463–466.
[CrossRef] [PubMed]
28. Jones, L.A. Perception of force and weight: Theory and research. Psychol. Bull. 1986,100, 29. [CrossRef]
29. Enoka, R.M.; Stuart, D.G. Neurobiology of muscle fatigue. J. Appl. Physiol. 1992,72, 1631–1648. [CrossRef]
30.
Goodwin, G.M.; McCloskey, D.I.; Matthews, P.B. Proprioceptive illusions induced by muscle vibration: Contribution by muscle
spindles to perception? Science 1972,175, 1382–1384. [CrossRef]
31.
Burke, D.; Hagbarth, K.E.; Löfstedt, L.; Wallin, B.G. The responses of human muscle spindle endings to vibration of non-
contracting muscles. J. Physiol. 1976,261, 673–693. [CrossRef]
32.
Martin, B.J.; Park, H.S. Analysis of the tonic vibration reflex: Influence of vibration variables on motor unit synchronization and
fatigue. Eur. J. Appl. Physiol. Occup. Physiol. 1997,75, 504–511. [CrossRef]
33. Johnson, K.O. The roles and functions of cutaneous mechanoreceptors. Curr. Opin. Neurobiol. 2001,11, 455–461. [CrossRef]
34.
Nakamoto, M.; Konyo, M.; Maeno, T.; Tadokoro, S. Reflective grasp force control of humans induced by distributed vibration
stimuli on finger skin with ICPF actuators. In Proceedings of the 2006 IEEE International Conference on Robotics and Automation,
Orlando, FL, USA, 15–19 May 2006; pp. 3899–3904.
35.
Sakurai, T.; Okamoto, S.; Konyo, M.; Tadokoro, S. Research of conditions of stimulus for inducing grasping force control reflex.
In Proceedings of the 2010 IEEE/SICE International Symposium on System Integration, Sendai, Japan, 21–22 December 2010;
pp. 408–413.
36.
Gescheider, A.; Bolanowski, S.; Hardick, K. The frequency selectivity of information-processing channels in the tactile sensory
system. Somatosens. Mot. Res. 2001,18, 191–201. [CrossRef]
37.
Macefield, V.G.; Häger-Ross, C.; Johansson, R.S. Control of grip force during restraint of an object held between finger and thumb:
Responses of cutaneous afferents from the digits. Exp. Brain Res. 1996,108, 155–171. [CrossRef]
38.
Okamoto, S.; Wiertlewski, M.; Hayward, V. Anticipatory vibrotactile cueing facilitates grip force adjustment during perturbative
loading. IEEE Trans. Haptics 2016,9, 233–242. [CrossRef]
39.
Delhaye, B.P.; Jarocka, E.; Barrea, A.; Thonnard, J.L.; Edin, B.; Lefèvre, P. High-resolution imaging of skin deformation shows that
afferents from human fingertips signal slip onset. Elife 2021,10, e64679. [CrossRef]
40.
Güçlü, B.; Öztek, C. Tactile sensitivity of children: Effects of frequency, masking, and the non-Pacinian I psychophysical channel.
J. Exp. Child Psychol. 2007,98, 113–130. [CrossRef]
41.
Terekhov, A.V.; Hayward, V. Minimal adhesion surface area in tangentially loaded digital contacts. J. Biomech.
2011
,44, 2508–2510.
[CrossRef]
42.
Delhaye, B.; Lefèvre, P.; Thonnard, J.L. Dynamics of fingertip contact during the onset of tangential slip. J. R. Soc. Interface
2014
,
11, 20140698. [CrossRef]
Appl. Sci. 2023,13, 2717 13 of 13
43.
Barrea, A.; Delhaye, B.P.; Lefèvre, P.; Thonnard, J.L. Perception of partial slips under tangential loading of the fingertip. Sci. Rep.
2018,8, 7032. [CrossRef]
44.
Chowdhury, M.A.; Helali, M.M. The effect of amplitude of vibration on the coefficient of friction for different materials. Tribol.
Int. 2008,41, 307–314. [CrossRef]
45.
Flanagan, J.R.; Wing, A.M. Effects of surface texture and grip force on the discrimination of hand-held loads. Percept. Psychophys.
1997,59, 111–118. [CrossRef]
46.
Rinkenauer, G.; Mattes, S.; Ulrich, R. The surface–weight illusion: On the contribution of grip force to perceived heaviness.
Percept. Psychophys. 1999,61, 23–30. [CrossRef]
47.
See, A.R.; Choco, J.A.G.; Chandramohan, K. Touch, texture and haptic feedback: A review on how we feel the world around us.
Appl. Sci. 2022,12, 94686. [CrossRef]
48.
Lee, J.; Lee, Y.; Park, S. Virtual gymnasium: Personalized weight perception interface in lifting virtual objects. Appl. Sci.
2022
,
12, 12414. [CrossRef]
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
