Representations of fabric hand attributes in the cerebral cortices based on the Automated Anatomical Labeling atlas

Abstract

Fabric hand is most frequently used by consumers and researchers to evaluate the touch feeling of textiles. Academically, many methods have been developed to characterize it psychologically and physically, and the relationship between the hand attributes of fabrics and their physical properties are well understood. However, in physiological terms, the cognitive mechanism of the brain on different attributes of fabric hand is not clear. Previous studies have shown that the sensory or discrimination information from fabric touch can be detected by the technology of functional magnetic resonance imaging (fMRI). In this study, further fMRI experiments were carried out, attempting to find the relationship between the cerebral cortices of various brain areas and different hand attributes of fabrics. The subtle atlas of Automated Anatomical Labeling (AAL) was used to display and analyze the blood oxygenation level dependent signals completely and conveniently. The results showed that when the subjects touched two samples with distinct fabric hand in a specified way, activation information and the index of the mean signal in every related brain areas can distinguish them, and several brain regions in the AAL atlas are linked to different fabric hand attributes. The technology of fMRI was proved again to be a promising tool for studying the cognitive mechanism of the brain on fabric touch.

Keywords

fabric hand finger touch brain responses functional magnetic resonance imaging Automated Anatomical Labeling

Fabric hand is used as a crucial element to characterize the touch feeling of textiles by individuals, and studies focusing on the topic have been carried out for almost a hundred years.¹ Up to now, methods used for fabric hand assessment can be divided into three categories: psychological (subjective), physical (objective), and physiological.^2,3

Psychological methods originated from the researches of Binns⁴ in 1925, in which 16 pairs of descriptors were used to discriminate the sense of touch from wool fabric. Physical methods can be dated to the study by Peirce,⁵ who evaluated the sensations of stiffness and hardness from fabric by testing the bending length of it using a simple device. Subsequently, many instruments were developed by researchers to evaluate fabric hand objectively, among which the KES (Kawabata Evaluation System)⁶ and FAST (Fabric Assurance by Simple Testing)⁷ are two of the most representative. Moreover, several other novel instruments have also been developed, such as the PhabrOmeter System,^8,9 FAMOUS,¹⁰ CHES-FY,^11–14 and so on. Using the above instruments, different hand attributes (e.g. soft, pliable, and coarse) of fabric can be linked to its physical properties (e.g. compressive property, bending property, and surface performance).

Physiologically, various physiological signals have been used to evaluate the touch feeling of fabrics.^15,16 The nature of fabric hand can be interpreted as the cerebral cognition on the physiological responses of skin receptors evoked by the tactile stimulus from different hand attributes.¹⁷ In addition, a neuromechanical fabric–skin-receptor model was built to study the relationship between neural responses of cutaneous mechanoreceptors and the fabric hand attribute of softness.¹⁸ However, the cognitive mechanism of the brain on different attributes of fabric hand is still not clear.

Fortunately, the technology of functional magnetic resonance imaging (fMRI)¹⁹ has evolved into one of the most advanced methods to detect the brain response evoked by external stimuli.^20,21 Magnetic resonance imaging (MRI) is a widely accepted technology for providing anatomical information of the human body. Based on this, the advanced technology of fMRI was developed to provide information about biological function, in addition to the concomitant anatomical information, which uses the naturally occurring paramagnetic deoxyhemoglobin in venous blood as a contrast agent.²² The technology of fMRI has been used by a large number of researchers to study brain activities related to various kinds of tactile perception on electrical stimulation, heat stimulation, sponge brush stimulation,²³ vibratory stimulation (a plastic cylinder),²⁴ and so on. For example, Hammeke et al.²⁵ studied the brain signal changes of subjects when their palms received tactile stimulation from the experimenter stroking in a tickling fashion. Correlated signal changes of 1–5% were detected by fMRI in the peri-rolandic region. These studies provided evidence of the high sensitivity of fMRI for detecting activation in the cerebral cortices related to tactile stimulation. It is a promising tool for studying the brain cognition on the tactile stimulation of fabrics.²⁶ Therefore, in this study, the technology of fMRI was used to detect the signals of brain responses to fabric touch, attempting to reveal the cognitive mechanism of the brain on tactile sense from fabric touch.

Previously, the technology of fMRI was used to compare the brain cognition of fabric touch on human glabrous and hairy skin,²⁷ showing that more emotional information arises from fabric touch on hairy skin (in which the unmyelinated C tactile afferents are contained), and more sensory or discriminating information on glabrous skin (in which more myelinated fast-conducting afferents are contained). So in the present study, further fMRI experiments for the fabric touch on glabrous skin were carried out in n attempt to link the cerebral cortices of various brain areas to different hand attributes of fabrics. In particular, the skin of fingertips was selected to touch fabrics in this study, because in comparison with other sites of the glabrous skin in the human hand, it contains the largest number of myelinated fast-conducting afferents.^28,29 There are two main contact methods used for perceiving the touch feeling of fabrics, active and passive. Related studies³⁰ have shown that the method of active touch is able to generate more physical stimuli than passive touch, as it involves the exertive manipulating of objects.³¹ Therefore, subjects in this study were required to touch fabrics in active methods, rather than receiving fabric touch passively.

The brain atlas of Automated Anatomical Labeling (AAL)³² is a well known tool for researchers of neuroscience to label or report the result of brain activations, which offers a detailed anatomical brain description within the Montreal Neurological Institute (MNI) space.³³ So in this study, the subtle AAL atlas was used to analyze and report the results in order to analyze the blood oxygenation level dependent (BOLD) signals in brain areas of the entire cerebrum completely and conveniently.

Materials and methods

Materials and performance evaluation

Two kinds of fabrics, silk and linen, were chosen in this study. It is well known that the fabric hand performance is distinctly different between these two kinds of fabrics. When the skin of human body contacts with them, it will arouse brain responses with representative differences, which can provide powerful evidence for judging brain functional areas related to fabric hand attributes. This is why these two kinds of fabrics were chosen.

The primary specifications of the two fabric samples are listed in Table 1. In order to confirm the difference of touch feeling (fabric hand) between the two fabric samples, the Kawabata Evaluation System for Fabrics (KES-F) was used to test the shearing, bending, and surface properties of them in both warp and weft directions, and the average value of the KES parameters in the two directions was calculated, which is listed in Table 1. A HI SCOPE KH-1000 digital video microscope was used to observe the surface morphologies of the two fabric samples.

Table 1.

The primary specifications of the two fabric samples

Parameters	Silk	Linen	Compare
Thickness (mm)	0.18	0.70	S<L
Mass (g/m²)	60.75	170.75	S<L
Structure	Plain	Plain	—
Density
warp (/10 cm)	400	200	S>L
weft (/10 cm)	550	200	S>L
Shearing
G (gf/[cm·(^o)])	0.25	3.22	S<L
2HG (gf/cm)	0.08	6.28	S<L
2HG5 (gf/cm)	0.23	11.75	S<L
Bending
B (gf·cm²/cm)	0.0105	0.1205	S<L
2HB (gf·cm²/cm)	0.0022	0.0925	S<L
Surface
MIU	0.169	0.170	S<L
MMD	0.018	0.098	S<L
SMD (μm)	0.575	9.570	S<L

G: shear rigidity; 2HG: shear hysteresis at 0.5° shear angle; 2HG5: shear hysteresis at 5° shear angle; B: bending rigidity; 2HB: bending hysteresis; MIU: mean friction coefficient (steel/fabric); MMD: mean deviation of MIU, surface frictional roughness; SMD: geometric roughness. The letters “S” and “L” represent silk and linen fabric, respectively.

Subjects

Eight healthy right-handed male subjects, with the age range of 24–35 years (average age of 28.6 years), took part in this fMRI experiment. Before the experiment, they were trained simply about the scanning procedure and gave written informed consent. The ethics committee of Shanghai Ruijing Hospital approved this study.

fMRI experiment

The purpose of this fMRI experiment is to study the brain responses of subjects when they touch the fabric using their fingers. The main principle of the fMRI experiment can be explained by the process of generation and detection of the brain BOLD²² signals evoked by given external stimulus, which is the tactile stimulus of fabrics by finger touch in this study, as shown in Figure 1.

Figure 1.

Sketch of generation and detection of the brain signals responsible for the task of touching fabrics. BOLD: blood oxygenation level dependent; MRI: magnetic resonance imaging.

The fMRI experiment was conducted using the block design,³⁴ as shown in Figure 2. One session (180 seconds) of the design is composed of six blocks (for every 30 seconds), three blocks for rest, and another three for the touching task. The two kinds of blocks are alternately arranged one by one.

Figure 2.

One session of the block-design functional magnetic resonance imaging experiment.

In particular, at the starting point of the “touching task” block, one piece of fabric (10 cm × 10 cm) was put into the interspace between the thumb and index finger of the subject's right hand, and the subject was required to perform the task of touching the fabric for 30 seconds. Because the touch sensations can be affected by the contact method,³¹ the subjects were required to touch the fabric in a specified way. Before the formal experiment, a piece of light cotton fabric was used to train all the subjects touching the fabric in the same way as follows: the subjects were required to gently hold (or pinch) the fabric with small pressure just using the thumbs and index fingers of their right hands, and then repeatedly rotate and rub the thumbs on the fabric surface in the anti-clockwise direction, at the frequency of about 1 Hz for 30 seconds, to consciously feel the touch feeling of the fabric. During the entire process of touching tasks, the subjects were required to hold and touch the fabric samples with small pressure, and to the greatest extent, keep the pressure constant for touching the two fabric samples.

At the starting point of the “rest” block, the subject was required to do nothing but open the thumb and index finger of the right hand to form an interspace (about 5 mm) between them. In addition, the fabric was removed from the interspace.

During the three blocks of the “touching task” in one session, the subject touched the same fabric. So, every subject took part in two sessions of the block-design fMRI experiment, one for the task of touching silk fabric, and another for linen. During the entire fMRI experiment, all the subjects were required to close their eyes but remain awake, with no head movement.

Neuroimaging data acquisition and processing

The fMRI experiment was performed at Shanghai Ruijin Hospital on a 3 T clinical whole body MRI scanner (GE 3.0T Signa HDxt superconductive MR system) using a standard head coil. The three-dimensional (3D) anatomical brain images of each subject were collected using the scanning sequence of dimension-spoiled gradient recalled echo (T1WI). The functional brain images, which are the fMRI time-series of BOLD signals, were acquired using BOLD sequence, which is the scanning sequence of single-shot EPI (echo planar imaging), with the following scanning parameters: repetition time (TR) = 2000 ms, echo time (TE) = 30 s, flip angle = 90°, field of view (FOV) = 24 cm × 24 cm, in-plane resolution = 3.44 mm × 3.44 mm.

The neuroimaging data acquired from the fMRI system were firstly preprocessed by using the most recognized method, Statistical Parametric Mapping,³⁵ and the software toolkit SPM12 (developed by members and collaborators of the Wellcome Trust Centre for Neuroimaging and available at http://www.fil.ion.ucl.ac.uk/spm/) was mainly used. The procedure of preprocessing contains motion correct, coregister, segment, normalize, and smooth, which are illustrated in Figure 3. In particular, in order to ensure that the results are accurate, the method of DARTEL (Diffeomorphic Anatomical Registration using Exponentiated Lie algebra)³⁶ was used for the processes of segment and normalize.

Figure 3.

Overview of the preprocessing procedure for the functional magnetic resonance imaging (fMRI) data.

Then, the General Linear Model³⁷ was used to carry out statistical analysis. In order to study the brain responses based on the AAL atlas, 116 brain masks or ROIs (regions of interest) were made from the 116 AAL regions, as shown in Figure 4. The mark numbers (NO.) and labels of the 116 AAL regions are available in the AAL software (available at http://www.gin.cnrs.fr/en/tools/aal-aal2/). Finally, activation information and brain signals in each of the 116 AAL masks were extracted and calculated by using SPM12 and Masbar.³⁸

Figure 4.

The three-dimensional model of the 116 brain regions in the Automated Anatomical Labeling atlas from different views.

Results and discussion

Comparison of fabric hand between the two samples

Figures 5(a) and (b) display the surface morphologies of the silk and linen fabric with the magnification of 50×, showing that yarns in the silk fabric are fine and closely arranged, and its surface is even and uniform. On the contrary, yarns in the linen fabric are thick and sparsely arranged, and the surface of the linen fabric is coarse and uneven. Moreover, Table 1 shows that the silk fabric is thinner than the linen fabric.

Figure 5.

The surface images of silk (a) and linen (b) fabrics (50×).

This feature can also be explained by the results of surface performances tested by the KES-F, as shown in Table 1. Although the two fabric samples almost have the same value of MIU, there are large differences in MMD and SMD between them. So, in comparison, it is definite that the touching hand on the surface of the linen fabric is coarser, less even, and less uniform than that of the silk fabric. Moreover, it is generally known that silk fabric has the hand of scroopy feel,^39–41 so that the fingers will suffer more resistance to move on its surface. Obviously, it also can be concluded from the KES results of shearing and bending shown in Table 1 that the silk fabric is more pliable with a more comfortable touch feeling than that of the linen fabric.

Therefore, according to the above analysis and comparison, four key attributes of fabric hand, “coarse,” “pliable,” “thin,” and “scroopy,” can be used to distinguish the two fabric samples.

Analysis of the activation information

After statistical analysis, group results (eight subjects) of activation information of clusters are reported at extent size P < 0.01, corrected for multiple comparisons using the cluster-wise false discovery rate (FDR).⁴² Tables 2 and 3 list the activation information in the AAL brain regions of the subjects when they touched the two fabric samples using the thumb and index finger of their right hands, with Table 2 showing the results of silk fabric and Table 3 those of linen fabric. Correspondingly, every activation cluster is presented in the slice of brain image, as shown in Figure 6 (silk fabric) and Figure 7 (linen fabric).

Figure 6.

Activations in Automated Anatomical Labeling (AAL) regions for the task of touching silk fabric. The number indexing each slice at the upper left-hand corner represents the Montreal Neurological Institute z-coordinate of the slice. The activation location is labeled below each slice by the AAL number.

Figure 7.

Activations in Automated Anatomical Labeling (AAL) regions for the task of touching linen fabric.

Table 2.

Brain activation in Automated Anatomical Labeling (AAL) regions for the task of touching silk fabric

AAL Regions		Peak			Cluster Level
No.	Label	x	y	z	Size	Z
14	Cerebelum_4_5_R	15	−54	−21	23	3.06
15	Cerebelum_6_L	−24	−51	−27	32	3.34
15	Cerebelum_6_L	−30	−51	−33	32	3.14
16	Cerebelum_6_R	30	−48	−33	32	3.73
		33	−57	−30		3.58
		27	−63	−27		2.87
19	Cerebelum_8_L	−24	−69	−51	25	3.57
29	Cingulum_Mid_L	0	0	39	47	4.19
29	Cingulum_Mid_L	−6	−6	48	47	3.44
35	Frontal_Inf_Oper_L	−60	6	9	42	3.14
		−54	12	24		3.14
		−48	9	15		3.10
36	Frontal_Inf_Oper_R	51	12	9	94	4.35
		57	12	3		3.05
		60	9	12		2.87
60	Insula_R	45	6	9	39	3.93
		51	6	−3		3.61
		45	0	−3		3.12
77	Parietal_Inf_L	−48	−36	51	106	3.50
		−51	−30	39		3.40
		−51	−24	45		3.36
78	Parietal_Inf_R	42	−45	51	28	3.13
		48	−39	45		2.70
		36	−39	48		2.58
81	Postcentral_L	−48	−24	51	284	4.16
		−33	−27	45		4.08
		−36	−33	54		3.63
83	Precentral_L	−24	−15	63	121	4.08
		−33	−24	54		3.79
		−30	−12	51		3.34
92	Rolandic_Oper_R	48	6	6	27	3.67
		57	12	0		3.56
		57	6	12		3.01
		57	−21	18	41	3.12
		57	−30	21		2.96
		45	−18	18		2.75
93	Supp_Motor_Area_L	−3	3	48	55	3.93
		−6	−9	51		3.34
		0	−3	57		3.07
95	SupraMarginal_L	−51	−24	42	45	3.44
		−54	−30	33		3.28
		−60	−21	42		3.15

No.: the labeled numbers of AAL regions; Label: the labeled names of AAL regions; Peak: the Montreal Neurological Institute coordinates (x, y, z) of peaks in the activated cluster; Size: the number of activated voxels contained in the cluster; Z: Z-value of activation intensity.

Table 3.

Brain activation in Automated Anatomical Labeling (AAL) regions for the task of touching linen fabric

AAL Regions		Peak			Cluster Level
No.	Label	x	y	z	Size	Z
35	Frontal_Inf_Oper_L	−51	9	12	20	3.43
35	Frontal_Inf_Oper_L	−54	12	3	20	2.94
81	Postcentral_L	−45	−21	51	117	4.61
		−45	−30	48		2.95
		−33	−30	45		2.89
83	Precentral_L	−30	−15	48	59	3.33
		−33	−12	60		3.10
		−27	−21	63		2.90

Tables 2 and 3 show that 16 activated clusters, distributing in 15 AAL regions, were detected for the task of touching the silk fabric. However, differently, only three clusters in three AAL regions were found when the subjects touched the linen fabric. In order to compare and analyze more conveniently and clearly, activations in the other 12 (15 minus 3) AAL regions for the task of touching linen fabric are reported with a lower threshold, P < 0.05, with no correction. All of the largest activated clusters in these AAL regions are listed in Table 4, and every activation cluster is presented in the slice of brain image in Figure 8.

Figure 8.

Activations in the other 12 Automated Anatomical Labeling (AAL) regions for the task of touching linen fabric (P < 0.05 with no correction).

Table 4.

Brain activation in the other 12 Automated Anatomical Labeling (AAL) regions for the task of touching linen fabric (P < 0.05 with no correction)

AAL Regions		Peak			Cluster Level
No.	Label	x	y	z	Size	Z
14	Cerebelum_4_5_R	12	−54	−24	20	2.25
15	Cerebelum_6_L	−18	−60	−30	28	2.82
		−27	−51	−33		2.39
		−21	−51	−27		1.84
16	Cerebelum_6_R	18	−63	−30	32	2.62
16	Cerebelum_6_R	24	−51	−33	32	2.57
19	Cerebelum_8_L	−21	−72	−51	31	3.26
29	Cingulum_Mid_L	−9	0	39	16	1.87
29	Cingulum_Mid_L	0	−3	39	16	1.82
36	Frontal_Inf_Oper_R	60	12	9	127	2.81
		45	6	24		2.77
		60	12	27		2.70
60	Insula_R	42	−3	0	27	2.94
60	Insula_R	39	0	12	27	2.27
77	Parietal_Inf_L	−54	−39	36	220	3.52
		−45	−42	42		3.44
		−45	−27	48		3.14
78	Parietal_Inf_R	36	−54	45	29	2.61
		45	−45	48		2.60
		36	−51	54		1.81
92	Rolandic_Oper_R	63	−21	15	83	2.98
92	Rolandic_Oper_R	54	−24	21	83	2.36
93	Supp_Motor_Area_L	0	−12	57	10	2.67
93	Supp_Motor_Area_L	−9	−12	48	10	1.87
95	SupraMarginal_L	−63	−24	15	85	2.70
		−51	−24	42		2.59
		−48	−24	24		2.46

The activated AAL regions will be analyzed in the following categories.

(1) Somatosensory areas

Somatosensory areas⁴³ in the brain are composed of two parts, the primary somatosensory cortex (SI) and the secondary somatosensory cortex (SII), with the SI being located in the postcentral gyrus⁴⁴ and SII in the parietal operculum on the ceiling of the lateral sulcus.^45,46 In this study, as shown in Tables 2 –4, activated clusters belonging to somatosensory areas are mainly located in the AAL regions of AAL-81 (SI) and AAL-92 (SII). In addition, there are few activated clusters in the region of AAL-77 belonging to SI.

For the SI, it is noted that the cluster size in AAL-81 for the situation of the silk fabric (284) is larger than that of the linen fabric (117) under the same threshold, P < 0.01, as shown in Tables 2 and 3, but it is just the reverse for the maximum activation intensity (Z-value) in AAL-81, 4.16 for the silk fabric versus 4.61 for the linen fabric. Combined with the above comparison results of fabric hand between the two samples, in can be inferred that the cluster size of AAL-81 is related to the two hand attributes, thin and pliable (fabrics that are thinner and more pliable could evoke more voxels), and the activation intensity is related to surface roughness (fabrics with a coarser surface could evoke a larger Z-value). As for the region of AAL-77, because only a small part of it belongs to the SI, the activation information can only provide references for the perception analysis of fabric touch in this study, and it needs further research to subdivide the region.

In the case of the SII, Table 2 shows that two clusters were activated in the brain region of AAL-92 for the situation of silk fabric, but under the same threshold, P < 0.01, there was no voxel survival in this AAL region for linen fabric, as shown in Table 3. The region was activated under the lower threshold P < 0.05 with no correction (Table 4), showing that the AAL-92 region may be a key area for distinguishing fabric touch.

The reason why more activations in AAL-92 were aroused for the task of touching the silk fabric is analyzed and concluded as follows. Early studies from neuroimaging researchers confirmed that the brain area of the SII is involved in higher level tactile perception, particularly playing an important role in discrimination for the surface morphology of objects.^47,48 In this study, on the one hand, in comparison with linen fabric, the surface of the silk fabric feels more scroopy, which belongs to a kind of high-level tactile perception; on the other hand, the above analysis of the surface morphology of the two samples shows that the surface morphology of the silk fabric is finer, with finer yarns more closely arranged than that of the linen fabric. Thus, more “effort” (activations) for the brain is needed to perceive the texture of the silk fabric.

Therefore, it is definite that the activation information in AAL-92, which belongs to the SII, is involved in the fabric hand attribute of scroopy, and is highly related to the surface texture of fabric.

(2) Motor-related areas

There are several regions in the brain involved in the movement of body parts,^49,50 such as the primary motor cortex in the precentral gyrus, the supplementary motor area⁵¹ just anterior to the primary motor cortex, the cerebellum,⁵² and so on. The functions of the motor cortex contain two types of processes, the initiation of movements involved in motor cortex activation and the ability to suppress undesired movements involved in motor cortex inactivation or suppression.⁵³ Obviously, the present study involved both of processes of volitional control. In the fMRI experiment, the subjects were required to touch the fabric by moving their fingers (the former process), and keep their heads motionless (the latter process). Only the former process (activation information) of the motor cortex is analyzed in this study, because the latter process is not the research emphasis.

In the primary motor cortex, activations were detected in the region of AAL-83 for both of the touching tasks under the same threshold of P < 0.01. Moreover, fewer voxels and smaller activation intensity were observed for the task of touching linen fabric (size = 59, Z = 3.33) than that of the silk fabric (size = 121, Z = 4.08). The worst situation was observed in the supplementary motor area (AAL-93), with the linen fabric (size = 10, Z = 2.67, P < 0.05) versus the silk fabric (size = 55, Z = 3.93, P < 0.01), as shown in Tables 2 and 4. A similar worst situation happened in the cerebellum, where four AAL regions, AAL-14, AAL-15, AAL-16, and AAL-19, belonging to the cerebellum, were activated for the task of touching silk fabric (P < 0.01) and linen fabric (P < 0.05). It is a universally consensus that the cerebellum plays an important role in motor control,⁵⁴ and it contributes to coordinating voluntary movements, precision, and accurate timing.⁵⁵

In this study, the subjects were required to touch the two fabric samples by moving their fingers in the same specified way (repeatedly circling theirs thumbs on the fabric surfaces in the anti-clockwise direction, at the frequency of about 1 Hz). The activation information in motor areas shows that, similar to AAL-92, more “effort” (activations) for the brain (motor areas) is needed to move their fingers on the surface of the silk fabric than on the linen fabric, because the surface of silk fabric is more scroopy than linen fabric, making it more difficult for the subjects to move their fingers on the surface of silk fabric than on linen fabric. Therefore, activation information in motor areas of the brain is suggested to be a representative reference for the scroopy feel of fabric surface.

(3) Emotion-related areas

Representations of emotions in the brain are complex, and one of the most important brain areas involved in the emotional information of touch is the insular cortex,⁵⁶ to which pleasant touch is highly related.^57,58 A previous study on the brain cognition of fabric touch shows that the insular cortex is a representation of the fabric performance of tactile comfort, especially for fabric touch on the hairy skin.²⁷ The performance of tactile comfort in this study reflects the degree of pleasant (or unpleasant) touch feelings from fabrics, meaning that fabrics with better performance of tactile comfort can give rise to a more pleasant touch.

In the present study, it is certain that the silk fabric feels more comfortable than the linen fabric. So, it can be predicted that the insular cortex will be more active for the task of touching the silk fabric than that of the linen fabric. In accordance with the prediction, Tables 2 and 4 show that activations were observed in the region of AAL-60 belonging to the insular cortex, with the following activation results: silk fabric (size = 39, Z = 3.93, P < 0.01) versus linen fabric (size = 27, Z = 2.94, P < 0.05). The results also show that for the two fabric samples, both of the cluster sizes and activation intensities are small, because the skin of the fingertips belongs is glabrous. Therefore, the insular cortex was demonstrated again to be responsible for the tactile comfort of fabrics.

(4) Other areas

Activations detected in the regions of AAL-35 and AAL-36 (Tables 2 –4) are located in Broca's area, which is involved in language comprehension and speech production.⁵⁹ Many studies have shown that the area is also linked to functions of working memory,⁶⁰ episodic long-term memory, action imitation,⁶¹ and so on. In this study, before the formal experiment, all the subjects were trained to touch fabrics in the same way. During the formal fMRI experiment, the subjects moved their fingers by recalling the language description of the training and imitating the finger action to touch the fabric in the required way. Then the above process evoked activations in the regions of AAL-35 and AAL-36.

Most of the activated clusters observed in the AAL regions of AAL-95, AAL-77, and AAL-78 (shown in Tables 2 and 4) belong to the supramarginal gyrus, which contributes to manipulation and finger positioning, especially for the left supramarginal gyrus.^62,63 Moreover, clusters observed in the region of AA-29 belong to the anterior cingulate cortex, which contributes to emotional processing, motor control, attention, and so on.^64,65

Therefore, according to results of the present study and the available references, activation information in the above three areas is supposed to have little to do with fabric hand attributes.

Analysis of the mean signals

The mean signal (the beta value in the General Linear Model) extracted from each of the 116 AAL brain regions of the subjects was plotted on a radar graph (Figure 9). Obviously, three kinds of important information can be acquired from Figure 9.

Figure 9.

The radar graph of mean signals (beta value) in all of the 116 AAL brain regions. The coordinates labeled in the circular direction represent the label number of the 116 AAL regions, and the coordinates marked in the diameter direction represent the mean signal (beta value). The main brain areas are also labeled in green font around the circumference (color online only).

Firstly, the shapes (or tracing patterns) of the two curves are very alike, indicating that, for the two tasks of touching different fabrics (silk and linen), brain responses in each AAL region are mainly consistent.

Secondly, the curve for the linen fabric is almost surrounded by that of the silk fabric, demonstrating that, in comparison, the brain is more active in most of the AAL regions for the task of touching the silk fabric than for the linen fabric. That is, the effect size of the tactile stimulation from touching the silk fabric is larger. The mean signal reflects the integrated effect of the task. So, it indicates that the effect size of the tactile stimulation from the comfortable touch feeling of pliable, thin, and scroopy of the silk fabric is larger than that from the uncomfortable touch feeling of coarse of the linen fabric.

Thirdly, the details of the curves can provide important information to distinguish fabric touch. Figure 10 shows curves of mean signals in the selected 17 AAL brain regions for the two fabric samples, including 15 AAL brain regions selected from Table 2, and the other two regions, AAL-59 (Insula_L) and AAL-91 (Rolandic_Oper_L). The latter two regions are selected because the insular cortex and the SII (AAL-91 is contained in) all belong to the brain areas of bilateral response to any unilateral stimulation.

Figure 10.

The radar graph of mean signals (beta value) in the selected Automated Anatomical Labeling (AAL) brain regions.

Figure 10 shows that the results of the mean signals in the selected AAL brain regions are in accord with the activation results, as shown in Tables 2 –4. When the subjects touched the fabric samples using their fingers, the largest mean signal was detected in the brain region of AAL-81, which belongs to the SI, and small value of mean signals were detected in the cortices of the cerebellum, cingulum, and insular. Therefore, the characteristic curves, as shown in the radar graph (Figures 9 and 10), can be applied as an important reference for the recognition and characterization of the material categories, texture, tactile properties, etc., of fabric stimuli. However, more fabric samples and targeted fMRI experiments are need to carry out further studies.

Conclusion

The technology of fMRI was used in this study to observe the brain responses of the subjects when they touched two fabric samples (silk and linen) with distinct hand using their fingers in a specified way. The results show that the touch feelings of the fabrics can be characterized by the BOLD-fMRI signals of the brain. Several brain regions in the AAL atlas are linked to different fabric hand attributes. The main brain areas highly related to fabric hand are the somatosensory cortices. The region of AAL-81, which belongs to the SI, is linked to three fabric hand attributes (thin, pliable, and coarse), and the AAL-92, which belongs to the SII, is linked to the fabric hand attribute of scroopy, and it is highly related to the surface texture of fabric. Moreover, activation information of the motor-related areas (AAL-83, AAL-93, AAL-14, AAL-15, AAL-16, and AAL-19) can provide references for the hand attribute of scroopy. Although the index of mean signals reflects the integrated effect of the stimulation, the beta values in related AAL regions can provide important references for the discrimination of fabric hand.

Footnotes

Acknowledgements

The authors acknowledge all of the subjects for their participation in the fMRI experiment. The authors also express their appreciation to the editors and anonymous reviewers for their editing works and beneficial comments for this paper.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Shandong Provincial Natural Science Foundation, China (Grant No. ZR2017BEM041) and Zibo City-Shandong University of Technology Cooperative Projects.

References

Ellis

Garnsworthy

. A review of techniques for the assessment of hand. Text Res J 1980; 50: 231–238.

Ciesielska-Wrobel

Van Langenhove

. The hand of textiles - definitions, achievements, perspectives - a review. Text Res J 2012; 82: 1457–1468.

Mitsuo

. Fabric handle and its basic mechanical properties. J Text Eng 2006; 52: 1–8.

Binns

. The discrimination of wool fabrics by the sense of touch. Br J Psychol Gen Sect 1926; 16: 237–247.

Peirce

. The “handle” of cloth as a measurable quantity. J Text Inst 1930; 21: 377–416.

Kawabata S. The Development of the objective measurement of fabric handle. In: Kawabata S, Postle R and Niwa M (eds) Proceedings of the first Japan-Australia Symposium on Objective Specification of Fabric Quality, Mechanical Properties and Performance, Kyoto, 10–12 May 1982, pp. 31–59. Osaka: The Textile Machinery Society of Japan.

Boos

AGD

Tester

The FAST approach to improved fabric performance. In: Stylios

(ed). Textile objective measurement and automation in garment manufacture, New York: Ellis Horwood, 1991, pp. 55–66.

Pan

Zeronian

Ryu

. An alternative approach to the objective measurement of fabrics. Text Res J 1993; 63: 33–43.

Yim

Yang

Kan

, et al. Analysis of fabric handle of warp knitted fabrics by PhabrOmeter system. AATCC Rev 2017; 17: 36–41.

10.

Stylios

. Fabric objective measurement; FAMOUS, a new alternative to low stress measurement. Int J Cloth Sci Tech 2000; 12: 1.

11.

. A comprehensive handle evaluation system for fabrics: I. Measurement and characterization of mass and bending properties. Meas Sci Technol 2007; 18: 3547–3554.

12.

Sun

Chen

, et al. Multivariate analysis of curve parameters to predict fabric stiffness handle from a pulling-out test. Text Res J 2018; 88: 863–872.

13.

Gao

Zhou

, et al. Effects of parameters on mass index of the CHES-FY system. Fiber Polym 2014; 15: 175–180.

14.

Gao

Zhou

, et al. Determination of featured parameters to cluster stiffness handle of fabrics by the CHES-FY system. Fiber Polym 2013; 14: 1768–1775.

15.

Yeung

In vivo physiological measurements. In: Li

Wong

ASW

(eds). Clothing biosensory engineering, Cambridge: Woodhead Publishing, 2006, pp. 285–300.

16.

Pan

. Applications of human physiological responses in the sensational evaluation of textile fabrics. J Suzhou Inst Silk Text Technol 2001; 21: 12–17.

17.

Jiyong

Xin

Rubin

. A review on the cognitive study of fabric hand. J Donghua Univ Nat Sci Ed 2007; 33: 677–681. 686.

18.

Zhao

Jiang

, et al. Responses of cutaneous mechanoreceptors within fingerpad to stimulus information for tactile softness sensation of materials. Cogn Neurodyn 2013; 7: 441–447.

19.

Ogawa

Tank

Menon

, et al. Intrinsic signal changes accompanying sensory stimulation: Functional brain mapping with magnetic resonance imaging. Proc Natl Acad Sci USA 1992; 89: 5951–5955.

20.

Logothetis

Pauls

Augath

, et al. Neurophysiological investigation of the basis of the fMRI signal. Nature 2001; 412: 150–157.

21.

Horwitz

Friston

Taylor

. Neural modeling and functional brain imaging: an overview. Neural Network 2000; 13: 829–846.

22.

Ogawa

Lee

Kay

, et al. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci USA 1990; 87: 9868–9872.

23.

Disbrow

Buonocore

Antognini

, et al. Somatosensory cortex: a comparison of the response to noxious thermal, mechanical, and electrical stimuli using functional magnetic resonance imaging. Hum Brain Mapp 1998; 6: 150–159.

24.

Gelnar

Krauss

Szeverenyi

, et al. Fingertip representation in the human somatosensory cortex: an fMRI study. Neuroimage 1998; 7: 261–283.

25.

Hammeke

Yetkin

Mueller

, et al. Functional magnetic resonance imaging of somatosensory stimulation. Neurosurgery 1994; 35: 677–681.

26.

Wang

Chen

, et al. Fabric tactile perception and its brain functional research. Int J Med Radiol 2014; 37: 323–327.

27.

Wang

Chen

, et al. Brain cognitive comparison of fabric touch on human glabrous and hairy skin. Text Res J 2016; 86: 318–324.

28.

Johansson

Vallbo

. Tactile sensibility in the human hand: relative and absolute densities of four types of mechanoreceptive units in glabrous skin. J Physiol Lond 1979; 286: 283–300.

29.

Vallbo

Johansson

. Properties of cutaneous mechanoreceptors in the human hand related to touch sensation. Hum Neurobiol 1984; 3: 3–14.

30.

Fernandes

Albuquerque

. Tactual perception: a review of experimental variables and procedures. Cognit Proc 2012; 13: 285–301.

31.

Liao

, et al. Effects of contact method and acclimation on temperature and humidity in touch perception. Text Res J 2018; 88: 1605–1615.

32.

Tzourio-Mazoyer

Landeau

Papathanassiou

, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 2002; 15: 273–289.

33.

Collins

Neelin

Peters

, et al. Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assisted Tomogr 1994; 18: 192–205.

34.

Shan

Wright

Thompson

, et al. Modeling of the hemodynamic responses in block design fMRI studies. J Cereb Blood Flow Metab 2014; 34: 316–324.

35.

Friston

Ashburner

Kiebel

, et al. Statistical parametric mapping: The analysis of functional brain images, Salt Lake City: Academic Press, 2007, pp. 1–8. .

36.

Ashburner

. A fast diffeomorphic image registration algorithm. NeuroImage 2007; 38: 95–113.

37.

Kiebel

Holmes

The General Linear Model. In: Friston

Ashburner

Kiebel

, et al.(eds). Statistical parametric mapping: The analysis of functional brain images, Salt Lake City: Academic Press, 2007, pp. 101–125. .

38.

Brett M, Anton J-L, Valabregue R, et al. Region of interest analysis using an SPM toolbox [abstract]. In: The 8th International Conference on Functional Mapping of the Human Brain, Sendai, Japan, 2–6 June 2002. Available on CD-ROM in NeuroImage, 16(2), abstract 497.

39.

Matsudaira

Kawabata

. A study of the mechanical properties of woven silk fabrics. Part I: fabric mechanical properties and handle characterizing woven silk fabrics. J Text Inst 1988; 79: 458–475.

40.

Matsudaira

Matsui

. Features of mechanical properties and fabric handle of silk weaves. J Text Inst 1992; 83: 133–143.

41.

Zhu

Pan

Wang

, et al. Hand feeling style of spun silk/chitosan blended fabric. J Silk 2017; 54: 23–30.

42.

Chumbley

Friston

. False discovery rate revisited: FDR and topological inference using Gaussian random fields. Neuroimage 2009; 44: 62–70.

43.

Drevets

Burton

Videen

, et al. Blood flow changes in human somatosensory cortex during anticipated stimulation. Nature 1995; 373: 249–252.

44.

Tame

Braun

Holmes

, et al. Bilateral representations of touch in the primary somatosensory cortex. Cogn Neuropsychol 2016; 33: 48–66.

45.

Eickhoff

Schleicher

Zilles

, et al. The human parietal operculum. I. Cytoarchitectonic mapping of subdivisions. Cereb Cortex 2006; 16: 254–267.

46.

Burton

Videen

Raichle

. Tactile-vibration-activated foci in insular and parietal-opercular cortex studied with positron emission tomography: mapping the second somatosensory area in humans. Somatosens Mot Res 1993; 10: 297–308.

47.

Kitada

Hashimoto

Kochiyama

, et al. Tactile estimation of the roughness of gratings yields a graded response in the human brain: an fMRI study. Neuroimage 2005; 25: 90–100.

48.

Simoes-Franklin

Whitaker

Newell

. Active and passive touch differentially activate somatosensory cortex in texture perception. Hum Brain Mapp 2011; 32: 1067–1080.

49.

Georgopoulos

Carpenter

. Coding of movements in the motor cortex. Curr Opin Neurobiol 2015; 33: 34–39.

50.

Svoboda

. Neural mechanisms of movement planning: motor cortex and beyond. Curr Opin Neurobiol 2018; 49: 33–41.

51.

Tanji

. The supplementary motor area in the cerebral cortex. Neurosci Res 1994; 19: 251–268.

52.

Schmahmann

Doyon

McDonald

, et al. Three-dimensional MRI atlas of the human cerebellum in proportional stereotaxic space. Neuroimage 1999; 10: 233–260.

53.

Ebbesen

Brecht

. Motor cortex - to act or not to act?. Nat Rev Neurosci 2017; 18: 694–705.

54.

Manto

Bower

Conforto

, et al. Consensus paper: roles of the cerebellum in motor control-the diversity of ideas on cerebellar involvement in movement. Cerebellum 2012; 11: 457–487.

55.

Fine

Ionita

Lohr

. The history of the development of the cerebellar examination. Semin Neurol 2002; 22: 375–384.

56.

Nieuwenhuys

The insular cortex: a review. In: Hofman

Falk

(eds). Evolution of the primate brain: From neuron to behavior, San Francisco: Elsevier Science, 2012, pp. 123–163. .

57.

Grandi

Gerbella

. Single neurons in the insular cortex of a macaque monkey respond to skin brushing: preliminary data of the possible representation of pleasant touch. Front Behav Neurosci 2016; 10: 90–90.

58.

Morrison

Bjornsdotter

Olausson

. Vicarious responses to social touch in posterior insular cortex are tuned to pleasant caressing speeds. J Neurosci 2011; 31: 9554–9562.

59.

Musso

Moro

Glauche

, et al. Broca's area and the language instinct. Nat Neurosci 2003; 6: 774–781.

60.

Ranganath

Johnson

D'Esposito

. Prefrontal activity associated with working memory and episodic long-term memory. Neuropsychologia 2003; 41: 378–389.

61.

Heiser

Iacoboni

Maeda

, et al. The essential role of Broca's area in imitation. Eur J Neurosci 2003; 17: 1123–1128.

62.

Lesourd

Osiurak

Navarro

, et al. Involvement of the left supramarginal gyrus in manipulation judgment tasks: contributions to theories of tool use. J Int Neuropsychol Soc 2017; 23: 685–691.

63.

Andres

Pelgrims

Olivier

, et al. The left supramarginal gyrus contributes to finger positioning for object use: a neuronavigated transcranial magnetic stimulation study. Eur J Neurosci 2017; 46: 2835–2843.

64.

Etkin

Egner

Kalisch

. Emotional processing in anterior cingulate and medial prefrontal cortex. Trends Cogn Sci 2011; 15: 85–93.

65.

Paus

. Primate anterior cingulate cortex: where motor control, drive and cognition interface. Nat Rev Neurosci 2001; 2: 417–424.