Presented is the tactile semiautomated passive-finger angle stimulator TSPAS, a new way to assess tactile spatial acuity and tactile angle discrimination using a computer-controlled tactile stimulus system that applies raised angle stimuli to a subject's passive fingerpad, while controlling for movement speed, distance, and contact duration.
Passive tactile perception is the ability to passively and statically perceive stimulus information coming from the skin; for example, the ability to sense spatial information is the strongest in the skin on the hands. This ability is termed tactile spatial acuity, and is measured by the tactile threshold or discrimination threshold. At present, the two-point threshold is extensively used as a measure of tactile spatial acuity, although many studies have indicated that critical deficits exist in two-point discrimination. Therefore, a computer-controlled tactile stimulus system was developed, the tactile semiautomated passive-finger angle stimulator (TSPAS), using the tactile angle discrimination threshold as a new measure for tactile spatial acuity. The TSPAS is a simple, easily operated system that applies raised angle stimuli to a subject's passive fingerpad, while controlling movement speed, distance, and contact duration. The components of the TSPAS are described in detail as well as the procedure to calculate the tactile angle discrimination threshold.
Touch perception is a fundamental form of the sensations processed by the somatosensory system, including haptic perception and tactile perception. Passive tactile perception, as opposed to active exploration, means that the object is moved to make contact with static skin1,2. As in other senses, spatial resolution in tactile perception, also termed tactile spatial acuity, is usually represented by the tactile threshold, detection threshold, or discrimination threshold2,3. In the past 100 years, the two-point threshold has commonly been used as a measure of tactile spatial acuity4. However, many studies have indicated that the two-point threshold is an invalid index of tactile spatial ability because two-point discrimination (TPD) cannot exclude nonspatial cues (e.g., if two points are too close, they may locate a single afferent receptive field, which readily evokes increased neural activity) and maintain a stable criterion for responses3,4,5. Owing to the number of drawbacks of TPD, several new and promising methods have been developed as replacements, such as tactile grating orientation (GO)3,6, two-point orientation discrimination5, raised letter recognition, gap detection7, dot patterns, Landolt C rings8, and angle discrimination (AD)9,10. At present, because of the advantages in operating GO, as well as the spatial structure and complexity of the stimulus used, GO is increasingly used to measure tactile spatial acuity11,12,13.
Although tactile GO is thought to rely on underlying spatial mechanisms, thereby yielding a reliable measure of tactile spatial acuity, it is still debated whether GO performance is partly affected by nonspatial cues14 (e.g., intensive signs that may provide a cue to identify the difference between orientation stimuli). Additionally, GO only consists of simple spatial orientation (i.e., horizontal and vertical) tasks and primarily involves sensory processing, which limits its use when exploring the hierarchical interplay between tactile primary processing in the primary somatosensory cortex and tactile advanced possessing involving the posterior parietal cortex (PPC) and supramarginal gyrus (SMG)15,16,17. To compensate for these drawbacks, tactile AD was developed to measure tactile spatial acuity9,10. In AD, a pair of angles passively slide across the fingertip. The angles vary in size, and the subject needs to determine which of the angles is larger. To consistently accomplish this task, spatial features of tactile angles must be represented and stored in the working memory and then compared and discerned. Therefore, tactile AD involves not only primary processing but also advanced cognition of tactile perception, such as working memory and attention.
As in a variety of line orientation perception tests, in tactile AD the subject is presented successively with one reference angle and one comparison angle and is asked to indicate which is the larger angle18,19,20,21. The lines composing the angles are equal in length and symmetrically distributed along an imaginary bisector. By symmetrically changing the spatial dimensions of the lines, all types of raised plane angles can be created. Therefore, a critical advantage of this method is that the angles being differentiated have similar spatial structures. In addition, the spatial representation gained in the AD is more sequential than that gained in GO. However, the AD threshold provides evidence that tactile spatial acuity is sufficient to allow spatial discrimination between objects22. Furthermore, the tactile spatial perception of the angle may be experienced from point to line and finally form a two-dimensional plane angle in which nonspatial cues may play only a small role.
The AD threshold was found to increase with increasing age, which might result from the need for high cognitive load in the tactile AD task. Thus, it may provide a monitoring mechanism in cognitive impairment diagnosis9,10. Although AD performance is affected by age-related decline, it can be significantly improved in young people by continuous training or similar tactile task training23. Furthermore, fMRI studies showed that a delayed match-to-sample tactile angle task activated certain cortical regions responsible for working memory, such as the posterior parietal cortex17,24. These findings suggest that tactile angle discrimination is a promising measure for tactile spatial acuity involving advanced cognition. Here, the tactile AD equipment and its use is described in detail. Other tactile researchers can reproduce the AD equipment and use it in their research.
The tactile AD equipment, or tactile semiautomatic passive-finger angle stimulator (TSPAS), uses an electronic slide to convey a pair of angle stimuli to slide passively across the skin (Figure 1). The subjects' arms lie comfortably, prostrate on a tabletop. The right hand sits on a hand plate in the table, and an index fingerpad is situated slightly below the opening of the plate. Computer software can control the slide, move it at a fixed speed, and move it forward and backward. As the slide moves forward, the angle stimuli slide passively across the skin at a fixed speed starting at the fingertip. When the slide moves backward to its starting position and changes to another pair of angle stimuli, the subject needs to lift the index finger up and wait for an order to lightly place it again at the opening. Thus, the equipment presents tactile angle stimuli at a controlled speed, stable contact duration, and constant interstimulus interval. The subject orally reports a sequence number, and the experimenter registers it as a response and proceeds to conduct the next trial.
Figure 1: Overview of the TSPAS.
The equipment consists of four parts: 1) tactile angle stimuli (i.e., the reference angle and ten comparison angles); 2) the hand plate that fixes the hand of the subject in place and keeps only the index finger in contact with the stimuli; 3) the electronic slider that carries the tactile stimuli; and 4) the personal computer (PC) control system that controls the speed and the movement distance of the electronic slide. Please click here to view a larger version of this figure.
Written informed consent was obtained from the subjects in compliance with the policies of the local medical ethics committee of Okayama University. The testing procedures gained review and consent from the local medical ethics committee of Okayama University.
1. Detailed composition and function of equipment
Figure 2: Example of tactile angle stimuli.
(A) An example of the reference angle (60°) and two (50° and 70°) of the ten comparison angles used in the experiment. In particular, detailed parameters of the reference angle were drawn. d represents the end point distance, R represents the radius of curvature in the local apex, and r represents the radius of curvature in the end point. (B) Example of a raised angle seen in 3D. The height of the raised line is 1.0 mm from the 3D view. Please click here to view a larger version of this figure.
Figure 3: Hand position of the subject and tactile angle stimuli movement direction.
The right hand of the subject was secured with nylon tape, and the subject was instructed to place his or her right index finger into the opening in the plate. The angle stimuli were clamped on the apparatus and were horizontally moved by the electronic slide to passively slide across the fingerpad. Please click here to view a larger version of this figure.
2. Running an experiment
In this study, the 3AFC (3-alternative forced-choice) technique and the logistic curve were used to estimate the tactile AD threshold. Participants were instructed to orally report the larger of the two angles perceived, or if they did not detect the difference, they could indicate the same. The equation of the logistic curve, which has been commonly applied to psychophysical experiments to measure thresholds27,28,29 is:
In this equation, there are two key parameters, α and β. β is representative of the logistic curve growth, and –α/β represents the X value of the logistic curve midpoint.
To apply the logistic curve to describe the AD threshold, the 3AFC result must be expressed as a frequency distribution, shown as a black square in Figure 4. Therefore, when the reference angle was less than the comparison angle, the same responses answered were divided into two: one half was added to correct judgment and the other to incorrect, and the revised correct responses were then transferred to the rate. When the reference angle was greater than the comparison angle, the same steps were taken as previously indicated, and the revised rate was reduced by 1. Through these steps, a coordinate system was set, with the degree of the angle representing the horizontal axis and the vertical axis representing the proportion of responses in which the comparison angle was perceived to be greater than the reference angle (Figure 4). In this coordinate, a logistic curve could be fitted by the least square method. The AD threshold was defined as half of the difference between the angle at accuracy rates of 25% and 75%.
Figure 4: Logistic curve fit.
The accuracy data of one subject in the AD task were used to fit the logistic curve using the least square method. The black squares represent the revised rates of one subject who completed the tactile AD task. The solid line is representative of the logistic curve acquired through the least square method when the residual was the smallest. Dashed lines indicate two points (A1, 0.25) and (A2, 0.75), and the AD threshold is (A2-A1)/2. After fitting the logistic curve, the specific parameters were obtained (α = 21.40, β = -0.35) and the AD threshold was calculated (3.51°). Please click here to view a larger version of this figure.
To test whether this curve was accurate, the goodness-of-fit for the logistic curve was evaluated using a chi-squared test, which was used to determine whether there was a significant difference between the observed rates and the expected rates (i.e., the values in the fitted logistic curves). Here, the null hypothesis states that there is no significant difference between the observed and expected values. The value of the chi-squared test was determined using the following formula:
In this equation, O = observed value, and E = expected value.
To test whether the null hypothesis could be rejected, a significance level of 15% was chosen as the cutoff criterion28 and the critical value was calculated (χ2 (8)0.15 = 12.03). Because there were 10 categories and the mean and standard deviation were used to fit the data to a logistic curve, there were 8 degrees of freedom (10–2). Thus, if the value from the chi-squared test of the logistic curve was larger than this critical value, the null hypothesis was rejected. The value (2.14) from the chi-squared test was smaller than this critical value (12.03), which indicates that the logistic curve fitting was suitable.
A new measure for tactile spatial acuity, tactile AD, is presented. In this system a pair of angles passively slides across the immobilized index fingerpad of a subject. AD combines the advantages of GO and TPD, reducing the impact of intensive cues and the neural peak impulse rate of a single point. This study shows that there is a gradual change in perceptual discrimination as the angle difference changes between the reference angle and the comparison angle4. In addition to the age effect, training effect, and cognitive impairment diagnosis monitoring of AD9,10,23, tactile AD is a valuable measure for tactile spatial acuity. Its variability needs to be verified in further studies, however. For example, tactile AD should correlate with other validated measures of tactile spatial acuity such as pattern or braille letter discrimination7,8.
Like other methods measuring tactile spatial perception, AD applies the threshold to measure angle discriminability. Unexpectedly, the smaller the angle discrimination threshold, the stronger the angle discriminability. In previous studies, an interpolation method was used to pinpoint the threshold value9,10. Although the method does not need to assume that the subject’s behavior is captured using a psychometric function, it only fits data of a half-size range of the comparison angles. In the current experiment, to cover the entire range of comparison angles, the logistic curve was used to calculate the threshold27,29. Because half of the comparison angles are smaller than the reference angle and the other half are larger than the reference angle, the current method can fit all data points once and calculate the angle discrimination threshold. The goodness-of-fit for the logistic curve was evaluated using a chi-squared test and the logistic curve fitting was found suitable28.
To conduct the AD experiments using the TSPAS system, the following points should be noted: First, because TSPAS is a semiautomatic system using PC software, it is necessary to verify again that the slide can move at the speed and distance set before the experiment. Second, it is necessary to determine whether or not the subject is awake during the experiment. Because the subject wears an eye mask during the experiment, he or she can easily become sleepy. In this case, the subject may miss some information and make an incorrect decision. Third, the enforced breaks are also necessary. If the fingerpad of the subject continues to be stimulated for a long time, the fingerpad may adapt to the raised angle stimulus and it may be hard for the subject to distinguish the difference between angles. Or the long period of stimulation may cause uncomfortable sensations in the fingerpad. Therefore, the number of trials and breaks should be strictly controlled.
The current characteristics of TSPAS and the range of tactile angles could limit the range of people tested. Therefore, TSPAS needs to use different ranges of tactile angles for different groups of people to measure their tactile spatial acuity. For example, because older people have a far bigger AD threshold than younger people9,10, the current range of tactile angles used in TSPAS cannot measure their AD threshold. Additionally, for those individuals whose fingerpads cannot completely feel the tactile angles, TSPAS is not valid at all, because they cannot envision the tactile angle by the passive sliding across their fingerpads. The difference in tactile spatial acuity between female and male subjects25 must be kept in mind as well. Future projects may need a lot of modification to determine the range of tactile angles to use for different groups of people in clinical use.
Although TSPAS can control the moving speed and distance of angle stimuli well, the manual delivery of angle stimuli is time-consuming and requires considerable attention and concentration on the part of the experimenter6. To eliminate these shortcomings with manual operations, a fully automatic tactile AD system was designed. The purpose of developing automatic equipment is to establish uncomplicated, efficient, and affordable equipment for controlled tactile angle applications. However, a remaining challenge is how the equipment can precisely and quickly adjust various angle sizes in a very short time. Hopefully the AD system described will be used and verified by others and promote the movement towards automatic tactile testing.
The authors have nothing to disclose.
This work was supported by the Japan Society for the Promotion of Science KAKENHI Grants JP17J40084, JP18K15339, JP18H05009, JP18H01411, JP18K18835, and JP17K18855. We also thank the technician (Yoshihiko Tamura) in our laboratory for helping us craft the raised angle.
Acrylic sheet (3 mm) | MonotaRO Co.,Ltd. | 33159874 | Good Material |
Acrylic sheet (1 mm) | MonotaRO Co.,Ltd. | 45547101 | Good Material |
EZ limo (easy linear motion motor) | ORIENTAL MOTOR CO., LTD. Made in Japan | EZS3 | Good Motorized Linear Slides |
Data Editing Software | ORIENTAL MOTOR CO., LTD. Made in Japan | EZED2 | easy to use |
Operating Manual (Orientalmotor) | ORIENTAL MOTOR CO., LTD. Made in Japan | HL-17151-2 | Good Guidebook |