Tactile Perception: Fine and Coarse Textures in Active Touch


The role of tactile texture information, especially roughness, and the role of movement in perceiving this roughness information are known to play a key role in recognizing objects as well as manipulating objects properly which leads to everyday task (e.g. picking up a cup). The idea of duplex theory refers that spatial cues (coarse) and the rate of vibration (fine) determines the perception of tactile texture. This therefore predicts that dynamic movement (i.e. sliding) will show better performance in fine texture than in coarse texture since dynamic movement causes vibration. The purpose of this study is to identify whether psychophysical threshold for fine and coarse texture discrimination differ during active touch (i.e. sliding or pressing) with constrained finger force. 10 female undergraduates participated age from 18 to 26 (Mage= 22.9 years) at a particular 260 um grating difference between the standard (S1) and the comparison stimulus (S2) in both fine (i.e. S1 = 220, S2 = 480) and coarse (i.e. S1 = 1100, S2 = 1360) conditions. ANOVA was conducted for statistical analysis. The main effect of roughness and movement on tactile texture discrimination performances showed significant result, however, interaction rejected our hypothesis which was with the movement fine condition would be more affected than coarse condition.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team


In many everyday tasks, we manipulate objects with our hands, for instance, controlling hand grips and identifying objects. As most natural objects have various shape and texture, tactile texture perception plays an important role in the objects’ tactile recognition (Lederman & Klatzky, 1987). A good illustration of this is, people instantly recognize familiar objects as soon as they perceive the tactile texture information of the objects without vision. For instance, finding a key in one’s pocket, as soon as he/she touches the object, the key, he or she instantly knows that the object he or she is touching or grasping is the one they are looking for. According to Johansson and Flanagan (2009), certain types of texture information especially roughness are essential in order to manipulate objects properly. For example, in the daily task of holding a glass of water, roughness information enables us to control the amount of force that needs to be applied either not to drop it (i.e. too little force) or break it (i.e. too much force). This is because roughness usually indicates higher friction and smooth, on the other hand, indicates lower friction. Therefore, the latter requires more grasp force at right angles to surface to prevent the object from slipping due to load force from object weight.

Our precise control over finger pressure derives from the highly sensitive tactile pads at the tips of the digits where a large number of cutaneous afferents provide the substrate of tactile sensation and play an important role in motor control. In the daily task of picking up a cup, for instance, tactile afferents signal key events such as the initial contact between the cup and the fingers. Thus, continuous cutaneous feedback is needed in manipulating objects to assist the motor dexterity of the hand. Also, Adams and his colleagues (2013) showed that tactile exploration to the hand movements are organized to optimize generating sense of touch, adjusting the contact and sliding forces as a function of the nature of the object and its surface features.

Early work (Katz, 1925) showed that without motion, there is no vibration therefore no perception of fine texture since vibration is the key for fine texture perception. When fingers explore and scan a fine texture surface, vibration occurs between the skin and the surface which signals the texture information. The perception of fine texture range is transferred by vibrotactile channels (Hollins & Risner, 2000). On the other hand, the perception of coarse texture range is transferred by slowly adapting type I mechanoreceptive afferents (SA I). Perceived roughness in coarse texture seemed to be a function of the spatial variation in SA I afferent firing rates (Yoshioka et al., 2001). These findings support duplex theory of texture perception which refers to the idea that fine textures are determined by rate of vibration and coarse textures are determined by spatial cues or distance between elements (Hollins & Risner, 2000).

In previous psychophysical studies, Lederman and Talyor (1972) conducted a study which put participants under conditions of active touch with unconstrained finger force in order to see the effect upon roughness perception, which showed that the greater the applied finger force was, the greater was the perceived roughness. They found that the hand speed also had a consistent effect on perceived roughness specifically fast hand speed increased the perception of roughness. However, Lederman and Tayler (1972) did not control force, speed and direction which leaves the question of roughness perception in controlled conditions unanswered. Hence, in our current study we will extend their findings by fixing the force, speed and direction in order to focus on roughness perception of fine and coarse textures in static and dynamic movements. Lederman (1981) also showed that there is no difference in coarse perception by dynamic and static movement, therefore, this will be expected in our hypotheses.

The current study aims to investigate tactile perception of fine and coarse textures with female young adults aged between 18 and 26. More specifically, the proposed study aims to discover whether thresholds for texture discrimination in fine (e.g. 220 – 480 microns) and coarse (e.g. 1100 – 1520 microns) ranges differ as a function of static (pressing) versus dynamic (sliding) touch with fixed (i) finger force, (ii) duration of applying movement and (iii) applying direction. It is hypothesized that static movement will have a higher 75% correct threshold than dynamic movement since pressing does not produce vibration which occurs impairment in texture perception. This leads to hypothesize that static movement condition in fine texture range will show the highest threshold.

Additionally, according to Peters and his colleagues (2009), physical difference between the fingers of women and men result in difference in somatosensory perception. Their study showed that tactile acuity improves with smaller finger size, which on average women has smaller fingers than men concluding that women is able to perceive more fine texture detail compared to men. Therefore, in current study, we focused on women to identify whether the previous finding (Peters et al, 2009) is applied to female population. Index fingertip size will be measured as a covariate based on the fact that people have the identical number of receptors on their hand (Bolton et al., 1966) and the previous study (Peters et al., 2009). This led us to hypothesize that the smaller the fingertip size, the high density of the receptors which will make participants be more sensitive in sense of touch, thereby lead to better performance level on tactile ability. Also, dexterity test of right hand and the measurement of cutaneous spatial resolution at the fingertip with grating (JVP) domes test will be measured as a covariate. A previous study showed a strong correlation between the grating resolution threshold and dexterity scores from one of the hand function tests (Tremblay et al., 2003). It also indicated that the strong relationship between the spatial resolution threshold and one’s dexterity shows an impaired spatial acuity at the fingertips which expects to face difficulties in detailed manipulation tasks. Therefore, these findings can lead to hypothesize a linear relationship with tactile texture discrimination performances with each of dexterity test scores and the JVP domes test scores. However, this study was conducted based on the elderly (age 60-95 years), the current study focuses on the young adults (age 18-26 years).



Gratings and Force Plate Tactile Discrimination Task

A set of rigid polyurethane gratings were used for the tactile stimuli, which were made using computer numerical control (CNC). Fine pairs with grating width of 220 um, standard stimulus (S1), and 320, 360, 400, 440, 480um, comparison stimulus (S2), and coarse pairs with grating width of 1100 um (S1) and 1200, 1280, 1360, 1440, 1520 um (S2) were presented. Both types of stimuli had ridge width of 100 um and the dimensions of each grating will be 35mm x 29.5mm. A set of gratings was positioned in parallel on a force plate which is designed to measure 3 forces and 3 moments, torques, applied to its surface as a person’s fingertip is in contact. Fx, Fy, and Fz are the 3 force components which act along the axes of orthogonal x, y, and z-coordinate system. Fx and Fy can be either horizontal or shear force component. In current study, as seen from figure 1, Fx is the horizontal force component, a force which acts in parallel direction with the line perpendicular to the gravity force direction. Fy is the shear force component, a unaligned force acting on a matter pushing it in one direction (+y) and another part in the opposite direction (-y). Lastly, Fz is the vertical force component, a force acting vertically which is orthogonal both to Fx and Fy (Chapple & Tullis, 1977). In current study, Fz is looking down axis. Mx, My, and Mz are the 3 moment components which are rotations around the x, y, and z axes. Based on the right hand rule (see appendix A for details; right hand rule – torque), positive moments are determined. Except Mx and My, Mz has a clockwise rotation around the positive direction of the z axis (Kim et al., 1999).

In order to prevent the gratings from moving, a cardboard filler was put at the leftover space on the force plate. Also, cardboard bridge was made and placed on top of the wooden plate with a pin aligned with each grating to guide the location of the stimulus since participants were not able to see the gratings (see Figure 1). Throughout the experiment, they heard white noise playing in their background from the headphones which blocked distraction and pulse sounds which helped them to get the right timing to start and finish their movement.

Cutaneous spatial resolution measurement – J.V.P domes

J.V.P domes test (Johnson & Phillips, 1981) will be used before the main experiment to look how sensitive participants’ sense of touch is which measures spatial acuity of skin surfaces. Participants will be told to report the orientation of the grooves and bars (see Appendix B). The aim in the grating orientation discrimination task is to determine the grating gap and bars widths that yield threshold performance (75% correct discrimination), a level midway between chance and perfect discrimination.

Dexterity Test – Nuts and Bolts Test

Hand Function Test (Fraser, 1981) was developed to provide an objective measurement on a standardized task with norms against which a participant’s performance can be compared. Also, it can provide a base line of performance against which the participant’s progress can be measured. It has unilateral and bilateral component; however, only unilateral components were tested (i.e. right hand). Two tasks, unscrewing (A1) and screwing (A2) the nuts from 10 fixed bolts were conducted in which their times for each task were recorded (see Appendix C).

Main Task Tactile Discrimination Task

In the main task participants were told to indicate which was rougher and the threshold for roughness discrimination was determined. First, participants were asked to wipe their hands with wet wipes we provided in order to remove dirt or liquid on their hand which can disturb their sense of touch and they were asked to wipe their hands after each block (i.e. each stimulus). Then, they were told to rest their right hand on sponges which is fixed on the wooden plate and also asked to close their eyes while doing the experiment. Throughout the experiment, background white noise was played over headphones, which last about 10 seconds per block including 8 pulse sounds. In every block they applied the controlled force (0.5 – 1.5 N) for a certain amount of time (1s) whenever they heard the high pitch, third and sixth pulses were high pitch and the others were low pitch. Cardboard screen was used to ensure to block their vision and focus on their sense of touch alone (see figure 3 & 4). As each block started, the participant conducted the discrimination task and while saying out loud the indicated rougher one, the experimenter switched the order of a set of gratings manually based on the random order provided by MATLAB programme.

Training Session

Before the actual experiment, participants were trained to get the right force (0.5N – 1.5N), the right timing (high pitch) and the duration of applying the force (1s per movement). A separate monitor screen was presented to participants which enabled them to see their force and the duration of applying force in live graph given as a training.


N = 13 were recruited from the University of Birmingham Psychology staff and students for Experiment 1 – 3, between the ages of 18 – 26 years. Prerequisites include right handed and no medical condition affecting the hands (e.g. injury, dermatitis, or palmoplantar hyperhidrosis, excessive uncontrollable sweating for palms). Their index finger of their right hand will be used to discriminate the tactile stimulus surfaces as their hand actively explore. Participants will also provide written informed consent and will be debriefed following the completion of the study.

Design & Analyses

A 2 x 2 repeated measures ANOVA (roughness x movement) will be carried out on the data for experiment 3 – main study, using variance to compare differences in means. For experiment 1 and 2 – pilot studies, statistical analysis was limited to descriptive. The independent variable will be 2 types of stimuli (fine vs coarse) and 2 types of movement (sliding vs pressing), which makes 4 conditions in total; Fine Sliding (FS), Fine Pressing (FP), Coarse Sliding (CS) and Coarse Pressing (CP). The dependent variable for the main experiment is the percentage of correct response and for the other two pilot studies is the 75% correct threshold which is the distance in the microns. Other extra measures (index finger size, J.V.P domes and dexterity test) will be used as covariates within the analyses of variation; ANCOVA.

In order to reduce the possible effects of practice or fatigue in this with-in subject design, the sample will be equally divided into two groups: one starting with fine stimuli condition and the other starting with coarse stimuli condition. Half of each group will start with dynamic movement condition followed by static movement and the other half in the opposite order. They will be asked to compare the two surfaces presented on each trial and discriminate which was rougher as stimulus pairs will be chosen randomly to the participants. Subjects will be given 10 seconds per trial to examine the surface and 12 trials per pair, which makes 60 trials for each condition (i.e. fine x sliding).

Experiment 1 – Pilot Study 1

Initial pilot study (n=1) was conductedstarting from measuring participant’s index finger size by electronic digital caliper followed by J.V.P domes test. Before the actual experiment, participantsdone the training session. Following this, participants then administered the tactile discrimination task. In the experiment, they will be asked to carry out two different tasks, sliding and pressing, in each of the two different types of gratings, fine and coarse, indicating which was rougher. Participants conducted 10 trials per stimulus which made in total 50 trials per condition. Between fine and coarse condition, they will complete the Dexterity test (A1 and A2) with a short break.

This initial pilot study led to several improvements since the total performance level was too low to able to identify the threshold at 75 percent, which indicates that tasks were too difficult for this participant in overall conclusion. On the basis of the data, the performance in coarse condition was overall better than the fine condition(percentage of MC: 60 >MF: 45), which supports our general background of this study. Also, in comparing the performance in fine and the performance in coarse in either sliding (dynamic) or in pressing (static), in coarse conditions there was not much difference (MCP: 62 andMCS: 58) which is similar as we expected. However, in the fine condition, fine pressing (MFP: 50) was higher than fine sliding (MFS: 40) which does not support our hypothesis. As in total, average % of pressing showed slightly higher than the sliding condition, however, it is not reliably different since the data has one participant.

Experiment 2 – Pilot Study 2

Second design (n=2) had improvements in four factorsin order to get the performance better. First, the grating range was changed to an easier range from fine (330 – 480 microns) and coarse (e.g. 880 – 1280 microns) ranges to fine (e.g. 220 – 480 microns) and coarse (e.g. 1100 – 1520 microns) ranges. To be specific, in both fine and coarse condition, S1 um was lowered, the step size was increased in fine condition and the range was increased in coarse condition (see Appendix E). Second, in order to prevent from fatigue and sensitization effect, the study was divided into 2-part study, specifically, first study was followed by second study within one day which they conducted 12 trials per stimulus which made in total 60 trials per condition.Third, participants were given feedback in training session in all 4 conditions so they can have an idea of which is right or wrong at the first time.

Figure 1 shows clearly that coarse is more accurate than fine condition. Also, the percentage of coarse conditions were performed higher compared to initial pilot study (MC2: 82 >MC1: 60), however, the fine conditions were yet not enough (MF2: 58) to estimate the 75% correct threshold since their performance level was too low.This result led to realise that the important part is in the middle of the range, 260 um, where participants’ performance level was close to 75% threshold and in the right region limited by 100% or by 50% chance.

Figure 1. Pilot Study 2_ Psychophysical Threshold

Experiment 3 – Main Study

10 female undergraduates were voluntarily participated (Mage= 22.9, SD =2.03) in third experiment. Procedure of the experiment was identical to the second pilot study except participants participated only at 260 um difference between S1 and S2 in both fine (i.e. S1 = 220, S2 = 480) and coarse (i.e. S1 = 1100, S2 = 1360) conditions.


The effects of roughness (coarse vs. fine) and movement (sliding vs. pressing) upon tactile texture discrimination performances at 260 um difference between S1 and S2, expressed as a proportion, was tested with a 2 x 2 repeated measures ANOVA. As predicted there was a highly significant main effect of roughness, F(1,9) = 59.145; p < .001; p2= .868, and a significant main effect of movement, F (1,9) = 555.025; p < .05; p2= .533. The interaction between roughness and movement was also significant,F (1,9) = 429.025; p < .05; p2= .704 (see Appendix F).Tactile texture performances showed better in coarse condition (M = 84) than fine condition (M = 66), while dynamic movement (i.e. sliding) led to perform better (M = 80) than when movement was static (i.e. pressing; M = 72) which supports our hypotheses. Sliding and pressing does not seem to make a difference in fine condition but does seem to make a difference in coarse condition. That is, the difference between dynamic and static movement depends on whether it is fine or coarse condition which is a two-way interaction (see Table 1; see Appendix G).However, it rejects our hypothesis which expected an interaction to show the opposite way with coarse showing no difference and fine showing difference. ANCOVA results to follow.

Table 1. Mean Percentage (%) of Correct Response out of 12 trials depending on coarse and movement conditions at 260 um difference between S1 and S2 (standard deviation in parentheses).






92.4 (8.4)

78.4 (8.9)


66.8 (8.8)

65.9 (9.1)


Adams, M. J., Johnson, S. A., Lefevre, P., Levesque, V., Hayward, V., Andre, T., & Thonnard, J.-L. (2013). Finger pad friction and its role in grip and touch. Journal of the Royal Society Interface, 10(80).

Bolton, C. F., Winkelmann, R. K., & Dyck, P. J. (1966). A quantitative study of Meissner’s corpuscles in man. Neurology, 16, 1-9.

Chapple, W. M., & Tullis, T. E. (1977). Evaluation of the forces that drive the plates. Journal of geophysical research, 82(14), 1967-1984.

Fraser, C. (1981). Hand Function Test Manual. The British Journal of Occupational Therapy, 44(8).

Hollins, M., & Risner, S. R. (2000). Evidence for the duplex theory of tactile texture perception. Perception & psychophysics, 62(4), 695-705.

Johansson, R. S., & Flanagan, J. R. (2009). Coding and use of tactile signals from the fingertips in object manipulation tasks. Nature Reviews Neuroscience, 10(5), 345-359.

Johnson, K. O., & Phillips, J. R. (1981). Tactile spatial resolution. I. Two-point discrimination, gap detection, grating resolution, and letter recognition. Journal of neurophysiology, 46(6), 1177-1192.

Katz, J. R. (1925). Rontgen spectographic testings on expanded rubber and its possible relevance for the problem of the extension characteristics of this substance. Naturwissenschaften, 13, 410-416.

Kim, G. S., Kang, D. I., & Rhee, S. H. (1999). Design and fabrication of a six-component force/moment sensor. Sensors and Actuators A: Physical, 77(3), 209-220.

Lederman, S. J., & Taylor, M. M. (1972). Fingertip force, surface geometry, and the perception of roughness by active touch. Perception & Psychophysics, 12(5), 401-408.

Lederman, S. J. (1981). The perception of surface roughness by active and passive touch. Bulletin of the Psychonomic Society, 18(5), 253-255.

Lederman, S. J., & Klatzky, R. L. (1987). Hand movements: A window into haptic object recognition. Cognitive psychology, 19(3), 342-368.

Peters, R. M., Hackeman, E., & Goldreich, D. (2009). Diminutive digits discern delicate details: fingertip size and the sex difference in tactile spatial acuity. Journal of Neuroscience, 29(50), 15756-15761.

Tremblay, F., Wong, K., Sanderson, R., & Cote, L. (2003). Tactile spatial acuity in elderly persons: assessment with grating domes and relationship with manual dexterity. Somatosensory & motor research, 20(2), 127-132.

Yoshioka, T., Gibb, B., Dorsch, AK., Hsiao, SS.,& Johnson, KO. (2001). Neural coding mechanisms underlying perceived roughness of finely textured surfaces. J Neurosci.


Appendix A

Right Hand Rule – Torque

Figure 2. It is conventional to choose the direction in the right hand rule direction along the axis of rotation (Further explanations are written in the figure).

Appendix B

Figure 4. JVP domes

Appendix C

Figure 5. Dexterity Test – Nuts and Bolts Test

Appendix D

Figure 6. Overview of the experiment. The participant is resting their right hand on sponges which is fixed on the wooden plate and asked to close their eyes while doing the experiment.

Finger at first pin (2) sliding the first stimulus

(3) sliding the second stimulus (4) sliding the second stimulus

(5) resting hand

Figure 7. Training Session: participants can see a live graph given as a feedback to get familiar with the right force (i.e. target force) which is given by a blue box and the right timing (1s). They were trained in all 4 conditions (FS, FP, CS and CP).

(1) resting hand

(3) pressing for 1s as they hear the high pitch (first one)

(5) pressing for 1s as they hear the high pitch (second one) (6) resting hand

Figure 8. Pictures are shown a procedure of the Main task – Tactile discrimination task. This case it is the pressing condition. (See Figure 7 for sliding condition).

Appendix E

Table 2. Changed Gratings Range from Pilot study 1 to Pilot Study 2















































Appendix F

2?2 repeated measures ANOVA

Within-Subjects Factors

Measure: MEASURE_1



Dependent Variable











Descriptive Statistics


Std. Deviation


Coarse_Sliding(%) (out of 12)
















Multivariate Testsa




Hypothesis df

Error df

You Might Also Like

I'm Alejandro!

Would you like to get a custom essay? How about receiving a customized one?

Check it out