towell@sztuowei.com
Get in touch.
Dear,I will reply in 12 hours. All your message are protected!
Integration of Length and Curvature in Haptic Perception
by：Tuowei
2019-09-09
We study whether and how length and curvature information are integrated when exploring objects on the one hand.
Subjects were asked to explore four objects between the thumb and the index finger.
Objects vary in terms of length, curvature, length, and curvature, which are related to circles, or inversecorrelated.
We found that performance was significantly better when both length and curvature were present than when only one of the two cues was available.
Therefore, we conclude that there are integral points in length and curvature.
Also, if the two threads are related in the loop crossover
Part not in one
In a related way, the performance is better than the combined prediction of two independent cues.
Our conclusion is that when the clues in the object are combined into a circle, the integration of curvature and length is very effective, which is the most common combination of curvature and length in daily life.
12 charging subjects (
Average age 21 ± 2 years, 4 males)
Participated in the experiment.
All subjects reported correctlyhanded.
All subjects were naive about the purpose of the experiment and agreed.
None of the subjects reported any known hand defects.
The project to carry out these experiments was approved by the ethics committee of the faculty of human movement sciences, VU University, Amsterdam.
Stimulus print in Z-
The company\'s Z450 3D printer, by combining gypsum composites with epoxy.
The resolution of this printer is 300 by 450 dpi and the layer thickness is 0. 089–0. 102u2005mm. The cross-
Part of the stimulus, I. e.
The curvature and length are printed on the horizontal plane, thus rendering at 450 resolution through 300 dpi.
This is more than 10 points per mm, which is enough for the measurement threshold.
The printed objects become stronger by soaking them briefly in liquid strong glue and then wiping them dry.
Finally, the surface to be felt in the experiment is slightly polished until smooth.
All stimuli are 30mm long and 20mm high.
The third dimension is described below.
To put them on the stand, there is a square hole in all the stimuli.
This ensures that the stimulus is fixed, but can be switched easily.
All the different stimuli can be seen inside.
We used four stimuli.
In Condition 1, the stimulus has a flat surface of 20. 5, 21. 0, 21. 5, 22. 0, and 22.
The interval between test stimulation and 20 was 5mm.
0 stimulus calculated by mm.
In condition 2, for reference stimuli and 97, the stimulus has a surface with a curvature of 100 µm. 6, 95. 2, 93. 0, 90. 9 and 88.
The test stimulus was 9 µm.
In this case, the maximum length of the stimulus is 20. 0u2005mm.
In Condition 3, the stimulus is part of a cylinder with a circular cross
The same part as the length diameter of the test and reference stimulus in condition 1.
The curvature is the same as in condition 2.
For Condition 4, we designed
Associated length and curvature.
This means that the curvature is also large for a large length.
This is completely contrary to condition 3, where small lengths are combined with large curvature.
This means that the length is 20 for the reference stimulus.
0 u2005 mm, but the curvature is equal to the curvature of the cylinder with a diameter of 22. 5u2005cm, i. e. 88. 9u2005m.
For the test stimulus, we combined the length of 20. 5, 21. 0, 21. 5, 22. 0, 22.
5mm with a curvature of 90. 9, 93. 0, 95. 2, 97.
6 and 100 m, respectively.
The length difference of the above stimuli is less than 0.
05mm with a difference of less than 0 between the indicated length and curvature.
The distance indicates a curvature of 5 µm.
Taking into account the differences required to measure the threshold, this accuracy is sufficient.
The exact method of verifying the stimulus size is described in.
Two instances are created for each stimulus.
This ensures that subjects cannot use minor differences in textures for example to distinguish between two shapes, but only on shapes.
The subject was blindfolded and sat in a chair.
The stimulus to be felt is placed on a stand that has a 45 degree angle to the edge of the table, 28 cm from the edge of the table (see ).
This ensures that the subject can catch the stimulus comfortably without having to bend his wrist.
The following process of presenting subjects with stimuli is based on the methods used by Durlach et al.
Determine the value that distinguishes small differences in length.
In this approach, we did not inform the subject of any difference in length or curvature to be concerned.
This ensures that we can directly compare the performance of subjects under different conditions.
This experiment is one-off. Forced interval-
Two stimuli were selected for the experiment and the subject (
Reference and testing)
At the beginning of the experiment, numbers 1 or 2 were assigned to two stimuli.
Whether the first one is a test or a reference stimulus is randomly assigned between blocks.
Subjects were asked to feel the stimulation between the thumb and the index finger and only focus on the shape of the stimulus.
They didn\'t get further tips on what would be different about the shape.
After feeling the two stimuli at the beginning of the experiment, the subjects felt one stimulus at a time and must indicate whether they thought it was number 1 or number 2.
After each answer, the experimenter will provide feedback.
In this type of experiment, feedback is important because it is easy for subjects to lose reference.
Give feedback to ensure that the subject is constantly reminded of which stimulus is named number 1 and which is named number 2.
Repeat 60 times for each pair of tests and reference stimuli for each condition.
The first 10 trials were practical and the last 50 were used for analysis.
The subject was not notified.
We tested five differences between reference and test stimuli under each condition.
Each condition was tested in an hour, with a total of 4 hours per subject.
By constructing a Latin rectangle, the order of the conditions and the Order of the test stimuli within the conditions are as balanced as possible in the participants.
This rectangle is created by first creating a Latin square of 4 by 4.
Another Latin square is constructed, and the first line of the previous Square moves to the left.
The same procedure was performed for the five differences between the reference and the test stimulus.
For the difference, a Latin square of 5 by 5 is constructed and the last two lines of the last Latin square are missed.
In 60 trials, the subjects received either of the two repeated reference stimuli (
Same two shapes)
Or one of the two repeated test stimuli.
This is random.
Two identical stimuli are presented 15 times respectively, which means 30 times for each shape.
In addition, 5 times were presented for each shape in the first 10 trials.
For each run of the 50 trials, we calculated the \'of each run \'.
Is a sensitivity measurement method that takes into account the situation where the subject indicates to detect the signal when there is no signal (false alarm).
In our case, they show that they actually feel the stimulation while they actually feel it.
By the following equation, the \"value\" is calculated from the correct recognition score of stimulus 1 and the incorrect recognition score of Stimulus 2: Where (), [0, 1]
It is the reciprocal of the cumulative Gaussian distribution, and the score subjects indicate sensory stimulation 1 at the time of sensory stimulation 1, and the score subjects indicate sensory stimulation 1 at the actual sensory stimulation 2.
We draw the mean values based on the reference differences for each condition.
Statistical analysis of \'values.
First, we check the degree of the ball.
If the ball is violated, we use the greenhouse-
Geisser fixed the value. A 4 (condition)by 5 (
Difference with reference)
Repeated measurement analysis of variance was performed.
Since the effect of the difference with the reference is insignificant, that is, as the length increases, \'will increase, we only looked more closely at the conditions under which Bonferroni corrected the planned paired comparisons between conditions 1-3, 2-3, 1-4, 2-4 and 3-4
When the subject senses an object containing two clues, the value of length and curvature is irrelevant and one can treat the two clues as orthogonal.
In this case, one can predict a\' by making a quadratic summation of the values of \'ofand\' to distinguish two objects containing length and curvature: in this equation, \'represents the prediction of an object containing length and curvature\', \'represents what is found in condition 1 \'()
And \'found in representative condition 2 \'().
We calculated this for each topic based on the differences in the references.
This leads to the five mean values of the predicted values, which can be compared with the measured values of conditions 3 and conditions 4.
In both cases, the length and curvature are related within the stimulus.
By comparing predictive values and measurements, we can check whether they are relevant in the subject\'s perception.
Subjects were asked to explore four objects between the thumb and the index finger.
Objects vary in terms of length, curvature, length, and curvature, which are related to circles, or inversecorrelated.
We found that performance was significantly better when both length and curvature were present than when only one of the two cues was available.
Therefore, we conclude that there are integral points in length and curvature.
Also, if the two threads are related in the loop crossover
Part not in one
In a related way, the performance is better than the combined prediction of two independent cues.
Our conclusion is that when the clues in the object are combined into a circle, the integration of curvature and length is very effective, which is the most common combination of curvature and length in daily life.
12 charging subjects (
Average age 21 ± 2 years, 4 males)
Participated in the experiment.
All subjects reported correctlyhanded.
All subjects were naive about the purpose of the experiment and agreed.
None of the subjects reported any known hand defects.
The project to carry out these experiments was approved by the ethics committee of the faculty of human movement sciences, VU University, Amsterdam.
Stimulus print in Z-
The company\'s Z450 3D printer, by combining gypsum composites with epoxy.
The resolution of this printer is 300 by 450 dpi and the layer thickness is 0. 089–0. 102u2005mm. The cross-
Part of the stimulus, I. e.
The curvature and length are printed on the horizontal plane, thus rendering at 450 resolution through 300 dpi.
This is more than 10 points per mm, which is enough for the measurement threshold.
The printed objects become stronger by soaking them briefly in liquid strong glue and then wiping them dry.
Finally, the surface to be felt in the experiment is slightly polished until smooth.
All stimuli are 30mm long and 20mm high.
The third dimension is described below.
To put them on the stand, there is a square hole in all the stimuli.
This ensures that the stimulus is fixed, but can be switched easily.
All the different stimuli can be seen inside.
We used four stimuli.
In Condition 1, the stimulus has a flat surface of 20. 5, 21. 0, 21. 5, 22. 0, and 22.
The interval between test stimulation and 20 was 5mm.
0 stimulus calculated by mm.
In condition 2, for reference stimuli and 97, the stimulus has a surface with a curvature of 100 µm. 6, 95. 2, 93. 0, 90. 9 and 88.
The test stimulus was 9 µm.
In this case, the maximum length of the stimulus is 20. 0u2005mm.
In Condition 3, the stimulus is part of a cylinder with a circular cross
The same part as the length diameter of the test and reference stimulus in condition 1.
The curvature is the same as in condition 2.
For Condition 4, we designed
Associated length and curvature.
This means that the curvature is also large for a large length.
This is completely contrary to condition 3, where small lengths are combined with large curvature.
This means that the length is 20 for the reference stimulus.
0 u2005 mm, but the curvature is equal to the curvature of the cylinder with a diameter of 22. 5u2005cm, i. e. 88. 9u2005m.
For the test stimulus, we combined the length of 20. 5, 21. 0, 21. 5, 22. 0, 22.
5mm with a curvature of 90. 9, 93. 0, 95. 2, 97.
6 and 100 m, respectively.
The length difference of the above stimuli is less than 0.
05mm with a difference of less than 0 between the indicated length and curvature.
The distance indicates a curvature of 5 µm.
Taking into account the differences required to measure the threshold, this accuracy is sufficient.
The exact method of verifying the stimulus size is described in.
Two instances are created for each stimulus.
This ensures that subjects cannot use minor differences in textures for example to distinguish between two shapes, but only on shapes.
The subject was blindfolded and sat in a chair.
The stimulus to be felt is placed on a stand that has a 45 degree angle to the edge of the table, 28 cm from the edge of the table (see ).
This ensures that the subject can catch the stimulus comfortably without having to bend his wrist.
The following process of presenting subjects with stimuli is based on the methods used by Durlach et al.
Determine the value that distinguishes small differences in length.
In this approach, we did not inform the subject of any difference in length or curvature to be concerned.
This ensures that we can directly compare the performance of subjects under different conditions.
This experiment is one-off. Forced interval-
Two stimuli were selected for the experiment and the subject (
Reference and testing)
At the beginning of the experiment, numbers 1 or 2 were assigned to two stimuli.
Whether the first one is a test or a reference stimulus is randomly assigned between blocks.
Subjects were asked to feel the stimulation between the thumb and the index finger and only focus on the shape of the stimulus.
They didn\'t get further tips on what would be different about the shape.
After feeling the two stimuli at the beginning of the experiment, the subjects felt one stimulus at a time and must indicate whether they thought it was number 1 or number 2.
After each answer, the experimenter will provide feedback.
In this type of experiment, feedback is important because it is easy for subjects to lose reference.
Give feedback to ensure that the subject is constantly reminded of which stimulus is named number 1 and which is named number 2.
Repeat 60 times for each pair of tests and reference stimuli for each condition.
The first 10 trials were practical and the last 50 were used for analysis.
The subject was not notified.
We tested five differences between reference and test stimuli under each condition.
Each condition was tested in an hour, with a total of 4 hours per subject.
By constructing a Latin rectangle, the order of the conditions and the Order of the test stimuli within the conditions are as balanced as possible in the participants.
This rectangle is created by first creating a Latin square of 4 by 4.
Another Latin square is constructed, and the first line of the previous Square moves to the left.
The same procedure was performed for the five differences between the reference and the test stimulus.
For the difference, a Latin square of 5 by 5 is constructed and the last two lines of the last Latin square are missed.
In 60 trials, the subjects received either of the two repeated reference stimuli (
Same two shapes)
Or one of the two repeated test stimuli.
This is random.
Two identical stimuli are presented 15 times respectively, which means 30 times for each shape.
In addition, 5 times were presented for each shape in the first 10 trials.
For each run of the 50 trials, we calculated the \'of each run \'.
Is a sensitivity measurement method that takes into account the situation where the subject indicates to detect the signal when there is no signal (false alarm).
In our case, they show that they actually feel the stimulation while they actually feel it.
By the following equation, the \"value\" is calculated from the correct recognition score of stimulus 1 and the incorrect recognition score of Stimulus 2: Where (), [0, 1]
It is the reciprocal of the cumulative Gaussian distribution, and the score subjects indicate sensory stimulation 1 at the time of sensory stimulation 1, and the score subjects indicate sensory stimulation 1 at the actual sensory stimulation 2.
We draw the mean values based on the reference differences for each condition.
Statistical analysis of \'values.
First, we check the degree of the ball.
If the ball is violated, we use the greenhouse-
Geisser fixed the value. A 4 (condition)by 5 (
Difference with reference)
Repeated measurement analysis of variance was performed.
Since the effect of the difference with the reference is insignificant, that is, as the length increases, \'will increase, we only looked more closely at the conditions under which Bonferroni corrected the planned paired comparisons between conditions 1-3, 2-3, 1-4, 2-4 and 3-4
When the subject senses an object containing two clues, the value of length and curvature is irrelevant and one can treat the two clues as orthogonal.
In this case, one can predict a\' by making a quadratic summation of the values of \'ofand\' to distinguish two objects containing length and curvature: in this equation, \'represents the prediction of an object containing length and curvature\', \'represents what is found in condition 1 \'()
And \'found in representative condition 2 \'().
We calculated this for each topic based on the differences in the references.
This leads to the five mean values of the predicted values, which can be compared with the measured values of conditions 3 and conditions 4.
In both cases, the length and curvature are related within the stimulus.
By comparing predictive values and measurements, we can check whether they are relevant in the subject\'s perception.
Custom message
