| | Development of the Hand Active Sensation Test: Reliability and ValidityPresented in part to the American Physical Therapy Association’s Combined Sections Meeting, February 13, 2004, Nashville, TN. Abstract Williams PS, Basso M, Case-Smith J, Nichols-Larsen DS. Development of the Hand Active Sensation Test: reliability and validity. ObjectiveTo develop and establish the reliability and validity of a new quantitative functional measure of haptic perception in the hand, the Hand Active Sensation Test (HASTe). DesignReliability was assessed by test-retest sessions. Validity was assessed via discriminant analysis, concurrent validity with 2-point discrimination and wrist position test, and receiver operating characteristic (ROC) curve construction. SettingSubject preference. ParticipantsHeterogeneous sample of 28 stroke survivors and 28 individually matched controls. InterventionSubjects used 1 hand to manipulate HASTe objects that vary by weight or texture to complete 18 match-to-sample trials. Main Outcome MeasuresTwo-point discrimination threshold, Wrist Position Sense Test (WPST) average error, and HASTe accuracy score. ResultsTest-retest reliability was strong (intraclass correlation coefficient model 3,1=.77). The HASTe score significantly discriminated the groups (t=8.3, P<.001) and correlated with 2-point discrimination (r=−.67, P<.001) and WPST (r=−.60, P<.001). ROC curve area was .94 for test 1 and .92 for the average of 2 tests. ConclusionsThe HASTe is a reliable and valid functional measure of haptic perception, appears to detect impairment of haptic perception even in stroke survivors with no reported sensory deficits, and may provide valuable quantitative clinical data about complex sensory loss and hand function after stroke. IF THE PRINCIPLE FUNCTIONAL use of the upper extremities is to reach for and to manipulate objects, then the hand can be considered as both an effector for the motor system and a receptor for the sensory system.1, 2, 3 When the hand acts as an effector, manipulation is used to perform a physical task with an object such as using a hammer to hit a nail. As a receptor, the hand operates as a perceptual unit by using manipulation to explore an object and extract salient object properties such as the handle’s texture or the hammer’s weight. Active sensation, or haptic touch, involves purposely using the actions of the hands, during object contact, for the purpose of somatosensory perception.3 The movements selected optimize the relevant somatosensory receptors to effectively gather the pertinent sensory qualities of the object being explored.4 Object properties refer to characteristics or dimensions of an object including its size, shape, weight, texture, surface compliance, and temperature. These particular object properties are reported to be the basis for object recognition but also influence the organization of upper-extremity movements with objects.5, 6 Weight and texture, in particular, have been reported to influence grip and load forces during grasp and lift.7 Haptic receptors include skin receptors that come into direct contact with objects as well as proprioceptors found in the skin, muscles, tendons, and joints that are activated during manual exploration.2, 8 Thus, active sensation is the synthesis of inputs from cutaneous and proprioceptive receptors acting as a single functional perceptual system.9, 10, 11 Exploratory procedures of the hand are stereotypical movements used for obtaining information about object properties.4 An exploratory procedure is associated with an object property when it is used preferentially under free exploration conditions and if it extracts that property accurately and quickly.12 A repetitive rubbing motion across the object surface or lateral motion exploratory procedure is optimal for texture properties, whereas hefting or lifting an object, the exploratory procedure of unsupported holding, is used for object weight. These specific motions are believed to be necessary to accurately detect the sensory qualities of weight and texture in objects. Although these motions are specific to particular object properties, gross sensory information about these and other object properties can also be gathered by haptic receptors active during initial grasp and lift.13 In the context of hand movement for sensory exploration, grasp and lift can be considered as the minimum motor ability required to explore an object; specific exploratory procedures, requiring more sophisticated hand movements, are incorporated to gather additional sensory information to define needed object properties for the desired task.3 Stroke is the leading cause of serious, long-term disability in the United States.14 Nearly 4.5 million stroke survivors are alive today, with up to 60% living with moderate to severe levels of disability. The specific losses of function after stroke depend on the site and size of the lesion. The primary motor and somatosensory impairments associated with the stroke lesion may disable upper-extremity interactions with objects in the environment. Because of the role somatosensory inputs play during object manipulation, most clinicians complete a somatosensory assessment to evaluate the contribution of sensory loss to dysfunction and to design treatment interventions for stroke survivors.15, 16, 17 Somatosensory deficits are reported in 53% to 60% of stroke survivors and include both proprioceptive and cutaneous inputs.15, 18 Stroke survivors frequently report a “clumsy hand” or a loss of dexterity because they are unable to modulate grasp and lift forces to the weight and texture properties of the object.15, 19, 20 Motomura et al21 reported that many tests of sensory perception are inadequate for identifying the complex sensory loss after stroke and that more people poststroke experience a sensory deficit than available tests can identify. Such a sensory deficit would impair the ability to perceive object properties regardless of the amount of manual exploration that can be performed and result in poor interaction with objects. Thus, a test that evaluates the hand as a receptor for the sensory system should provide much needed information about upper-extremity function poststroke. The purpose of this study was to develop and evaluate the reliability and validity of the Hand Active Sensation Test (HASTe), a clinical test of haptic touch after stroke. Development included design and fabrication of appropriate test objects and testing procedures. Methods  Participants A consecutive sample of 28 stroke survivors was recruited from local rehabilitation facilities, concurrent research studies at Ohio State University, and stroke-survivor support groups. An equal number of noninjured controls were recruited from the community and individually matched to a stroke survivor by sex, hand dominance, and age ±3 years (table 1). The mean age for the stroke group was 60.18±14.46 years (range, 32−82y), and for the control group, it was 60.07±14.64 years (range, 33−84y), with a mean absolute age difference between matched subject pairs of 1.04±1.07 years (range, 0−3y). In addition to having a reliable form of communication, stroke survivors were enrolled if they could grasp (with all fingers and the thumb), lift, and release a 3.81cm (1.5in) diameter cylinder, weighing 224g (8oz) (test object size and maximum weight), similar to a small juice can. To promote heterogeneity in the stroke-survivor sample, subjects were enrolled regardless of time from injury, lesion location, or personal report (yes-no question) of sensory changes in the paretic hand. The mean time from injury for the stroke survivors was 17±21 months (range, 8d to 7y), with an equal number of left and right-sided injuries (14/14). Twelve subjects reported no sensory loss, and 16 subjects reported some degree of sensory loss in the paretic hand. All subjects were excluded if they reported any past or current diagnosis of peripheral nervous system, central nervous system, skin, medical (including diabetes without recognized neuropathy), or orthopedic condition that could alter sensation. | | |  | | Female | Male | Right Dom | Right Test | Left Test | Left Dom | Right Test | Left Test | Break 0 to <1h | Break 1 to <2h | Break 2 to 4h | |
|---|
| N/group | 11 | 17 | 23 | 11 | 12 | 5 | 2 | 3 | 9 | 9 | 10 | | | | |
Instrumentation Hand Active Sensation Test To complete the HASTe, subjects used 1 hand to manually explore objects that vary by weight and texture by using a match-to-sample forced choice recognition task without visual assistance.22 Across 18 trials, 9 test objects are matched twice, once to texture and once to weight. Subjects indicate their choice without identifying or describing the matching object property. The total number of accurate matches (0−18) is recorded. To optimize the HASTe to daily life, objects mimicked a common familiar functional object that typically varies by weight and texture and is successfully explored by 1 hand, a drink container. Weight and texture properties were chosen because they are preferentially explored by touch and because of their influence on grip and load forces during manipulation.7, 22 Every effort was made to ensure that the only differences between the objects were intended, so the objects were the same shape and size, cylindrical 3.81cm (1.5in) in diameter and 4.75cm (3 in) in height, made from a noncompliant material. The following weights used were appropriate to a juice can: .18, .21, and .24L (6, 7, and 8oz). To prevent assistance from potential feedback about the object’s contents, packing material was selected so that even if the object was violently shaken or dropped, the contents would not shift or move. Objects were weighed on a balance scale to ensure that they were indeed identical matches. Weights were chosen to match the weight of a full 154-g (5.5-oz) small juice container (.24L [8oz]) and then the smallest increment that differentiated stroke and control subjects in a pilot study (28g [1oz]). The texture materials mimicked, as closely as possible, the common surfaces found on drinking cups including plastic, paper, and styrofoam. Textures were applied to the vertical surface of the cylinder so that the seams did not provide additional or conflicting feedback. A threshold detection pilot study, performed by using both mildly sensory impaired (n=5) and nonimpaired subjects (n=7), was conducted to find 3 weights and 3 textures that at least 75% of the nonimpaired subjects could accurately discriminate. During the test, subjects were seated at a table with the height adjusted such that they could rest and move the tested arm comfortably on the table. The test arm was placed under a curtain to prevent subjects from seeing the tested arm, the objects, or the examiner. Instructions included the following: (1) use 1 hand to first manually explore the target object and then each possible match that vary by weight or texture (never both) and “find the match,” (2) you can touch each object as many times as you need to determine your answer, and (3) you have a maximum of 5 minutes for each trial. Matching among 3 objects minimizes the potential for accurate matches to be made because of chance or guessing.4, 5 Although subjects were informed that the objects differed by weight or texture with all other object properties the same (size, shape, temperature, surface compliance), it is important to note that subjects were not instructed to match to a particular object property (eg, “find the object that has the same texture”). If elbow or shoulder movement prevented a subject from moving between objects, the examiner slid the objects over to and away from the subject’s hand per the subject’s request. The examiner did not assist the subject with manual exploration of the objects. Once the subject found the match, he/she indicated to the examiner either verbally or by pointing to the number on the curtain. Subjects were not required nor encouraged to describe in any manner the object property identity or to explain why they chose the match. Before scoring started, subjects were given 2 demonstration trials of the test with sample objects. The sample objects were not the same size nor did they have the same weights and textures as the test objects. The examiner did not offer feedback about performance during the sample or HASTe trials. Annett Hand Preference Questionnaire Hand dominance is considered an external manifestation of lateral asymmetries in the cerebral hemispheres.23 Because many people who are left-handed were trained to write right-handed, the assessment of hand dominance requires more than writing preference.23, 24 Questions on the Annett Hand Preference Questionnaire (APHQ)23 are easy to understand, and the ordinate scale classification system was most appropriate for matching subjects between groups.25 Stroke subjects answered the questions based on their prestroke hand use and were assigned to 1 of 6 categories (right/left: consistent, inconsistent, ambidextrous) that coupled hand dominance with hand tested (ie, right dominant, left tested). Control subjects were matched to the stroke survivor by hand dominance in addition to age and then had the same hand tested as the paretic hand of the stroke-survivor match. Wrist Position Sense Test Testing followed the design and procedure created by Carey et al26 to evaluate proprioception at the wrist. The Wrist Position Sense Test (WPST) has strong test-retest reliability (r range, .88−.92) with a defined criterion of impairment (11°±4.8°), is valid for identifying proprioceptive deficit at the wrist in stroke survivors, and is responsive to genuine change (±8°). The author of the WPST evaluated the examiner through videotape review before the study onset. Two-point discrimination Two-point discrimination thresholds were evaluated for the distal phalanx of the index and pinky fingers. The DiskCriminator and the associated procedure, as developed by Mackinnon and Dellon,27 were used. Reports on the reliability coefficient values for static 2-point discrimination are disparate: .18 to .92.28, 29 The third edition of the diagnosis and examination manual published by the American Society for Surgery of the Hand defines normal threshold as less than 6mm.30 Procedure To examine test-retest reliability and to minimize the potential effect of recovery, therapeutic intervention, or any other form of practice on HASTe performance as well as to permit stroke survivors with acute injuries (0−6mo) to enroll, subjects participated in 2 test sessions completed on the same day. The same examiner performed all test sessions. When convenient, subjects were tested in the Human Performance Laboratory at Ohio State University. Subjects were also tested in various therapy departments, subject homes, and other community locations per subject preference. Before participation, all subjects provided informed consent; the protocol was approved by the Biomedical Sciences Institutional Review Board. All subjects completed the APHQ first and then the WPST and 2-point discrimination tests. The HASTe was the last test of the first session. Stroke survivors used the more-affected upper extremity, and the matched control subjects used the corresponding upper extremity based on the AHPQ classification. The second session was completed after a break of 45 minutes to 4 hours, dictated by convenience for the stroke-survivor subjects. Control subjects’ break time was the same as the break time of their matched stroke survivor. Retest sessions did not always occur in the same location as the first test session. Because the reliability of the DiskCriminator and the WPST has already been established, only the HASTe was administered during the second session. The order of the 18 HASTe trials as well as the order of the 3 match objects within a trial were randomly assigned for each of the 2 test sessions and were the same across subjects. The average testing time was 25±1 minutes (range, 11−55min) for the stroke group and 14±3 minutes (range, 10−20min) for the control group. Data Analysis Reliability and any potential threats were assessed by Pearson r, paired t test, and single measure intraclass correlation coefficient model 3,1 (ICC3,1). Standard error (SE) of measurement was calculated with the ICC3,1.31 Internal consistency was computed by the Cronbach α.31 To evaluate validity, independent sample t tests were calculated by using test 1 data for the 3 sensory tests. One-way analysis of variance (ANOVA) and t test analyses were used to investigate potential differences in HASTe scores caused by age, sex, hand dominance (right vs left), and APHQ (right or left: consistent, inconsistent, ambidextrous) category. Subjects were grouped by decade for age group comparisons; group size varied from 5 to 17. Pearson r correlations were used to determine if there were relations between HASTe score and measures of 2-point discrimination and average error on the WPST. Receiver operating characteristic (ROC) curves were created by using sensitivity and specificity values from test 1 scores and average (test 1, test 2) test scores and the area under the curve calculated.31, 32 All statistical procedures were performed by using SPSS software.a Results Reliability Reliability measures along with mean test 1 and 2 scores are presented in table 2. Test-retest reliability for the HASTe across all subjects was strong (ICC3,1=.77, Pearson r=.78), supporting that the scores are both correlated and in agreement despite a significant difference found for the paired t test analysis (P<.03).31 Paired t test results were not significant for controls (P=.57) but were significant for stroke survivors (P=.002). The values for Pearson r and ICC for each group were expected to be lower in magnitude because of the smaller sample size and decreased range of scores, especially in the control group, within the subject sets.31 SE of measurement values suggest that changes in score greater than 3 are most likely not caused by measurement error. Break time duration and test-retest location did not influence reliability measures for HASTe accuracy score as seen in table 3. Internal consistency of the 18-item instrument was also strong with a Cronbach coefficient α of .82. | | | | HASTe Accuracy | Test 1 | SEM 1 | Test 2 | SEM 2 | Mean Change | Paired t (P) | r | ICC3,1 | | |
|---|
| Total sample (N=56) | 11.66±4.20 | 1.96 | 12.48±3.55 | 1.70 | 0.82±2.58 | .2.31 (.024) | .78 | .77 | .87 | | | Stroke survivors (n=28) | 8.46±3.51 | | 10.32±3.29 | | 1.86±2.88 | 3.42(.002) | .64 | .64 | .78 | | | Controls (n=28) | 14.86±1.53 | | 14.63±2.28 | | −0.21±1.97 | .576(.570) | .53 | .49 | .65 | | | | |
| ⁎ ICC for the average of test 1 and test 2. |
| ⁎ Significant difference reflects effect of time not location as indicated by high correlation and ICC values. |
Validity HASTe accuracy discriminated between stroke survivors and controls (P<.001) with mean accuracy scores of 14.86±1.53 for controls and 8.46±3.51 for stroke survivors (fig 1). Both groups of subjects scored significantly higher (P<.001) on the 9 texture trials (6.36±2.50) than the 9 weight trials (5.30±2.31); however, the HASTe discriminated between stroke survivors and controls for discrimination of weight and texture (P<.001) as well as by specific object property (ie, plastic, P<.001). Both groups showed variability in performance (stroke survivors greater than controls), supporting the HASTe as a continuous scale able to distinguish degrees of impairment with lower scores indicative of greater impairment. The 1-way ANOVA for age and APHQ category and the t tests for sex and hand dominance were not significant (table 4). Sensitivity and specificity values were calculated for each HASTe test 1 score and each average of test 1 and test 2 scores, and both were plotted on an ROC curve (fig 2). The area under the curve was .941 (95% confidence interval [CI], 0.871−1.011) for test 1 and .922 (95% CI, 0.843−1.001) for average scores. By using an accuracy score of 13, sensitivity was .857 and specificity was 1.0 for test 1. | ⁎ Hand dominance is right or left. †APHQ is 1 of 6 categories (right/left: consistent, inconsistent, ambidextrous). |
The HASTe score showed moderate and negative correlations with WPST average error and 2-point discrimination measures. These 2 established measures of somatosensation also discriminated between the 2 groups (table 5). | ⁎ Average of test 1 and test 2. |
Discussion The quality of a clinical instrument is determined by its dependability and diagnostic accuracy.33, 34 The HASTe score showed good test-retest reliability by ICC (.77) and very good diagnostic accuracy as quantitatively expressed by the area under the ROC curve (.94 for test 1, .92 for the mean of tests 1 and 2) and the relative sensitivity and specificity calculations by using 13 as the cut score. The HASTe successfully discriminated between stroke survivors and controls and had moderate, negative correlations with established measures of somatosensation. Unlike other studies that had identified age and hand dominance as influences on sensory capacity,34, 35 these factors did not contribute to the differences in scores within or between the subject groups. However, it should be noted that the group sizes for the age comparison varied from 2 to 9 for subject group and 5 to 17 when subject groups were combined; the small and varied group size may have contributed to the lack of significance in this comparison. Test construction evaluated by the Cronbach α and independent sample t tests per item support that the 18 items were homogenous and collectively evaluate haptic perception. The ROC plot provides a complete picture of the ability of the HASTe to make distinctions at each score value.31, 32 A continuous scale like the HASTe provides clinicians more precise information about performance than a dichotomous or categorical measure; scores on the HASTe will provide both measures.31 On a dichotomous scale, subjects scoring between 0 to 12 can be considered impaired and those scoring 13 to 18, nonimpaired with expected sensitivity and specificity of .857 and 1.0, respectively. A categorical scale around 13 can also be made with 9 to 10 labeled as moderately impaired and 11 to 12 mildly impaired. It is expected that the clinician will use the sensitivity and specificity values in combination with other evaluation data to make decisions about his/her patient. Additionally, the SE of measurement values in table 2 suggest a change in HASTe score greater than 3 can most likely be attributed to genuine change in performance as opposed to measurement error, but this needs further evaluation in a larger group of subjects. Although it was assumed that sensory changes would be more prevalent in the stroke-survivor population, it was not expected that all stroke survivors would be found impaired on the HASTe. Indeed, 4 stroke survivors scored 13 or greater (see fig 2) consistent with typical function. Yet, not all of the control subjects had a perfect score, supporting the HASTe as a challenging test with a range of performance in nonimpaired subjects. It should be noted that no control subject scored below 13, thereby justifying our impairment classification of 12 or lower. The HASTe was intentionally designed to be rater independent, thereby eliminating threats to reliability and validity reported for other sensory measures that are dependent on examiner skill for stimulus control and judgment of performance.14, 27, 29, 33, 36 The source of the significant difference on the paired t test for test-retest reliability was not attributable to break duration or testing location (see table 3). Same-day testing obviates the influence of spontaneous recovery and evaluates the influence of test practice. If the items on the HASTe were learnable, then the group with the shortest break time (<1h) might be expected to have the greatest improvement; this was not the case. Location (laboratory vs nonlaboratory setting and/or changing between settings) was also not an influence on test performance (see table 5). These results suggest that the HASTe is inherently reliable; the change in scores among the stroke survivors is most likely attributable to improvements in the ability or strategy used to search for object qualities rather than learning the test objects themselves or could be because of a decrease in fatigue during the second testing session when only the HASTe was performed. Thus, further reliability testing with a longer interval (days not hours) between tests should be conducted to clarify the stability of scores. Also, further evaluation of object exploration strategies would contribute to a better understanding of this improvement. The HASTe accuracy score had moderate to good correlations (see table 5) with the established measures of tactile sensation (2 point) and proprioception (WPST), supporting that it is a test of sensation in the hand. The fact that the values were not greater suggests that the HASTe is a distinct measure of sensation. There are multiple established somatosensory tests available to the clinician26, 27, 37 as well as new measures in the literature35, 38; thus, one can question whether it is prudent to add another test to the list. Despite the quality of results, to be useful, the HASTe needs to provide unique information about hand function not currently available to the clinician. Collectively, sensory tests form a spectrum of sensibility from those that measure detection of isolated tactile or proprioceptive modalities (2-point, WPST) to those that test functional use of sensory information (stereognosis) or the impact of sensory loss on manual dexterity rather than sensory function itself. Between these endpoints are tests that evaluate object property perception and discrimination, including the HASTe, and address stimulus detection or discrimination threshold,36, 39 passive texture discrimination,34, 35, 38 or sustained weight perception.38 Each test is attempting to tease out the ability to perceive and discriminate object properties. However, the HASTe remains distinct from these measures in 5 important ways. First, the HASTe score is not a discrimination threshold; instead, it is a measure of the ability to use the hand to obtain sensory information. Second, in each of the aforementioned tests, the object properties are assessed in isolation and are frequently object properties not common to handheld objects such as carpet or sand paper.34 The HASTe not only combines weight and texture properties because both must be successfully managed in day-to-day interactions with objects but also uses property distinctions common to handheld objects like drink containers. Third, the HASTe evaluates the sensory capacity of the entire hand, whereas most of the object property tests assess the fingertip.27, 34, 35, 38 Fourth, of those tests that do evaluate the entire hand, subjects are required to identify the object or the object property either verbally34 or by matching to a visual comparison.35, 38 Both verbal and visual comparison can be confounded by cognitive and language limitations unrelated to sensory capacity. During the HASTe, subjects could indicate their preference by touching the object, thus eliminating reliance on language or visual discrimination. Finally, the HASTe was able to identify a sensory loss in 66% of stroke survivors with no previously identified sensory impairments. Of those that scored below 13 on the HASTe, 11 had an average error score on the WSPT within normal limits (≤11°). Normal scores for 2-point discrimination for the distal phalanx range from 0.5 to 6mm,40 depending on the study; if the 6mm criterion is used, 15 subjects that were impaired on the HASTe scored within the normal range for 2-point discrimination. This suggests that the HASTe is a highly sensitive evaluation tool that is providing information about aspects of sensory perception that are not evaluated by these traditional measures. Additional comparisons with other sensory discrimination measures are warranted to identify relative sensitivity and specificity. Conclusions This study created the HASTe and evaluated its reliability and validity. The HASTe was designed to be a standardized, quantitative measure of haptic perception of the whole hand. The objects and their properties can be successfully explored with 1 hand and are optimally related to the daily use of the hand during grasp, lift. and release tasks. Together, the results support our findings that the HASTe is both a reliable and valid measure of haptic perception providing information to the clinician and patient that are not currently available. The HASTe is distinct from other tests by the sensory function it measures (haptic touch) and the clinical information it offers to both patients and clinicians; yet it only measures 2 aspects of haptics (weight, texture). Therefore, it does not provide information about other haptic components (temperature, surface compliance, size, or shape), which are measured by other tools. Nevertheless, it may be a more sensitive indicator of sensory hand function than more commonly used clinical measures. Use of the HASTe, however, is limited to stroke survivors who have sufficient hand function to grasp, lift, and release the test objects. In addition, the relation of scores on the HASTe to hand function was not a component of this study; this analysis is currently underway. Supplier References 1. 1Ada L, Canning CG, Carr JH, Kilbreath SL, Shepherd RB. Task specific training of reaching and manipulation. In: Bennett KM, Castiello U editor. Insights into the reach to grasp movement. Amsterdam: Elsevier Science; 1994;p. 239–265. 2. 2Gibson JJ. The senses considered as perceptual systems. Boston: Houghton Mifflin; 1966;. 3. 3Lederman SJ, Klatzky RL. Haptic aspects of motor control. In: Boller F, Grafman J editor. Handbook of neuropsychology. Amsterdam: Elsevier Science; 1997;p. 131–147. 4. 4Lederman SJ, Klatzky RL. Hand movements: a window into haptic object recognition. Cognit Psychol. 1987;19:342–368. MEDLINE |
CrossRef
5. 5Klatzky RL, Lederman SJ, Metzger VA. Identifying objects by touch: an “expert system”. Percept Psychophys. 1985;37:299–302. MEDLINE 6. 6Krantz M. Haptic recognition of objects in children. J Genet Psychol. 1972;120:121–133. MEDLINE 7. 7Johansson RS, Cole KJ. Grasp stability during manipulative actions. Can J Physiol Pharmacol. 1994;72:511–524. MEDLINE 8. 8Klatzky RL, Lederman SJ. The haptic glance: a route to rapid object identification and manipulation. In: Gopher D, Koriat A editor. Attention and performance XVII: cognitive regulation of performance: interaction of theory and application. Cambridge: MIT Pr; 1999;p. 165–196. 9. 9Loomis JM, Lederman SJ. Tactual perception. In: Boff K, Kaufman I, Thomas J editor. Handbook of human perception and performance. New York: Wiley; 1986;p. 31.1–31.41. 10. 10Schiffman HR. Sensation and perception. 5th ed.. New York: John Wiley; 2001;. 11. 11Voisin J, Lamarre Y, Chapman CE. Haptic discrimination of object shape in humans: contribution of cutaneous and proprioceptive inputs. Exp Brain Res. 2002;145:251–260. MEDLINE |
CrossRef
12. 12Klatzky RL, Lederman SJ. Identifying objects from a haptic glance. Percept Psychophys. 1995;57:1111–1123. MEDLINE 13. 13Klatzky RL, Lederman SJ. Stages of manual exploration in haptic object identification. Percept Psychophys. 1992;52:661–670. MEDLINE 14. 142002 heart and stroke statistical update. Dallas: American Heart Association; 2001;. 15. 15Carey LM. Somatosensory loss after stroke. Crit Rev Phys Rehabil Med. 1995;7:51–91. 16. 16Dellon AL, Kallman CH. Evaluation of functional sensation in the hand. J Hand Surg [Am]. 1983;8:865–870. MEDLINE 17. 17Winward CE, Halligan PW, Wade DT. Current practice and clinical relevance of somatosensory assessment after stroke. Clin Rehabil. 1999;13:48–55. MEDLINE |
CrossRef
18. 18In: Gresham GE, Duncan PW, Stason WB, et al. editor. Post-stroke rehabilitation. Clinical practice guideline no. 16. Rockville: U.S. Department of Health and Human Services, Agency for Health Care Policy and Research; 1995;. 19. 19Shumway-Cook A, Woollacott MH. Motor control: theory and practical applications. 2nd ed.. Philadelphia: Lippincott Williams & Wilkins; 2001;. 20. 20Parry RH, Lincoln NB, Appleyard MA. Physiotherapy for the arm and hand after stroke. Physiotherapy. 1999;85:417–425. Abstract | Full Text |
Full-Text PDF (286 KB)
|
CrossRef
21. 21Motomura N, Yamdori A, Asaba H, Sakai T, Sawada T. Failure to manipulate objects secondary to active touch disturbance. Cortex. 1990;26:473–477. MEDLINE 22. 22Klatzky RL, Lederman SJ, Matula DE. Haptic exploration in the presence of vision. J Exp Psychol Hum Percept Perform. 1993;19:726–743. MEDLINE |
CrossRef
23. 23Annett M. A classification of hand preference by association analysis. Br J Psychol. 1970;61:303–321. MEDLINE 24. 24Fess EE. Defining handedness: a critical variable. J Hand Ther. 1997;10:314. MEDLINE 25. 25McMeekan ER, Lishman WA. Retest reliabilities and interrelationship of the Annett Hand Preference Questionnaire and the Edinburgh Handedness Inventory. Br J Psychol. 1975;66:53–59. 26. 26Carey LM, Oke LE, Matyas TA. Impaired limb position sense after stroke: a quantitative test for clinical use. Arch Phys Med Rehabil. 1996;77:1271–1278. Abstract |
Full-Text PDF (2354 KB)
|
CrossRef
27. 27Mackinnon SE, Dellon AL. Two-point discrimination tester. J Hand Surg [Am]. 1985;10(6 Pt 1):906–907. MEDLINE 28. 28Dellon AL, Mackinnon SE, Crosby PM. Reliability of two-point discrimination measurements. J Hand Surg [Am]. 1987;12(5 Pt 1):693–696. MEDLINE 29. 29Lincoln NB, Crow JL, Jackson JM, Waters GR, Adams SA, Hodgson P. The unreliability of sensory assessments. Clin Rehabil. 1991;5:273–282.
CrossRef
30. 30American Society for Surgery of the Hand. The hand: examination and diagnosis. 3rd ed.. New York: Churchill Livingstone; 1990;. 31. 31Portney LG, Watkins MP. Foundations of clinical research: applications to practice. 2nd ed. Upper Saddle River: Prentice Hall; 2000;. 32. 32Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem. 1993;39:561–577. MEDLINE 33. 33Bell-Krotoski JA, Buford WL. The force/time relationship of clinically used sensory testing instruments. J Hand Ther. 1997;10:297–309. MEDLINE 34. 34Robertson SL, Jones LA. Tactile sensory impairments and prehensile function in subjects with left-hemisphere cerebral lesions. Arch Phys Med Rehabil. 1994;75:1108–1117. MEDLINE |
CrossRef
35. 35Byl N, Leano J, Cheney LK. The Byl-Cheney-Boczai Sensory Discriminator: reliability, validity, and responsiveness for testing stereognosis. J Hand Ther. 2002;15:315–330. Abstract | Full Text |
Full-Text PDF (1525 KB)
|
CrossRef
36. 36Winward CE, Halligan PW, Wade DT. The Rivermead Assessment of Somatosensory Performance (RASP): standardization and reliability data. Clin Rehabil. 2002;16:523–533. MEDLINE |
CrossRef
37. 37Gaubert CS, Mockett SP. Inter-rater reliability of the Nottingham method of stereognosis assessment. Clin Rehabil. 2000;14:153–159. MEDLINE |
CrossRef
38. 38Dannenbaum RM, Michaelsen SM, Desrosiers J, Levin MF. Development and validation of two new sensory tests of the hand for patients with stroke. Clin Rehabil. 2002;16:630–639. MEDLINE |
CrossRef
39. 39Weinstein S. Fifty years of somatosensory research: from the Semmes-Weinstein monofilaments to the Weinstein Enhanced Sensory Test. J Hand Ther. 1993;6:11–22. MEDLINE 40. 40Dellon AL. Evaluation of sensibility and re-education of sensation in the hand. Baltimore: Williams & Wilkins; 1981;. a School of Physical Therapy, Ohio University, Athens, OH b Occupational Therapy Division, School of Allied Medical Professions, Ohio State University, Columbus, OH c Physical Therapy Division, School of Allied Medical Professions, Ohio State University, Columbus, OH.  Reprint requests to Deborah S. Nichols-Larsen, PhD, 106 Atwell Hall, 453 W 10th Ave, Columbus, OH 43210
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. PII: S0003-9993(06)00974-9 doi:10.1016/j.apmr.2006.08.019 © 2006 the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT
