One of the goals in the field of prosthetics is to design prosthetic limbs that resemble the functional characteristics of the anatomical limb as closely as possible. This is indeed a challenging goal in the case of the upper-extremity prosthetic limb because of the long-recognized complexity of the hand and its multiple uses in everyday life. 1,2 The anatomical hand is a remarkable organ that allows for not only the grasping and manipulation of objects, but also gesturing and other functional behaviors for exploring the environment. 3 One important feature of the hand is its participation in the perceptual judgments of the weight of hand-held objects. Recent research suggests a functional link between the ability to judge the weight of hand-held objects and the control of grip pressure. 4,5 Thus, it would appear that efficient use of grip pressure exerted by the anatomical hand is at least partially dependent upon the ability to perceptually judge the weight of the object. If a goal of prosthetic research is to develop prosthetic hands that resemble the functional capabilities of the anatomical hand, it would make sense to develop a body of knowledge on weight discrimination in the prosthetic hand.
The ability to accurately judge the weight of a lifted object with the hand and arm appears to depend on both peripheral (sensory) and central (motor) influences. Sensory information about the object’s weight comes from a variety of sources including cutaneous mechanoreceptors and proprioceptors located in the hand and arm. These receptors provide information about the friction and pressure exerted by the object on the skin and about the forces generated by the muscles and joints to lift the object against gravity. 6
Evidence for the peripheral contributions to the discrimination of hand-held weight has been provided by a variety of studies on participants using their anatomical hand. 7,8,9,10 For example, Cole and Sedgwick (1992) observed significant impairments of weight estimation in a deafferented participant, leading them to conclude that afferent information is important in this type of task. However, other sources of sensory information, such as visual cues, may also contribute to one’s ability to estimate the weight of a hand-held object. 8
In addition to sensory cues, a second major source of information available for weight discrimination is central information. Central/motor influences may include the following: 1) the amount of grip pressure;4 2) total muscular contraction associated with the increased weight of the object;11 3) how the lift of the object is executed;12 4) the physical size or shape of the object that may bias the perception of weight;13 and 5) neuromuscular fatigue. 14 All of these influences on the motor outflow to the hand and arm have been shown to affect one’s ability to accurately judge the weight of a hand-held object. What this evidence suggests is that the motoric aspects of object manipulation during weight estimation may influence the perception of weight, supporting the view that perception and action are intricately and reciprocally linked. 15,16 In addition, a hypothesis has been put forward, called the weighted fusion model, that suggests it is the total amount of neuronal activity arising from grip and lift forces that determines the perceived heaviness of an object. 17
Based upon the evidence about sensory and motor cues used in concert to estimate weight, a person with an amputation of the upper extremity wearing a prosthetic arm may be at a disadvantage in assessing the weight of a gripped object for several reasons. An individual with a below the elbow amputation wearing a body-powered prosthesis suffers from a lack of cutaneous mechanoreceptors such as the slow adapting Ruffini corpuscles (moderate, static touch), rapid adapting Meissner corpuscles (light, dynamic touch involving flutter), rapid adapting Pacinian corpuscles (light, dynamic touch involving vibration), slow adapting Merkel disks, and rapid adapting hair follicles (light, dynamic touch). 18,19 Whether or not a person with a congenital limb deficiency (PWCLD) can compensate for this loss of proprioceptive cues with visual information is not known. In addition, the ‘sense of effort’ used by the PWCLD may be altered because of presumed changes in the motor outflow associated with operating the prosthesis. Relatedly, and depending on the extent of the amputation, the weight of the prosthesis may require more or less motor outflow than that for the anatomical hand. The resulting change in motor outflow may cause a perceptual bias, possibly predicted by the weighted fusion model. 17 The present paper reports two experiments that explore the capability for weight discrimination with a prosthetic limb.
In the first experiment we compared a PWCLD, who was an experienced prosthetic user, to nonamputees who used their anatomical hand and arm, or who used a prosthetic simulator, in estimating weight of hand-held objects. Recent work in our laboratory has used a simulated upper extremity, body-powered prosthesis to study the early development of prosthetic coordination and control. 20 The simulator closely resembles the cosmetic and functional features of a standard prosthesis worn by a PWCLD below the elbow. Given the reciprocal nature of perception and action, 15,16 examining the ability to pick up information specifying properties such as weight might provide further insights into the functional limitations of current prosthetic designs.
For the first experiment, we hypothesized that the ability to accurately discriminate weight of the hand-held objects using a prosthetic device, assessed by a standard psychophysical measure—the difference threshold (limen), would be poorer compared to the anatomical hand. The difference limen (DL) is the amount of change in a stimulus that is required to produce a just noticeable difference in the sensation of that stimulus. 21 This hypothesis was based on three reasons: 1) The added weight of the prosthetic simulator would increase ‘stimulus intensity,’ as defined by Weber’s Law, 20,22 which predicts a higher DL as stimulus intensity increases. The predictions for the PWCLD in this regard were unclear because the added weight of the prosthetic device may be somewhat offset by the weight loss of the amputated limb; 2) Cutaneous mechanoreceptors in the hand could not be used as effectively using the prosthetic simulator, and not at all by the PWCLD. However, it is unclear whether experience using the prosthesis by the PWCLD could offset the loss of cutaneous mechanoreceptors. In addition, there is some evidence of increased sensitivity in the skin of the residual limb following an amputation, 23 as well as spinal 24 and cortical reorganization 25,26 following an amputation; and 3) Poorer control over the prosthetic simulator might compromise the ability to sense the effort associated with lifting the weighted objects. All of these factors might allow the PWCLD to discriminate weight more effectively than the less experienced participants using the prosthetic simulator. However, it was unclear whether weight discrimination by the PWCLD would be less effective than that of the participants using their anatomical hand.
MATERIALS AND METHODS
Sixteen females and 3 males, right-hand dominant, age range, 20–40 years, were recruited from undergraduate classes at San Francisco State University. In addition, one 34-year-old male, quadruple PWCLD from birth who is right hand dominant, and uses a voluntary opening prosthesis with a split-hook prehensor, also participated. The Institutional Review Board (IRB) approved the study. All participants provided written informed consent prior to participation in the study.
Ten identical canisters (55 mm high × 35 mm diameter) made of capped cylinders filled with lead pellets were used for the weight discrimination. Participants sat comfortably near a standard table with their anatomical hand (or prosthesis) on the table, with the nonparticipating hand in their lap.
The method of constant stimuli originally developed by Fechner was used. 21 In this method, a comparative judgment for determining just noticeable differences requires two stimuli (in our case, two identically shaped objects of different weights). One of the weights is called the ‘standard’, and one is called the ‘comparison.’ The standard remains fixed, while the comparison is varied across a range of different weights bracketing the standard. The participant is given experience lifting either the standard or the comparison, followed by the other, and is asked which is heavier. In our study, a 125 g standard weight was compared randomly to 105, 110, 115, 120, 125, 130, 135, 140, or 145 g. Weight presentation was randomized such that each weight was tested eight times resulting in 72 trials. The presentation order of the standard and comparison weight was randomized across trials. The presentation order of the effector (simulator or anatomical hand) was counter-balanced across nonamputee participants. On each trial, the participant lifted the two weights in succession and was permitted to wield each weight for no longer than 10 seconds. After wielding the second weight, the participant was immediately asked to determine which was heavier. Mid-way through the testing session, a 5-minute break was given. Participants performed two testing sessions on separate days, one with the prosthetic simulator (P) and one with the anatomical hand (A). The order of testing was randomized. The PWCLD performed one session only with his dominant prosthetic limb.
Following conventional psychophysical methods, 21 the proportion of trials for which the comparison weight was judged heavier than the standard were converted to z-scores and plotted against the comparison weights. A line of best fit was then imposed on these data points. From the equation associated with the line of best fit, it was possible to calculate three dependent variables, the point of subjective equality (PSE), the constant error (CE), and the difference limen (DL). The PSE is the value at which the comparison stimulus is deemed equal to the standard stimulus. It represents the point at which the comparison weight is judged heavier than the standard weight on 50% of the trials. It corresponds to a z-score of 0. The CE is the difference between the PSE and the standard stimulus (125 g in this case). The DL is the amount of change in a stimulus that is required to produce a just noticeable difference in the sensation of that stimulus. It represents the average of two differences—that between the PSE and the weight judged heavier than the standard on 25% of trials (corresponding to a z-score of −0.67) and that between the weight judged heavier than the standard on 75% of trials (corresponding to a z-score of 0.67) and the PSE.
The calculation for the DL is shown in the following equation: EQUATIONEQUATION
The DL can also be assessed qualitatively by examining the slope of the line of best fit. A steeper slope indicates a smaller DL and therefore better discrimination.
The second experiment attempted to shed further light on the issue of weight discrimination by having the PWCLD participant from the first experiment discriminate weights with and without vision. Manipulating vision may provide further insight into what perceptual cues assist a PWCLD in discriminating weight, since it is still unclear as to what types of perceptual information are used by a PWCLD when operating a prosthetic device. The premise behind this type of manipulation is that visual information (e.g., kinematic parameters of the visual flow field) might specify force parameters correlated with the wielding of weight. There is evidence that the limitation of judging weight imposed by the loss of important cutaneous mechanoreceptors in the hand and arm can be compensated for by the use of vision during the perceptual judgment. 8 How vision may compensate for proprioceptive loss is not clear, but presumably the observer can use vision to determine the kinematic characteristics (i.e. velocity, acceleration) of the hand and arm while the object is wielded. 12 The manner in which the hand and arm change trajectory during the wielding of different weights may inform perceptual judgments. Given this, we hypothesized that the ‘vision condition’ would produce a more accurate discrimination of weights than would the ‘no vision condition’.
The PWCLD from the previous experiment participated in this study. He is right hand dominant and uses a voluntary-opening, split hook prehensor. The study was approved by the IRB. Prior to testing, the participant signed a written informed consent.
Ten identical canisters (55 mm high × 35 mm diameter) made of capped cylinders filled with lead pellets were used for weight discrimination. The participant sat comfortably at a standard table.
Following the procedures described in Experiment 1, the method of constant stimuli originally developed by Fechner was used 21 using the same standard weight (125 g) and comparison weights (105, 110, 115, 120, 125, 130, 135, 140, or 145 g).
The vision and no vision trials were randomly presented throughout the session. Prior to the start of each trial, the participant was informed if vision or no vision would be employed. If ‘no vision’, the participant was asked to close his eyes; if vision, the participant was asked to keep his eyes open. Pilot testing revealed that with vision, participants would focus their eyes on the wall, or tabletop, or other external object instead of directing their attention to the task. Given this observation, the participant was instructed to look at the weight in the vision condition.
The weights used were kept from view of the participant and were presented only when the participant’s eyes were closed or opened depending on the condition of the trial. Weight presentation was randomized such that each weight was tested eight times for both the vision and no vision conditions resulting in 144 trials. The order of presentation of the standard and comparison weight was randomized across trials.
Because vision was removed prior to each no vision trial, the researcher placed the weight in the open prehensor and then removed it when the trial was finished. Upon removal of the second weight, the participant was asked immediately to verbally indicate which was heavier. This practice was upheld for the vision trials to ensure consistency among conditions.
One-third of the way through the testing session, and again two-thirds of the way through the testing session, a 2-minute break was given. Only one testing session was conducted, during which the dominant hand was used.
As in the first experiment, the frequency counts were z-transformed prior to the calculation of the PSE, CE, and DL. These dependent measures were determined from the line of best fit that was associated with the data.
The z-transformed frequencies and their corresponding lines of best fit for each group as well as the PWCLD are presented in Figure 1. The mean PSEs for the anatomical hand, the simulator, and the PWCLD were 126.48, 126.24, and 122.26 g, thus yielding CEs of 1.48, 1.24, and -2.74 g, respectively. Unexpectedly, the difference in CE between the anatomical hand and the simulator was not significant, t (18) < 1, likely because of the high variability among the participants in the use of the prosthetic simulator (sd = 4.53g) as compared to the anatomical hand (sd = 2.03g). Relatedly, the CE for the PWCLD was outside of the 95% confidence limits for the anatomical condition, but inside the 95% confidence limits for the prosthetic simulator. The latter result stems from the much-increased variability in perceptual judgments when using the prosthetic simulator.
The means and standard deviations (in parentheses) of the DLs for the anatomical hand and the prosthetic simulator were 6.35g (1.40g) and 14.81g (5.75g) respectively. The DL for the PWCLD was 8.23g. The DLs for the prosthetic simulator were significantly higher than the DLs for the anatomical hand, t (18) = 6.272, p < 0.001, and the DL for the PWCLD was within the 95% confidence limits for the anatomical hand and the stimulator. Once again, it would seem that the latter result holds only because of the much higher variability in the DLs for the prosthetic simulator than for the anatomical hand.
Figure 2 depicts the z-transformed data and their associated lines of best fit for the vision and no-vision condition. The coefficients (betas) for the slopes of the regression lines were 0.0677 and 0.0513 for the no-vision and vision conditions respectively. The correlations and R2 values (in parentheses) for the no-vision and vision conditions were 0.71 (0.4975) and 0.93 (0.8581). Both correlations were significantly different from zero at the .05 level, however, only the vision correlation was significantly different from zero at the .01 level. A two-tailed dependent t-test on the two correlations 27 (p. 53) revealed that they were not significantly different from each other, t (5) = 1.89, p > .05. As there is not an appropriate test for the significance of the difference between dependent slopes (betas), the absolute residuals from the two regression lines were compared using a two-tailed, paired t-test. The test showed that the residuals were not significantly different from each other, t (8) = 1.43, p > .05. For the vision condition, the PSE was 126 g and for the no vision condition the PSE was 120 g. For the vision condition, the DL was 11 g and for the no vision condition the DL was 13 g.
The z-transformation permits a standard interpretation of the data across various studies using various weight increments and frequencies of paired comparisons. Earlier studies have indicated a DL of about 2g for a standard weight of 50g (i.e. ≈ 5%). Following Weber’s Law, which states that the proportional change in intensity of a stimulus for the detection of a just noticeable difference is a constant, it follows that the standard weight of 125g would be expected to yield a DL of about 6g. Thus, results from our study are similar to results from other studies on weight discrimination. 21
Our study supported the hypothesis that naïve participants using an upper-extremity prosthesis were significantly less accurate in the discrimination of weight of hand-held objects compared to discrimination with the anatomical hand. However, it is not possible at this time to determine whether the inferior performance with the simulator was a function of the added weight of the prosthesis, the lack of information from cutaneous receptors, inexperience, or a combination of these factors. We also provided evidence that an experienced PWCLD was capable of making perceptual judgments that were similar to those made by participants using their anatomical hand. Presumably, experience using the prosthesis was the mediating factor. Thus, the ability to perceive weight using a prosthesis might develop with practice. At question, though, is what information the PWCLD uses to compensate for proprioceptive loss. A logical source of information that the PWCLD may learn to depend on to assist in weight discrimination is vision. The possibility of dependence on vision as a compensatory mechanism aiding in weight discrimination led to the second experiment.
The results revealed little difference between the ‘vision’ and ‘no vision’ conditions and may suggest that vision does not play a key role in weight perception by an experienced PWCLD. However, Runeson, and Frykholm (1981) 28 and Bingham (1987) 29 showed that vision of the anatomical limb allowed for an accurate judgment of the mass of objects. By using visual perceptions of displacement (of both the body and the object), velocity, object trajectory, etc., participants made judgments of the mass of the objects. These studies took a different approach in that the participants were asked to observe the lifting of the objects by another person to determine the mass rather than actually lifting the object themselves. Even though participants did not lift the objects themselves, the results of these studies give some insight as to what types of visual cues are used in the perceptual judgments of weight discrimination.
Of course, it could be that even though the PWCLD participant was provided visual information in one of the conditions, he chose to ignore it. In fact, during the debriefing, our PWCLD participant told us that the visual condition was somewhat distracting compared to the nonvisual condition. Apparently, he was concentrating more on the proprioceptive cues from certain parts of his residual arm during the perceptual judgments, and he attempted to focus on these cues instead of the visual cues when they were available. However, he was unable to pinpoint the source of the proprioceptive cues. There is evidence of increased sensitivity in the skin of the residual limb 23 and neural reorganization following amputation. 24,25,26 Whether this reorganization is a mere function of time or actual use is not known. However, the PWCLD had several years of experience using a prosthetic limb and this experience may have contributed to increased sensitivity due to neural reorganization so that proprioceptive cues from the residual arm were heightened. All of these factors may have contributed to his enhanced perceptual judgments compared to the novel participants using a prosthetic simulator. In addition, his strategy of ignoring the visual cues when they were available may have led to the nonsignificance between the vision and no vision conditions.
The hypothesis that the DL for weight discrimination would be higher using the prosthetic simulator than the anatomical hand was supported in this study (Experiment 1). One possibility is that this finding might be a result of the increase in the arm’s moment of inertia as a function of wearing the simulator. An increased moment of inertia in the simulated prosthetic condition may have increased the total amount of neural stimulation from the hand and arm. According to the weighted fusion model, 17 increased neural stimulation determines the perceived heaviness of a hand-held object. Thus, the DL difference between the simulated prosthetic condition and the anatomical condition may be at least partially accommodated by this model. Future studies could test this model by equating the moments of inertia of the anatomical arm and the prosthetic simulator; this is not trivial problem, however. A similar, though alternative, explanation is that judging weight with the simulator posed an additional problem because the naïve participant had to take into account both the weight of the simulator and the weight of the object when attempting to determine if one object was heavier than another. Thus, according to Weber’s Law (1834), a greater change in the object’s weight would be required to produce a just noticeable difference. If this explanation is tenable or if the weighted fusion model is correct, the ability to discriminate weight might be similar in the two conditions if the arm’s and the simulator’s moments of inertia are equated.
Another possibility is that the sensation arising from the prosthetic simulator was unfamiliar to the participants, leading to increased DLs and increased variability. A contributing factor to this unfamiliarity may be reduced motor control over the hand and arm while using the prosthetic simulator. Presumably, several sources of sensory information are needed to both control the prosthesis and to perceive the weight of an object. However, the situation is complicated because control of the prosthesis is also required to generate the movements that reveal the information specifying an object’s weight. Lederman and Klatzky (1987) 30 highlighted the relationship between movement control and haptic perception some time ago. They noted that specific hand movements maximize the sensory input corresponding to certain object properties and they have documented clear links between profiles of hand movements and perception of those properties. For example, individuals use static contact to determine temperature, lateral motion to determine texture, contour following to determine shape, and unsupported holding and wielding to determine weight. Moreover, Bushnell and Boudreau (1993) 31 have suggested that the developmental progression in infant’s ability to detect shape at approximately 4-months-of-age, temperature, hardness, and texture at approximately 6-months-of-age, weight at approximately 9-months-of-age, and configurational shape at about 12-months-of-age is paced by developmental changes in control over the arm, hand, and fingers that are needed to make the movements that reveal such object properties.
A logical possibility is that participants in the first experiment were less able to scale grip pressure relative to perceived heaviness of the weight when using the prosthetic simulator. The link between grip pressure and perceived heaviness of an object reinforces the reciprocity between perception and action, where, in this case, grip pressure must be scaled to object weight to effectively lift and wield the object but object weight must be apprehended by holding and wielding so that grip pressure can be scaled appropriately. If one assumes then that the sensory and motor elements of active touch and manipulation by the hand are inextricably linked, as proposed by several researchers, 3,30,31 then the perceptual deficits exhibited by participants in the simulated prosthetic condition are not surprising. One would expect though, that as control over the prosthesis improves, so to should the ability to discriminate object properties such as weight.
Results from the second experiment suggest that vision is not required by an experienced prosthetic user to accurately discriminate the weight of hand-held objects. It is hypothesized that PWCLDs with considerable experience using a prosthesis may undergo increased kinesthetic sensitivity in the residual limb, neural re-organization and better control over the prosthesis, all of which could contribute to more accurate weight discrimination without the need for additional confirmation by vision. It is relevant to note here that a controversial but long-held assumption among many who study perception is that vision calibrates the other perceptual systems so that motor control becomes increasingly less reliant on vision over time. 32 If true, we would expect novice prosthetic users to be more reliant on vision to control the prosthesis than experienced prosthetic users. However, the novice should become less reliant on vision and more capable of controlling movement on the basis of kinesthetic information as practice continues. Similar results have been found for able-bodied adults learning novel motor skills. 33
It is our view that a body of knowledge be developed based on research that investigates the many factors related to weight discrimination. Finally, we would like to emphasize our belief that the prosthesis and supporting anatomical structures (i.e. residual limb, upper torso) are both perceptual and motor devices. How perception and action are inextricably linked in a functional manner during the use of the prosthesis is a major goal of our work. In addition, how the person with an amputation who uses a prosthesis learns perceptual and motor skills is also of great importance to the field of prosthetics.
The authors would like to thank Robert Radocy, Lawrence Carlson, and Maurice LeBlanc for their contributions to the development of the simulated prosthesis used in the present studies.
This investigation was supported by a Research Infrastructure in Minority Institutions award from the National Center for Research Resources with funding from the Office of Research on Minority Health, National Institutes of Health #5 P20 RR11805, and by the National Institute on Disability and Rehabilitation Research, Department of Education, #H133G000024– 01.
1. Bell C. The Hand: Its Mechanism and Vital Endowments as Evincing Design. London: William Pickering; 1833.
2. Jones LA. The Assessment of Hand Function: A Critical Review of Techniques. J Hand Surg 1989; 14A: 221–228.
3. Turvey MT. Dynamic Touch. Am Psychol 1996; 51: 1134–1152.
4. Flanagan JR, Wing AM. Effects of Surface texture and Grip Force on the Discrimination of Hand-Held Loads. Percept Psychophys 1997; 59: 111–118.
5. Weeks DL, Wallace SA, Noteboom, JT. Precision-Grip Force Changes in the Anatomical and Prosthetic Limb During Predictable Load Changes. Exp Brain Res 2000; 132: 404–410.
6. Rothwell J. Control of Voluntary Movement. 2nd Ed. New York: Chapman & Hall; 1994.
7. Cole JD, Sedgwick EM. The Perception of Force and of Movement in a Man without Large Myelinated Sensory Afferents Below the Neck. J Physiol 1992; 449: 503–515
8. Fleury M, Bard C, Teasdale, et al. Weight Judgment: The Discrimination Capacity of a Deafferented Subject. Brain 1995; 118: 1149–1156.
9. Sanes JN, Shadmehr R. Sense of Muscular Effort and Somesthetic Afferent Information in Humans. Can J Physiol Pharmacol 1995; 73: 223–233.
10. Miall RC, Ingram HA, Cole JD, Gauthier GM. Weight Estimation in a “Deafferented” Man and in Control Subjects: Are Judgments Influenced by Peripheral or Central Signal? Exp Brain Res [serial on-line]
2000;133:DOI10. 1007/s002210000433. http://link.springer.de/link/service/journals/00221/bibs/0133004/01330491.htm
11. Aniss AM, Gandevia SC, Milne RJ. Changes in Perceived Heaviness and Motor Commands Produced by Cutaneous Reflexes in Man. J Physiol 1988; 397: 113–126.
12. Roberts WL, Davis CM. The Consistent Acceleration of Lifted Objects: Implications for Kinesthetic Illusions and the Perception of Weight. J Mot Behav 1978; 10: 287–293.
13. Ellis RR, Lederman SJ. The Material-Weight Illusion Revisited. Percept Psychophys 1999; 61: 1564–1576.
14. Gandevia SC, McCloskey DI. Interpretation of Perceived Motor Commands by Reference to Afferent Signals. J Physiol 1978; 283: 493–499.
15. Gibson JJ. The Senses Considered as Perceptual Systems. 1st Ed. Boston: Houghton Mifflin Company; 1966.
16. Gibson JJ. The Ecological Approach to Visual Perception. 1st Ed. Boston: Houghton-Mifflin; 1979.
17. Rinkenauer G, Mattes S, Ulrich R. The Surface-weight Illusion: On the Contribution of Grip Force to Perceived Heaviness. Percept Psychophys 1999; 61: 23–30.
18. Chapman CE, Tremblay F, Ageranioti-Belanger SA. Role of Primary Somatosensory Cortex in Active and Passive Touch. In: Wing, AM, Haggard, P, Flanagan, JR, (Eds.). Hand and Brain: The Neurophysiology and Psychology of Hand Movements. San Diego: Academic Press; 1996: 329–347.
19. Latash ML. Neurophysiological Basis of Movement. 1st Ed. Champaign, IL: Human Kinetics; 1998.
20. Wallace SA, Anderson DI, Anderson D, Mayo AM, Nguyen KT, et al. Motor Performance with a Simulated Artificial Limb. Percept Mot Skills 1999; 88: 759–764.
21. Gescheider GA. Psychophysics. Method, Theory and Application. 2nd Ed. Hillsdale NJ: Lawrence Erlbaum Associates, Publishers; 1985.
22. Weber EH. De Pulsu, Resorpitione, Auditu et Tactu: Annotationes Anatomicae et Physiologicae. 1st Ed. Leipzig: Koehlor, 1834.
23. Haber WB. Reactions to Loss of Limb: Physiological and Psychological Aspects. Ann NY Acad Sci 1958; 74: 14–24.
24. Florence SL, Kaas JH. Large-Scale Reorganization at Multiple Levels of the Somatosensory Pathway Follows Therapeutic Amputation of the Hand in Monkeys. J Neurosci 1995; 15: 8083–8096
25. Cohen LG, Bandinelli S, Findley TW, Hallett M. Motor Reorganization After Upper Limb Amputation in Man. Brain 1991; 114: 615–627.
26. Roricht S, Meyer B-U, Niehaus L, Brandt SA. Long-Term Reorganization of Motor Cortex Outputs After Arm Amputation. Neurol 1999; 53: 106–111.
27. Cohen J, Cohen P. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. 1st Ed. Hillsdale, NJ: Lawrence Erlbaum; 1975.
28. Runeson S, Frykholm G. Visual Perception of Lifted Weight. J Exp Psychol, Hum Percept Perform 1981; 7: 733–740.
29. Bingham, GP. Kinematic Form and Scaling: Further Investigations on the Visual Perception of Lifted Weight. J Exp Psychol, Hum Percept Perform 1987; 13: 155–177.
30. Lederman SJ, Klatzky RL. Hand Movements: A Window into Haptic Object Recognition. Cognit Psychol 1987; 19: 342–368.
31. Bushnell EW, Boudreau JP. Motor Development and the Mind: The Potential Role of Motor Abilities as a Determinant of Aspects of Perceptual Development. Child Dev 1993; 64: 1005–1021.
32. Stratton G M. Some Preliminary Experiments on Vision
without Inversion of the Retinal Image. Psychol Rev 1896; 3: 611–617.
33. Fleishman E A, Rich S. Role of Kinesthetic and Spatial-Visual Abilities in Perceptual-Motor Learning. J Exp Psychol 1963; 66: 6–11.