Loss of haptic feedback impairs control of hand posture: a study in chronically deafferented individuals when grasping and lifting objects

Experimental Brain Research https://doi.org/10.1007/s00221-019-05583-2 RESEARCH ARTICLE Loss of haptic feedback impairs control of hand posture: a study in chronically deafferented individuals when grasping and lifting objects R. Chris Miall1 · Orna Rosenthal1 · Kristin Ørstavik2 · Jonathan D. Cole3 · Fabrice R. Sarlegna4 Received: 14 February 2019 / Accepted: 12 June 2019 © The Author(s) 2019 Abstract Previous work has highlighted the role of haptic feedback for manual dexterity, in particular for the control of precision grip forces between the index finger and thumb. It is unclear how fine motor skills involving more than just two digits might be affected, especially given that loss of sensation from the hand affects many neurological patients, and impacts on everyday actions. To assess the functional consequences of haptic deficits on multi-digit grasp of objects, we studied the ability of three rare individuals with permanent large-fibre sensory loss involving the entire upper limb. All three reported difficulties in everyday manual actions (ABILHAND questionnaire). Their performance in a reach–grasp–lift task was compared to that of healthy controls. Twenty objects of varying shape, mass, opacity and compliance were used. In the reach-to-grasp phase, we found slower movement, larger grip aperture and less dynamic modulation of grip aperture in deafferented participants compared to controls. Hand posture during the lift phase also differed; deafferented participants often adopted hand postures that may have facilitated visual guidance, and/or reduced control complexity. For example, they would extend fingers that were not in contact with the object, or fold these fingers into the palm of the hand. Variability in hand postures was increased in deafferented participants, particularly for smaller objects. Our findings provide new insights into how the complex control required for whole hand actions is compromised by loss of haptic feedback, whose contribution is, thus, highlighted. Keywords Human · Proprioception · Somatosensation · Grasp · Haptics · Manipulation Introduction When healthy, we may take for granted our ability to hold everyday objects easily and securely, with minimal attention required to keep the object in hand during movement. We are normally unaware of the complex control issues required to reach and grasp the object, adopting in advance a hand posture shaped to the object itself. Nor are we aware of the changing forces required to securely grip and lift the object, * R. Chris Miall 1 School of Psychology, University of Birmingham, Birmingham B15 2TT, UK 2 Department of Neurology, Oslo University Hospital, Oslo, Norway 3 Centre of Postgraduate Research and Education, Bournemouth University, Bournemouth, UK 4 Aix Marseille University, CNRS, ISM, Marseille, France when grip force is precisely coordinated with the voluntary lift to compensate for load and rotational forces (for a review, Johansson and Flanagan 2008). The complexity of such tasks becomes evident only when we detect errors, such as a slip, or complete failure, for example, if our fingers are too cold to provide good sensory information (Cheung et al. 2003), or after a stroke (Nowak et al. 2003) or after numbing the skin over the fingertips (Johansson and Westling 1984; Witney et al. 2004). In the field of robotics, grasp and object manipulation are recognised as significant challenges (Suárez-Ruiz and Pham 2016). Despite enormous technical advances (see review by Yousef et al. 2011), robots are seriously limited by lack of proprioception (the sense of position and movement of body segments), of touch, and force feedback (Soter et al. 2018). Some pundits even suggest that mundane tasks like picking up and moving small objects will remain for humans for quite some time, while robots can take over other apparently complex tasks (computation, planning, driving, medical diagnosis and the like; Manyika et al. 2017; D’Anna 13 Vol.:(0123456789) Experimental Brain Research et al. 2017). On the other hand, a challenge remains for bioengineers interested in developing prostheses, for instance, for amputees who complain about the lack of haptic feedback necessary for everyday actions with upper-limb prostheses (Hochberg et al. 2006; Aflalo et al. 2015; D’Anna et al. 2017). In this article, we refer to grasp as the fixed hand posture used to hold an object (Feix et al. 2016), manipulation as handling or control of an object (i.e. with changing hand postures) and haptic feedback (Grafton 2010) as sensory inputs arising during the interaction with objects, from touch and proprioception. Human grasp of objects is highly stereotypical (Reilmann et al. 2001; Castiello 2005). A number of classification schemes have been proposed that capture this finite range of postures [see Schlesinger 1919, Taylor and Schwartz 1955 cited in Cutkosky (1989); Napier 1956; Kamakura et al. 1980; Kapandji 1989; Cutkosky 1989; Cesari and Newell 2000]. These taxonomies have recently been rationalised into a single classification of 33 grasp postures (Feix et al. 2016). One example of a mechanically simple grasp is achieved by a precision grip, a pinch using just index finger and thumb, which can be behaviourally characterised by the grip aperture (distance between thumb and index fingers) and by grip force. Precision grip aperture has been studied extensively in reach-and-grasp tasks, and experiments with individuals with haptic loss have revealed the importance of sensory feedback for accurate control of aperture during the reach towards the object (Jeannerod et al. 1984; Jeannerod 1986; Gentilucci et al. 1994; Simoneau et al. 1999; Jackson et al. 2000). Grip and load forces between the thumb and one (or all 4) fingers are also accurately controlled as the object is moved (Danion and Sarlegna 2007; Johansson and Flanagan 2008; Hermsdörfer et al. 2008). Again, the impact of haptic loss is well documented, for instance, when the finger pads are numb after anaesthesia, grip force is excessive and coordination is poor (Westling and Johansson 1984; Augurelle et al. 2003; Monzee et al. 2003). Excessive grip force and impaired coordination have also been reported in patients deprived of haptic feedback because of a sensory neuropathy (Gentilucci et al. 1994; Hermsdörfer et al. 2008; Thonnard et al. 1997). Using more than two digits can provide greater stability than the precision grip between thumb and index finger (Napier 1956; Cutkosky 1989; Saudabayev et al. 2018), albeit at the cost of controlling more muscles and joints. Studies have shown that haptic loss impairs handwriting and manipulating small objects (Gentilucci et al. 1994; Rothwell et al. 1982; Hepp-Reymond et al. 2009; Danna and Velay 2017). A patient with a severe haptic loss was found to perform a grooved pegboard test in ~ 14 min (Cuadra et al. 2019), which is ten times longer than controls (Ruff and Parker 1993). Augurelle et al. (2003) report (...truncated)