Two-digit grasping is a prototype of human motor coordination. Yet, there is an ongoing debate about how such movements are controlled (see for example Kleinholdermann et al., 2007; Smeets et al., 2010; van de Kamp and Zaal, 2007; Zaal and Bongers, 2014). Three main theories have been put forward. Jeannerod (1984) proposed that grasping consists of a transport component (movement of the wrist) that brings the hand to a location near the object and a grip component that ensures that the distance between the digits (grip aperture) changes in a manner that is appropriate for the size of the object. Others argue that the thumb is transported to a position on the object, with the index finger moving relative to the thumb to match grip aperture to the object’s size (Mon-Williams and McIntosh, 2000; Wing and Fraser, 1983). Finally, we have proposed that the trajectories of the tips of the digits when grasping is simply the combination of the trajectories of the same tips in goal-directed single-digit movements (‘pointing’), whereby each tip is moved to a specific position on the object (Schot et al., 2011; Smeets and Brenner, 1999; Smeets and Brenner, 2001; Verheij et al., 2012; Voudouris et al., 2013).

We have previously shown that people can simultaneously adapt movements of the finger and thumb of one hand to opposite prismatic displacements (Schot et al., 2014). In the current study, we let people grasp a cube and a cuboid before and after a period in which their single-digit movements were adapted in opposite directions. The three theories on the control of grasping make different predictions for how aftereffects of opposite adaptation during the single-digit movements will transfer to the grip aperture and the grip position during subsequent grasping movements (Table 1).

If, as Jeannerod (1984) suggested, grasping is controlled by planning the movement of the wrist and the opening and closing of the grip aperture, one would not expect opposite adaptation acquired during single-digit movements to transfer to grasping: the aftereffects of the opposite displacements of the wrist when adapting single-digit movements of the index finger and thumb will cancel each other during subsequent movements of both together when grasping, leaving the grip position unaltered. The grip aperture will also not be influenced by adapting pointing movements because it is irrelevant for single-digit movements.

If it is not the wrist movement that is controlled, but the movement of the thumb, as proposed by Mon-Williams and McIntosh (2000) and Wing and Fraser (1983), adaptation of single-digit thumb movements should lead to an aftereffect in the transport of the thumb. This aftereffect will be in the opposite direction than the visual displacement of the thumb in the adaptation phase. The movement of the index finger relative to the thumb will remain unaffected, again because grip aperture is irrelevant during a single-digit movement. The prediction is thus that grip position will show the aftereffect adaptation of the thumb, without any alteration of grip aperture.

Our own theory, based on the control of single-digit movements, predicts transfer of adaptation during pointing movements to the grip aperture, but not to the grip position, because when single-digit movements that are adapted in opposite directions are executed simultaneously, the aftereffects will cancel each other for the grip position, but will summate for grip aperture. So, if movements of the index finger are performed under leftward visual displacement and movements of the thumb are performed under rightward visual displacement during adaptation (thumb right), the predicted aftereffect is a rightward deviation of the index finger and a leftward deviation of the thumb, leading to a larger grip aperture. When the visual displacements are reversed, grip aperture should decrease.

The adaptation of pointing with the tip of each of the two single digits (central part of Figure 1A) was similar to the equivalent adaptation in our previous experiment (Schot et al., 2014). After removing the prisms, the digits’ movements during grasping showed a clear aftereffect of the adaptation during pointing. Interpreted in terms of a transport- and grip-component, we found a clear aftereffect of adapting to prisms during single-digit pointing movements on the grip aperture on subsequent grasping trials. In the thumb right adaptation, the target for pointing with the tip of the thumb was displaced to the right, and that for pointing with the tip of the index finger was displaced to the left. Adapting to these displacements made the thumb move more leftward and the index finger more rightward to a certain visual location. When done simultaneously, this should move the digits further apart. In accordance with our theory that grasping can be regarded as simultaneous movements of the individual digits, grip aperture did increase after thumb right adaptation (and decrease after thumb left adaptation), both for the small and the large block (Figure 1B). As the two digits were always adapted in opposite directions, we expected their average (which we use as a measure of grip position) to remain unaffected by the adaptation. Our results show that this is the case (Figure 1C).

Figure 1

Download asset Open asset

Figure 1—source data 1

Positions at 1 cm before crossing the board for of all trials underlying Figure 1 and the summary data underlying the central parts of Figure 1B,C.

https://doi.org/10.7554/eLife.21440.004

Download elife-21440-fig1-data1-v2.xlsx

Statistical analysis of the effects of session, adaptation phase, and target block size on grip aperture showed a significant phase by session interaction (F(1,7)=7.3, p=0.03) indicating that grip aperture indeed changed differently after thumb left and thumb right adaptation (Figure 1B). This difference is 1.2 cm. Besides the phase by session interaction, there were also significant effects of block size (the target cuboid was grasped with a 2.3 cm larger grip aperture than the target cube; F(1,7)=324.1, p<0.001) and of session (F(1,7)=24.8, p=0.002). Statistical analysis of the effects of session, adaptation phase, and target block size on the grip position did not show any significant effects (all p>0.1, Figure 1C). The difference corresponding to the non-significant tendency towards an adaptation phase by session interaction for grip position was only 0.1 cm, an order of magnitude smaller than the 1.2 cm that we found for grip aperture.

To give an indication of the relationship between our measure of grip aperture and the time course of changes in grip aperture during the movement, we plotted the time course of grip aperture during the last movement towards the target cube in one of the pre-adaptation phases and during the first movements towards the target cube in both post-adaptation phases for one representative participant (inset in Figure 1B). During the first movement towards the small block after thumb left adaptation (solid red curve), this participant’s digits came so close together that they touched each other, giving her haptic feedback that they should be further apart. By the time her digits crossed the far end of the board, she had already partly corrected her movement. Thus, our measure of grip aperture gives an underestimation of the initial aftereffect. For thumb right adaptation something similar may have happened in some trials when the digits move so far apart that grip aperture reaches its anatomical limit, but this is not the case for the example in the dashed blue curve in the figure.

We designed our experiment around a set of predictions that could distinguish between three theories of grasping (Table 1), ignoring the joints and muscles involved in making the movements. One might argue that the ability to adapt in opposite directions with two digits is related to transfer more easily occurring from proximal to distal muscles than from distal to proximal muscles (Krakauer et al., 2006). We have ascertained that adaptation does not only transfer to grip formation, which primarily involves distal muscles, but also transfers to grip position, which primarily involves proximal muscles. We did this by adapting both digits in the same direction (see Appendix 1 ‘Transfer to grip position’). Although such transfer is not distinctive for our theory, it shows that adapting single-digit pointing movements does not only influence grip aperture, but involves the complete reach-to-grasp movement.

Some readers of our original adaptation study (Schot et al., 2014) have argued that the opposite adaptation of the two digits might be related to the eye rather than to the digit that was adapted, as we used a fixed digit-eye pairing during adaptation. If this objection would be substantiated, it might also affect the interpretation of the present study. We therefore decided to repeat the original experiment (with new participants in a new set-up, full description in Appendix 2 ‘Effect of digit-eye pairing’) but using monocular single-digit pointing to measure the aftereffect. We did so using both the digit-eye pairing that we used in the adaptation phase and the opposite pairing. We replicated the results that we originally obtained with binocular viewing for monocular viewing. Most importantly, the aftereffect for each digit was equally large for both eyes, irrespective of whether the eye that was viewing was the one used to adapt the digit.

We show that adapting the thumb and index finger in opposite directions during single-digit movements gives rise to an aftereffect in grip aperture but not in grip position during subsequent grasping movements. Thus, the pattern of results is only consistent with the predictions based on our theory that grasping can be regarded as movements of the individual digits: adapting each digit during single-digit pointing movements influences the grip aperture during grasping movements. The two other theories predict quite different patterns of results (Table 1). Transfer of prism adaptation has been shown to be absent or only partial when tested with a different movement speed (Kitazawa et al., 1997), throwing style (Martin et al., 1996) or load (Fernández-Ruiz et al., 2000). This shows that adaptation can be very specific to a particular movement (van der Kooij et al., 2016). The fact that we find transfer from single-digit pointing to the grip aperture in grasping is therefore not at all trivial but suggests that grasping is indeed the combination of two single-digit movements.

All three theories limit themselves to describing the movements of the task-relevant elements. They differ in what they consider to be the relevant elements for grasping. None of them is concerned with how the movements are achieved in terms of the roles of all muscles and joints involved, so none predicts how the many task-irrelevant elements are orchestrated (Latash et al., 2002; van Beers et al., 2013). Only the transport-grip theory is concerned with the movement of the wrist (used as a proxy for the position of the hand), which is therefore task-relevant according to that theory. We have previously noted that using the wrist as a proxy for grip position makes it necessary to add a third controlled variable because the position to which the wrist should be transported to grasp an object depends on the orientation of the hand and the size of the object (Smeets and Brenner, 1999). We therefore use ‘grip position’ (the average of the positions of the two digits) as a proxy for hand position.

How the wrist, elbow and shoulder move (or, more generally, how the redundancy problem is solved) is beyond the scope of any of the mentioned theories. We therefore interpret our findings in terms of the movements of the endpoints of the digits, completely ignoring how these movements are achieved in terms of muscles and joints. In doing so, we ignore the evidence that adaptation is specific for the joints involved in the pointing movements (Krakauer et al., 2006). Such evidence makes our findings even more surprising, because as can be seen in the example in Video 1, the participants mainly performed the pointing movements and their corrections from the elbow and shoulder, whereas changes in grip aperture must involve changing the angles of the digits relative to the hand. This suggests that the transfer that we found to grip aperture in grasping is not based on transfer of a joint-specific adaptation, but on a higher-level control of the trajectory of the digit in space.

Video 1

Download asset

Finding transfer from transporting the separate digits in pointing to grip aperture in grasping (Figure 1B) contradicts the idea that grip aperture is controlled separately from transport of the hand or thumb (see Table 1). Finding that grip position was not systematically further to the left after thumb right adaptation (Figure 1C) contradicts the theory that the thumb determines the transport of the hand. Both the presence of an effect on grip aperture and the lack of effect on grip position are consistent with similar but independent adaptation of the movements of finger and thumb. Finding that the grip aperture can increase as well as decrease depending on the direction of the adaptation shows that the effect is due to the adaptation itself, rather than for instance uncertainty about the size of the object leading to a larger hand opening.

The observation that the digits sometimes touch each other during the grasping movement (as in the example in the inset of Figure 1B) is also consistent with independent adaptation of the movements of finger and thumb. With independent control of the thumb and index finger, it is possible that people plan digit trajectories that cannot be executed simultaneously for anatomical reasons. This seems indeed to have happened in this study. Neither of the other theories can account for the digits touching each other during the grasping movement because if the aperture is controlled it is only expected to decrease (after reaching maximal grip aperture) until it matches the size of the object.

Together, these results provide strong evidence that grasping can be considered to be built up of the same movements as single-digit pointing: grasping is the combination of two goal-directed single-digit movements (Smeets and Brenner, 1999).

The methods employed in this study, are very similar to those of our original adaptation study (Schot et al., 2014). The main difference is the use of grasping instead of single-digit pointing in the pre- and post-adaptation phases. This study was part of a research programme that has been approved by the Ethische Commissie Bewegingswetenschappen (ECB 2006–02). All participants gave their informed consent before participating in this study.

Procedure

Request a detailed protocol

Eight participants (25–29 years of age, two males) took part in the experiment after giving informed consent. They wore PLATO shutter glasses. Movements of Infrared emitting diodes attached to the fingernails of the thumb and index finger were recorded at 250 Hz using an Optotrak 3020 system. A target object was attached to a wooden board at one of three possible target locations (5 cm apart). The board obstructed the participants’ vision of the hand until just before contact with the target object.

The experiments consisted of three phases: pre- and post-adaptation phases which each consisted of 30 grasp trials, and an adaptation phase which consisted of 30 pointing trials with the thumb and 30 pointing trials with the index finger. Three objects were used: a starting cube (2.3 cm), a target cube (2.3 cm), and a target cuboid (2.3 × 2.3×4.6 cm). The starting cube and the target cube were used in all three phases. The target cuboid was only used in the pre- and post-adaptation phases. It was included to force participants to use visual information to guide their grip aperture and to ensure that any observed adaptation effects are not specific to the object used in the adaptation phase.

Participants started all trials by grasping the starting cube with the glasses shut (Figure 2, Video 1). In the pre- and post- adaptation phases (i.e., on grasp trials), participants viewed the target binocularly without prism glasses. Once the shutter glasses opened, they grasped either the target cube or the target cuboid by its left and right surfaces. The cuboid was oriented such that it was grasped by the surfaces that were twice as far apart as those of the cube. In the adaptation phase (i.e., on single-digit pointing trials), participants viewed the target monocularly through 10 diopter prism glasses that were worn over the shutter glasses. The prism in front of the left eye shifted the image of the target about 5 cm to the right. The prism in front of the right eye shifted the image of the target about 5 cm to the left. Before each pointing trial, participants were told either to touch the left side of the target cube with their thumb or to touch the right side of the cube with their index finger. Once the shutter glasses in front of the corresponding eye opened, participants moved the appropriate digit to the appropriate side of the target cube.

Figure 2

Download asset Open asset

The experiment consisted of two sessions that were performed on separate days with their order counterbalanced across participants. In one session, vision was displaced to the right when touching the cube with the index finger and to the left when touching the cube with the thumb (thumb left). In the other session, the displacements were reversed (thumb right). Within each session, trials were presented in pseudo-random order that ensured that all trial-types were distributed evenly: in the pre- and post-adaptation phase, each combination of target size and target location was presented once every six trials; in the adaptation phase, each combination of digit and target location was presented once every six trials.

1. 1. Fernández-Ruiz J
   2. Hall-Haro C
   3. Díaz R
   4. Mischner J
   5. Vergara P
   6. Lopez-Garcia JC

   (2000) Learning motor synergies makes use of information on muscular load

   Learning & Memory 7:193–198.

   https://doi.org/10.1101/lm.7.4.193

   • PubMed
   • Google Scholar
2. 1. Jeannerod M

   (1984) The timing of natural prehension movements

   Journal of Motor Behavior 16:235–254.

   https://doi.org/10.1080/00222895.1984.10735319

   • PubMed
   • Google Scholar
3. 1. Kitazawa S
   2. Kimura T
   3. Uka T

   (1997)

   Prism adaptation of reaching movements: specificity for the velocity of reaching

   Journal of Neuroscience 17:1481–1492.

   • PubMed
   • Google Scholar
4. 1. Kleinholdermann U
   2. Brenner E
   3. Franz VH
   4. Smeets JBJ

   (2007) Grasping trapezoidal objects

   Experimental Brain Research 180:415–420.

   https://doi.org/10.1007/s00221-007-0867-6

   • PubMed
   • Google Scholar
5. 1. Krakauer JW
   2. Mazzoni P
   3. Ghazizadeh A
   4. Ravindran R
   5. Shadmehr R

   (2006) Generalization of motor learning depends on the history of prior action

   PLoS Biology 4:e316.

   https://doi.org/10.1371/journal.pbio.0040316

   • PubMed
   • Google Scholar
6. 1. Kuling IA
   2. Brenner E
   3. Smeets JBJ

   (2016) Errors in visuo-haptic and haptic-haptic location matching are stable over long periods of time

   Acta Psychologica 166:31–36.

   https://doi.org/10.1016/j.actpsy.2016.03.011

   • PubMed
   • Google Scholar
7. 1. Latash ML
   2. Scholz JP
   3. Schöner G

   (2002) Motor control strategies revealed in the structure of motor variability

   Exercise and Sport Sciences Reviews 30:26–31.

   https://doi.org/10.1097/00003677-200201000-00006

   • PubMed
   • Google Scholar
8. 1. Martin TA
   2. Keating JG
   3. Goodkin HP
   4. Bastian AJ
   5. Thach WT

   (1996) Throwing while looking through prisms. II. Specificity and storage of multiple gaze-throw calibrations

   Brain 119 ( Pt 4:1199–1211.

   https://doi.org/10.1093/brain/119.4.1199

   • PubMed
   • Google Scholar
9. 1. Mon-Williams M
   2. McIntosh RD

   (2000) A test between two hypotheses and a possible third way for the control of prehension

   Experimental Brain Research 134:268–273.

   https://doi.org/10.1007/s002210000479

   • PubMed
   • Google Scholar
10. 1. Schot WD
    2. Brenner E
    3. Smeets JBJ

    (2011) Grasping and hitting moving objects

    Experimental Brain Research 212:487–496.

    https://doi.org/10.1007/s00221-011-2756-2

    • PubMed
    • Google Scholar
11. 1. Schot WD
    2. Brenner E
    3. Smeets JB

    (2014) Simultaneous adaptation of the thumb and index finger of the same hand to opposite prism displacements

    Journal of Neurophysiology 111:2554–2559.

    https://doi.org/10.1152/jn.00326.2013

    • PubMed
    • Google Scholar
12. 1. Smeets JBJ
    2. Brenner E

    (1999) A new view on grasping

    Motor Control 3:237–271.

    https://doi.org/10.1123/mcj.3.3.237

    • PubMed
    • Google Scholar
13. 1. Smeets JBJ
    2. Brenner E

    (2001) Independent movements of the digits in grasping

    Experimental Brain Research 139:92–100.

    https://doi.org/10.1007/s002210100748

    • PubMed
    • Google Scholar
14. 1. Smeets JBJ
    2. Martin J
    3. Brenner E

    (2010) Similarities between digits' movements in grasping, touching and pushing

    Experimental Brain Research 203:339–346.

    https://doi.org/10.1007/s00221-010-2236-0

    • PubMed
    • Google Scholar
15. 1. van Beers RJ
    2. Brenner E
    3. Smeets JBJ

    (2013) Random walk of motor planning in task-irrelevant dimensions

    Journal of Neurophysiology 109:969–977.

    https://doi.org/10.1152/jn.00706.2012

    • PubMed
    • Google Scholar
16. 1. van de Kamp C
    2. Zaal FTJM

    (2007) Prehension is really reaching and grasping

    Experimental Brain Research 182:27–34.

    https://doi.org/10.1007/s00221-007-0968-2

    • PubMed
    • Google Scholar
17. 1. van der Kooij K
    2. Overvliet KE
    3. Smeets JBJ

    (2016) Temporally stable adaptation is robust, incomplete and specific

    European Journal of Neuroscience 44:2708–2715.

    https://doi.org/10.1111/ejn.13355

    • PubMed
    • Google Scholar
18. 1. Verheij R
    2. Brenner E
    3. Smeets JBJ

    (2012) Grasping kinematics from the perspective of the individual digits: a modelling study

    PLoS One 7:e33150.

    https://doi.org/10.1371/journal.pone.0033150

    • PubMed
    • Google Scholar
19. 1. Voudouris D
    2. Smeets JBJ
    3. Brenner E

    (2013) Ultra-fast selection of grasping points

    Journal of Neurophysiology 110:1484–1489.

    https://doi.org/10.1152/jn.00066.2013

    • PubMed
    • Google Scholar
20. 1. Wing AM
    2. Fraser C

    (1983) The contribution of the thumb to reaching movements

    The Quarterly Journal of Experimental Psychology Section A 35:297–309.

    https://doi.org/10.1080/14640748308402135

    • PubMed
    • Google Scholar
21. 1. Zaal FT
    2. Bongers RM

    (2014) Movements of individual digits in bimanual prehension are coupled into a grasping component

    PLoS One 9:e97790.

    https://doi.org/10.1371/journal.pone.0097790

    • PubMed
    • Google Scholar

Author details

1. Willemijn D Schot

   Department of Human Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam, Netherlands

   Present address

   Educational Development and Training, Utrecht University, Utrecht, Netherlands

   Contribution

   Conceptualization, Investigation, Writing—original draft

   Competing interests

   No competing interests declared

2. Eli Brenner

   Department of Human Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam, Netherlands

   Contribution

   Conceptualization, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3611-2843

3. Jeroen BJ Smeets

   Department of Human Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam, Netherlands

   Contribution

   Conceptualization, Funding acquisition, Writing—review and editing

   For correspondence

   J.B.J.Smeets@vu.nl

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3794-0579

• 1,057

  views

• 154

  downloads

• 12

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.