Bud I Savlje Vic 2018

CORTEX2260_proof ■ 22 March 2018 ■ 1/16
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
Available online at www.sciencedirect.com

65
66
1
2
ScienceDirect 67
68
3 69
4 Journal homepage: www.elsevier.com/locate/cortex
70
5 71
6 72
7 73
8 74
9 Clinical neuroanatomy 75
10 76
11
12
Cross-talk connections underlying dorsal and 77
78
13
14 Q8 ventral stream integration during hand actions 79
80
15 81
16 82
17 Q7 Sanja Budisavljevic a,*, Flavio Dell'Acqua b and Umberto Castiello a,c 83
18 a 84
Department of General Psychology, University of Padova, Padova, Italy
19 b 85
Natbrainlab, Department of Neuroimaging, Institute of Psychiatry, Psychology and Neuroscience, King's College
20 86
21 London, London, United Kingdom
c 87
22 Centro Linceo Interdisciplinare, Accademia Nazionale dei Lincei, Roma, Italy
88
23 89
24 90
25 article info abstract 91
26 92
27 Article history: According to the dual-stream theory, the processing of visual information is divided into a 93
28 Received 25 August 2017 ventral pathway mediating object recognition, and a dorsal pathway supporting visuo- 94
29 Reviewed 16 November 2017 motor control. Increasing evidence suggests that these streams are not independent, but 95
30 Revised 15 January 2018 where this dorsal-ventral stream integration occurs, in the context of hand actions, re- 96
31 97
Accepted 21 February 2018 mains unknown. We explored the candidate white matter pathways linking dorsal and
32 98
Action editor Georg Goldenberg ventral visual streams in 30 right-handed participants performing hand movements of
33 99
Published online xxx varying complexity (reaching, grasping and lifting), using advanced diffusion imaging
34 100
35 tractography based on the spherical deconvolution and kinematical analysis. We provided
101
36 Keywords: for the first time a direct evidence of cross-communication between dorsal and ventral
102
37 Arcuate fasciculus visual streams in humans, through vertical occipital fasciculus and temporo-parietal fibers
103
38 Dorsal stream of the arcuate fasciculus during on-line control of skilled object-oriented hand actions. We
104
39 Skilled actions showed that individual differences in the microstructure of these cross-talk connections 105
40 Ventral stream were associated with the variability of arm deceleration, the grip opening phase and the 106
41 Vertical occipital fasciculus grasp accuracy. This study suggests that hand kinematics, in skilled hand actions where 107
42 high degree of online control is required, is related to the anatomy of the dorsal-ventral 108
43 109
networks, bringing important insights to the dual-stream theory and the sensorimotor
44 110
organization of hand actions.
45 111
46 © 2018 Elsevier Ltd. All rights reserved.
112
47 113
48 114
49 115
50 116
51 117
52 pathway extending to the inferotemporal cortex that medi- 118
53 Q1 1. Introduction 119
ates object recognition, and a dorsal ‘action’ pathway projec-
54 120
ting to the posterior parietal cortex that supports visuomotor
55 When we reach, grasp and lift an object, fast integration of 121
56 control (Goodale, 2014; Milner & Goodale, 2008; Rizzolatti &
sensory information is crucial for performing the movement
Matelli, 2003; Ungerleider & Mishkin, 1982). While a degree 122
57 successfully. According to a dual-stream theory, the process- 123
58 of functional specialization and segregation invariably exists
ing of visual information is divided into a ventral ‘perception’ 124
59 125
60 126
61 * Corresponding author. University of Padova, Via Venezia 8, 35131, Padova, Italy. 127
62 E-mail address: sanja.budisavljevic@gmail.com (S. Budisavljevic). 128
63 https://doi.org/10.1016/j.cortex.2018.02.016 129
64 0010-9452/© 2018 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Budisavljevic, S., et al., Cross-talk connections underlying dorsal and ventral stream integration
during hand actions, Cortex (2018), https://doi.org/10.1016/j.cortex.2018.02.016
2 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 between the two streams, information processed by the two target brain regions (e.g., prefrontal cortex), ii) feedback loops, 66
2 networks must closely interact in everyday life, especially for and iii) by ‘continuous cross-talk’ at multiple stages and lo- 67
3 more complex behavior such as skilled hand actions cations through direct lateral connections between the two 68
4 (Cloutman, 2013; Milner, 2017; van Polanen & Davare, 2015). streams (Cloutman, 2013). This integration might be especially 69
5 70
Nevertheless, little attention has been payed to how, where, important for more complex actions, when the dorsal stream
6 71
and when the dual streams interact. needs to retrieve detailed information about object identity
7 72
Significant insights first emerged from anatomical tract stored in the ventral stream areas, while the ventral stream
8 73
9 tracing studies in monkeys, describing direct reciprocal in- receives up-to-date grasp-related information from dorsal 74
10 terconnections between the two visual streams. For example, areas to refine the object internal representation (van Polanen 75
11 ventral inferior temporal area TE has strong bidirectional & Davare, 2015). Up to date, no study has specifically investi- 76
12 projections to the inferior parietal lobe (Zhong & Rockland, gated the ‘continuous cross-talk’ possibility. Our study aims to 77
13 2003) and prefrontal cortex (Borra, Ichinohe, Sato, Tanifuji, & fill this gap by exploring the candidate cross-talk connections 78
14 Rockland, 2010; Gerbella, Belmalih, Borra, Rozzi, & Luppino, in humans and their role in skilled actions. 79
15 2010), while the anterior intraparietal sulcus is inter- Based on the current neuroanatomical models of dual 80
16 connected to the superior and middle temporal gyri of the stream processing (Distler, Boussaoud, Desimone, & 81
17 82
Q2 ventral stream (Borra et al., 2008). Similarly, diffusion mag- Ungerleider, 1993; Felleman & Van Essen, 1991; Ungerleider,
18 83
netic resonance imaging (MRI) studies in humans reported Galkin, Desimone, & Gattass, 2008) we hypothesized that the
19 84
20 white matter tracts between superior/middle temporal and vertical occipital fasciculus (VOF) and the posterior segment of 85
21 inferior parietal regions (Budisavljevic et al., 2015; Catani et al., the arcuate fasciculus (pAF), both connecting the lateral sur- 86
22 2007, 2005) and between dorsal and ventral occipital visual faces of the ventral and dorsal streams, enable the continuous 87
23 areas (Yeatman, Rauschecker, & Wandell, 2013, 2014). Thus, cross-talk between the two visual streams (Fig. 1A). The VOF is 88
24 anatomy points to inter-stream communication that could the only major white matter pathway allowing the commu- 89
25 underlie integration between the two visual pathways. From a nication between dorsal and ventral visual maps with full 90
26 functional point of view, studies of the time course and hemifield representations (Takemura et al., 2016; Weiner, 91
27 laminar activation profiles in the monkey brain, showed that Yeatman, & Wandell, 2016; Yeatman et al., 2014, 2013). The 92
28 93
the two networks engage in a direct cross-talk at multiple VOF likely carries signals from the ventral regions that encode
29 94
stages, at least within the visual domain (Chen et al., 2007; object properties such as form, identity, and color (Cohen
30 95
Givre, Schroeder, & Arezzo, 1994; Oram & Perrett, 1996; et al., 2000; Malach et al., 1995; McKeefry & Zeki, 1997; Wade,
31 96
32 Schroeder, Mehta, & Givre, 1998). Similarly, functional MRI Brewer, Rieger, & Wandell, 2002; Zeki et al., 1991) to dorsal 97
33 studies in humans observed that the two streams are not in- regions that map spatial location to action plans (Fischer, 98
34 dependent, since they exhibit strong functional connectivity Bulthoff, Logothetis, & Bartels, 2012; Merriam, Gardner, 99
35 during object recognition processes (Freud, Rosenthal, Ganel, Movshon, & Heeger, 2013; Tootell et al., 1997). Parallel and 100
36 & Avidan, 2015; Sim, Helbig, Graf, & Kiefer, 2015) and likely anterior to the VOF, the neighbouring pAF is a temporo- 101
37 communicate regarding the visual and motor dimensions parietal pathway intermingled within the arcuate fasciculus 102
38 relevant for the execution of hand actions (Bracci & Peelen, fibers, connecting the inferior parietal lobe with the superior 103
39 2013; Fabbri, Stubbs, Cusack, & Culham, 2016; Konen & and middle temporal gyri (Budisavljevic et al., 2015; Catani 104
40 105
Kastner, 2008; Mahon, Kumar, & Almeida, 2013; Oosterhof, et al., 2007, 2005; Ramayya, Glasser, & Rilling, 2010; Thiebaut
41 106
Wiggett, Diedrichsen, Tipper, & Downing, 2010). de Schotten, Cohen, Amemiya, Braga, & Dehaene, 2014). The
42 107
43 Overall, evidence suggests that the integration between the pAF connections could represent part of a ventro-dorsal 108
44 two streams occurs at different levels, including i) shared network of the extended dual-stream processing model 109
45 110
46 111
47 112
48 113
49 114
50 115
51 116
52 117
53 118
54 119
55 120
56 121
57 122
58 123
59 124
60 125
61 126
62 127
63 Fig. 1 e Descriptive example of A) the cross-talk connections (tracts of interest) including the VOF (in blue) and the pAF (in 128
64 yellow); and B) the control tract, ILF in the left hemisphere (in green). 129
65 130
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 3
1 (Rizzolatti & Matelli, 2003), which serves as an integration 66

2 node and plays a role in both perception and action (Milner,
2. Materials and methods 67
3 1997; Rizzolatti & Matelli, 2003; Shapiro, Hillstrom, & Husain, 68
4 2.1. Participants 69
2002; Singh-Curry & Husain, 2009).
5 70
In this study we tested the role of these cross-talk con-
6 A sex and age-balanced sample of 30 healthy participants (13 71
nections in hand actions of increasing complexity: i) simple
7 males; mean age 24.6 ± 2.8, age range: 20e31 years) was 72
reaching movements, which rely on the spatial information
8 recruited. All participants were right-handed according to the 73
9 about the body and the target object, ii) reach-to-grasp 74
Edinburgh Handedness Inventory (Oldfield, 1971). No history
10 movements (referred to as ‘grasping’), which require addi- 75
of neurological and psychiatric disorders was present in the
11 tional sensorimotor processing of 3D intrinsic object proper- 76
study sample. All participants gave informed written consent
12 ties, and lastly iii) reach-to-grasp-to-lift actions (referred to as 77
13 in accordance with the ethics approval by the Institutional 78
‘lifting’), which compared to grasping also integrate non-
14 Review Board at the University of Padova, in accordance with 79
visual object's weight information in order to scale fingertip
15 the Declaration of Helsinki (Sixth revision, 2008). 80
forces accordingly. Diffusion imaging tractography based on
16 the spherical deconvolution (SD) was chosen over the classical 81
17 2.2. Behavioral experiment 82
diffusion tensor model because it can better resolve the
18 83
crossing fibers problem, affecting both the VOF and the pAF,
19 84
and improve tractography reconstructions by reducing the 2.2.1. Task and stimulus
20 85
21 presence of false negatives (Tournier, Calamante, Gadian, & Participants were requested to perform three tasks: i) a 86
22 Q3 Connelly, 2004; Dell'Acqua et al., 2010; Dell'Acqua, Simmons, reaching task, in which they were asked to perform a move- 87
23 Williams, & Catani, 2013). The VOF and the pAF were ment toward the stimulus and touch the stimulus frontal 88
24 dissected in 30 right-handed healthy participants, whose surface with their knuckles, maintaining the hand in a closed 89
25 hand kinematics was separately recorded for reaching, fist, ii) a reach-to-grasp task (referred to as ‘grasping’), in 90
26 grasping and lifting movements. In order to test the specificity which they were asked to reach toward and grasp the stim- 91
27 of our findings, we also dissected a control tract e the left ulus with a precision grip (using the index finger and thumb), 92
28 and iii) a reach-to-grasp-to-lift task (referred to as ‘lifting’), in 93
inferior longitudinal fasciculus (ILF) (Fig. 1B). This ventral
29 94
temporo-occipital bundle, connecting visual areas to the which they reached toward and grasped the stimulus with a
30 95
amygdala and the hippocampus (Catani, Jones, Donato, & precision grip in order to move the object to an empty
31 96
ffytche, 2003), plays a role in a variety of visual functions, container (Fig. 2). We maintained the same physical re-
32 97
33 including visual perception (ffytche, 2008; ffytche & Catani, quirements between our tasks, and for grasping and lifting 98
34 2005), visual memory (Ross, 2008), and face recognition (Fox, movements only the first ‘reach-to-grasp’ part was kinemat- 99
35 Iaria, & Barton, 2008). The hindrance modulated orienta- ically analyzed. The fist's posture was chosen as to minimize 100
36 tional anisotropy (HMOA) was used as a measure of white the distal involvement. Participants fixated the target object 101
37 matter microstructure, representing a tract-specific index, during all actions. The stimulus consisted of a spherical object 102
38 and better reflecting the microstructural organization (e.g., (2 cm diameter) that would normally be grasped with a pre- 103
39 myelination, axonal density, axon diameter and fiber disper- cision grip. The size of the target object was kept the same, in 104
40 order to maintain the same physical setting for the three 105
sion) than the traditional voxel-specific diffusion indices (e.g.,
41 106
fractional anisotropy) (Dell'Acqua et al., 2010, 2013). tasks. All participants were explicitly asked to use a precision
42 107
Increasing task complexity may require an increased grip for grasping and lifting tasks. Trials in which the partic-
43 108
44 contribution of cortical processing carried along the ventral ipants did not comply with the task or did not fixate the 109
45 visual stream areas and transfer of object-related informa- stimulus were not included in the analysis. 110
46 tion between ventral and dorsal streams. Thus, we hy- 111
47 pothesized that the nature of the task would modulate the 2.2.2. Procedure 112
48 correlation between visuomotor behavior and white matter Each participant sat on a height-adjustable chair in front of a 113
49 microstructure. Associations between the cross-talk con- table (900 900 mm) with the elbow and wrist resting on the 114
50 nections and the hand kinematics should be observed dur- table surface and the right hand in the designated start posi- 115
51 tion. The hand was pronated with the palm resting on a pad 116
ing grasping and lifting movements where more detailed
52 117
(ventral) information about the object is needed, compared (60 70 mm), which was shaped to allow for a comfortable and
53 118
to reaching actions that rely predominantly on the dorsal repeatable posture of all digits, i.e., slightly flexed at the
54 119
stream. It should be noted that this is the first explorative metacarpal and proximal interphalangeal joints. The starting
55 120
56 study to test the ‘continuous cross-talk’ hypothesis in pad was attached 90 mm away from the edge of the table sur- 121
57 humans, and it was not designed to confirm or falsify spe- face. The object was placed on a platform located at a distance 122
58 cific predictions on the nature of the expected correlations. of 300 mm between the platform and the sagittal plane of the 123
59 We could though expect that the arm deceleration phase, hand's starting position on the right side of the table (Fig. 2). 124
60 which depends on visual feedback and involves sensori- 125
61 motor adjustments for an efficient and accurate grasping 2.2.3. Kinematics recording 126
62 and lifting of the object, together with specific measures A 3D-Optoelectronic SMART-D system (Bioengineering Tech- 127
63 nology and Systems, BTS) was used to track the kinematics of 128
concerned with the formation of the grasp, would be
64 129
significantly associated with the anatomy of the lateral the participant's right upper limb. Three light-weight infrared
65 130
cross-talk connections. reflective markers (.25 mm in diameter; BjTjSj) were taped to
4 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 66
2 67
3 68
4 69
5 70
6 71
7 72
8 73
9 74
10 75
11 76
12 77
13 78
14 79
15 80
16 Fig. 2 e (A) Descriptive example of the experimental set up showing designated start position, target location, container location 81
17 (white circle) and infrared cameras position. (B) Schematic drawing representing reaching, grasping and lifting movements. 82
18 83
19 84
the following points: (i) thumb (ulnar side of the nail); (ii) index movement onset (Time to Peak Wrist Velocity), the time at
20 85
21 finger (radial side of the nail); and (iii) wrist (dorsodistal aspect which the acceleration of the wrist was maximum from 86
22 of the radial styloid process). Six video cameras (sampling rate movement onset (Time to Peak Acceleration), the time at 87
23 140 Hz) detecting the markers were placed in a semicircle at a which the deceleration of the wrist was maximum from 88
24 distance of 1e1.2 m from the table. The camera position, roll movement onset (Time to Peak Deceleration). For grasping 89
25 angle, zoom, focus, threshold, and brightness were calibrated and lifting tasks additional ‘grasp’ kinematics was assessed, 90
26 and adjusted to optimize marker tracking, followed by static namely the time at which the distance between the thumb 91
27 and dynamic calibration. For the static calibration, a three- and index finger was maximum, between movement onset 92
28 axes frame of five markers at known distances from each and hand contact time (Time to Maximum Grip Aperture) and 93
29 94
other was placed in the middle of the table. For the dynamic its amplitude (Amplitude of Maximum Grip Aperture); the
30 95
calibration, a three-marker wand was moved throughout the standard deviation of the Amplitude of Maximum Grip Aper-
31 96
32 workspace of interest for 60s. The spatial resolution of the ture (Variability of the Maximum Grip Aperture) e a measure 97
33 recording system was .3 mm over the field of view. The used to characterize the uncertainty or perceptual and/or 98
34 standard deviation of the reconstruction error was below motor inconsistency (i.e., accuracy) of the grasp formation 99
35 .2 mm for the x-, y-, and z-axes. (Flindall, Doan, & Gonzalez, 2014); and the time and amplitude 100
36 of the maximum opening and closing grip velocity. 101
37 2.2.4. Data processing 102
38 Following data collection, each trial was individually checked 2.3. MRI data acquisition 103
39 for correct marker identification and the SMART-D Tracker 104
40 Diffusion imaging data was acquired using a Siemens Avanto 105
software package (BjTjSj) was used to provide a 3-D recon-
41 106
struction of the marker positions as a function of time. The 1.5T scanner housed in Padova University Hospital with actively
42 107
data were then filtered using a finite impulse response linear shielded magnetic field gradients (maximum amplitude 45 mT/
43 108
44 filter (transition band ¼ 1 Hz, sharpening variable ¼ 2, cut-off m1). The body coil was used for RF transmission, and an 8- 109
45 frequency ¼ 10 Hz; D'Amico & Ferrigno, 1990, 1992). Movement channel head coil for signal reception. Protocol consisted of a 110
46 onset was defined as the time at which the tangential velocity localizer scan, followed by a single-shot, spin-echo, EPI 111
47 of the wrist marker crossed a threshold (5 mm/sec) and sequence with the following parameters: TR ¼ 8500, TE ¼ 97, 112
48 remained above it for longer than 500 msec. For the grasping FOV ¼ 307.2 307.2, matrix size ¼ 128 128, 60 slices (no gaps) 113
49 and lifting tasks the end of movement was defined as the time with isotropic (2.4 2.4 2.4 mm3) voxels. The maximum 114
50 at which the hand made contact with the stimulus and diffusion weighting was 2000 sec/mm2, and at each slice loca- 115
51 quantified as the time at which the hand opening velocity tion 7 images were acquired with no diffusion gradients applied 116
52 117
crossed a threshold (5 mm/sec) after reaching its minimum (b ¼ 0 sec/mm2), together with 64 diffusion-weighted images in
53 118
value and remained above it for longer than 500 msec. For the which gradient directions were uniformly distributed in space
54 119
reaching task the end of movement was defined as the time at and repeated 3 times, in order to increase signal to noise ratio
55 120
56 which the hand made contact with the stimulus and quanti- (SNR). Gains and scaling factors were kept constant between 121
57 fied as the time at which the wrist velocity crossed a threshold acquisitions. Scanning lasted approximately 30 min. 122
58 (5 mm/sec) after reaching its minimum value and remained 123
59 above it for longer than 500 msec. For each of the three tasks, 2.4. Correction of motion and eddy current distortion, 124
60 12 trials were administered in a randomized order and the and estimation of the fiber orientation distribution (FOD) 125
61 following kinematic ‘reach’ parameters were extracted using 126
62 a custom protocol run in Matlab 2014b (The 4 MathWorks, Each subject's raw image data were examined before pro- 127
63 ceeding on to further analyses to detect any outliers in the 128
Natick, MA, USA): the time interval between movement onset
64 129
and end of movement (Movement Time), the time at which data, including signal drop-outs, poor signal-to-noise ratio,
65 130
the tangential velocity of the wrist was maximum from and image artifacts such as ghosts. Any subject whose raw
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 5
1 data contained volumes with significant image quality issues vertically oriented principal diffusion direction (color-coded 66
2 was removed from further analyses. The remaining 30 par- in blue). We retained the candidate fibers within the occipital 67
3 ticipants were processed as follows. lobe that are primarily vertical and removed fibers that 68
4 DWI datasets were concatenated and corrected for subject belonged to the pAF. This method has been previously 69
5 70
motion and geometrical distortions using ExploreDTI (http:// described in Yeatman et al. (2013, 2014).
6 71
www.exploredti.com; Leemans, Jeurissen, Sijbers, & Jones,
7 72
2009). SD (Tournier, Calamante, & Connelly, 2007, 2004; 2.6.2. pAF
8 73
9 Dell'Acqua et al., 2007) approach was chosen to estimate We used a two regions of interest (ROIs) approach to isolate 74
10 multiple orientations in voxels containing different pop- the pAF, according to a dissection method previously 75
11 ulations of crossing fibers. SD was calculated applying the described in Catani, Jones, and ffytche (2005, 2007) and 76
12 damped version of the Richardson-Lucy algorithm with a fiber Budisavljevic et al. (2015). Two separate ROIs were manually 77
13 response parameter a ¼ 1.5, 200 algorithm iterations and delineated on the fractional anisotropy (FA) maps of each 78
14 h¼.15 and n ¼ 15 as threshold and geometrical regularization subject in the inferior parietal lobule and the posterior tem- 79
15 parameters (Dell'Acqua et al., 2010). An example of the poral region of each hemisphere. All streamlines passing 80
16 recovered FOD profiles obtained with these settings is pro- through these two ROIs were considered to belong to the pAF. 81
17 82
vided in the Supplementary Material (Figure S1). Fiber orien-
18 83
tation estimates were obtained by selecting the orientation 2.6.3. ILF
19 84
20 corresponding to the peaks (local maxima) of the FOD profiles. We used a two ROIs approach to dissect the ILF that acted as a 85
21 To exclude spurious local maxima, we applied both an abso- control tract, according to a method previously described in 86
22 lute and a relative threshold on the FOD amplitude Catani et al. (2003). The first ROI was manually placed around 87
23 (Dell'Acqua, Simmons, Williams, & Catani, 2013). The first the white matter of the anterior temporal lobe, while the sec- 88
24 “absolute” threshold corresponding to a HMOA threshold of ond ROI was defined around the white matter of the occipital 89
25 .015 was used to exclude intrinsically small local maxima due lobe, with both ROIs drawn on consecutive axial FA slices. 90
26 to noise or partial volume effects with isotropic tissue. This Lastly, although novel multi fiber methods like SD improve 91
27 threshold was set to select only the major fiber orientation the visualization of anatomical connections not visible with 92
28 93
components and exclude low amplitude spurious FOD com- traditional diffusion tensor methods, SD can produce more
29 94
ponents obtained from gray matter and cerebro-spinal fluid false positives. To control for this bias we visually inspected
30 95
isotropic voxels. The second “relative” threshold of 5% of the all the tracts, in addition to using an absolute threshold based
31 96
32 maximum amplitude of the FOD was applied to remove on HMOA, as explained previously. 97
33 remaining unreliable local maxima with values greater than 98
34 the absolute threshold but still significantly smaller than the 2.7. Statistical analysis 99
35 main fiber orientation (Dell'Acqua et al., 2013). 100
36 Statistical analysis was performed using SPSS software (Version 101
37 2.5. Tractography algorithm 21) (SPSS, Chicago, IL). Gaussian distribution was confirmed for 102
38 kinematic and tractography HMOA variables using the 103
39 Whole brain tractography was performed selecting every ShapiroeWilk test (a-level: p < .05) (Shapiro & Wilk, 1965) 104
40 brain voxel with at least one fiber orientation as a seed voxel. 105
allowing the use of parametric statistics. The mean value for
41 From these voxels, and for each fiber orientation, streamlines 106
each kinematic parameter of interest was determined based on
42 were propagated using a modified Euler integration with a 107
12 individual observations for each participant. Repeated mea-
43 step size of 1 mm. When entering a region with crossing white 108
44 sures ANOVA analyses were performed separately for each ki- 109
matter bundles, the algorithm followed the orientation vector nematic measurement (i.e., total movement time, time of peak
45 110
of the least curvature. Streamlines were halted when a voxel
46 velocity, acceleration, deceleration) to examine kinematical 111
without fiber orientation was reached or when the curvature
47 differences between the three movement tasks (i.e., the fixed 112
48 between two steps exceeded a threshold of 45 . All SD and 113
factor). Paired-samples t-tests were conducted for comparing
49 tractography processing was performed using StarTrack, a 114
grasp-specific measures of grasping versus lifting movements.
50 freely available Matlab software toolbox developed by one of
Post hoc comparisons for specific tasks were considered sta- 115
51 the authors (F.D. NatBrainLab, King's College London), and tistically significant if they survived Bonferroni correction for 116
52 based on the methods described in Dell'Acqua et al. (2013). 117
multiple comparisons (a-level: p < .05).
53 118
2.6. Tractography dissections of dorsal-ventral white The mean HMOA was calculated for each tract in each
54 119
55 matter connections subject, defined as the absolute amplitude of each lobe of the 120
56 FOD, and representing a quantitative index of the degree of 121
57 To visualize fibre tracts and quantify tract-specific measures tract anisotropy. HMOA was chosen over voxel-based FA as it 122
58 we used TrackVis software (http://www.trackvis.org; Wang, is considered to be a tract-specific index highly sensitive to 123
59 Benner, Sorensen, & Wedeen, 2007). Examples of tractography axonal myelination, fiber diameter and axonal density 124
60 (Dell'Acqua et al., 2013). The main difference between FA and 125
reconstructions in representative subjects are shown in Fig. 3.
61 HMOA is that FA is a traditional voxel-based metrics providing 126
62 2.6.1. VOF an average measure of anisotropy of the entire voxel derived 127
63 We used a one region of interest (ROI) drawn on consecutive 128
from fitting the data according to the diffusion tensor model.
64 129
axial slices around a cluster of occipital voxels posterior to the On the contrary, HMOA is a new tract-specific metrics based
65 130
arcuate fibers and immediately lateral to the ILF, with a on the SD approach. The main advantages of HMOA are: i)
6 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 66
2 67
3 68
4 69
5 70
6 71
7 72
8 73
9 74
10 75
11 76
12 77
13 78
14 79
15 80
16 81
17 82
18 83
19 84
20 85
21 86
22 87
23 88
24 89
25 90
26 91
27 92
28 93
29 94
30 95
31 96
32 97
33 98
34 99
35 100
36 Fig. 3 e A) pAF (yellow) and VOF (blue) tractography reconstructions overlaid on a post-mortem dissection figure by Curran 101
37 (1909). The course of the fasciculus is viewed from the lateral aspect of the hemisphere, B) example of tractography 102
38 reconstructions of the pAF and VOF in the left and the right hemisphere in a representative single subject, C) inter- 103
39 individual variability of the pAF and VOF in the left hemisphere for six additional subjects (S1eS6). Abbreviations: pAF, 104
40 posterior segment of the arcuate fasciculus; VOF, vertical occipital fasciculus; IFOF, inferior fronto-occipital fasciculus; UF, 105
41 uncinate fasciculus; STG, superior temporal gyrus; F.trans.o., posterior part of the fasciculus transversus occipitalis (VOF), 106
42 most of which has been removed in image A) to show a large window through which the IFOF can be observed. 107
43 108
44 109
45 110
46 111
47 112
48 resolving partial volume contamination of different white using the q-value of .05 for significant results (FDR p < .05), which 113
49 matter tracts crossing within the same voxel or brain region, treated 26 different kinematic parameters and four HMOA 114
50 and ii) providing a distinct and therefore a more “true” tract- values for four different fiber bundles (VOF and pAF, bilaterally) 115
51 specific quantification of anisotropy and microstructural or- as instances of multiple testing. As a posteriori analysis, we 116
52 ganization along each white matter tract (i.e., if two fibers are extracted and performed the correlation analysis also on the 117
53 crossing the same voxel, two distinct and independent HMOA control tract e the left ILF, to test the specificity of our findings. 118
54 values are assigned to each tract) (Dell'Acqua et al., 2013). 119
55 120
However, in order to test our a priori assumed higher sensi-
56 121
tivity of the tract-specific HMOA index to detect significant 3. Results
57 122
associations, compared to the traditional voxel-based FA
58 123
measure, we have performed a posteriori correlation analysis 3.1. Behavioral results
59 124
60 also on the FA measures of our tracts of interest. 125
61 Pearson bivariate correlation analysis was used to detect the Movement times and the mean kinematic values of reaching, 126
62 strength of the correlation between the tract-specific HMOA grasping and lifting actions are shown in Table S1 of the 127
63 measure and the kinematic markers of reaching, grasping and Supplementary Material. Although we have used a mixed 128
64 lifting movements. We employed a false discovery rate (FDR) sex group for the study, there was no significant effect of sex 129
65 correction (Benjamini & Hochberg, 1995) for 104 comparisons on the kinematic measures (all p > .066). A one-way repeated 130
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 7
1 66
2 67
3 68
4 69
5 70
6 71
7 72
8 73
9 74
10 75
11 76
12 77
13 78
14 79
15 80
16 81
17 82
18 83
19 84
20 85
21 86
22 87
23 88
24 89
25 90
26 Fig. 4 e Kinematical markers showing statistically significant differences between (A) reaching, grasping and lifting 91
27 92
movements and (B) between grasping and lifting movements. Errors bars: 95% CI. ***p < .001, *p < .05.
28 93
29 94
30 95
31 96
measures analysis of variance (ANOVA) indicated significant 3.2. Relating inter-individual differences in movement
32 97
33
differences in reach kinematics between the three tasks for kinematics to the anatomy of the cross-talk connections 98
34 the total movement time (F (2, 28) ¼ 130.088, p < .0001), time to 99
35 peak velocity (F (2, 28) ¼ 96.552, p < .0001), time to peak ac- Tractography results for the VOF and the pAF (Fig. 3) revealed 100
36 celeration (F (2, 28) ¼ 146.793, p < .0001) and time to peak that the dorsal-ventral networks showed statistically sig- 101
37 deceleration (F (2, 28) ¼ 75.716, p < .0001). Post-hoc compari- nificant associations between their microstructural integrity, 102
38 sons (Fig. 4A) showed that movement duration was shorter for indexed by HMOA, and the kinematics of the grasping (VOF) 103
39 reaching than for grasping (t (29) ¼ 16.415, p < .0001) and and lifting movements (VOF and pAF), but not for the 104
40 lifting (t (29) ¼ 13.035, p < .0001). Also, time to peak velocity, reaching actions (Fig. 5; Table S2 of the Supplementary 105
41 acceleration and deceleration was reached earlier for reaching 106
Material).
42 107
than for grasping (tVEL (29) ¼ 13.987, p < .0001; tACCEL Since these pathways lie in close proximity to each other, it
43 108
(29) ¼ 17.436, p < .0001; tDECEL (29) ¼ 11.328, p < .0001) and is important to mention that their tissue properties, in terms
44 109
45 lifting (tVEL (29) ¼ 13.279, p < .0001; tACCEL (29) ¼ 17.199, of HMOA, were highly correlated between the right VOF and 110
46 p < .0001; tDECEL (29) ¼ 12.298, p < .0001). Time to peak ac- the right (r ¼ .593, p < .0001) and left pAF (r ¼ .629, p < .0001); 111
47 celeration was reached significantly earlier (t (29) ¼ 2.666, however the left VOF was independent of the left and right 112
48 p ¼ .012) for lifting compared to grasping movement (Fig. 4A). pAF (p > .096) but significantly correlated with its right coun- 113
49 Regarding the grasp kinematics (Fig. 4B), the time of terpart (r ¼ .617, p ¼ .0003). Furthermore, just like for the ki- 114
50 maximum grip opening velocity was reached earlier for nematic measures, there was no effect of sex on the HMOA 115
51 grasping compared to lifting (t (29) ¼ 9.091, p < .0001), while tract measures (all p > .564), hence the correlation analyses 116
52 the amplitudes of the maximum grip opening (t (29) ¼ 10.343, were done on a mixed sex group. 117
53 118
p < .0001) and grip closing velocity (t (29) ¼ 6.848, p < .0001) Our a posteriori analysis based on the FA measure revealed
54 119
were significantly higher for grasping compared to lifting. The similar trends in data, reflected in weak associations with the
55 120
amplitude of the maximum grip aperture was not significantly kinematics of grasping (VOF) and kinematics of lifting (VOF
56 121
57 different between grasping and lifting tasks (t (29) ¼ 1.191, and pAF), which failed to reach significance after the FDR 122
58 p ¼ .243). This parameter is rather stable and the changes in correction (p > .025). Although the indices of FA and HMOA 123
59 grip aperture usually occur due to changes in object size, were significantly correlated in all the tracts, i.e., the left 124
60 which in our tasks was identical. However, the variability of (r ¼ .464, p ¼ .010) and right pAF (r ¼ .559, p ¼ .001), the left 125
61 the maximum grip aperture was significantly lower for lifting (r ¼ .540, p ¼ .002) and right VOF (r ¼ .551, p ¼ .002), as hy- 126
62 compared to grasping (t (29) ¼ 9.327, p < .0001), suggesting that pothesized the tract-specific HMOA measure seemed more 127
63 grasping for lifting task was implemented more precisely sensitive to detect the associations with the kinematic 128
64 and accurately than grasping the object when no subsequent 129
markers compared to the classical voxel-based FA measure.
65 130
action was requested. Lastly, we observed no significant correlations for HMOA or FA
8 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 66
2 67
3 68
4 69
5 70
6 71
7 72
8 73
9 74
10 75
11 76
12 77
13 78
14 79
15 80
16 81
17 82
18 83
19 84
20 85
21 86
22 87
23 88
24 89
25 90
26 91
27 92
28 93
29 94
30 95
31 96
32 97
33 98
34 99
35 100
36 101
37 102
38 103
39 104
40 105
41 106
42 107
43 108
44 Fig. 5 e Graphical overview of the correlation analyses between HMOA and the kinematics of reaching, grasping and lifting 109
45 movements. Dashed gray lines indicate that no associations were found (non-involved pathways), colored dashed lines 110
46 denote an association was found but it did not survive the FDR correction (weakly involved pathways), while colored solid 111
47 lines signify strong associations (strongly involved pathways) that survived the FDR correction. For VOF, correlations with 112
48 the amplitude of the maximum opening grip velocity are shown, except for the left VOF and the variability of the maximum 113
49 grip aperture in grasping task; for the pAF, correlations with the time to peak deceleration are shown. Abbreviations: pAF, 114
50 posterior segment of the arcuate fasciculus; VOF, vertical occipital fasciculus. Significance codes represent FDR adjusted 115
51 p-values: *p < .05; **p < .01; ***p < .001. 116
52 117
53 118
54 119
55 120
56 121
57 values of our control tract e the left ILF (Table S2 of the 3.2.1. Kinematics and the VOF 122
58 Supplementary Material) and kinematic markers, even Our results showed strong association of the bilateral 123
59 before performing the FDR correction (all p > .070). The inter- VOF with the grasp phase of both grasping and lifting actions, 124
60 individual variability of the HMOA measure of the left ILF was indicating a contribution of the left VOF in the transfer of 125
61 higher (SD ¼ .013, or 14.7% of the mean value) than that of the accuracy information for the formation of hand choreography 126
62 bilateral VOF and pAF (all SD < .007, or up to 13.2% of the mean 127
during grasping, and a contribution of the right VOF in the grip
63 128
value), thus the lack of correlation shown in the control tract opening phase during both grasping and lifting actions (Fig. 6;
64 129
could not be due to the lack of variation. Table S2 of the Supplementary Material).
65 130
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 9
1 66
2 67
3 68
4 69
5 70
6 71
7 72
8 73
9 74
10 75
11 76
12 77
13 78
14 79
15 80
16 81
17 82
18 83
19 84
20 85
21 86
22 87
23 88
24 89
25 90
26 91
27 92
28 93
29 94
30 95
31 96
32 97
33 98
34 Fig. 6 e Correlation plots showing the associations between the bilateral VOF (in blue), the bilateral pAF (in yellow), and the 99
35 left ILF control tract (in green) in terms of HMOA and the kinematics of grasping and lifting movements. Significance codes 100
36 show corresponding FDR adjusted p-values at '***' p < .001 and '**' p < .01. 101
37 102
38 103
39 104
40 105
HMOA of the left VOF was correlated with the level of r ¼ .450, p ¼ .012) (Fig. 6), there was no significant difference
41 106
variability of the maximum grip aperture. For the grasping between the correlations of the left and right VOF with the
42 107
movements higher HMOA corresponded to greater variability amplitude of maximum opening grip velocity during grasping
43 108
44 of the maximum grip aperture (r ¼ .527, p ¼ .002) and, (z ¼ 1.012, p ¼ .311) or lifting (z ¼ .779, p ¼ .435). Thus, it is 109
45 consequently, to less accuracy. This relationship was not likely that the bilateral VOF is associated with the grip opening 110
46 found for lifting movements (r ¼ .194, p ¼ .302). The differ- phase as the hand is approaching the target object in order to 111
47 ence between the correlations of HMOA and the time of either grasp or lift it. 112
48 maximum grip aperture for the two movements was not 113
49 significant (z ¼ 1.532, p ¼ .125). The possible reason for 3.2.2. Kinematics and the pAF 114
50 the lack of significant correlation between HMOA and the We observed a statistically significant relationship between 115
51 maximum grip aperture for the lifting task might lie in the the bilateral pAF and the transport (reach) component of lift- 116
52 117
fact that the variability of the maximum grip aperture for ing actions (Fig. 6; Table S2 of the Supplementary Material). In
53 118
lifting was significantly lower compared to grasping e thus particular, higher HMOA of the bilateral pAF was associated
54 119
55 not variable enough to detect differences between the sub- with a later time to peak deceleration in lifting movements 120
56 jects. Lastly, although the right VOF was not significantly (left pAF: r ¼ .589, p < .001; right pAF: r ¼ .698, p < .0001), a 121
57 related to the variability of the maximum grip aperture kinematic marker indexing a careful arm deceleration phase 122
58 during grasping, this correlation was not significantly for guiding the hand to a stable final fingers' positioning. 123
59 different from the one observed for the same movement Although we found correlations between the time to peak 124
60 within the left VOF (z ¼ 1.453, p ¼ .146). deceleration and the right pAF in reaching (r ¼ .451, p ¼ .012) 125
61 Secondly, higher HMOA of the right VOF corresponded to and the bilateral pAF in grasping (left pAF: r ¼ .424, p ¼ .019; 126
62 higher amplitude of the maximum opening grip velocity for right pAF: r ¼ .393, p ¼ .032) movements, these correlations did 127
63 128
both grasping (r ¼ .567, p ¼ .001) and lifting actions (r ¼ .559, not survive the FDR correction but they were not statistically
64 129
p ¼ .001). Although correlations involving the left VOF failed to different from the above-mentioned significant correlations
65 130
survive the FDR correction (grasping: r ¼ .425, p ¼ .019, lifting: for the lifting kinematics (all p > .123).
10 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 The grasp-specific kinematics was significantly associated Our findings confirmed that a close collaboration exists 66
2 with the pAF during lifting, but not during grasping move- between the ventral and dorsal visual streams in skilled ac- 67
3 ments. Higher HMOA of the right pAF corresponded to longer tions, in line with the functional neuroimaging literature. For 68
4 time of the maximum opening grip velocity during lifting example, an involvement of the ventral stream was shown for 69
5 70
movements (r ¼ .595, p < .001). Although the correlation with processing grasp-related properties such as object size, shape,
6 71
the left pAF failed to survive the FDR correction (r ¼ .431, and weight (Cavina-Pratesi, Goodale, & Culham, 2007;
7 72
p ¼ .018), there was no significant difference between the Gallivan, Cant, Goodale, & Flanagan, 2014; Grill-Spector,
8 73
9 correlations of the left and right pAF (z ¼ 1.011, p ¼ .312) and Kushnir, Edelman, Itzchak, & Malach, 1998; Monaco et al., 74
10 the time of the maximum opening grip velocity, pointing to 2014), while object-selective responses were observed in dor- 75
11 the fact that bilateral pAF plays a specific role for lifting ac- sal stream regions (Fang & He, 2005; Grill-Spector, Kushnir, 76
12 tions given that the correlations for the grasping condition Hendler, & Malach, 2000; James, Humphrey, Gati, Menon, & 77
13 were significantly different from the lifting ones (for the right Goodale, 2002; Konen & Kastner, 2008; Vinberg & Grill-Spector, 78
14 pAF: z ¼ 3.773, p ¼ .0001; for the left pAF: z ¼ 2.478, 2008). Similarities were also noted between the representa- 79
15 p ¼ .013). tions of the viewed actions in the ventral and dorsal visual 80
16 In sum, although the bilateral pAF could be related to the streams (Bracci & Peelen, 2013; Fabbri et al., 2016; Konen & 81
17 82
arm deceleration phase for all the considered movements, it Kastner, 2008; Mahon et al., 2013; Oosterhof et al., 2010; Roth
18 83
seems to be particularly associated with the timing of the & Zohary, 2015). There seems to be no absolute distinction
19 84
20 opening grip aperture when the object needs to be lifted and between the two streams, and the linear hierarchical dual 85
21 not solely grasped. stream model has been recently modified towards a more 86
22 interconnected network model (De Haan & Cowey, 2011). 87
23 Thus, despite their functional specialization, our findings 88
24 4. Discussion corroborate the notion that significant transfer of information 89
25 exists between the two visual streams, enabling both to 90
26 Despite the apparent segregation, neurophysiological, contribute to complex visuomotor behavior. 91
27 neuroimaging and lesion studies indicate that dorsal and However, in a simple action context such as reaching (i.e., 92
28 ventral visual streams need to communicate in everyday 93
transport of the arm in space to the location of a target object),
29 94
life for successful execution of complex behaviors after correction for multiple comparisons no significant cor-
30 95
(Cloutman, 2013; De Haan & Cowey, 2011; Goodale, 2014; relation between the anatomy of the considered cross-talk
31 96
Milner, 2017; Schenk & McIntosh, 2009; van Polanen & connections and kinematics was observed. This finding is
32 97
33 Davare, 2015). Our study tried to answer the question of consistent with the notion that the dorsal ‘action’ stream 98
34 whether and how these two visual streams interact with mediates the visual control of reaching actions (Milner & 99
35 one another to support visually guided skilled hand actions. Goodale, 2008; Rizzolatti & Matelli, 2003; Ungerleider & 100
36 We explored whether the performance of hand actions Mishkin, 1982). In this respect, reaching activates the dorsal 101
37 characterized by different levels of complexity, namely stream regions in functional MRI studies (Cavina-Pratesi et al., 102
38 reaching, grasping and lifting, is modulated by the anatomy 2010; Filimon, Nelson, Huang, & Sereno, 2009; Prado et al., 103
39 of cross-talk connections between the dorsal and ventral 2005) and it becomes selectively disrupted by transcranial 104
40 visual streams. 105
magnetic stimulation over the posterior parietal cortex
41 106
We demonstrate, for the first time, that lateral occipital (Desmurget et al., 1999) and in patients with severe optic
42 107
43
and temporo-parietal white matter pathways mediate the ataxia whose dorsal stream is damaged (Buxbaum & Coslett, 108
44 cross-talk between the two visual streams in order to support 1998; Gre a et al., 2002; Pisella et al., 2000; Rossetti et al., 109
45 skilled hand actions. We suggest that the VOF and the pAF are 2005). Our group has recently showed that dorsal white mat- 110
46 at the basis of the anatomical coupling between dorsal and ter networks, such as fronto-parietal (Budisavljevic et al., 111
47 ventral regions, and that the nature (i.e., complexity) of the 2016) and premotor pathways (Budisavljevic et al., 2017), 112
48 task modulates the correlation between visuomotor behavior significantly modulate the kinematics of reaching actions (see 113
49 and their white matter microstructure. We observed that the also Catani et al., 2017, for newly described intra-parietal 114
50 anterior pAF was significantly associated with the arm connections that could also underly reaching movements). 115
51 deceleration phase of lifting movements. It is during the 116
Nevertheless, it should be acknowledged that the inter-
52 117
deceleration phase that an efficient control and the execution individual variability of the time to peak deceleration for
53 118
of small trajectory adjustments occur as the hand approaches reaching (unlike other reaching measures) was lower than for
54 119
the target object. Furthermore significant associations with grasping and lifting actions, which could potentially reflect
55 120
56 the cross-talk connections were also found at the level of the the observed lack of correlation. 121
57 grasp component with specific reference to the speed (pAF) 122
58 and amplitude (VOF) adopted to reach the maximum grip 4.1. Relating the kinematics of grasping and lifting to 123
59 aperture and the associated variability (VOF) indexing accu- the anatomy of the VOF 124
60 racy of the grasp. These parameters are vital for a successful 125
61 grasp given that it is immediately after the time maximum Individual differences in the microstructure of the bilateral 126
62 grip aperture is reached that the honing phase of the fingers VOF were significantly associated with the variability of grasp- 127
63 upon the object starts. This phase must allow for enough time 128
specific kinematics, i.e., the level of variability during grip
64 129
to determine and implement contact points and minimize formation for grasping, and the grip opening phase during
65 130
variability. both grasping and lifting actions.
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 11
1 Higher HMOA of the left VOF corresponded to higher areas, are used as priors by the dorsal visuomotor stream to 66
2 variability of the grasp aperture during grasping task, while specify goal-directed grasping actions and are activated 67
3 higher HMOA of the right VOF corresponded to higher am- earlier in time than dorsal regions involved in specifying ac- 68
4 plitudes of the opening grip velocity during both grasping and tion plans (Zimmermann, Verhagen, de Lange, & Toni, 2016). 69
5 70
lifting, i.e., faster opening of the hand up to maximum grip It is argued that formation of the grasp is planned to take
6 71
aperture. In other words, it seems that VOF transfers infor- into account not only the perceived visual characteristics of
7 72
mation relevant to the unfolding of the hand opening as the the object, but also internalized information based on the past
8 73
9 hand reaches the maximum grip aperture. Maximum grip experience about the likely accuracy of the reaching compo- 74
10 aperture is the key kinematical parameter indexing the nent (Wing et al., 1986), processed by the ventral stream re- 75
11 beginning of a hand closure (Jeannerod, 1984), and it is scaled gions. Thus, the VOF fibers may support object-directed 76
12 mainly depending on the visual availability of objects and skilled actions by fine-tuning the opening of the hand in order 77
13 their intrinsic visual properties such as size and shape. The to decrease transport errors and accomplish an efficient and 78
14 computation of these visual object properties is crucial for accurate grasp of a target object. 79
15 successful hand preshaping in grasping movements 80
16 (Castiello, 2005), and when visual feedback is not available 4.2. Relating the kinematics of lifting to the anatomy of 81
17 82
exaggerated opening of the hands occurs (Hesse, Miller, & the pAF
18 83
Buckingham, 2016; Jakobson & Goodale, 1991).
19 84
Also, the variability of the maximum grip aperture, The pAF fibers connect the cortical nodes of a ventro-dorsal
20 85
21 commonly used as an index of grasp accuracy and transport network (Rizzolatti & Matelli, 2003), e.g., the inferior parietal 86
22 error (Flindall et al., 2014), was significantly associated with lobe and the superior temporal regions that play a crucial role 87
23 the anatomy of the VOF. Higher HMOA of the left VOF was in both perception and action, especially for skilled object use 88
24 associated with greater and more variable grip apertures. (Binkovski & Buxbaum, 2013; Iacoboni et al., 2001; Martin et al., 89
25 Kinematical studies showed that more variable and less ac- 2016; Milner, 1997; Rizzolatti & Matelli, 2003; Shapiro et al., 90
26 curate grip apertures, expected to occur when reaching faster, 2002; Singh-Curry & Husain, 2009). The parietal terminations 91
27 are often compensated with bigger opening of the hands of the bilateral pAF were often specifically implicated in lifting 92
28 93
(Wing, Turton, & Fraser, 1986). Similarly, in the present study actions: i) the left intraparietal sulcus in reactive online ad-
29 94
the wider opening of the hand gives increased tolerance for justments of grip force for lifting (Dafotakis, Sparing, Eickhoff,
30 95
positioning errors just prior to contact, and improves the Fink, & Nowak, 2008), and ii) the right inferior parietal lobe in
31 96
32 chances of the hand encompassing the object before collision. updating of predictive force scaling (Jenmalm, Schmitz, 97
33 We may speculate that subjects who had higher HMOA of the Forssberg, & Ehrsson, 2006), corrective reactions and updat- 98
34 VOF had faster transfer of information between dorsal and ing of memory representations of object's weight (Schmitz, 99
35 ventral streams, leading to bigger and more variable grip ap- Jenmalm, Ehrsson, & Forssberg, 2005) and dynamic object 100
36 ertures. Support for this suggestion comes from the assump- manipulation and anticipatory coordination of grip and load 101
37 tion that an increase in HMOA reflects an increase in forces maintaining grasp stability during lifts (Ehrsson, 102
38 myelination level and axonal density, leading to faster signal Fagergren, Johansson, & Forssberg, 2003). This is why it was 103
39 conduction time (Dell'Acqua et al., 2010), possibly bringing a not surprising that the pAF connections were strongly asso- 104
40 105
faster and greater fingers' opening. This is in line with a recent ciated with the kinematics of the lifting actions.
41 106
study showing that higher speed of visual information pro- Lifting actions require an additional factor e object's weight,
42 107
cessing is associated with a higher HMOA of the fronto- to be incorporated into the motor plan and used to calculate the
43 108
44 parietal white matter networks (Chechlacz, Gillebert, necessary grip and lift fingertip forces (Nowak, Glasauer, & 109
45 Vangkilde, Petersen, & Humphrey, 2015). It must be said, Hermsdo € rfer, 2013). Properties such as weight are not directly 110
46 however, that it is difficult to interpret differences in HMOA in specified in the visual information, but are part of the object's 111
47 a linear fashion since HMOA also increases with decreasing identity for which both perception and memory are important 112
48 axonal radius and radial diffusivity (Dell'Acqua et al., 2010, (Kentridge, 2014). In other words, weight is relevant to skilled 113
49 2013), possibly due to decreased myelination. object manipulation and is estimated based on stored knowl- 114
50 Thus, the VOF supports the transfer of object-related edge and more conceptual identity information processed by 115
51 the ventral stream. Evidence that expectations about the ob- 116
properties necessary for efficient grasp formation by allow-
52 117
ing the communication between ventral regions that process ject's weight are represented in the ventral stream regions
53 118
intrinsic object properties and dorsal regions that map spatial (Gallivan et al., 2014) and that visual information from the
54 119
55 location to action plans (Takemura et al., 2016; but see Freud, ventral pathway can influence weight estimations during lift- 120
56 Plaut, & Behrmann, 2016 for dorsal stream in object percep- ing (Brenner & Smeets, 1996; Jackson & Shaw, 2000), may 121
57 tion). This is in line with studies showing a crucial role of the explain the strong correlations of lifting kinematics with the 122
58 VOF fibers in the visual shape processing (Lee Masson, cross-talk pAF (and VOF) fibers. Furthermore, people can adjust 123
59 Wallraven, & Petit, 2017) and the ventral stream cortical their lifting movements online after a sudden visible change in 124
60 areas in processing object size during grasping (Hesse et al., object's weight despite a short time frame (Brouwer, Georgiou, 125
61 2016; Schenk, 2012) e explaining why patients with optic Glover, & Castiello, 2006). It is argued that this online adjust- 126
62 ataxia and dorsal stream damage are able to use object size for ment, thought to rely on the dorsal stream, is influenced by 127
63 128
grasping (Cavina-Pratesi, Ietswaart, Humphreys, Lestou, & visual cues about weight information processed by the ventral
64 129
Milner, 2010). Furthermore, recent evidence suggests that stream. Our findings corroborate the notion that although
65 130
perceptual features, processed by the occipital ventral stream weight might be processed at the level of the ventral perceptual
12 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 stream, this information could be made readily available to the 4.3. Further implications 66
2 visuomotor dorsal stream for online control through direct 67
3 cross-talk connections. As mentioned above, object weight is part of a high-level, non- 68
4 More specifically, we found that the bilateral pAF fibers visual object property that requires an elaborate conceptual 69
5 70
were significantly related to movement deceleration, which processing. The fact that lifting actions were significantly
6 71
reflects the control phase during which comparisons between related to the more anterior cross-talk pAF connections is in
7 72
motor commands and sensory feedback occur and any line with observations that the object's conceptual encoding
8 73
9 necessary corrections to the original movement plan are occurs more in the anterior ventral areas, whereas the pos- 74
10 made (Paulignan, MacKenzie, Marteniuk, & Jeannerod, 1990, terior ventral regions might respond more to specific visual 75
11 1991). Participants with higher HMOA of the bilateral pAF properties, both in the monkey (Goda, Tachibana, Okazawa, & 76
12 spent less time decelerating towards the target object during Komatsu, 2014) and human (Peelen & Caramazza, 2012) 77
13 lifting, possibly implying faster transfer of object-related in- brains. This notion is further supported by the evidence that 78
14 formation, and less corrections made to the original move- the anterior areas in the ventral stream process weight rep- 79
15 ment plan. It is possible that the pAF, just like its cortical resentations (Gallivan et al., 2014). Moreover, conceptual as- 80
16 endpoint, the inferior parietal lobe (Schmitz et al., 2005; de pects of actions requiring knowledge of skilled object use were 81
17 82
Lange, 2006; Jenmalm et al., 2006), is part of a neural circuit previously associated with the cortical terminations of the
18 83
involved in fast corrective responses triggered in lifts, moni- pAF fibers, the inferior parietal lobe (Binkovski & Buxbaum,
19 84
20 toring a sensory mismatch and updating sensorimotor 2013; Martin et al., 2016) and the superior temporal regions 85
21 memories related to weight. (Martin et al., 2016). Consistent with these observations we 86
22 Previous research showed that movement deceleration suggest that the anterior cross-talk connections, i.e., the pAF, 87
23 affects the finger opening phase prior to contact, and sug- might be specialized for higher-level, dynamic object-use 88
24 gested that the demands for precise control gradually increase representations (object weight). Instead, the posterior cross- 89
25 when the fingers are almost at contact with the object (Chieffi talk VOF fibers might be more involved in visual representa- 90
26 & Gentilucci, 1993; Corradini, Gentilucci, Leo, & Rizzolatti, tion of objects and crucial for grasp formation according to 91
27 1992; Lemon, Johansson, & Westling, 1995; Paulignan, properties such as size, shape, and orientation in both 92
28 93
Jeannerod, MacKenzie, & Marteniuk, 1991). This is reconcil- grasping and lifting movements.
29 94
able with our finding that the grip opening phase, i.e., the
30 95
speed at which the hand was opened while approaching the
31 5. Conclusions 96
32 target object before the lift, was also strongly associated with 97
33 the pAF fibers. Higher microstructural organization of the 98
This is the first explorative study to demonstrate that the
34 right pAF corresponded to later time to maximum opening 99
cross-talk between dorsal and ventral visual streams is sup-
35 grip velocity. Longer time to reach maximum grip velocity 100
ported by the vertical occipital (VOF) and temporo-parietal
36 may serve to extend the time window within which contact 101
37 (pAF) white matter fibers. We observed a direct relationship 102
points can be established more precisely and firmly. During
38 between the anatomy of the bilateral cross-talk connections 103
lifting it is necessary to carefully control finger positioning and
39 the applied forces in order to avoid the tilting or falling of the
and dominant hand kinematics of skilled actions (i.e., 104
40 grasping and lifting) requiring a high degree of online control. 105
target object (Lukos, Ansuini, & Santello, 2007). To do this,
41 These pathways played an important role during the arm 106
participants need to coordinate the ‘braking’ phase of the arm
42 deceleration phase and the specification of a key parameter 107
43 (i.e., deceleration phase) as to accompany the hand towards 108
for a successful grasp, i.e., maximum grip aperture. However,
44 the object with the gradual opening of the fingers. The rela- 109
the cross-talk integration is not the only form of communi-
45 tionship between pAF fibers and the grip formation is in line 110
cation between the dual processing streams, and important
46 with studies noting a crucial role of the human intraparietal 111
contributions from top-down feedback connections or addi-
47 sulcus in hand preshaping (Davare, Andres, Clerget, 112
48 tional common target brain areas should not be under- 113
Thonnard, & Olivier, 2007; Tunik, Frey, & Grafton, 2005).
49 estimated. We hope that future studies, combining 114
These two kinematical features, arm deceleration and hand
50 multimodal approaches with larger data cohorts, will bring 115
opening phase mark the on-line control processes (Glover,
51 additional insights into the neural basis of dorsal-ventral 116
2004). The dual role of the pAF in grip formation and arm
52 stream integration. 117
deceleration reflects the prominent finding in the functional
53 118
literature that the anterior intraparietal sulcus of the inferior
54 119
55 parietal lobe has two main roles: i) the integration of intrinsic 120
Conflict of interest Q4
56 object properties, such as size and shape into a motor plan, 121
57 and ii) the updating of grasp formation by online detection 122
None declared.
58 and correction of errors in motor execution (Dafotakis et al., 123
59 2008; Rice, Tunik, & Grafton, 2006; Tunik et al., 2005). We 124
60 suggest that the pAF fibers are important in integrating Uncited references 125
61 visuomotor-specific information from the dorsal stream with 126
62 the object-relevant information from the ventral stream in Begliomini et al., 2008,Della Sala et al., 2006,Goldenberg et al., 127
63 2001,Hermsdo € rfer et al., 2001,Marteniuk et al., 1987,Naito et al., 128
order to control the on-line the movement trajectory and the
64 2005,Rossetti et al., 2003,Tretriluxana et al., 2008. Q6
129
grip formation during skilled hand actions such as lifting.
65 130
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 13
1 Catani, M., Jones, D. K., & ffytche, D. H. (2005). Perisylvian 66

2 Acknowledgements language networks of the human brain. Annals of Neurology, 67
3 57(1), 8e16. 68
4 This work was supported by the Progetto Strategico, Uni- Catani, M., Robertsson, N., Beyh, A., Huynh, V., de Santiago 69
5 Requejo, F., Howells, H., et al. (2017). Short parietal lobe 70
di Padova (N. 2010XPMFW4) to Umberto Castiello. We
versita
6 connections of the human and monkey brain. Cortex, 97, 71
would like to thank NeMo laboratory (http://nemolaboratory. 339e357.
7 72
com) for helpful comments on the manuscript. Cavina-Pratesi, C., Goodale, M. A., & Culham, J. C. (2007). FMRI
8 73
reveals a dissociation between grasping and perceiving the
9 74
size of real 3D objects. Plos One, 2(5), e424.
10 75
Supplementary data Cavina-Pratesi, C., Ietswaart, M., Humphreys, G. W., Lestou, V., &
11 76
Milner, A. D. (2010). Impaired grasping in a patient with optic
12 77
ataxia: Primary visuomotor deficit or secondary consequence
13 Supplementary data related to this article can be found at
of misreaching? Neuropsychologia, 48(1), 226e234.
78
14 https://doi.org/10.1016/j.cortex.2018.02.016. Cavina-Pratesi, C., Monaco, S., Fattori, P., Galletti, C., 79
15 McAdam, T. D., Quinlan, D. J., et al. (2010). Functional 80
16 magnetic resonance imaging reveals the neural substrates 81
17 references of arm transport and grip formation in reach-to-grasp 82
18 actions in humans. The Journal of Neuroscience, 30(31), 83
19 10306e10323. 84
20 Begliomini, C., Nelini, C., Caria, A., Grodd, W., & Castiello, U. Chechlacz, M., Gillebert, C. R., Vangkilde, S. A., Petersen, A., & 85
21 (2008). Cortical activations in humans grasp-related areas Humphrey, G. W. (2015). Structural variability within 86
22 depend on hand used and handedness. PLoS One, 3(10), e3388. frontoparietal networks and individual differences in 87
23 Benjamini, Y., & Hochberg, Y. (1995). Controlling the false attentional functions: An approach using the theory of visual 88
24 discovery rate: A practical and powerful approach to multiple attention. The Journal of Neuroscience, 35(30), 10647e10658. 89
25 testing. Journal of the Royal Statistical Society Series B Chen, C. M., Lakatos, P., Shah, A. S., Mehta, A. D., Givre, S. J., 90
26 (Methodological), 57(1), 289e300. Javitt, D. C., et al. (2007). Functional anatomy and interaction 91
27 Binkovski, F., & Buxbaum, L. J. (2013). Two action systems in the of fast and slow visual pathways in macaque monkeys. 92
28 human brain. Brain and Language, 127(2), 222e229. Cerebral Cortex, 17(7), 1561e1569. 93
Borra, E., Ichinohe, N., Sato, T., Tanifuji, M., & Rockland, K. S. Chieffi, S., & Gentilucci, M. (1993). Coordination between the
29 94
(2010). Cortical connections to area TE in monkey: Hybrid transport and the grasp components during prehension
30 95
modular and distributed organization. Cerebral Cortex, 20(2), movements. Experimental Brain Research, 94(3), 471e477.
31 96
257e270. Cloutman, L. L. (2013). Interaction between dorsal and ventral
32 97
Bracci, S., & Peelen, M. V. (2013). Body and object effectors: The processing streams: Where, when and how? Brain and
33 Language, 127(2), 251e263.
98
organization of object representations in high-level visual
34 Cohen, L., Dehaene, S., Naccache, L., Lehericy, S., Dehaene- 99
cortex reflects body-object interactions. The Journal of
35 Lambertz, G., Henaff, M. A., et al. (2000). The visual word form 100
Neuroscience, 33(46), 18247e18258.
36 Brenner, E., & Smeets, J. B. (1996). Size illusion influences how we area: Spatial and temporal characterization of an initial stage 101
37 lift but not how we grasp an object. Experimental Brain Research, of reading in normal subjects and posterior split-brain 102
38 111(3), 473e476. patients. Brain, 123(Pt 2), 291e307. 103
39 Brouwer, A. M., Georgiou, I., Glover, S., & Castiello, U. (2006). Corradini, M. L., Gentilucci, M., Leo, T., & Rizzolatti, G. (1992). 104
40 Adjusting reach to lift movements to sudden visible changes Motor control of voluntary arm movements. Kinematic and 105
41 in target's weight. Experimental Brain Research, 173(4), 629e636. modelling study. Biological Cybernetics, 67(4), 347e360. 106
42 Budisavljevic, S., Dell'Acqua, F., Djordjilovic, V., Miotto, D., Curran, E. J. (1909). A new association fiber tract in the cerebrum. 107
43 Motta, R., & Castiello, U. (2017). The role of the frontal aslant With remarks on the fiber tract dissection method of studying 108
44 tract and premotor connections in visually guided hand the brain. Journal of Comparative Neurology, 19, 645e656. 109
45 movements. Neuroimage, 146, 419e428. Dafotakis, M., Sparing, R., Eickhoff, S. B., Fink, G. R., & 110
46 Budisavljevic, S., Dell'Acqua, F., Rijsdijk, F. V., Kane, F., Nowak, D. A. (2008). On the role of the ventral premotor cortex 111
47 Picchioni, M., McGuire, P., et al. (2015). Age-related differences and anterior intraparietal area for predictive and reactive 112
48 and heritability of the perisylvian language networks. The scaling of grip force. Brain Research, 1228, 73e80. 113
Journal of Neuroscience, 35(37), 12625e12634. Davare, M., Andres, M., Clerget, E., Thonnard, J. L., & Olivier, E.
49 114
Budisavljevic, S., Dell'Acqua, F., Zanatto, D., Begliomini, C., (2007). Temporal dissociation between hand shaping and grip
50 115
Miotto, D., Motta, R., et al. (2016). Asymmetry and structure of force scaling in the anterior intraparietal area. The Journal of
51 116
the fronto-parietal networks underlie visuomotor processing Neuroscience, 27(15), 3974e3980.
52 117
in humans. Cerebral Cortex, 27(2), 1532e1544. De Haan, E. H., & Cowey, A. (2011). On the usefulness of ‘what’
53 and ‘where’ pathways in vision. Trends in Cognitive Sciences,
118
Buxbaum, L. J., & Coslett, H. B. (1998). Spatio-motor
54 15(10), 460e466. 119
representations in reaching: Evidence for subtypes of optic
55 Dell'Acqua, F., Rizzo, G., Scifo, P., Clarke, R. A., Scotti, G., & 120
ataxia. Cognitive Neuropsychology, 15(3), 279e312.
56 Castiello, U. (2005). The neuroscience of grasping. Nature Reviews Fazio, F. (2007). A model-based deconvolution approach to 121
57 Neuroscience, 6(9), 726e736. solve fiber crossing in diffusion-weighted MR imaging. IEEE 122
58 Catani, M., Allin, M. P., Husain, M., Pugliese, L., Transactions on Biomedical Engineering, 54(3), 462e472. 123
59 Mesulam, M. M., Murray, R. M., et al. (2007). Symmetries in Dell'Acqua, F., Scifo, P., Rizzo, G., Catani, M., Simmons, A., 124
60 human brain language pathways correlate with verbal Scotti, G., et al. (2010). A modified damped Richardson-Lucy 125
61 recall. Proceedings of the National Academy of Sciences USA, algorithm to reduce isotropic background effects in spherical 126
62 104(43), 17163e17168. deconvolution. Neuroimage, 49(2), 1446e1458. 127
63 Catani, M., Jones, D. K., Donato, R., & ffytche, D. H. (2003). Dell'Acqua, F., Simmons, A., Williams, S. C., & Catani, M. (2013). 128
64 Occipito-temporal connection in the human brain. Brain, 126, Can spherical deconvolution provide more information that 129
65 2093e2107. fiber orientations? Hindrance modulated orientational 130
14 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 anisotropy, a true-tract specific index to characterize white Glover, S. (2004). Separate visual representations in the planning 66
2 matter diffusion. Human Brain Mapping, 34(10), 2464e2483. and control of action. The Behavioral and Brain Sciences, 27(1), 67
3 Della Sala, S., Faglioni, P., Motto, C., & Spinnler, H. (2006). 3e24. 68
4 Hemisphere asymmetry for imitation of hand and finger Goda, N., Tachibana, A., Okazawa, G., & Komatsu, H. (2014). 69
5 movements, Goldenberg's hypothesis reworked. Representation of the material properties of objects in the 70
6 Neuropsychologia, 44(8), 1496e1500. visual cortex of nonhuman primates. The Journal of 71
7 Desmurget, M., Epstein, C. M., Turner, R. S., Prablanc, C., Neuroscience, 34(7), 2660e2673. 72
Alexander, G. E., & Grafton, S. T. (1999). Role of the posterior Goldenberg, G., Laimgruber, K., & Hermsdo € rfer, J. (2001). Imitation
8 73
parietal cortex in updating reaching movements to a visual of gestures by disconnected hemispheres. Neuropsychologia,
9 74
target. Nature Neuroscience, 2(6), 563e567. 39(13), 1432e1443.
10 75
Distler, C., Boussaoud, D., Desimone, R., & Ungerleider, L. G. (1993). Goodale, M. A. (2014). How (and why) the visual control of action
11 76
Cortical connections of inferior temporal area TEO in macaque differs from visual perception. Proceedings Biology Sciences,
12 77
monkeys. Journal of Comparative Neurology, 334(1), 125e150. 281(1785), 20140337.
13 D'Amico, M., & Ferrigno, G. (1990). Technique for evaluation of Grea, H., Pisella, L., Rossetti, Y., Desmurget, M., Tilikete, C., 78
14 derivates from noisy biomechanical displacement data using Grafton, S., et al. (2002). A lesion of the posterior parietal 79
15 model-based bandwidth-selection procedure. Medical and cortex disrupts on-line adjustments during aiming 80
16 Biological Engineering and Computing, 28(5), 407e415. movements. Neuropsychologia, 40(13), 2471e2480. 81
17 D'Amico, M., & Ferrigno, G. (1992). Comparison between the more Grill-Spector, K., Kushnir, T., Edelman, S., Itzchak, Y., & 82
18 recent techniques for smoothing and derivative assessment in Malach, R. (1998). Cue-invariant activation in object-related 83
19 biomechanics. Medical and Biological Engineering and Computing, areas of the human occipital lobe. Neuron, 21(1), 191e202. 84
20 30(2), 193e204. Grill-Spector, K., Kushnir, T., Hendler, T., & Malach, R. (2000). The 85
21 Ehrsson, H. H., Fagergren, A., Johansson, R. S., & Forssberg, H. dynamics of object-selective activation correlate with 86
22 (2003). Evidence for the involvement of the posterior recognition performance in humans. Nature Neuroscience, 3(8), 87
23 parietal cortex in coordination of fingertip forces for grasp 837e843. 88
24 stability in manipulation. Journal of Neurophysiology, 90(5), Hermsdo € rfer, J., Goldenberg, G., Wachsmuth, C., Conrad, B., 89
25 2978e2986. Ceballos-Baumann, A. O., Bartenstein, P., et al. (2001). Cortical 90
26 Fabbri, S., Stubbs, K. M., Cusack, R., & Culham, J. C. (2016). correlates of gesture processing: Clues to the cerebral 91
27 Disentagling representations of object and grasp properties in mechanisms underlying apraxia during imitation of 92
28 the human brain. The Journal of Neuroscience, 36(29), 7648e7662. meaningless gestures. Neuroimage, 14(Pt 1), 149e161. 93
Fang, F., & He, S. (2005). Cortical responses to invisible objects in Hesse, C., Miller, L., & Buckingham, G. (2016). Visual information
29 94
the human dorsal and ventral pathways. Nature Neuroscience, about object size and object position are retained differently in
30 95
8(10), 1380e1385. the visual brain: Evidence from grasping studies.
31 96
Felleman, D. J., & Van Essen, D. C. (1991). Distributed hierarchical Neuropsychologia, 91, 531e543.
32 97
processing in the primate cerebral cortex. Cerebral Cortex, 1(1), Iacoboni, M., Koski, L. M., Brass, M., Bekkering, H., Woods, R. P.,
33 1e47. Dubeau, M. C., et al. (2001). Reafferent copies of imitated
98
34 ffytche, D. H., & Catani, M. (2005). Beyond localization: From actions in the right superior temporal cortex. Proceedings of the 99
35 hodology to function. Philosophical Transactions of the Royal National Academy of Sciences USA, 98(24), 13995e13999. 100
36 Society of London B Biological Sciences, 360, 767e779. Jackson, S. R., & Shaw, A. (2000). The Ponzo illusion affects grip- 101
37 ffytche, D. H. (2008). The hodology of hallucinations. Cortex, 44, force but not grip-aperture scaling during prehension 102
38 1067e1083. movements. Journal of Experimental Psychology Human Perception 103
39 Filimon, F., Nelson, J. D., Huang, R. S., & Sereno, M. I. (2009). and Performance, 26(1), 418e423. 104
40 Multiple parietal reach regions in humans: Cortical Jakobson, L. S., & Goodale, M. A. (1991). Factors affecting higher- 105
41 representations for visual and proprioceptive feedback during order movement planning: A kinematic analysis of human 106
42 on-line reaching. The Journal of Neuroscience, 29(9), 2961e2971. prehension. Experimental Brain Research, 86(1), 199e208. 107
43 Fischer, E., Bulthoff, H. H., Logothetis, N. K., & Bartels, A. (2012). James, T. W., Humphrey, G. K., Gati, J. S., Menon, R. S., & 108
44 Human areas V3A and V6 compensate for self-induced planar Goodale, M. A. (2002). Differential effects of viewpoint on 109
45 visual motion. Neuron, 73(6), 1228e1240. object-driven activation in dorsal and ventral streams. Neuron, 110
46 Flindall, J. W., Doan, J. B., & Gonzalez, C. (2014). Manual 35(4), 793e801. 111
47 asymmetries in the kinematics of a reach-to-grasp action. Jeannerod, M. (1984). The timing of natural prehension 112
48 Later Asymmetries of Body, Brain and Cognition, 19(4), 489e507. movements. Journal of Motor Behavior, 16(3), 235e254. 113
Fox, C. J., Iaria, G., & Barton, J. J. S. (2008). Disconnection in Jenmalm, P., Schmitz, C., Forssberg, H., & Ehrsson, H. H. (2006).
49 114
prosopagnosia and face processing. Cortex, 44, 996e1009. Lighter or heavier than predicted: Neural correlates of
50 115
Freud, E., Plaut, D. C., & Behrmann, M. (2016). ‘What’ is happening corrective mechanisms during erroneously programmed lifts.
51 116
in the dorsal visual pathway. Trends in Cognitive Sciences, 20(10), The Journal of Neuroscience, 26(35), 9015e9021.
52 117
773e784. Kentridge, R. W. (2014). Object perception: Where do we see the
53 Freud, E., Rosenthal, G., Ganel, T., & Avidan, G. (2015). Sensitivity weight? Current Biology, 24(16), R740eR741.
118
54 to object impossibility in the human visual cortex: Evidence Konen, C. S., & Kastner, S. (2008). Two hierarchically organized 119
55 from functional connectivity. Journal of Cognitive Neuroscience, neural systems for object information in human visual cortex. 120
56 27(5), 1029e1043. Nature Neuroscience, 11(2), 224e231. 121
57 Gallivan, J. P., Cant, J. S., Goodale, M. A., & Flanagan, J. R. (2014). de Lange, F. P. (2006). Weight lifting in the human brain. The 122
58 Representation of object weight in human ventral visual Journal of Neuroscience, 26(41), 10327e10328. 123
59 cortex. Current Biology, 24(16), 1866e1873. Lee Masson, H., Wallraven, C., & Petit, L. (2017). "Can touch 124
60 Gerbella, M., Belmalih, A., Borra, E., Rozzi, S., & Luppino, G. (2010). this": Cross-modal shape categorization performance is 125
61 Cortical connections of the macaque caudal ventrolateral associated with microstructural characteristics of white 126
62 prefrontal areas 45A and 45B. Cerebral Cortex, 20(1), 141e168. matter association pathways. Human Brain Mapping, 38(2), 127
63 Givre, S. J., Schroeder, C. E., & Arezzo, J. C. (1994). Contribution of 842e854. 128
64 extrastriate area V4 to the surface-recorded flash VEP in the Leemans, A., Jeurissen, B., Sijbers, J., & Jones, D. K. (2009). 129
65 awake macaque. Vision Research, 34(4), 415e428. ExploreDTI: A graphical toolbox for processing, analyzing, and 130
c o r t e x x x x ( 2 0 1 8 ) 1 e1 6 15
1 visualizing diffusion MR data. In 17th Annual Meeting of Intl Soc Peelen, M. V., & Caramazza, A. (2012). Conceptual object 66
2 Mag Reson Med Hawaii, USA (p. 3537). representations in human anterior temporal cortex. The 67
3 Lemon, R. N., Johansson, R. S., & Westling, G. (1995). Corticospinal Journal of Neuroscience, 32(45), 15728e15736. 68
4 control during reach, grasp, and precision lift in man. The a, H., Tilikete, C., Vighetto, A., Desmurget, M.,
Pisella, L., Gre 69
5 Journal of Neuroscience, 15(9), 6145e6156. Rode, G., et al. (2000). An 'automatic pilot' for the hand in 70
6 Lukos, J. R., Ansuini, C., & Santello, M. (2007). Choice of contact human posterior parietal cortex: Toward reinterpreting optic 71
7 points during multidigit grasping: Effect of predictability of ataxia. Nature Neuroscience, 3(7), 729e736. 72
8 object center of mass location. The Journal of Neuroscience, van Polanen, V., & Davare, M. (2015). Interactions between dorsal 73
27(14), 3894e3903. and ventral streams for controlling skilled grasp.
9 74
Mahon, B. Z., Kumar, N., & Almeida, J. (2013). Spatial frequency Neuropsychologia, 79(Pt B), 186e191.
10 75
tuning reveals interactions between the dorsal and ventral Prado, J., Clavagnier, S., Otzenberger, H., Scheiber, C.,
11 76
visual systems. Journal of Cognitive Neuroscience, 25(6), 862e871. Kennedy, H., & Perenin, M. T. (2005). Two cortical systems for
12 77
Malach, R., Reppas, J. B., Benson, R. R., Kwong, K. K., Jiang, H., reaching in central and peripheral vision. Neuron, 48(5),
13 Kennedy, W. A., et al. (1995). Object-related activity revealed 849e858.
78
14 by functional magnetic resonance imaging in human occipital Ramayya, A. G., Glasser, M. F., & Rilling, J. K. (2010). A DTI 79
15 cortex. Proceedings of the National Academy of Sciences of the USA, investigation of neural substrates supporting tool use. Cerebral 80
16 92(18), 8135e8139. Cortex, 20(3), 507e516. 81
17 Marteniuk, R. G., MacKenzie, C. L., Jeannerod, M., Athenes, S., & Rice, N. J., Tunik, E., & Grafton, S. T. (2006). The anterior 82
18 Dugas, C. (1987). Constraints on human arm movement intraparietal sulcus mediates grasp execution, independent or 83
19 trajectories. Canadian Journal of Psychology, 41(3), 365e378. requirement to update: New insights from transcranial 84
20 Martin, M., Beume, L., Kümmerer, D., Schmidt, C. S., Bormann, T., magnetic stimulation. The Journal of Neuroscience, 26(31), 85
21 Dressing, A., et al. (2016). Differential roles of ventral and 8176e8182. 86
22 dorsal streams for conceptual and production-related Rizzolatti, G., & Matelli, M. (2003). Two different streams form the 87
23 components of tool use in acute stroke patients. Cerebral dorsal visual system: Anatomy and functions. Experimental 88
24 Cortex, 26(9), 3754e3771. Brain Research, 153(2), 146e157. 89
25 McKeefry, D. J., & Zeki, S. (1997). The position and topography of Ross, E. D. (2008). Sensory-specific amnesia and hypoemotionality 90
26 the human colour centre as revealed by functional magnetic in humans and monkeys: Gateway for developing a hodology 91
27 resonance imaging. Brain, 120(Pt 12), 2229e2242. of memory. Cortex, 44, 1010e1022. 92
28 Merriam, E. P., Gardner, J. L., Movshon, J. A., & Heeger, D. J. (2013). Rossetti, Y., Pisella, L., & Vighetto, A. (2003). Optic ataxia revisited. 93
Modulation of visual responses by gaze direction in human Experimental Brain Research, 153(2), 171e179.
29 94
visual cortex. The Journal of Neuroscience, 33(24), 9879e9889. Rossetti, Y., Revol, P., McIntosh, R., Pisella, L., Rode, G.,
30 95
Milner, A. D. (1997). Vision without knowledge. Philosophical Danckert, J., et al. (2005). Visually guided reaching: Bilateral
31 96
Transactions of the Royal Society of London B Biological Sciences, posterior parietal lesions cause a switch from fast
32 97
352(1358), 1249e1256. visuomotor to slow cognitive control. Neuropsychologia, 43(2),
33 Milner, A. D. (2017). How do the two visual streams interact with 162e177.
98
34 each other? Experimental Brain Research, 235, 1297e1308. Roth, Z. N., & Zohary, E. (2015). Position and indentity information 99
35 Milner, A. D., & Goodale, M. A. (2008). Two visual systems re- available in fMRI patterns of activity in human visual cortex. 100
36 viewed. Neuropsychologia, 46(3), 774e785. The Journal of Neuroscience, 35(33), 11559e11571. 101
37 Monaco, S., Chen, Y., Medendorp, W. P., Crawford, J. D., Fiehler, K., & Schenk, T. (2012). No dissociation between perception and action 102
38 Henriques, D. Y. (2014). Functional magnetic resonance imaging in patient DF when haptic feedback is withdrawn. The Journal 103
39 adaptation reveals the cortical networks for processing grasp- of Neuroscience, 32(6), 2013e2017. 104
40 relevant object properties. Cerebral Cortex, 24(6), 1540e1554. Schenk, T., & McIntosh, R. D. (2009). Do we have independent 105
41 Naito, E., Roland, P. E., Grefkes, C., Choi, H. J., Eickhoff, S., visual streams for perception and action? Cognitive 106
42 Geyer, S., et al. (2005). Dominance of the right hemisphere and Neuroscience, 1(1), 52e62. 107
43 role of area 2 in human kinesthesia. Journal of Neurophysiology, Schmitz, C., Jenmalm, P., Ehrsson, H. H., & Forssberg, H. (2005). 108
44 93, 1020e1034. Brain activity during predictable and unpredictable weight 109
45 Nowak, D. A., Glasauer, S., & Hermsdo € rfer, J. (2013). Force control changes when lifting objects. Journal of Neurophysiology, 93(3), 110
46 in object manipulation-A model for the study of sensorimotor 1498e1509. 111
47 control strategies. Neuroscience and Biobehavioral Reviews, 37(8), Schroeder, C. E., Mehta, A. D., & Givre, S. J. (1998). A 112
48 1578e1586. spatiotemporal profile of visual system activation revealed by 113
Oldfield, R. C. (1971). The assessment and analysis of handedness: current source density analysis in the awake macaque.
49 114
The Edinburgh inventory. Neuropsychologia, 9(1), 97e113. Cerebral Cortex, 8(7), 575e592.
50 115
Oosterhof, N. N., Wiggett, A. J., Diedrichsen, J., Tipper, S. P., & Shapiro, K., Hillstrom, A. P., & Husain, M. (2002). Control of
51 116
Downing, P. E. (2010). Surface-based information mapping visuotemporal attention by inferior parietal and superior
52 117
reveals crossmodal vision-action representations in human temporal cortex. Current Biology, 12(15), 1320e1325.
53 parietal and occipitotemporal cortex. Journal of Shapiro, S. S., & Wilk, M. B. (1965). Analysis of variance test for
118
54 Neurophysiology, 104(2), 1077e1089. normality (complete samples). Biometrika, 52(3/4), 591e611. 119
55 Oram, M. W., & Perrett, D. I. (1996). Integration of form and Sim, E. J., Helbig, H. B., Graf, M., & Kiefer, M. (2015). When action 120
56 motion in the anterior superior temporal polysensory area observation facilitates visual perception: Activation in visuo- 121
57 (STPa) of the macaque monkey. Journal of Neurophysiology, motor areas contributes to object recognition. Cerebral Cortex, 122
58 76(1), 109e129. 25(9), 2907e2918. 123
59 Paulignan, Y., Jeannerod, M., MacKenzie, C., & Marteniuk, R. Singh-Curry, V., & Husain, M. (2009). The functional role of the 124
60 (1991). Selective perturbation of visual input during inferior parietal lobe in the dorsal and ventral stream 125
61 prehension movements. 2. The effects of changing object size. dichotomy. Neuropsychologia, 47(6), 1434e1448. 126
62 Experimental Brain Research, 87(2), 407e420. Takemura, H., Rokem, A., Winawer, J., Yeatman, J. D., 127
63 Paulignan, Y., MacKenzie, C., Marteniuk, R., & Jeannerod, M. Wandell, B. A., & Pestilli, F. (2016). A major human white 128
64 (1990). The coupling of arm and finger movements during matter pathway between dorsal and ventral visual cortex. 129
65 prehension. Experimental Brain Research, 79(2), 431e435. Cerebral Cortex, 26(5), 2205e2214. 130
16 c o r t e x x x x ( 2 0 1 8 ) 1 e1 6
1 Thiebaut de Schotten, M., Cohen, L., Amemiya, E., Braga, L. W., & Retinotopy and colour. Philosophical Transactions of the Royal 34
2 Dehaene, S. (2014). Learning to read improves the structure of Society of London B Biological Sciences, 357(1424), 963e973. 35
3 the arcuate fasciculus. Cerebral Cortex, 24(4), 989e995. Wang, R., Benner, T., Sorensen, A. G., & Wedeen, V. J. (2007). 36
4 Tootell, R. B., Mendola, J. D., Hadjikhani, N. K., Ledden, P. J., Diffusion toolkit: A software package for diffusion imaging 37
5 Liu, A. K., Reppas, J. B., et al. (1997). Functional analysis of V3A data processing and tractography. Proceedings International 38
6 and related areas in human visual cortex. The Journal of Society Magazine Resonance Medical, 15, 3720. Q5 39
7 Neuroscience, 17(18), 7060e7078. Weiner, K. S., Yeatman, J. D., & Wandell, B. A. (2016). The 40
8 Tournier, J. D., Calamante, F., & Connelly, A. (2007). Robust posterior arcuate fasciculus and the vertical occipital 41
determination of the fibre orientation distribution in diffusion fasciculus. Cortex. https://doi.org/10.1016/j.cortex.2016.03.012.
9 42
MRI: Non-negativity constrained super-resolved spherical Wing, A. M., Turton, A., & Fraser, C. (1986). Grasp size and
10 43
deconvolution. Neuroimage, 35(4), 1459e1472. accuracy of approach in reaching. Journal of Motor Behavior,
11 44
Tournier, J. D., Calamante, F., Gadian, D. G., & Connelly, A. (2004). 18(3), 245e260.
12 45
Direct estimation of the fiber orientation density function Yeatman, J. D., Rauschecker, A. M., & Wandell, B. A. (2013).
13 from diffusion-weighted MRI data using spherical Anatomy of the visual word form area: Adjacent cortical
46
14 deconvolution. Neuroimage, 23(3), 1176e1185. circuits and long-range white matter connections. Brain and 47
15 Tretriluxana, J., Gordon, J., & Winstein, C. J. (2008). Manual Language, 125(2), 146e155. 48
16 asymmetries in grasp-shaping and transport-grasp Yeatman, J. D., Weiner, K. S., Pestilli, F., Rokem, A., Mezer, A., & 49
17 coordination. Experimental Brain Research, 188(2), 205e315. Wandell, B. A. (2014). The vertical occipital fasciculus: A 50
18 Tunik, E., Frey, S. H., & Grafton, S. T. (2005). Virtual lesions of the century of controversy resolved by in vivo measurements. 51
19 anterior intraparietal area disrupt goal-dependent on-line Proceedings of the National Academy of Sciences U S A, 111(48), 52
20 adjustments of grasp. Nature Neuroscience, 8(4), 505e511. E5214eE5223. 53
21 Ungerleider, L. G., Galkin, T. W., Desimone, R., & Gattass, R. (2008). Zeki, S., Watson, J. D., Lueck, C. J., Friston, K. J., Kennard, C., & 54
22 Cortical connections of area V4 in the macaque. Cerebral Frackowiak, R. S. (1991). A direct demonstration of functional 55
23 Cortex, 18(3), 477e499. specialization in human visual cortex. The Journal of 56
24 Ungerleider, L. G., & Mishkin, M. (1982). Two cortical visual Neuroscience, 11(3), 641e649. 57
25 systems. In D. J. Ingle, M. A. Goodale, & R. J. W. Mansfield Zhong, Y. M., & Rockland, K. S. (2003). Inferior parietal lobule 58
26 (Eds.), Analysis of visual behavior (pp. 549e586). Cambridge, MA: projections to anterior inferotemporal cortex (area TE) in 59
27 MIT Press. macaque monkey. Cerebral Cortex, 13(5), 527e540. 60
28 Vinberg, J., & Grill-Spector, K. (2008). Representation of shapes, Zimmermann, M., Verhagen, L., de Lange, F. P., & Toni, I. (2016). 61
edges, and surfaces across multiple cues in the human visual The extrastriate body area Computes Desired goal states
29 62
cortex. Journal of Neurophysiology, 99(3), 1380e1393. during action planning. eNeuro, 3(2). https://doi.org/10.1523/
30 63
Wade, A. R., Brewer, A. A., Rieger, J. W., & Wandell, B. A. (2002). ENEURO.0020-16.2016.
31 64
Functional measurements of human ventral occipital cortex:
32 65
33 66

Bud I Savlje Vic 2018

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bud I Savlje Vic 2018

Uploaded by

Copyright:

Available Formats

CORTEX2260_proof ■ 22 March 2018 ■ 1/16

Available online at www.sciencedirect.com

1 (Rizzolatti & Matelli, 2003), which serves as an integration 66

1 Catani, M., Jones, D. K., & ffytche, D. H. (2005). Perisylvian 66

You might also like