Bimanual grasping adheres to Weber's law. [Dataset]

Abstract

The perception–action model (PAM; Milner and Goodale (1995, 2006)) suggests two uses for vision that are independent and governed by anatomically and functionally distinct visual streams in the extrastriate cortex. The ventral stream is assumed to process vision-for-perception, whereas the dorsal stream is assumed to process vision-for-action. The model was originally inspired by studies on patients who suffered from circumscribed damage to either dorsal or ventral stream areas and consequently seemed to show behavioral impairments restricted to either visuomotor or perceptual tasks (e.g., Goodale et al., 1991, 1994). A range of different paradigms has been employed in recent decades to provide evidence for segregated functions also in neurologically healthy humans, with many of them being discussed controversially in the current literature (for reviews see Bruno et al., 2008; Franz & Gegenfurtner, 2008; Schenk & McIntosh, 2010; Smeets & Brenner, 2006). One of the more recent paradigms used to substantiate the claim that vision for action and vision for perception follow different rules also in neurologically intact individuals is the finding by Ganel et al. (2008) that vision for perception but not vision for action adheres to Weber's law.

Citation

HESSE, C., HARRISON, R.E., GIESEL, M. and SCHENK, T. 2021. Bimanual grasping adheres to Weber's law. [Dataset]. Hosted on Zenodo [online]. Available from: https://doi.org/10.5281/zenodo.5115810

• Acceptance Date: Oct 3, 2021
• Online Publication Date: Nov 25, 2021
• Publication Date: Nov 30, 2021
• Deposit Date: Feb 23, 2024
• Publicly Available Date: Feb 23, 2024
• DOI: https://doi.org/10.5281/zenodo.5115810
• Keywords: Perception-action; Dissociation; Psychophysics; Two-visual streams
• Public URL: https://rgu-repository.worktribe.com/output/2250077
• Related Public URLs: https://rgu-repository.worktribe.com/output/2092475 (related Journal article)
• Type of Data: TXT files.
• Collection Date: Jul 21, 2021
• Collection Method: Twenty-two neurologically healthy volunteers with normal or corrected to normal visual acuity participated in the experiment. The data from two participants was excluded from analysis as they did not follow the instructions (i.e., placing their hands at the respective start positions and grasping the boxes on their sides). The remaining 20 participants (15 female) were between 18 and 37 years old and right-handed by self-report. The sample size of N=20 was based on a G*Power 3.1 analysis (Faul et al., 2009). Using a mixed design (with task, manual estimation vs. grasping, as between subject factor) and object size as a within-subject factor with four levels, Ganel et al. (2017) observed a large effect of f=.70 (ηp 2=.34) for the interaction between the linear components of bimanual grasping and perceptual estimation for the JNDs in their study. A TrakStar™ electromagnetic motion tracker (Ascension Technology Corporation, NDI) with a sampling rate of 240 Hz was used to record participants' hand movements. One marker was placed on the fingernail of participants' index finger on each hand. It was kept in place with whitetac and medical tape, exposing the pad of the finger so that participants received normal haptic feedback. Small Velcro strips were attached at the base of the index finger and at the wrist to keep the wire from each marker out of the way during the trials. Participants wore liquid-crystal shutter glasses (PLATO Translucent Technologies, Toronto, Ontario; Milgram (1987)) to control vision occlusion. The experiment was programmed in Matlab (Mathworks, Natick, MA, USA). The 3D position data of the two markers were filtered offline using a second-order Butterworth filter with a cut-off frequency of 15 Hz. Resultant velocity was calculated from the filtered 3D position data of the markers. Movement onset was defined as the first frame where the resultant velocity of one of the markers exceeded 0.05 m/s. The end of the grasping movement corresponded to the moment the box was lifted off the release button. Grasping trials in which the start or end of the movement could not be reliably determined (e.g., participants moved before the go-signal) were excluded from further analysis. This was the case for 16 out of 1,600 trials across all participants. MGA was calculated as the maximum 3D Euclidean distance between the hands reached during the grasping movement. The hand aperture in the bimanual estimation trials was determined as the 3D Euclidean distance between the hands at the time of the keypress. Estimation trials in which participants moved their hands at the moment the aperture was recorded were excluded from further analysis. This was the case for four out of 1,600 trials across all participants. In line with previous literature (e.g., Ganel et al., 2008, 2017), JND values were defined as the SDs of the MGAs for each participant in each condition (box width, task type, start position combinations).1 Both MGA and JND data were initially analyzed using a 2 (task type: grasping vs. estimation) × 2 (start position: fixed vs. adapting) × 4 (box width: 160, 240, 320, and 400 mm) repeated-measures ANOVA.