Computational protocol: Distinction between hand dominance and hand preference in primates: a behavioral investigation of manual dexterity in nonhuman primates (macaques) and human subjects

Similar protocols

Protocol publication

[…] The data of the behavioral tasks were analyzed manually from the recorded video sequences. The software VirtualDubMpeg2® (Developper Avery Lee, free software, http://www.virtualdub.org) allowed visualizing the video sequences frame by frame, corresponding to a time resolution of 40 msec (acquisition at 25 frames per second). The data were processed first in Excel® worksheets, before they were transferred to Sigmastat®/Sigmaplot® (Systat Software Inc., http://www.sigmaplot.com) and SPSS® (SPSS Inc., Chicago, IL) allowing more elaborated graphic representation and statistical analysis.The hand dominance was determined based on a single task, the modified Brinkman board task performed with one hand imposed at a time. Two types of data were analyzed for the monkeys (Schmidlin et al. ). (i) The score, defined as the number of pellets correctly retrieved during the first 30 sec; (ii) The contact time (CT), defined as the time interval between the first contact of a finger (most often the index finger) with the pellet and the moment when the fingers left the slot with the reward. The CT is a pertinent parameter in addition to the score, as the latter can sometimes be biased. Indeed, the animal may be disturbed by external noises, or may exhibit a lack of motivation or concentration. In such cases, the monkey may interrupt the test, leading to a distortion of the score. Moreover, the CT truly measures the actual manipulation of the pellets with the fingers. The CT was measured for the first five horizontal and the first five vertical slots in the 20 last daily sessions at plateau, whereas the score was calculated for every daily session. The onset of the plateau was defined, when the learning curve tended to saturate (as estimated by visual inspection), as the first value in the nearly flat curve of the score that was not exceeded by one of the five following score values. For human subjects, the analysis of hand dominance was based mainly on the score in 30 sec, although the CT was also established for comparison in a sample of subjects.The hand preference for monkeys was determined based on four tests: the modified Brinkman board task, when the animal was free to use both hands simultaneously, the reach and grasp drawer task, the tube task, and the bimanual Brinkman board task. For human subjects, two tests were considered, the tube task and the bimanual Brinkman board task, as well as the questionnaire indicating their self-assessed hand preference. For the tube task, the preferred hand was defined as the hand used to grasp the reward into the tube, playing the manipulative role, whereas the other hand, holding the tube, played the postural role. The preferred hand (left hand or right hand) was determined for each tube task trial performed by the subject (humans and monkeys), in order to calculate the handedness index (HI) (see below). For the bimanual board task, the subjects (humans and monkeys) used two different strategies to retrieve the reward. In the first one, the hand above the board pushed the reward while the other hand collected it below the board. In the second one, the hand positioned below the board pushed up the reward using one finger (usually the index finger) and the other hand grasped it above the board, performing the precision grip. In the first strategy (adopted in more than 98% of trials in five out of eight monkeys), the preferred hand is the one pushing the reward. Indeed its role is manipulative, whereas the role of the other hand is postural. For the second strategy, the preferred hand is the one retrieving the reward, as its action is more manipulative and more challenging (precision grip), as compared to the role of the other hand (one finger used). Additionally, the board has an inclination, making this movement still more difficult. This second strategy was used in about half of the trials in one monkey (Mk-MI) and it was predominant in two other monkeys (Mk-CA and Mk-AN; 68% and 98%, respectively). For the reach and grasp drawer task (in monkeys only), the preferred hand is the hand grasping the reward (manipulative role) while the other hand, the postural one, holds the drawer.For these three tasks (bimanual Brinkman board task, reach and grasp drawer task, tube task), we computed the HI (Westergaard et al. ; Spinozzi et al. ; Hopkins et al. ; Schmitt et al. ), defined as follows: the number of trials the right hand (R) was used as preferred hand minus the number of times the left hand (L) was used as preferred hand, divided by the total number of trials: Consequently, a negative HI reflects a left bias whereas a positive HI reflects a right bias. The HI (lateralization) ranges between +1 (strongly right-handed) and −1 (strongly left-handed).For the modified Brinkman board task, we measured the score in 30 sec when the animal was free to use both hands, and counted the number of pellets grasped with each hand. The hand with the highest score is considered as the preferred hand.For the questionnaire, we calculated a handedness score by using the criteria of MacManus ():“Laterality scores (laterality indices):Score all the items as −1 = Always left, −0.5 = Usually left, 0 = Either, +0.5 = Usually right and +1 = Always right. For items 4 (dish), 6 (jar), and 9 (potato) a strong right-hander would answer left. These three items should therefore be reverse scored by changing the sign on the values given previously (i.e., +1 = Always left, etc.). Having done this, then one can obtain the overall laterality score, an average of all 11 items.”The score was then transformed into percentage (−100% indicating strongly left-handed and +100%, strongly right-handed).The statistical analysis was conducted as follows. For the tube task, the reach and grasp drawer task, and the bimanual Brinkman board task, we used a binomial test (SPSS®; see ). For the scores of the modified Brinkman board task, we used either the paired t-test or the Wilcoxon signed-rank test (Sigmastat®). Finally, for the CT derived from the modified Brinkman board task, we used either the unpaired t-test or the Mann–Whitney U test (Sigmastat®).In order to limit the duration of the behavioral session with human subjects to a reasonable extent, the modified Brinkman board task using both hands simultaneously, as well as the reach and grasp drawer task, were not performed with human subjects. These tests, aimed in the monkeys to determine their preferred hand, were considered redundant for human subjects with the handedness questionnaire. […]

Pipeline specifications

Software tools SigmaPlot, SPSS
Application Miscellaneous
Organisms Homo sapiens, Macaca fascicularis