The neural mechanism of biomechanical constraints in the hand laterality judgment task: A near-infrared spectroscopy study

Highlights

• Biomechanical constraints (BC) effect is tested in hand laterality judgment task.

• Each stimulus is showed once in the practice and experimental phase, respectively.

• The activation of the left superior parietal lobe (SPL) varies with BC.

• The left SPL is involved in the BC effect.

• The BC effect is obtained for the (awkward) left hand of right-handers.

Abstract

The mental rotation (MR) task is defined as a discrimination task between mirror-reversed images involving discrepancy in angular orientation. Various studies have shown that the MR task likely causes mental imagery, that is, visual and/or motor imagery, depending on stimulus types. When figures of rotated hands are presented to be identified as a left or right hand, reaction times (RTs) usually show an effect of biomechanical constraints (BC): a hand in a position difficult to reach with a real movement results in longer RTs. The BC effect as a marker of motor imagery has been investigated by brain function measures (fMRI, PET, EEG and MEG) as well as by RTs. Unlike other neuroimaging techniques, NIRS (near-infrared spectroscopy) imposes few physical constraints on participants and is relatively unaffected by motion artifact, which permits serial assessments of tasks in relaxed and natural environment. Focusing on these advantages, a NIRS study on motor imagery in HLJ was carried out in which we measured the brain activation during the HLJ task and a single character judgment task. In the HLJ task, both the RTs and the activity of the left superior parietal lobe (SPL) showed an interaction between Hand (left, right) and Orientation (135°, 225°) i.e., the BC effect, but not in the character judgment task. More specifically, in the analysis of BC-related activity of SPL, although the Hand × Orientation interaction was significant, the left SPL for the left hand significantly increased from 135° to 225°, but the reversed increase (from 225° to 135°) was not found for the right hand. These results suggest that left SPL is involved in the BC effect and NIRS differentiates left hand awkwardness of right-hander in the HLJ task.

Introduction

Mental rotation (MR) is an example of a cognitive process that is assumed to involve an analog transformation of an analog representation [1]. Shepard & Metzler [2] introduced the concept of MR into cognitive psychology with what has become one of the most well-accepted experiments in the field. In their study, participants were asked to judge whether two three-dimensional cube figures, which appeared in different orientations, were identical or mirror images. The reaction time (RT) required to determine whether the pairs of the two cubes were the same increased linearly with increases in the angular difference between the two cubes, indicating that participants mentally rotated an image of one object until it matched the other [3]. Subsequently, this MR effect has been reported in studies with various objects, such as alphanumeric characters [4], letter-like characters [5], and two-dimensional unfamiliar shapes [6]. In contrast, stimuli pertaining to the human body might lead to different patterns of results related to specific characteristics of the body parts involved, such as the biomechanical constraints (BC) underlying the actual movement.

A behavioral task typically used for testing MR of a body part is the “laterality judgment” task in which participants are asked to judge the laterality (left or right) of depicted body parts, such as hands or feet, presented in different views and angular orientations. Several studies found that the time required for a hand laterality judgment (HLJ) is similar to the time taken to execute a corresponding hand movement [7], [8], [9]. These results suggest that to perform the HLJ task, participants mentally rotate their upper limbs toward the position of the displayed stimulus in a way that is consistent with the BC underlying the actual movement [10]. Various studies on the HLJ task have shown that the judgment during the task depends on the BC [7], [8], [9], [10], [11]. These studies demonstrated that RTs depend on the participants’ posture or differ for hand pictures rotated away or toward the mid-sagittal plane (i.e., lateral or medial rotation, respectively). On the basis of these behavioral results, the HLJ task is taken as an example of motor imagery, while the MR of an object task is taken as an example of visual imagery [11].

Vannuscorps et al. [10] reviewed previous studies and proposed that “The prominent explanation for the BC effect is that in order to perform the HLJ task, subjects mentally rotate their upper limbs toward the position of the displayed stimulus in a way that is consistent with the BC underlying the actual movement”. The “motor imagery” account of the BC effect indicates that the motor imagery relies on the same representations and processes as those involved in action planning and control, i.e., central motor areas may play a role in the motor imagery. Actually, a parieto-frontal network, known for its contribution to the planning and execution of hand movements, is activated during motor imagery [9], [11], [12], [13], [14]. For example, de Lange et al. [9] using functional magnetic resonance imagining (fMRI) found that the activity of the parieto-frontal network, including bilateral intraparietal sulcus and dorsal PM (PMd), varied according to the arms’ BC, showing different orientation effects for left and right hands (Hand × Orientation interaction) during the HLJ task. That is, the “Left hand stimuli were easier to judge in the clockwise orientations than in the counterclockwise ones, whereas the reverse was true for right hand stimuli [9].” The PMd activation indicates that the participants mentally rotate their own hand to compare it with the stimulus.

However, the BC effect was also found in the responses of motor-impaired individuals performing the HLJ task [10], [15]. Vannuscorps et al. [10] reported that response latencies of an individual (DC) with a congenital absence of upper limbs for hand drawings in the HLJ task were influenced by the BC of hand movements in the same way as control participants’ response latencies. They concluded that the BC effect in the HLJ task is not strictly dependent on “motor imagery” and can arise from the visual processing of body parts being sensitive to such constraints. They explain their finding, i.e., the BC effect of motor-impaired individuals by means of knowledge of BC, which is an intrinsic component of body part of visual perception processes. Presumably, such knowledge may induce higher cognitive functions and the postulated higher cognitive efforts evoke a question why some motor-impaired individuals show the BC effect and others does not. For instance, Funk and Brugger [15] found that one individual (AZ), with a congenital absence of upper limbs who experienced phantom limb sensations, showed BC effect in the HLJ task; in contrast, the other individual (CL), who did not experience phantom limb sensations, did not show BC effect. The authors interpreted their finding from the perspective of the phantom limb sensations in which visual experience is assumed to activate pre-existing limb representations common to both action observation and execution. This allows the AZ to engage in a process of motor imagery, and CL’s performance represents the necessity of sensorimotor experiences with reference to the hands. Taken together, although these findings show that motor imagery has a critical role to perform the HLJ task, the neural correlates underlying the BC effect are controversial due to the participation of multi-factors in the HLJ task.

Some fMRI and PET studies showed that the parietal lobe and PMd are related to the MR of body parts [9], [11], [12], [13], [14], whereas, some Electroencephalogram studies of MR of body parts reported that event-related potential correlates of MR of body parts are primarily distributed over parietal and occipital electrodes [3], [16]. Thayer and Johnson [3] suggest that the neural computations underlying MR of hand are carried out primarily in bilateral parietal and occipital lobes, while motor regions may play a role in providing ongoing kinesthetic feedback during MR or in checking the results of imagined transformation.

Unlike other neuroimaging techniques such as fMRI and positron emission tomography (PET), NIRS has relatively fewer physical constraints on the participant and is tolerant to motion artifact, which permits serial assessments of tasks in relaxed and natural settings [17]. Focused on the advantage of NIRS, the present study was designed to investigate the neural mechanism of BC effect in the HLJ task using NIRS.

Furthermore, in the previous studies using HLJ task, for the purpose of enhancing statistical power, each stimulus was repeated at least several times regardless of normal individuals or patient(s) as participants [9], [10], [11], [15]. The present study was partly motivated to develop a simple HLJ test and to relieve the burden on normal participants or patients during the measurement of their brain activities using NIRS. To achieve the goal in the present study, each stimulus was presented only once in the practice and experimental phase, respectively. We used the simple HLJ task as an experimental task and a character task as a control task. In the HLJ task, we used three hand forms (viewed from back, palm and thumb-side) of the five hand forms of Sekiyama [8] as the three stimuli which have similar RT functions. She found that the orientation of 135° and 225° corresponded to the “least manageable (most difficult) directions in the actual movements” in the rotation for these three hand forms. According to her results, these two orientations were used to test the BC effect. Therefore, during the data analysis, we conducted a two way-repeated ANOVA with Hand (left, right) and Orientation (135°, 225°) as within factors. We aimed to examine activations of parietal lobe and PM in the simple HLJ task without extensive stimulus repetition. We hypothesized that activation of parietal lobe and PM show a hand × orientation interaction depending on “visual imagery” [10], [15] and “motor imagery” [9], [11].