Investigating the dependencies between visual and kinesthetic motor imagery

Motor imagery (MI) is considered a cognitively driven process which involves – but is not limited to – the imagination of visual and kinesthetic properties of movement (Kosslyn et al., 2010). Visual MI is believed to provide information concerning spatial coordinates of a movement (from the first- or third-person perspective) and kinesthetic MI provides biomechanical information about the movement from the first-person perspective (e.g., muscle force; Féry, 2003). Frameworks for developing and delivering motor imagery in sports, such as Layered Stimulus Response Training (LSRT; Cumming et al., 2017) and the PETTLEP model (i.e., consideration of Physical, Emotion, Task, Timing, Learning, Environment, Perspective related aspects of tasks; Holmes & Collins, 2001; Scott et al., 2022) are based on the integration of different sensory modalities to enhance the imagery experience (i.e., visual, kinesthetic and auditory imagery). An additional commonality between the LSRT and PETTLEP approaches is the emphasis on the generation – or on prioritization – of a visual image early in the imagery process, providing a skeleton to inform and integrate other modalities such as kinesthetic MI. However, whether imagery modalities (visual, kinesthetic) are independent or dependent remains untested. Here we aim to disentangle visual and kinesthetic components of MI, providing evidence of a potential hierarchy within imagery modalities. Researchers typically instruct and measure visual and kinesthetic MI modalities separately (e.g., Guillot et al., 2008; Stinear et al., 2006). Yet, individuals often experience visual and kinesthetic MI simultaneously, regardless of instruction. For example, during completion of the Vividness of Motor Imagery Questionnaire-2, athletes reported experiencing both modalities concurrently, or experienced visual MI before kinesthetic MI (Callow & Roberts, 2010). When individuals are given only limited experiences of a task (i.e., just visual or non-visual physical practice), asymmetries in the use of visual and kinesthetic MI occur. Peters et al. (submitted) gave groups either visual or no-visual physical practice of novel hand gesture sequences. This isolated exposure resulted in greater accuracy for the visual vs. physical practice group (i.e., greater similarity between imagined and actual movement times/MTs and perception of imagery quality), for both visual and kinesthetic MI. The advantage for visual practice suggests a benefit and stronger contribution of a visual representation during both visual and kinesthetic MI. In terms of cortical neurophysiology, there is evidence both for similarities in the areas involved in visual and kinesthetic MI (i.e., somatosensory cortex, premotor cortex, supplementary motor areas and cerebellum), as well as differences (Filgueiras et al., 2017). Visual MI has been associated with greater activity in the visual cortex than kinesthetic MI (Guillot et al., 2008; Mizuguchi et al., 2017; yet see Kilintari et al., 2016). There is also evidence that kinesthetic MI results in greater activation of the motor cortex than visual MI, through Transcranial Magnetic Stimulation (TMS) methods to assess corticospinal excitability (Stinear et al., 2006). Despite neurophysiological evidence that visual and kinesthetic MI are somewhat separate and that they should be instructed and measured separately, there is still uncertainty regarding the general independence of these modalities and the potential dependence of kinesthetic MI on visual MI (Klatzky, 1994; Stevens, 2005). Indirect evidence that kinesthetic MI may be dependent on what might be thought of as a visual scaffold can be inferred from studies where action observation (AO) and motor imagery (MI) have been combined. Studies have shown positive benefits when individuals engage in, and synchronize, kinesthetic MI with a video of the imagined action, than when they do only imagery or observation (this combination has been termed “AOMI”). For example, the instruction to use kinesthetic MI with a visual display of the same action results in greater corticomotor excitability and favorable, or additive, behavioral outcomes than either method alone (Wright et al., 2014; Scott et al., 2023; see Chye et al., 2023 for meta-analysis). It is proposed that the visual display replaces the requirement for visual MI, providing a structure to inform kinesthetic MI (Eaves et al., 2016; Wright et al., 2022). However, the presence of such visual displays may not necessarily negate the generation of visual MI and factors such as experience, visual perspective and agency in the task can influence the use of visual MI (Mizuguchi et al. 2016; Scott et al., 2022; Wright & Holmes, 2023). According to the visual guidance hypothesis (Meers et al., 2020), the visual display during AOMI simply provides a guide for kinesthetic MI, allowing a stronger generation of this latter imagery type, implying the need for a visual representation to generate kinesthetic MI. MI can be considered a multi-dimensional process requiring the generation, manipulation and maintenance of an image (Cumming & Eaves, 2018; Kraeutner et al., 2020). One way in which generation and maintenance properties can be captured is through rating scales quantifying imagery quality and chronometric difference scores, respectively (e.g., Collet et al., 2011; Moran et al., 2012; Williams et al., 2015). In chronometric difference scores, the duration of an imagined movement is compared to the time taken to actually (i.e., physically) perform the movement, with smaller discrepancies indicating greater accuracy during MI. This measure provides insight regarding the temporal aspect of the imagined movement, thought to be informed mostly by kinesthetic MI (Decety et al., 1989; Féry, 2003; Munzert et al., 2015). For example, imagining walking with a weighted backpack significantly increased imagined movement times compared to imagining walking without (Decety et al., 1989; Munzert et al., 2015). Perhaps relating to the manipulation of imagined content, gaze metrics have been shown to give insight into visual-spatial content-related aspects of imagery (e.g., through saccades and fixations; Heremans et al., 2008; Lanata et al., 2020; see Causer et al., 2013 for a review). Ocular patterns during visual MI of a task are similar to those observed when actually performing the same task (Heremans et al., 2008). Poiroux et al. (2015) investigated the congruence between independently instructed visual and kinesthetic MI after physically practicing a task which involved laterally moving 15 wooden blocks. More task-based saccades occurred during visual than kinesthetic MI, indicating some distinction between these processes related to spatially-oriented action components. However, there were also more task-related saccades during kinesthetic MI than control conditions, where participants looked at the stimuli (without imagining) or performed a counting task while looking at the stimuli. These latter gaze-related saccades even for kinesthetic MI, point to the involvement of visual-spatial representations for this type of imagery, potentially serving to guide the generation of kinesthetic MI. The overarching aim of this study is to determine the dependencies between visual and kinesthetic MI modalities. We will measure mental chronometry (MI accuracy) and eye gaze metrics (visual spatial aspects) to give insight into both the temporal and spatial properties of imagined content, respectively. In addition, we will measure subjective ratings of ease and quality of imagery for the different modalities. During each type of MI, we will probe the involvement of visual MI through the restriction of gaze. Specifically, a fixation cross will be used to hinder visual-based MI (see Stevens, 2005). Presumably, if kinesthetic MI does not require or involve visual MI, restricting gaze should not impact mental chronometry times or other subjective measures of quality for this imagery modality. Participants will gain experience performing a hand gesture sequence, with each gesture performed in a different spatial location, involving vertical and horizontal trajectories, but within an individual’s field of vision. To determine the requirement and presence of visual-spatial representations in MI, all participants will then complete three different conditions; 1) control (no MI), 2) visual MI, and 3) kinesthetic MI. Each of these conditions will be completed with and without instructions to fixate gaze on a stationary cross, resulting in a 3 (Imagery condition) X 2 (Visual fixation) repeated measures’ design. During each condition, participants will have eye gaze recorded, and mental chronometry times will be measured during imagery conditions. The control condition will be completed before practice of the gesture task and will involve participants just looking at a four-quadrant grid (with and without a fixation cross present). Participants will also have their eye gaze tracked at the end of the study, when they will be asked to observe a video of an actor performing the same gestural task as practiced. The video will be filmed from the first-person perspective captured by the eye-tracking glasses, so it matches what the participant would actually see if they were performing. This final condition will allow us to determine differences in gaze patterns as a result of visual perception versus imagining. This project added to this OSF will address the contributions of visual MI during both visual and kinesthetic MI (Experiment 1). On completion of this experiment we will update this OSF with details of a second experiment to test the role of kinesthetic MI during visual and kinesthetic MI.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.