Probing the deployment of peripheral visual attention during obstacle-crossing planning

Abstract

Gaze is directed to one location at a time, making peripheral visual input important for planning how to negotiate different terrain during walking. Whether and how the brain attends to this input is unclear. We developed a novel paradigm to probe the deployment of sustained covert visual attention by testing orientation discrimination of a Gabor patch at stepping and non-stepping locations during obstacle-crossing planning. Compared to remaining stationary, obstacle-crossing planning decreased visual performance (percent correct) and sensitivity (d') at only the first of two stepping locations. Given the timing of the first and second steps before obstacle crossing relative to the Gabor patch presentation, the results suggest the brain uses peripheral vision to plan one step at a time during obstacle crossing, in contrast to how it uses central vision to plan two or more steps in advance. We propose that this protocol, along with multiple possible variations, presents a novel behavioral approach to identify the role of covert visual attention during obstacle-crossing planning and other goal-directed walking tasks.

Keywords: locomotion; movement planning; obstacle avoidance; peripheral vision; visual attention.

Figure 1

Real and virtual environments. Images of the real and virtual environment as a participant approached and stepped over a virtual obstacle. Participants received visual feedback of leg position throughout the experiment. In the virtual world, but not in the real world, the walkway had a textured floor surface and an obstacle with a fixation cross positioned two steps away from the participant. The four rows show the participant (A) at the starting position, (B) taking the first step, (C) stepping over the obstacle with the lead limb, and (D) at the end position. *The black markings over the legs are shown for visual clarity in this figure but were not present in the VR environment.

Figure 2

Experimental setup and procedures. (A) Participants stood on a 3-m platform between parallel bars while wearing a head-mounted display (HTC Vive Pro) embedded with a Pupil Labs eye-tracker. In the virtual world, but not in the real world, the floor surface had texture (not displayed here, see Figure 1), and a 5 cm obstacle placed two steps away from the participants’ starting position. Participants fixated on a cross placed on the obstacle to have them gaze two steps ahead. (B) Sequence of events in each trial. Fixation: Participants began by fixating on the cross for 500 ms. Cue onset: Following fixation, a cue presented at fixation prompts the preparation of obstacle crossing or to remain stationary. In this example, a rightward arrow cue is presented, indicating the participant should take their first step with the right leg as quickly as possible after cue onset. The cue could also be a leftward arrow, indicating the participant should begin stepping with their left leg first, or a horizontal bar indicating that they remain stationary. The cue consequently dictated the stepping and non-stepping locations along the walkway. Gabor Onset: The test stimulus (Gabor patch) was presented 300 ms after cue onset (cue-test interval of 300 ms). The Gabor patch appeared with equal probability at one of the four potential step locations (R1, L1, R2, L2, see C); in this example, the Gabor patch appears in R2. Gabor offset: The offset of the Gabor patch occurred 77.8 ms after its onset. Cross and cue offset: The offset of the cross and cue occurred 44.4 ms after the offset of the Gabor patch. On stationary trials, this is the end of the trial. On obstacle-crossing trials, participants initiated their first step within 1,000 ms after cue onset and approached and stepped over the obstacle. Participants also needed to verbally report the orientation of the Gabor patch as quickly as possible. (C) Potential stepping and Gabor patch locations determined by baseline step locations. There were two possible step locations in the first (step 1) and second (step 2) step, represented as the first: (1) and second (2) potential step locations for the Right (R) and Left (L) leg. For obstacle-crossing trials, the leftward arrow cue led to steps into positions L1 and R2 (L cue steps), and the rightward arrow cue led to steps into positions R1 and L2 (R cue steps), followed by stepping over the obstacle. (D) Orientation discrimination of the test stimulus (Gabor patch). Participants verbally reported the Gabor patch’s orientation relative to vertical (Right or Left) as quickly as possible following its presentation.

Figure 3

Obstacle-crossing behavior. (A) Foot trajectories of the lead (the first leg to cross obstacle) and trail (the second leg to cross obstacle) limb from a representative participant as they approached and stepped over the obstacle. Lead and trail horizontal distance are shown and defined as the horizontal distance between the obstacle and the foot in the step before crossing the obstacle. Lead and trail foot clearance is depicted and defined as the vertical distance between the obstacle and the foot at the point of crossing the obstacle. (B) Stacked density plots for lead and trail horizontal distance for each participant. (C) Stacked density plots for lead and trail foot clearance for each participant. Dots and error bars represent the mean and 95% CIs for each participant.

Figure 4

Step performance (step onset and foot placement). (A) Stacked density plots of step onset time for each participant; dots and error bars represent the mean and 95% CI. Vertical dotted lines represent important time points from our study. The vertical lines (left to right) represent Gabor patch onset, Gabor patch offset, and mean step onset relative to cue onset. The purple, blue, and pink shaded regions represent the timings for indicators of movement planning extracted from the literature. Changes in the center of pressure (purple), referred to as anticipatory postural adjustments, and cortical (EEG) activity (blue) over the parietal and occipital lobes occur shortly after the onset of a movement cue (~100–450 ms) and indicate movement planning. Step onset (pink) occurs later (~650–710 ms) and represents the end of the planning phase and the beginning of the execution phase. The step onset times in our study were similar to step onset times in previous studies (right most vertical line vs. pink shaded region), making it likely that cortical activity and anticipatory postural adjustments in our participants began shortly after cue onset. The overlap between the indicators of movement planning and the onset and offset of the Gabor patch show that our participants likely visually discriminated the Gabor patch during the planning phase of the movement. (B) Foot placement relative to test stimulus (Gabor patch) location. The starting position and foot placement for every left cue trial from a representative participant. The Gabor patch depicts the test stimulus location. The colored asterisks represent the left (purple) and right (blue) steps, and the yellow dot and error bars represent the mean and standard deviation (SD). EEG: electroencephalography. *1: Braquet et al. (2020). *2: Uemura et al. (2013a). *3: Uemura et al. (2013b).

Figure 5

Orientation discrimination. This figure shows raw and change (Δ) scores (obstacle-crossing scores minus stationary scores) factored by relevance and step for (A) percent correct (visual performance); (B) d’ (visual sensitivity); (C) β (response bias); (D) Δ percent correct (Δ visual performance); (E) Δ d’ (Δ visual sensitivity); and (F) Δ β (Δ response bias). In the bottom panel (D–F), the horizontal dashed line represents scores from the stationary baseline. Filled symbols represent values that were statistically different (p < 0.05) from the stationary baseline, determined by whether the 95% confidence intervals (error bars) crossed zero. We used linear mixed-effects models for statistical analyses on the Δ scores (Table 2). Pairwise comparison p values in (D) were evaluated at a Bonferroni corrected alpha value of 0.013 (see “Material and methods” Section).