Postural constraints alter the motor effort of visual orienting

Visual exploration in naturalistic environments is an active, embodied process that requires coordinated movement of the eyes, head, and body. Because these movements incur physical effort, people face a cost-minimizing trade-off between repeatedly sampling the external environment and encoding task-relevant information into visual working memory (VWM). Prior research establishes that memory reliance increases with the motor effort cost of information acquisition, but how physical constraints such as posture modulate these costs remains poorly understood. We manipulated posture and viewing angle and measured working memory reliance. Twenty-seven participants completed a block-copying task in a 3 (posture: Sit, Swivel, Stand) x 4 (viewing angle: 45°, 90°, 135°, 180°) within-participants design. The model screen was placed at these angles relative to the workspace, so that wider angles required larger and more effortful gaze shifts. We recorded gaze with mobile eye tracking and whole-body movement with motion capture and measured how long participants inspected the model on each look (look duration) and how often they looked (number of looks). Posture modulated memory use at the wider viewing angles (135° and 180°), where orienting required effectors beyond the head and eyes. At 180°, look durations were longest in the sit condition, intermediate in the swivel condition, and shortest when standing. The kinematic analyses showed distinct orienting strategies: seated participants pushed upper-body effectors to their biomechanical limits, whereas standing participants distributed rotation across the whole kinematic chain. Memory reliance was also elevated in the swivel condition, which suggests that offloading rotation to an external implement carries its own mechanical and control costs. These results support two conclusions. Postural constraints alter the recruitment and coordination of visual effectors and change the motor effort cost of external sampling, and participants compensate by relying more on VWM. Shifts in memory reliance can therefore index underlying motor effort costs, which offers a way to compare physical load across tasks in naturalistic settings.