Linkenauger, Witt, and Proffitt (Journal of Experimental Psychology: Human Perception and Performance, 37(5), 1432–1441, 2011, Experiment 2) reported that right-handers estimated objects as smaller if they intended to grasp them in their right rather than their left hand. Based on the action-specific account, they argued that this scaling effect occurred because participants believed their right hand could grasp larger objects. However, Collier and Lawson (Journal of Experimental Psychology: Human Perception and Performance, 43(4), 749–769, 2017) failed to replicate this effect. Here, we investigated whether this discrepancy in results arose from demand characteristics. We investigated two forms of demand characteristics: altering responses following conscious hypothesis guessing (Experiments 1 and 2), and subtle influences of the experimental context (Experiment 3). We found no scaling effects when participants were given instructions which implied the expected outcome of the experiment (Experiment 1), but they were obtained when we used unrealistically explicit instructions which gave the exact prediction made by the action-specific account (Experiment 2). Scaling effects were also found using a context in which grasping capacity could seem relevant for size estimation (by asking participants about the perceived graspability of an object immediately before asking about its size on every trial, as was done in Linkenauger et al., 2011; Experiment 2). These results suggest that demand characteristics due to context effects could explain the scaling effects reported in Experiment 2 of Linkenauger et al. (2011), rather than either hypothesis guessing, or, as proposed by the action-specific account, a change in the perceived size of objects. © 2017, The Author(s).