Haptic exploration, the process by which humans gather sensory information through physical contact with objects, is inherently active and adaptive. We hypothesized that humans optimize the duration of their explorations, and that this optimization is shaped by an interplay of valuation and predictive processes that also take aversive consequences into account, e.g. motor effort. Motor control literature demonstrates that energetic costs are central determinants of human behavior (Shadmer et al., 2016). To study this in the context of haptic exploration, we assessed exploration duration and performance in a two-alternative forced-choice spatial frequency discrimination task under varying conditions of task demands and motor effort. Task demands were manipulated by varying the discriminability of the virtual grating stimuli. Motor effort was manipulated by implementing forces counteracting the participants’ movements while switching between stimuli. Participants were instructed to switch between stimuli after each single swipe movement. We observed that higher task demands lead to higher numbers of movements (= longer exploration duration), likely reflecting a compensatory mechanism that allows to attain a certain level of task performance. This effect though is reduced when motor effort is increased: while low and medium task demands yield similar numbers of movements regardless of the related motor effort, higher demands do not lead to more movements when motor effort is very high. We conclude that the extent to which increased task demands are compensated via the extension and exploitation of an exploration depends on the costs that the agent is being confronted with.