Empirical literature on digital technologies for student learning is generally unable to identify separately whether learning gains arise from reciprocity in response to the gift of a valuable gadget (the 'gadget effect') or from increasing exposure to relevant materials (the 'content effect'). This paper attempts to disentangle these mechanisms using a randomized control trial in junior secondary schools in Lagos, Nigeria. It estimates three contrasts: (i) the effect of just receiving an eReader with non-curriculum content, (ii) the marginal effects of receiving an eReader with curriculum text books, and (iii) the marginal effects (relative to ii) of receiving curriculum with supplementary current and remedial instructional content. The findings show that six to eight months of exposure to eReaders led to modest positive impacts on learning, but only if the devices had curriculum material and were filling input gaps resulting from a lack of textbooks. Consistent with other recent findings, even six to eight months of exposure to eReaders with non-curriculum recreational material reduced student learning outcomes. These results demonstrate that the promise of digital solutions to improve learning depends largely on the extent that these solutions address unmet access to instructional material. The paper also finds that exposure to eReaders improved student retention. However, these impacts are not very robust and could be achieved much more cost-effectively through the provision of information about the economic returns to education.