Participatory sustainability appraisal methods are increasingly being used in the science, environment and natural resource management domains. The quality of the learning process has importance to the creation of outcomes, including policies, which deal effectively with complexity and uncertainty. While generating empirical evidence of individual learning outcomes may be considered desirable in order to monitor progress or significant change, it is important to consider how this empirical data is generated and interpreted. For example, in questionnaire format, as presented in this research, it would be difficult to assess whether learning resulted directly from the selected cases or from broader individual experiences unrelated to the case studies. A longitudinal study, including ‘before’ and ‘after’ assessment may be able to provide additional insight. However, at a deeper level, it might also be argued that the questionnaire approach reflects the learning that is expected by the designers of the questionnaire. In so doing, it creates the possibility of missing critical learning outcomes defined by the ‘learner’ themselves. This draws attention to the fact that individual learning is deeply personal and multi-layered. It may be a trap to conceptualise learning as a largely cognitive process, creating the possibility of missing other equally important aspects of being and knowing. If the need for empirical testing of learning outcomes is accepted as valuable, the lesson for adaptive learning would be the need to strongly consider the means by which this is undertaken. For example, beyond individual learning, it may also be worth considering the value of empirical testing to identify the existence, structure and dynamics of learning nodes at which system-wide learning is exchanged across nested adaptive hierarchies.