Table 2 Key messages from conceptualising involvement as a conversation that supports learning

 • Involvement in research is essentially a conversation where researchers and the public share their knowledge, values and opinions to learn from each other

 • Researchers’ learning from others’ experiential knowledge is subjective and unpredictable, making measurement of this outcome difficult

 • The conversation needs to be ongoing, formal and informal, to avoid bias in researchers’ thinking in every decision they make about their research

 • The approach doesn’t need to be standardised – it doesn’t always mean setting up a group in the same way every time

 • The quality of the interaction between researchers and the public may be more important to support learning than the precise process

 • Objective reporting of the outcomes and impact of involvement do not provide the full picture because researchers’ learning is missed out

 • New approaches to evaluation and reporting impact of involvement may be required to include researchers’ personal accounts of their experience and the wider impacts on the research agenda and culture