The paper from Hickman et al. [1] on the need for a translational approach in the addictions field must be seen as timely. Research agencies all over the world are trying to overcome difficulties in bringing scientific discoveries from the bench to the bedside in times where the pace of scientific discovery in the life sciences is accelerated, and the need for rapid translation into practice has increased. The NIH Roadmap for medical research [2] is probably a good example of the relevance given to translational research today. In the addictions field, difficulties in replicating reality in the laboratory are probably far more complex than in other domains, and hence the need for a translational approach is even greater. The example chosen by the authors (alcohol and the risk of opiate overdose death) illustrates the complexities faced by addiction researchers, as well as the potential for improvement through a translational approach. The same would apply to similar arenas (i.e. the interactions between alcohol and tobacco), with an enormous potential public health impact. One could anticipate that the more difficult it is to reproduce a disease in the laboratory, the more helpful a translational approach would be. There is no doubt that it is difficult to replicate human addictive behaviours in laboratory rats, but it is also true that animal models of addiction have proved useful in some areas. For example, the development of pharmacological treatments highlights why a translational approach is needed. In the alcohol field, for example, animal models still have relevant limitations concerning anticraving drugs, and the results obtained at the bench are quite often far from what happens at the bedside. This is no surprise for clinicians, who know that the bed is not the preferred place for alcoholics who live and interact in a world far more complex than a Skinner box; but those clinicians have little contact with researchers who develop animal models. This is part of the large gap which needs to be bridged. I agree completely with the authors that there are a number of hypotheses in the addictions field that rest ‘on an evidence base which is not as strong as generally assumed’ and would therefore benefit from a translational approach. I would add that some of them are quite crucial: one interesting example was published in a recent issue of Addiction where the alcohol dependence syndrome was reviewed [3]. More than 30 years after it was first described [4], the alcohol dependence syndrome is still an evolving concept which benefits from a translational approach: animal models have tested criteria such as tolerance [5] and epidemiological studies such as the National Epidemiological Survey on Alcohol and Related Conditions (NESARC) [6] have been used to challenge the bi-axial concept of the alcohol dependence syndrome and complex statistical analyses have helped to check the validity of the construct [7, 8]. The gap between basic and applied research in the addiction field lies not only in the difficulties of replicating the complexity of human addictive behaviour in experimental settings, but also in the difficulties in measuring the basic addictive phenomena properly. Even though progress has been made, we are far from being able to measure accurately and reliably core elements such as neuroadaptation or craving. These reflections should not be taken as reluctance to adopt a translational approach, but as proof of the tremendous efforts needed to push research in this direction. Instead of being disappointed we should focus upon the progress which has already been made, and in the windows of opportunity that new technologies are opening. Neuroimaging techniques are one of the areas where translational research appears to be more promising [9], and it is already contributing to a deeper understanding of the neurobiological pathways of reward, craving and withdrawal. Translational research is essential to promote the validation of knowledge acquired through basic research, but it is not enough to ensure implementation of new findings. Knowledge transfer and dissemination, a final step to improve population health, is also a large challenge in the addictions field. Because population health is our final aim, we should keep an eye on the difficulties we face in the transfer and dissemination of knowledge, while we insist upon the need for more translational research.