In our previous article on the question whether heroin addicts are able to give informed consent voluntarily to research on heroin‐assisted treatment, we criticized the ongoing bioethical discussion of a flawed conceptualization of heroin addicts' options. As a participant in this discussion, Edmund Henden defends the conceptualization as sufficient for determining whether heroin addicts are able to give informed consent to the research on heroin‐assisted treatment voluntarily. This discussion on research on heroin‐assisted treatment seems to go astray in several respects. In his reply to our article Henden maintains some of the biases, such as the necessity of abstinence in recovery, that seem to prevail in addiction research on a more general level as well. These biases run the danger of having implausible ethical implications on stakeholders in addiction research and treatment. In our reply to him, we will further clarify and discuss the importance of describing the relevant issues in plausible terms that do justice to the realities of the cases of informed consent in research on heroin‐assisted treatment and also raise a wider issue of the ethics of wording as well as of the narrow scope, or ‘tunnel vision’, in addiction research as currently conducted.