1. The United States Code of Federal Regulations (henceforth, the Regulations) requires that research conducted with minors be minimal risk unless participation offers a prospect of direct medical benefit to the child. According to the Regulations, research qualifies as minimal risk when “…the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.” As such, the minimal risk standard is consistent with the possibility that a child experience minor pain or discomfort, for example the minor pain felt from venipuncture. However, whether a procedure poses only minimal risk to a child depends on facts about that particular child. For example, if a child is very afraid of needles, this constitutes a reason not to enroll her in a study that requires venipuncture, even if the study is generally considered to be minimal risk. See the Department of Health and Human Services: Protection of human subjects. Washington: US Government Printing Office; 1991: 45 CFR 46 Subparts A and D.

2. This is per the Regulations. See note 1, Department of Health and Human Services 1991.

3. Herein I am concerned with whether a choice is made autonomously and not with whether the agent making the choice is autonomous, full stop. Following Tom Beauchamp, I understand an autonomous action as an action that is, (1) intentional, (2) adequately informed (made with sufficient understanding), and voluntary (free of controlling influences). See Beauchamp, T L. Who deserves autonomy and whose autonomy deserves respect? In: Taylor, J, ed. Personal Autonomy: New Essays in Personal Autonomy and its Role in Contemporary Moral Philosophy. Cambridge: Cambridge University Press; 2005, at 310–29.CrossRefGoogle Scholar In the second section of the paper I discuss what an individual must understand to make an autonomous choice about enrolling, or staying enrolled, in research.

4. I distinguish these classes of minors based on capacity, not age. I will not address the question of how investigators should determine what class a particular child belongs in.

5. See note 1, Department of Health and Human Services 1991.

6. For example, see Wendler, D, Shah, S. Should children decide whether they are enrolled in nonbeneficial research? American Journal of Bioethics 2003;3(4):1–7, andCrossRefGoogle ScholarPubMed Baines, P. Assent for children’s participation in research is incoherent and wrong. Archives of Diseases in Childhood 2011;96:960–2.CrossRefGoogle ScholarPubMed

7. See note 6, Wendler and Shah 2003, at 1.

8. This view is widely accepted in the philosophical literature on action theory. The locus classicus here is Intention, by G.E.M Anscombe. Anscombe GEM. Intention. Cambridge, MA: Harvard University Press; 1957.

9. See note 6, Wendler, Shah 2003, at 2.

10. Wendler and Shah explicitly qualify that they are not requiring that the child be altruistically motivated, only that she possess the concept of altruism. See Wendler, D, Shah, S. A response to commentators on “Should children decide whether they are enrolled in nonbeneficial research?” American Journal of Bioethics 2003;3(4):37–8, at W37.CrossRefGoogle Scholar

11. Critics of Wendler and Shah’s suggestion that possessing the concept of altruism is necessary to make an autonomous choice about participation in nonbeneficial research include: Baylis, F, Downie, J. The limits of altruism and arbitrary age limits. American Journal of Bioethics 2003;3(4):19–21;CrossRefGoogle ScholarPubMed Diekema, DS. Taking children seriously: What’s so important about assent? American Journal of Bioethics 2003;3(4):25–6;CrossRefGoogle ScholarPubMed Fisher, C. A goodness-of-fit ethic for child assent to nonbeneficial research. American Journal of Bioethics 2003;3(4):27–8;CrossRefGoogle ScholarPubMed Halila, R, Lotjonen, S. Why shouldn’t children decide whether they are enrolled in nonbeneficial medical research. American Journal of Bioethics 2003;3(4):35–6; andCrossRefGoogle ScholarPubMed Robinson, WM. What’s altruism got to do with it? American Journal of Bioethics 2003; 3(4):23–4.CrossRefGoogle Scholar

12. This is clear from how the assent requirement is discussed in the National Commission Report. The report suggests that assent should be obtained from “any child capable of some degree of understanding” and it is suggested that minors as young as seven-years-old can give assent. The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Report and Recommendations: Research Involving Children 1978, at 129.

13. Wendler and Shah also make this point. See note 6, Wendler, Shah 2003, at 5.

14. Wendler and Shah also argue that allowing intermediate minors to choose may well do more harm than good. See note 6, Wendler, Shah 2003, at 3.

15. See Dworkin, G. Is more choice better than less? Midwest Studies in Philosophy 1982;7(1):47–56.CrossRefGoogle Scholar

16. For a classic defense of this view, see Mill, JS. On Liberty. Cambridge: Cambridge University Press; 2011.CrossRefGoogle Scholar

17. This addresses the concern that by denying children a choice about participation on the basis that they cannot make an autonomous choice, we are holding them to higher standards than we sometimes hold adults. This concern is raised by Botkin JR. Preventing exploitation in pediatric research. American Journal of Bioethics 2003;3(4):31–2, at 31.

18. Thanks to David Wendler for pointing to this additional concern.

19. For example, Bartholome WG. Informed consent, parental permission, and assent in pediatric practice. Pediatrics 1995;96:981–2; Diekema (see note 11, Diekema 2003); King, NMP, Cross, AW. Children as decision makers: Guidelines for pediatrics. Journal of Pediatrics 1989;115(10:10–6).CrossRefGoogle Scholar

20. See note 6, Wendler, Shah 2003, at 3, and Wendler, D. Assent in paediatric research: Theoretical and practical considerations. Journal of Medical Ethics 2006;32(4):229–34.CrossRefGoogle ScholarPubMed

21. Some examples include Bartholome (See note 18, Bartholome 1995); Baylis and Downie (See note 10, Baylis, Downie 2003, at 20); Diekema (See note 10, Diekema 2003); and King and Cross (See note 19, King, Cross 1989).

22. For a defense of this idea, see Millum, J. The Moral Foundations of Parenthood. Oxford: Oxford University Press; 2017, at chapter 6.Google Scholar

23. For a defense of the idea that participation in nonbeneficial research is an imperfect duty, see Shapshay S, Pimple K. Participation in biomedical research is an imperfect moral duty: A response to Harris, John. Journal of Medical Ethics 2007;(33)414–7; and Millum (see note 22, Millum 2017, at chapter 6).Google Scholar

24. As Wendler and Shah point out, there are many other opportunities for children to exercise agency by making choices. See note 6, Wendler, Shah 2003, at 5–6.

25. Those who have argued in this vein include Diekema (see note 11, Diekema 2003); McGee, E. Altruism, children, and nonbeneficial research. American Journal of Bioethics 2003;3(4):21–2;CrossRefGoogle ScholarPubMed Nelson, RM, Reynolds, WW. We should reject passive resignation in favor of requiring the assent of younger children for participation in nonbeneficial research. American Journal of Bioethics 2003;3(4):11–3; andCrossRefGoogle ScholarPubMed Sibley, A, Pollard, AJ, Fitzpatrick, R, Sheehan, M. Developing a new justification for assent. BMC Medical Ethics 2016;17(1):1–9.CrossRefGoogle ScholarPubMed

26. Wendler argues that participation in nonbeneficial research is likely to (indirectly) benefit children. See Wendler, D. A new justification for pediatric research without the potential for clinical benefit. American Journal of Bioethics 2012;12(1):23–31.CrossRefGoogle ScholarPubMed

27. See Nuffield Council on Bioethics. Children and Clinical Research: Ethical Issues 2015.

28. See note 27, Nuffield Council on Bioethics 2015, at recommendation 14. See also note 25, Sibley, et al. 2016, for a defense of Nuffield’s proposal concerning intermediate minors and assent.

29. For example, Diekema writes, “At its best, assent is an interactive, instructional process that includes the investigator, the parents, and the child with the goal of assuring the child has at least a simple understanding of the study purpose, the procedures, that will directly involve the child, and the possible harms and benefits of participation.” See note 11, Diekema 2003, at 26. Numerous commentators have suggested that the parental consent process should not be entirely separate from the assent process. For example, see Joffe, S. Rethink “affirmative agreement,” but abandon “assent”. American Journal of Bioethics;2003.3(4):9–11;CrossRefGoogle ScholarPubMed Nelson and Reynolds (see note 25, Nelson, Reynolds 2003); and Sibley, Pollard, Fitzpatrick, and Sheehan (see note 25, Sibley et al. 2016).

30. Sibley, Pollard, Fitzpatrick, and Sheehan,. explicitly note this and use “assent*” to refer to the assent requirement as proposed by Nuffield. In their paper they defend Nuffield’s “assent*” requirement. They explicitly reject the idea that assent should be understood as “obtaining affirmative agreement.” See note 25, Sibley et al. 2016.

31. Notably, Joffe emphasizes that the reason a child does not want to participate should bear on whether she gets her way. See note 29, Joffe 2003, at 10.

32. See note 27, Nuffield Council on Bioethics 2015, at 155.

33. Again, Sibley, Pollard, Fitzpatrick, and Sheehan are explicit about this. See notes 25 and 30, Sibley et al. 2016.

34. In fact, Wendler and Shah explicitly state that some children who are not capable of making an autonomous choice should still be given age-appropriate information and explanation about the study (perhaps written as well as verbal, when appropriate), that should be allowed to make small decisions that arise during their participation, if appropriate (e.g., which elevator to take, what to wear, which task to do first, etc.), and that, in general, researchers should work to make the child’s experience as beneficial and enjoyable as possible. See note 6, Wendler, Shah 2003, at 5–6.

35. Dissent is sometimes understood as an objection to what one is presently experiencing only. However, it seems that it can also be anticipatory without blurring the line between assent and dissent. This should be clear in what follows. Thanks to Jane Johnson for drawing my attention to this point.

36. For discussions of whether or why investigators should respect the dissent of animals used in research, see Fenton A. Can a chimp say “no”? Reenvisioning chimpanzee dissent in harmful research. Cambridge Quarterly of Healthcare Ethics 2014;23(2):130–9; Kantin, H, Wendler, D. Is there a role for assent or dissent in animal research? Cambridge Quarterly Healthcare Ethics 2015;24(4):459–72;CrossRefGoogle ScholarPubMed and Johnson J. Animal dissent in research: What then must we do? (unpublished manuscript).

37. I assume that it is permissible to expose minors to minimal harm or distress in the context of participation in nonbeneficial research, consistent with the minimal risk standard set by the Regulations. See note 1, Department of Health and Human Services 1991, at 45 CFR 46, Subpart A.

38. Joffe also makes this point. See note 29, Joffe 2003, at 10.

39. I understand dissent in a way such that not every indication of discontent constitutes genuine dissent. Genuine dissent is expressed in response to what the individual is experiencing, or anticipates she might experience, due to her participation, or possible participation, in research. The idea is, roughly, that she would not experience the discontent if she were not participating. For example, some very young children (e.g., a one-year-old) cry all the time. If such a child is crying during a research procedure, this might not constitute genuine dissent to the research procedure.

40. Fisher explicitly raises this concern in response to Wendler and Shah. See note 11, Fisher 2003, at 27.

41. In a brief response to Baines (see note 6, Baines 2011) Wilkinson suggests that assent is perhaps usefully understood as non-dissent. See Wilkinson, D. Dissent about assent in paediatric research. The Journal of Medical Ethics 2012;(38)1:2.CrossRefGoogle ScholarPubMed

42. Notice that if a child’s parents can override her choice, then it is unclear that she has been given a choice in any meaningful sense. Further, telling the child she has a choice and then failing to honor it, or even to express reluctance to honor it, would be disempowering and undermine trust. Paul Baines makes a similar point. See note 6, Baines 2011, at 961.