Cognitive limitations, choice, and children

To consider the merits of giving intermediate minors a choice about research participation, we need to know exactly what the child would be making a choice about. The obvious answer, and the one given by the Regulations, is ‘participation in research.’ But a little thought reveals that this does not make sense when it comes to intermediate minors. Two general points about making a choice are relevant here. First, when one agrees to participate in an activity, she agrees to participate in that activity under some description. For example, I can agree to participate in a recreational soccer game without agreeing to participate in a fundraiser for a political campaign, even if the soccer game is a fundraiser. Second, an individual cannot agree to something unless she possesses the concepts required to mentally represent that thing as an option.^{Footnote 8} For example, a typical five-year-old cannot agree to participate in a campaign fundraiser at least because she lacks an adequate grasp of the concepts candidate, campaign, and political office. While she might be able to agree to participate in the event under a different description—e.g., playing in a soccer game—this would not amount to agreeing to participate in the fundraiser.

As Wendler and Shah discuss, an individual cannot agree to participate in nonbeneficial research unless, at the very least, she has an adequate grasp of the concepts of research and risk. They point out that in order for a subject to be able to weigh the risks, she must be able to appreciate how those risks are relevant in the context of her life, values, and specific circumstances.^{Footnote 9} Wendler and Shah also argue that when it comes to deciding whether to participate in nonbeneficial research, an individual must possess the concept of altruism—acting for the good of others with no promise of personal gain.^{Footnote 10} While many commentators disagreed with Wendler and Shah on this point, I think that it is correct.^{Footnote 11} A child who lacks the concept of altruism cannot understand the social significance or value of participating. Thus, she cannot understand that or why there is a nontrivial reason in favor of her participating. This point is especially pertinent in the context of nonbeneficial research. I say more about this below.

The problem with asking intermediate minors to make a choice about participation in nonbeneficial research is that most of them will lack a sufficient grasp of the concepts required to mentally represent this proposition, and so will be unable to genuinely entertain it as an option. Those on the older end of the spectrum may be capable of grasping the requisite concepts to the degree required to at least entertain participation as an option and so make a choice about it. But this choice will not be meaningful in virtue of being ill-informed.

In sum, many intermediate minors literally cannot choose to participate in research because they lack the concepts required to entertain the relevant propositions, and those who can will be incapable of making an autonomous choice, at least in virtue of its not being sufficiently informed.

However, a proponent of ‘they should choose’ might argue that allowing intermediate minors to choose is valuable, even if that choice is not autonomous. As many of Wendler and Shah’s critics pointed out, the idea behind the assent requirement was never to set the bar this high.^{Footnote 12} Rather, the idea is that there is value in allowing a child to decide whether to participate in research as she understands what it would involve. So just as a five-year-old can agree to participate in a soccer game, even if not a campaign fundraiser, an intermediate minor might agree to ‘staying at the clinic for a few hours, having blood drawn, and answering some questions.’

However, the value of obtaining an intermediate minor’s agreement to participate in research as she understands what it would involve is dubious. If a child’s understanding is sufficiently impoverished or inaccurate, then her decision might not reflect her values, desires, or interests. Consider an analogy. Normally I would not choose to wake up at six a.m. to go to my local coffee shop and get fresh coffee. But suppose I find out that the shop is giving out free coffee this morning, so I decide to wake up early and get some free coffee. I am, however, missing a crucial piece of information: To qualify for the giveaway, I must commit to buying one pound of coffee from them per month for the next six months. If I knew about that qualification, I would not have made the choice I did.

Similarly, suppose an intermediate minor is under the mistaken impression that her participation in a research study will have a particular benefit (e.g., she once saw a television program about a research study that looked fun, and she incorrectly assumes that the study she would participate in involves similar activities). In this case, the child is not misinformed. Rather, she misunderstood the information she received, perhaps in a similar way to how I mistakenly assumed that the free coffee came without strings attached. In that case, her agreement would be no better an indication of her values, desires, or interests than my decision is an indication of my preferences regarding the relative merits of free coffee and sleeping in.

The problem is exacerbated by the practical and epistemic difficulties involved in judging how well a child understands the information she is given. While the investigators can try to gauge the child’s understanding, it will not always be possible to tell exactly whether, how, or to what extent her conception of what participation involves is incomplete or inaccurate. Or even if an intermediate minor understands the activities required by participation, lack of experience might render her a poor predictor of how she will react to them. This concern is especially pertinent when it comes to nonbeneficial research with healthy intermediate minors, as it is likely that they will have minimal experience as subjects in a clinical or research setting.

Another difficulty is that children often want to please adults or show that they are mature enough to handle responsibilities. This might lead some intermediate minors to participate only to please their parents (making it unclear that they are empowered by having the choice). Related considerations might make a child who wishes to withdraw from a study reluctant to do so, even though she would have that option (she might think withdrawing would show than she couldn’t handle the responsibility of choosing after all).^{Footnote 13}

This last point shows that not only is the value of giving intermediate minors a choice dubious, but that it might be harmful. ^{Footnote 14} Another concern along these lines is that giving an intermediate minor a choice might itself be a source of anxiety. The child might well realize that she does not have a sufficient grasp of what participation involves, and so is choosing in the dark. Knowing that one is ill-prepared to make a choice can be stressful. More generally, it is a mistake to think that having a choice about something is always empowering, or otherwise beneficial. As Gerald Dworkin has pointed out, choices come with the burden of being (or at least feeling) responsible for the choice and its outcome.^{Footnote 15}

So, one problem for the proposal that intermediate minors should decide whether to participate in nonbeneficial research is that, due to their cognitive limitations and lack of experience, they are not in the right epistemic position to do so autonomously, and the value of making the choice non-autonomously is dubious. Of course, the fact that an individual is not well positioned to make an autonomous choice need not imply she should not be given that choice. A cognitively normal adult might choose to purchase a house without doing adequate research on the area, integrity of the building, or other relevant issues. There is no relevant information that she is cognitively unable to understand; rather, she simply did not do her homework. Even so, it is arguable that the value of autonomy justifies noninterference in this case.^{Footnote 16} But such considerations do not straightforwardly apply when the agents are intermediate minors and not adults, and the decision in question concerns matters that the intermediate minors do not adequately understand.^{Footnote 17}

A related concern is that when an intermediate minor non-autonomously chooses to participate in research, this might seem to provide justification for exposing the child to risks in the same way that a person’s providing valid consent does. But if the child’s choice is nonautonomous, this appearance is illusory.^{Footnote 18}

Considerations of beneficence

So far, the problems I have raised for the ‘they should choose’ answer to the Input Question center on considerations about minors’ cognitive limitations and lack of experience. However, one might appeal to considerations of beneficence in support of giving them a choice. In particular, some have argued that giving intermediate minors a choice is beneficial in virtue of teaching them to make decisions.^{Footnote 19}

But this argument is unconvincing. As Wendler and Shah point out, it is doubtful that giving intermediate minors a choice about something they do not well understand would help them learn to make good decisions.^{Footnote 20} After all, part of making a good decision is ensuring that one is sufficiently well informed. When one realizes that she is not sufficiently informed, the right thing to do may well be to defer to someone who is. So, it seems doubtful that giving intermediate minors a choice about whether to participate will improve their decision-making skills. In fact, to the extent that allowing intermediate minors to decide influences their development as decision-makers, it might hinder more than help.

Further, it is unclear that there is any reason to think that giving intermediate minors this particular choice would be beneficial to their development as decision-makers. Typically, one has ample opportunities throughout childhood to decide about a wide variety of matters, some of which are likely more momentous than whether to participate in a nonbeneficial research study. To make the beneficence argument work, it would need to be shown that there is something distinctive about the potential benefits of giving intermediate minors this particular choice. It is hard to see what this distinctive feature might be.

But even if we set these doubts aside, beneficence cannot justify allowing intermediate minors to decide unless the potential benefits outweigh the potential harms discussed above. It seems doubtful that this is the case.

Respect for developing autonomy and rational agency

A common argument in favor of giving intermediate minors a choice about participation appeals to respect for their developing autonomy or capacity.^{Footnote 21} However, it is unclear how this argument is supposed to go. On one interpretation, it collapses in the argument just discussed—that allowing intermediate minors to decide aids this development by teaching them to make good decisions. On another, the fact that they will be autonomous in the future provides a reason to treat them as if they are autonomous now. But this is clearly wrong. While there may be reason to treat an individual who was previously autonomous and has now lost capacity (e.g., a demented adult) in a way that respects her past autonomy, this does not seem apt when it comes to minors with developing capacities.

However, the appeal to developing autonomy might be getting at something else. While intermediate minors are not capable of making an autonomous choice about whether to participate in research, they are still rational agents. That is, they are individuals who are capable of self-directed, intentional action, and are rational in the sense that they are engaged in an active effort to understand the world and make sense of what is happening to and around them. Both rationality and agency are distinct from autonomy. While intermediate minors are not capable of making an autonomous choice about participation in research, they are rational agents, in the sense described.

With this in mind, there is a plausible line of reasoning from the claim that intermediate minors are rational agents to the conclusion that they should be allowed to decide whether to participate in nonbeneficial research. The argument might be put as follows:

While there is a lot right in this argument, it relies on the assumption that an intermediate minor’s medical wellbeing is the only consideration that would justify overriding her agency and choosing for her. I think this assumption is incorrect.

Since nonbeneficial research stands to benefit society in nontrivial ways, participation has moral significance. This means that there is a moral reason for intermediate minors to participate in nonbeneficial research. Due to this, not giving an intermediate minor this choice need not be a morally problematic, or unjustified, failure to respect her agency. This point, in conjunction with the arguments that follow, support Wendler and Shah’s claim that possessing the concept of altruism is relevant to whether a child should be given a choice about whether she participates in nonbeneficial research.

Consider an analogy. Suppose the father of an intermediate minor makes her come with him to help in a soup kitchen. He reasons that her participation in this charity work will further a good cause (i.e., feeding the hungry) and positively influence her moral development. Arguably, parents have an obligation to teach their children to become morally good people. It is also plausible, even if controversial, that children have moral and societal duties, even if fewer than adults.^{Footnote 22} Volunteering at the soup kitchen is one way, but of course not the only way, of fulfilling an imperfect moral duty to society.^{Footnote 23} Plausibly, considerations of this sort justify the father’s decision not to give his daughter a choice about whether she helps at the soup kitchen and so, does not constitute a morally problematic failure to respect her agency. It would be such a failure if he never (or only very rarely) allowed her to make choices about what to do. But occasionally failing to give her a choice about whether to participate in morally valuable activities seems unobjectionable.

Similar considerations apply to the case of intermediate minors and nonbeneficial research. Choosing for an intermediate minor is not a morally problematic failure to respect her agency if she is allowed to exercise her agency sufficiently often in other aspects of life, and there are significant moral reasons in favor of her participation.^{Footnote 24}

Arguably, underlying the argument from agency is the worry that depriving intermediate minors of the choice concerning their participation in nonbeneficial research studies is akin to treating them like lab rats. In other words, depriving them of a choice amounts to treating them as mere means to some end (albeit, a socially worthy one) rather than as ends in themselves. This worry underlies the objection, raised by many commentators, that failing to give intermediate minors a choice fails to respect them as persons, or as beings with moral value, or as nonobjects. ^{Footnote 25} The soup kitchen analogy shows that such worries are misplaced. Good (or at least morally permissible) parenting involves making some choices for one’s children. Doing so is sometimes justified by moral reasons that might not benefit the child directly. If these moral reasons are good enough, then choosing for the child is consistent with treating her respectfully.

One way to bring out the force of the above points is to consider the following (imagined) exchange between a twenty-year old and her parent:

• Parent: When you were seven, you almost enrolled as a healthy volunteer in a research study on childhood leukemia. You didn’t participate because when we told you about what participation would involve—being at the hospital for half a day, have blood drawn, and answering some questions from doctors—you decided that you preferred to play computer games.

• Child: Was I scared to participate? Or did I just want to play computer games?

• Parent: You didn’t seem scared or anxious. But we gave you a choice, and you preferred to go home and play computer games.

• Child: Wait, why did you let me decide? I didn’t understand what I was doing! I had ample time to play computer games in my childhood and you gave me the opportunity to make all sorts of other choices. Participating in that study would have been socially valuable and important. I didn’t understand that when I made the choice to play computer games.

If the twenty-year-old has a strong sense of social responsibility and an appreciation for the value of medical research, she might well see the choice of her younger self as a bad one and wish she had chosen differently—or wish that she had not been allowed to make this particular choice.^{Footnote 26} While not every child would react in this way, the point is that this reaction seems reasonable. The above discussion explains why it is reasonable.

In this section I have argued that ‘they should choose’ is the wrong answer to the Input Question, and that rejecting this answer is consistent with treating intermediate minors respectfully. Before defending an answer to the Input Question, I explain why the question matters, and must be addressed in discussions of the assent requirement.

What nonmaleficence requires when it comes to giving intermediate minors a say

Earlier, I argued that nonmaleficence does not justify giving intermediate minors a choice about whether to participate in nonbeneficial research. However, I think that it does justify giving them a say of some sort. While I do not claim that nonmaleficence is the only consideration that justifies giving intermediate minors a say, I argue that to the extent that there are additional such considerations, they do not require anything more in practice than what is already required by nonmaleficence. I return to this point after I present and defend my answer to the Input Question.

I claim that nonmaleficence requires giving intermediate minors a say for two distinct reasons. The first is that an individual’s perspective can often be a guide to whether or how she might be harmed by participation. I argue that this provides a reason not only for investigators to respect intermediate minors’ dissent, but also to obtain affirmation of their non-dissent to participation. The second is that failing to actively solicit affirmation of an individual’s non-dissent to participation can itself be harmful. This provides an additional reason for obtaining affirmation of intermediate minors’ non-dissent. In the following two subsections, I explain each of these ideas.

Respecting dissent

Many commentators have proposed that regulatory and guidance documents require that investigators respect the dissent of all minors involved in research. In the context of research participation, I understand ‘dissent’ as an expressed objection, either verbal or behavioral, to what an individual is experiencing, or anticipates that she will experience, in response to a research procedure.^{Footnote 35} Here I understand ‘objection’ and ‘object’ in an atypically broad way that departs from normal usage. For example, objections might take the form of refusing to stay still or crying out. Notably, this makes it possible for an individual to express dissent without intending to do so. It also makes it possible for an individual to dissent without understanding anything about the research procedure, or even that she is participating in research. This is important because it makes it possible to apply the dissent requirement to very young minors, individuals with severe cognitive impairments, and nonhuman animals.^{Footnote 36}

To implement the dissent requirement with intermediate minors, the child should be provided with age-appropriate information and explanation concerning the study and what participation would require of her. She should be engaged in discussion and given ample opportunity to express thoughts and ask questions. However, at no point should she be told or given the impression that it is ‘up to her’ whether she participates, or that she should participate only if she wants to. Rather, the investigators should frame the situation in a way that makes it clear that she is expected to participate unless participation is causing her to experience what I will henceforth refer to as ‘discontent.’ One experiences discontent when she experiences more-than-minimal pain, discomfort, anxiety, or distress of any sort.^{Footnote 37}

It is important that investigators do not put undue pressure on an intermediate minor to participate—it must be clear that no one wants her to participate if she is discontented. However, at the same time, the range of legitimate reasons for objecting must be clear. In particular, only reasons that are indicative of discontent are legitimate. For example, boredom, a desire to rebel, or the fact that research participation is not as much fun as playing computer games, are not legitimate reasons for objecting.^{Footnote 38} The balance is delicate, and investigators should be trained in talking with intermediate minors to ensure that they do not convey mixed messages with respect to whether the child is being given a choice about participating.

Putting a limit on the range of legitimate reasons for objection does not guarantee, of course, that the child will not object for reasons that she is not supposed to consider relevant. Indeed, a child might well realize that she can avoid participating by telling the investigator that she is scared when, in fact, she is bored, wanting to go home and play computer games, or just rebelling. Doing this will get her what she wants; and to this extent, she can choose to take action that will lead to her nonparticipation. But this does not mean that she was given a choice. Just as a child can malinger to avoid going to school or throw a noisy tantrum to force her parent to take her out of the library, a child can take action that will guarantee her nonparticipation in a research study. But this does not mean that she was given a choice any more than children are typically given a choice about going to school or staying in the library with their parents.

In response to any indication that an intermediate minor is dissenting, the investigator should pause (or delay beginning) the relevant procedure and check in with her. While an intermediate minor’s dissent will often take the form of a verbally articulated objection, investigators should also respond to any behavioral indication of discontent even absent a verbally articulated objection.

Importantly, respecting dissent does not necessarily require investigators to terminate an individual’s participation or decide that she should not be enrolled (in the case that dissent is expressed prior to enrollment) at the first indication of dissent. Rather, when an intermediate minor dissents, the investigator should begin by talking with the child to determine the source of her discontent and see whether it can be addressed. For example, perhaps reexplaining something will quell the child’s anxiety. Or, if it turns out that an intermediate minor is bothered by very bright lights in the exam room, then it might be possible to dim them (unless their brightness is essential to conducting the research). In some cases, the child might just need a break.^{Footnote 39} However, investigators should take care that they do not pressure an intermediate minor to continue if her discontent is not eased.

If it is not possible to make modifications that will appease an intermediate minor without compromising the research, participation should be terminated. In cases where the child dissents prior to enrollment, she should not be enrolled unless the source of her unwillingness can be adequately addressed.

In sum, respecting dissent requires that an investigator honor an intermediate minor’s sustained, or unresolvable, objection to participation that indicates more than minor discontent. With respect to the Input Question, this gives her a say to the extent that she can object to, or veto, participation. However, the fact that she cannot object for just any reason at all, and that the range of legitimate reasons for objecting is not circumscribed by the intermediate minor herself, distinguishes respecting an individual’s dissent from giving her a choice.

Obtaining affirmation of an individual’s non-dissent to participation

The dissent requirement is motivated by the fact that an individual’s perspective can be a guide to whether or how she might be harmed by participation in research. This fact implies that investigators should do what they can, within reason, to obtain adequate evidence that the individual is not discontented, and that beginning or continuing participation will not cause her to become discontented.

The fact that an intermediate minor does not volunteer an objection (or give a behavioral indication that she is discontented) will not always suffice to provide investigators with adequate evidence in this respect.^{Footnote 40} On the contrary, many intermediate minors, even those on the older end of the spectrum, may be hesitant to express their reluctance or raise a concern for fear of being ‘difficult,’ admitting weakness, or disappointing their parents or the investigators. For these reasons the dissent requirement, as described above, is insufficient to ensure that investigators have adequate evidence of an intermediate minor’s non-dissent. I understand ‘non-dissent’ as a state that is characterized by the absence of certain beliefs, thoughts, and qualitative experiences. Namely, those that would make dissent appropriate.^{Footnote 41} More precisely,

The motivation for (b) is that an intermediate minor who is mildly concerned about beginning or continuing participation is not discontented, but her concern might make dissent appropriate.

There are two distinct reasons that nonmaleficence requires that investigators obtain affirmation of an intermediate minor’s non-dissent. First is the reason already discussed: obtaining affirmation of non-dissent is sometimes necessary in order for an investigator to obtain adequate evidence that the research is not causing, nor will cause, discontent.

The second reason is that failing to actively solicit affirmation of an intermediate minor’s non-dissent can itself be harmful. If an investigator initiates or continues a procedure without soliciting affirmation of an intermediate minor’s non-dissent, this might well might cause anxiety, frustration, or fear. The child might feel as if she is being shuffled around without sufficient concern for her thoughts or feelings, or as if she lacks control over what is happening to her. Further, while the investigator will have told the intermediate minor that she should not hesitate to speak up if she has concerns or is feeling discontented, this might seem insincere if the investigator then plows forward with the procedures without checking in. In other words, the intermediate minor might interpret the investigator’s failure to solicit affirmation of her non-dissent as suggesting that the investigator does not really care about what she is thinking or feeling, and just wants to get on with things.

How to solicit affirmation of an intermediate minor’s non-dissent without giving her a choice

Earlier, I explained that one important difference between allowing intermediate minors to object to or veto participation on one hand, and giving them a choice about participating on the other, is that in the former case the child cannot object for just any reason. Rather, there is a limited range of legitimate reasons for objection, and the child has no say in determining what those reasons are. A second important difference is that in obtaining affirmation of non-dissent, the intermediate minor is being asked to confirm that she is neither discontented nor concerned that participation will cause her to become discontented. She is not being asked to positively endorse proceeding with participation.

As with implementing the dissent requirement, soliciting affirmation of non-dissent requires care. In soliciting affirmation of non-dissent, the investigator must ensure that she does not make it sound as if the intermediate minor is being offered a choice. For example, an investigator should not say, prior to enrollment, “Would you like to proceed?” or, “Does this sound like something you want to do?” Rather, she should say things like, “How are you feeling? Are you worried about anything we’ve talked about? Does anything sound scary? If not, are you ready to start?” After participation is underway, the investigator should regularly check in by saying things like, “Has everything felt okay so far? Are you ready to continue?” The intent is to firmly, but gently, make it clear that the default is that she participate, but that the investigators really want to know if she has concerns or is discontented.

In sum, soliciting affirmation of an intermediate minor’s non-dissent to participation differs substantively, and in a way that is ethically and practically significant, from giving intermediate minors a choice about participation in nonbeneficial research. In soliciting affirmation of non-dissent, the investigator is basically seeking confirmation that she is okay. In giving the child a choice, the investigator is telling the child that it is up to her whether she participates (assuming that parental consent has been provided); the child can decline to participate for any reason and her choice will be honored.^{Footnote 42}

Back to nonmaleficence as the basis for giving intermediate minors a say

One might suggest that nonmaleficence alone cannot ground the requirement to obtain affirmation of non-dissent, or that there are additional all-things-considered reasons that require giving intermediate minors a say. With respect to the former, I think this concern derives from an overly narrow understanding of nonmaleficence—as though the nonmaleficence requires only that investigators avoid causing participants physical pain. In fact, nonmaleficence also requires that investigator avoid causing psychological harm. This includes trying to avoid causing the sort of emotional distress that can result if a child believes that her perspective is not valued, or that she lacks control over the pace at which the investigator is proceeding. Giving the child more of a say can help researchers achieve that goal, along with the goal of discovering other ways that participating might harm the child.

My response to the latter point is related: Nonmaleficence, in fact, requires quite a lot. While it is plausible that considerations such as beneficence, respect for dignity (understood as respect for personal and bodily integrity), and respect for rational agency may provide additional all-things-considered reasons for soliciting or obtaining affirmation of intermediate minors’ non-dissent in the way I describe, it is hard to see how they could require something more without requiring that intermediate minors be given a choice about whether they participate. I have already argued that these considerations do not justify giving intermediate minors a choice.