Abstract
The two main issues in this case are the level of risk to which the child will be exposed and the deception involved. Research that is not expected to benefit the child participants should entail no more than minimal risk to those participants. Research involving deception requires justification even with adult subjects: deception must be methodologically necessary. In this case the issues are linked – any harm will be caused by the child’s lack of awareness of the true nature of the situation in which they have been placed. So our questions are: must the child be kept unaware of that situation; is the risk of psychological harm minimal; and, if not, can it be ethically justified?
There is much debate as to what minimal risk means. European legislation defines minimal risk as the ‘probability of harm or discomfort not greater than that ordinarily encountered in daily life or during the performance of routine physical or psychological examination’ (Sammons, 2009). So is the anxiety or embarrassment that a child might experience on discovering that they have ‘lost’ a kitten equivalent to a harm or discomfort ordinarily encountered in daily life? It is unlikely that a child will, in the ordinary course of events, lose (as in misplace) a kitten, puppy or some other pet. The situation in which the research will place the child is artificial and unusual. Nevertheless, the researchers are likely to argue that what matters is whether any harm or discomfort that the child might experience in the research setting is greater than that encountered in everyday life. They state that any anxiety or embarrassment would last no more than a few minutes. It is plausible that children do experience short-lived anxiety or embarrassment in everyday life, from a variety of different causes.
However, I find the claim that any discomfort caused by the ‘loss’ of the kitten would be of this short-lived kind to be implausible. As an adult I think I would feel more than such momentary anxiety if I lost a kitten that had been placed in my care. If it was my own kitten I would feel quite upset, possibly at my carelessness to protect the well-being of the vulnerable animal. If it was someone else’s kitten I would, in addition, feel guilt about the distress I would cause the owner. Even if I was forgiven, I anticipate that I would still experience self-blame and regret. These are not momentary feelings of a few minutes’ duration. I would hope that as an adult I would, nevertheless, be able to cope with such feelings and they would not cause any lasting psychological damage. However, if this is how I would react – and I do not believe I am unusual in this respect – how would an 8 year old child react?
Research guidelines stress the need to consider age-specific risk. For example, the RCPCH (2000) guidelines state that: ‘childhood is a vulnerable, formative time, when harms can have serious impact as well as be potentially long lasting’. Similarly, Williams (2012) says of social research that involves asking children about their experiences and feeling: ‘It is easy, as adults, to forget how intensely worrying [these things] were to us as children’. So we need to be mindful that a given degree or type of harm may affect children more than it affects an adult – they need special protection.
My view is that a child will experience more than momentary or short-lived self-reproach, fear of adult reaction, anxiety or discomfort. It is easy as adults to forget how much it matters to children that they live up to adult expectations and how difficult it can be to rationalize (or get into perspective) their perceived failings. Furthermore, a sensitive child, one who lacks confidence, is anxious or who has strict parents is likely to have a more adverse reaction. This is, therefore, not minimal-risk research.
However, the risks could and should be reduced by debriefing the child. According to the British Psychological Society’s Code of Human Research Ethics: ‘Where an essential element of the research design would be compromised by full disclosure to participants … explicit procedures should be stated to obviate any potential harm arising from such withholding’ and there should be an ‘appropriate risk management and harm alleviation strategy’ (BPS, 2010).
It is not sufficient to tell the child that the kitten will probably turn up and that they should not worry. The child needs to know that the kitten is fine and that they are not at fault. This will not undermine the purpose of the research or the nature of the data collected because it occurs after the child’s reactions have been observed. But this raises a further problem – should the child be told that they have been deceived or tricked?
Telling the child that they had been tricked may undermine their trust in adults, even if it is explained that this was done to see how they would react. It may also have effects on the child’s view of the acceptability of deception. A child who has been told that lying to or tricking others is wrong may become confused or pick up a mixed message of the ‘don’t do as I do, do as I tell you’ variety. This would be harmful psychologically. However, it may be possible to provide reassurance without risking these harmful consequences. The child could be told that there was a hole in the wall through which the kitten had escaped but that it had now been found, possibly even having someone come in with the kitten to reassure the child that the animal was safe. But this, of course, still involves deception, albeit benign.
Your view on whether this is ethically acceptable will depend on your view of the morality of deception. The justification for this benign deception will be consequentialist: the research has scientific merit, is likely to benefit other (future) children, it cannot be conducted without deception, and the ‘reassurance’ strategy, although deceptive, minimizes expected harm. The deception is not, therefore, wrong. This is unlikely to satisfy a non-consequentialist, who will object that people, including children, should not be deceived for the benefit of others, which is essentially what would happen with the ‘reassurance’ strategy. The issue from this perspective is not whether the child will be harmed – they would not be – but whether the child will be wronged.
Research ethics guidelines are a combination of consequentialist and non-consequentialist considerations, as is our ‘ordinary’ moral reasoning. Few of us are ‘pure’ consequentialists or non-consequentialists, which makes this a difficult dilemma to solve. My view is that the decisions of research ethics committees should not turn on which moral theory their members advocate. If a research proposal conforms to ethical guidelines but there is reasonable disagreement as to whether it is unethical on other grounds and that disagreement can be traced to theoretical differences, the research should be approved. In this case we might reasonably disagree about what makes deception wrong, but as BPS guidelines permit deceptive research when this is methodologically justified and harm alleviation strategies are in place, the proposed ‘reassurance’ strategy makes the research ethically justifiable.
We may remain uneasy about deceiving young children, or think that parents should not consent for their children to be involved in such research, but these are separate issues to the question of whether research with children that involves deception is justifiable according to existing guidelines. If we object to deceptive research, then we should work to change guidelines. Some valuable research may, of necessity, require deception, however.