Ethics of animal research

monkey-3512996_1280A few months ago, we posted about how we could use animals for research.  Today we are looking at the ethics surrounding the decision to conduct research using nonhuman animals. Most students can reel off the ethics involved in conducting research on humans (informed consent, lack of harm, right to withdraw, privacy, etc.) but when we talk about the ethics of conducting research using nonhuman animals as proxies for humans, they are less clear. Often the argument gets stuck at the level of “It’s OK for medical research, but not for cosmetics.”  This is not good enough for an understanding of the complexities (nor for an exam answer).  For students that wish to argue that conducting research on nonhuman animals in order to avoid causing pain or distress to humans can never be ethical, point out that this is a worthy philosophical question, and could even be a counter-argument in a psychology debate on the topic, but again, it cannot constitute the main argument of an exam essay on ethics.

The APA,  BPS and Australian government publish guidelines for conducting nonhuman animal research ethically. What emerges from the guidelines are the ‘3 Rs’ of animal research:

  • Replace animals with other alternatives – such as computer simulations, use of lesser species (such as single‐cell amoebae and nematode worms),  use micro-dosing, CRISPR DNA editing, or human cell cultures – known more colloquially as ‘patient in a dish’ or ‘body on a chip’.  But animals are used to generate new hypotheses, so CRISPR editing was tried out on animals first, as was stem cell research. 
  • Reduce the number of individual animals used, by using data from other researchers, or by repeated micros-sampling on one animal in a repeated measures design.
  • Refine procedures to minimise suffering, by using appropriate anaesthetics and painkillers, and training animals to cooperate with procedures to minimise any distress. Imaginative research, where faecal matter is analysed to investigate stress levels, rather than drawing blood from an obviously stressed animal, has a part to play here.

In Psychology Sorted Book 1, we provide summaries of studies by Xu et al. (2015) and Stanton et al. (2015) which show how nonhuman animals may be used more ethically, to contrast with others such as Barr et al. (2004) and Weaver et al. (2004) which cause more stress to the animals used. These will help to keep your students more closely focused on the complexities of whether and how we should use nonhuman animals in psychological research.

We’re all just animals, aren’t we?

ape-1000270_640

The question of when to use non-human animal studies as evidence for human behaviour is a tricky one. Because it remains unethical to lesion the brain of a live human to look for a correlation between brain damage and behaviour (at the moment!), animal studies are used in large numbers at the biological approach. However, many people are becoming more disturbed by this than previously as over the years we have come to realise that animals also suffer pain, fear and anxiety as we do, and maybe other ways should be sought to conduct animal studies.

In Psychology Sorted, this is part of the Biological extension: the British Psychological guidelines for working with animals (2012) state that researchers should: Replace animals with other alternatives. Reduce the number of animals used. Refine procedures to minimise suffering.  But isn’t how they are used a large part of the problem?  After all, observation under natural conditions should be no problem.  Xu et al. (2015) researched naturally-occurring depression in macaque monkeys by observing monkeys living at a research base in China in environmental conditions that closely resembled what they would experience in the wild, for nearly 3 years.  The monkeys were housed in colonies, usually of two males and 16-22 females, with offspring of under six months.

Instead of unnaturally separating baby chimpanzees from their mothers, as Bowlby and others have done, causing distress,  Stanton et al. (2015) ‘picked up poo’: they investigated the effect of maternal stress on the glucocorticoid levels of infant chimpanzees by examining and measuring faecal glucocorticoid metabolite (FGM) concentrations of mothers and babies in the wild.  Much less stress for the monkeys, though maybe not for the researchers! 

Bearing in mind that we are animals too, it is time empathy stretched to our non-human cousins, and these methods seem to be a first step on the way.  See Psychology Sorted for more examples of ethical animal research.

 

The new IA process for IB Diploma – get started!

road-1668916_640

Let’s get started! This is a useful summary for teachers and students of the process for the new IA (internally-assessed student-conducted experiment…now you see why the name is shortened 🙂  This will first be assessed in May 2019, and I’m sure some of you are getting started soon.

Group work is mandatory.  Up to 4 students in a group, and preferably each group conducting a different experiment, so you don’t run out of participants.  The experiment is run together by the group to collect the raw data, but every section, and all data calculations, have to be performed and written about individually.  

Statistical Analysis must be conducted by everyone.  Descriptive statistics identify if there is a difference between the two conditions and inferential statistical analysis tells you whether or not this difference is significant at the p<0.05 level.  Unless you are an expert statistician, it is easier to just manipulate the independent variable once to give two conditions under which you measure the dependent variable. Plan how you are going to do this, and which tests you need to use before even starting your experiment.

Ethical Considerations – be sure that your experiment will cause no harm or stress to the participants, who may not be animals or young children.  Conformity experiments are not allowed, because they are stressful, and you may not ask your participants to eat or drink anything in order to test the effects. Neither may you deprive them of sleep.  Your appendices at the very end of your report should contain a blank copy of the informed consent form, a copy of your briefing and debriefing notes, raw data tables and your calculations for the analysis.

IA Report – This needs a header containing the following information:  title; your IB candidate code and the codes of all group members; date, month and year of submission; no. of words.

__________________________

The Report should be between 1800 and 2200 words and split into the following 4 sections:

Introduction  (6 marks) –  Contains the aim of the experiment, and explains the link between the experiment and the model or theory on which it is based.  (Most likely your experiment will be based on another study or experiment, but you need to know the underlying theory and show the link).  The hypotheses should be written out carefully, and contain the operationalised independent and dependent variable.  It is probably easier to write these separately first and then combine them to make the hypothesis.

Exploration (4 marks) – This is where you describe your procedure, including the design, sampling technique, participant characteristics, controlled variables and materials.  Write it very carefully, as you will want to refer back to it later in your last (Evaluation) section.

Analysis (6 marks) – Consists of correctly chosen and applied descriptive and inferential statistics.  The descriptive statistical analysis results should be shown in a bar chart (graph) that is carefully labelled.  The inferential statistics results need to be interpreted in terms of what they show about the hypothesis.  Do you have to accept or reject your null hypothesis, and why?

Evaluation (6 marks) – This is where you explain your results, in relation to the theory/model and study on which you based your experiment.  You need to explain the strengths and limitations of your design, sample and procedure and suggest how you could have improved upon what you did.  We cannot always anticipate the effect of decisions we made earlier when deciding how to conduct the experiment, but we can explain their effect at the end.

All IAs need a list of references at the back, and the appendices follow this.  They do not count towards the word count.

__________________________

Remember – it doesn’t have to be a complex experiment.  The simpler the better.  Old ‘favourites’ from Cognitive Psychology always do well: Loftus and Palmer, Stroop, Peterson & Peterson and Bransford & Johnson are all tried and tested studies from the area of memory.