Theme 3: Ethical practice of science

1. Universalism

science is to be judged on the quality of the science all over the world, regardless of where it originates (doesn't depend on who did the work, where it was done + where it was published)

Code of practice

series of guidelines for what behaviour Is ok and not ok, avenues of appeal, advice / regulate members about ethical issues

Research misconduct

-includes fabrication, falsification, plagiarism or deception in proposing, carrying out / reporting the results of research + failure to declare or manage a serious conflict of interest ~0.01% of researchers per year (confirmed cases) ~1% of researchers per year (anonymous self reports) David Resnik- misconduct occurs due to "stressful" or "imperfect" environment

David resnick- reasons to adhere to ethical norms

1. "Norms promote the aims of research - knowledge, truth, avoidance of error." 2. "Ethical standards promote the values essential to collaborative work, such as trust, accountability, mutual respect, and fairness. Guidelines for authorship, copyright and patenting policies, data sharing and confidentiality in peer review, are designed to protect intellectual property interests while encouraging collaboration."

5 freedoms- relate to replacement (enhance wellbeing)

1. From Hunger & First -access to fresh water + good diet 2. From Discomfort -appropriate environment including shelter, resting area 3. From pain, injury + disease - prevention / treatment 4. To Express Normal Behaviour -space, facilitates, company of the animals own kind 5. From Fear & Distress -conditions + treatment which avoid mental suffering

Procedures for commercial lab notebooks

1. GENERAL o Notebooks numbered, dated and name of scientist o Each page should be signed, dated and witnessed 2. ENTRIES o Write in black permanent ink... o Record in order of date...One day per page. ... o Title, aim, then record results and interpretations... o Methods should be included the first time... 3. SAMPLE ANALYSIS o Data printouts should be fixed permanently into notebook or cross reference to computer file o May have a methods book cross reference to that

Humans as subjects in research

1. Origins of codes governing ethical research on humans 2. Informed consent in drug trials + all experiments with human subjects 3. Famous psychosocial experiments

The 3 Rs- factors affect whether using animals in experimentation is appropriate

1. Reduction 2. Replacement 3. Refinement

5 common mistakes for new scientists

1. Rule out data that doesn't support my hypothesis 2. Extrapolating beyond range of data 3. Hypothesis doesn't equal explanation? 4. Does it need testing? 5. Correlation vs causation?

Self regulating through 3 ideals of science

1. Universalism 2. Peer review 3. Repetition

3 ideals of science and how they aim to limit research misconduct - URR

1. Universalism: bell labs highly regarded, he had many prestigious publication with respected co-authors , all of the papers were originally accepted for review 2. Review process: all of the papers were originally accepted for review + publication, system failed initially with lack of checks 3. Repetition: several scientists reported they were unable to reproduce the same results scepticism

Nuremburg code; 10 key points

1. voluntary 2. fruitful results 3. design based on prior knowledge 4. no unnecessary suffering 5. not allowed if death/disability likely 6. humanitarian importance (risk/benefit) 7. adequate facilities to protect subject 8. scientifically qualified persons 9. subject can opt out 10. must be stopped - likely to result in injury, disability, or death to the experimental subject

Vancouver protocol for authorship-3 key requirements

1.Data - idea, concept, design + analysis 2.Writing - draft / editing 3.Overall responsibility - should have seen all content, intellectual responsibility for your part

Famous psychological experiments; Obedience and the individual Stanley Milgram

1963 • experiment stemmed form WW2, his parents were holocaust survivors • looked at if people are really evil, and whether perfectly normal people can commit evil acts by just following orders of an authortive figure • set up experiment where the subject didn't realise they were a subject of the experiment, were in control of a device that gave eclectic shocks to other subjects (just thought they were an actor) • subjects are tricked into thinking they chose their role as teachers / learner in the experiment • learner is told he will be asked memory questions from teacher and if answer is incorrect - he receives electric shock (highest is lethal of ~400 volts) • voltage increased each time they got a question wrong • to see if the teachers (the subjects) could put a stop the experiment / stop following what the scientist is instructing them to do • the electric shocks were real but they were fake.

David Resnick- reasons to adhere to ethical norms

3. "Many ethical norms e.g. policies on research misconduct, conflicts of interest, human subject protection and animal care help to ensure that researchers who are funded by public money can be held accountable to the public." 4. "Ethical norms in research also help to build public support for research." 5. "Many of the norms of research promote a variety of other important moral and social values, such as social responsibility, human rights, animal welfare, compliance with the law, and public health and safety."

Famous psychological experiments; Stanford prison experiment Phillip Zimbardo

Stanford uni 1971 • college student = subjects • aimed to test that the inherent personality traits of prisoners + guards are the main cause of abusive behaviour in prison • participants = guards, enforced authorial measures + subjected prisoners to psychological torture • prisoners accepted psychological abuse, harassed other prisoners who tried to stop it (at the request of the guards) • Zimbardo allowed abuse to continue • 2 prisoners left mid experiment • after objection from graduate student that Zimbardo was dating- experiment was terminated after 6 days

3. Repetition

for other scientists to repeat the methodology of this study, if many labs aren't able to repeat it, there will be more doubt cast on the study and may be regarded as wrong need to be able to repeat the study and get the same results!

Informed consent

includes telling patient the following about the study given as much info about the study as possible so they can decide if they want to participate or not, by looking at possible benefits + risks and weighing these up. -purpose, all procedures, funding, risks, benefits

Famous psychological experiments; Stanley Milgram in violation of

o violated Nuremburg code 9 - 'subject can opt out' o violated informed consent were not told about the true nature of the study could not make an informed decision o violated the ability of the subjects to opt out were discouraged not to withdraw, every time they asked to stop the experimenter wouldn't really let them / said the whole study would be terminated if he dropped out. o didn't allow them to stop even in spite of causing risk, harm, pain + lasting damage

2. Peer

process in publication - every scientific paper should be sent out to 2-3 experts in the field for the experimental data to e reviewed and they decide if it Is valid + important enough to be published

Registering trials in advance seems to help

• % trials published that reporting positive results registration of trials increases the likelihood of negative results being reported

Case study; Harry harlow

• 1950s • infamous for work on primates that affect the infant psychologically • Experiments on love + depression • Results drove the shift in ideas of what is ok and what's not regarding animals • Impacted societies undemanding of brain, behaviour • Revolutionized how parents raised children taught them to love and hug children • Study provided new definition on love + personalities • Found the key as his work showed the mother-child relationship is described as love • Performed distressing experiments on animals • Hurt animals to help us - was this worth it? • Thin boundary between scientific research + cruelty • Separated baby monkeys from mothers - put into cages with diapers

Pressures in science

• Authorship • Pressure • Data selection • Management issues

1. Reduction

• Comparable amounts of data from fewer animals • More info for the same number of animals

Unethical behaviours

• Discussing with your colleagues confidential data from a paper that you are reviewing for a journal • Stretching the truth on a grant application in order to convince reviewers that your project will make a significant contribution to the field • Overworking, neglecting, or exploiting graduate or post-doctoral students

E.g. Facebook emotion experiment

• For 1 week in 2012, FB altered the algorithms it used to determine which status updates appeared in the news feed of thousands or randomly selected users. • The algorithm filtered content based on emotional content - posts were either negative or positive • Experiment 1: reduced positive content of news feeds • Experiment 2: reduced the negative content of news feeds • Facebook quantifies percentage of words in peoples status updates that were neg or pos • Results were published based on if you re subjected to pos/neg content, and what people are likely to post having being exposed to that content for a week • without informed consent from fb users but argue that they have a terms& conditions page its peoples fault for not reading it

Intellectual context

• Logical contradiction: animals are like us + they are not like us. Is it one or the other? Deontological Approach: rule based ethics Utilitarian approach: "greatest good to be the greatest number" or the "ends justify the means" • Context: effect of the type of animal, what the research if for, and how the research is conducted? affect our perception of what is acceptable?

Ethics case study 2

• Ph.D Konstanz Germany • Worked in Bell labs • ~100 papers in 6 years • peer review process accepted all of his papers • others working in the same field started to get suspicious (due to the huge number of publications coming from him people started looking at the papers in more detail • found 2 graphs from 2 different papers to be the same / very similar people scrutinized his papers more, launched an enquiry into his work, co-authors of the papers were interviewed, drafts + raw data was looked at for each paper • he didn't keep the raw data • he couldn't present any of the original samples (they had all been destroyed) • admitted to falsifying some of the data "to make it more convincing" used programs to generate data instead of present it via experiments • reused data from experiments • 28 papers were withdraws, 9 still in question • ramifications: sacked from the lab, Phd revoked • found he was a plagiarist from his undergraduate days • no longer able to be a peer reviewer / apply for funds through main German research fund

Keeping a record to prevent issues

• Record everything - cross reference to data files • Retain raw data for 5 years after publication • Data should stay in lab where the work was done • Keep sample until 5 years after publication

Climate change research

• Scientists are conservative, tend to be cautious don't like to make bold proclamations about specific theories (eg climate change)- as they are likely to be shot down by political + social views political + social pressure • Alarmist statements - don't sound objective -when public think there is no scientific consensus for climate change they are less likely to support

Publication bias

• Tendency of certain trials with largest effect size are more likely to be published • Issue with publication bias: Increases the risk that the observed effect doesn't reflect the true effect may negatively impact consistency, precision, magnitude of effect • Reliance on meta-analysis to compile results from many experiments can exacerbate this bias

Ethics Case study 1

• Tony Lasaga, Yael professor • Suspected by his graduate student + research assistant of being a pedo • Convicted + sent to prison in 2002 for 15 years • Executive editor of Editor of Geochimica et Cosmochimica Acta refused to send the paper he submitted for review bc he was a convicted felon • Does the private life of a scientists affect whether their research should be published? scientists are part of society, a community with moral repsosnbilties should have to uphold moral, ethical standard • However: there were many co-authors who were not involved in bad behaviour why couldn't they progress their careers and have their paper and hard work / years be submitted?

2. Replacement

• Use methods performed without the use of animals: cell + tissue culture, computer simulations • Use "lower organisms" ie drosophila flies

3. Refinement

• Use methods that alleviate /minimise potential pain + distress • Enhance wellbeing of animals kept

Reasons why some scientists behave unethically

• don't believe that ethics are important / apply to them • believe that the public needs to hear certain info even if it may not be ethical • to improve / assist their study • government funding - may want to see certain results scientist sways results to fit that • falsifying data • biased • spent years developing a theory and the experiment doesn't agree with the theory hard of reverse the years of engrained conditioning • student and doesn't want to "let down the boss" bc he has been working on the study / theory for 20 years • fame / prizes financial gain • vested interests ie shares in a company, or doing work with a company want good outcomes for the company and you can benefit from it (eg: drug company) • career / promotional reasons

Famous psychological experiments; Stanford prison experiment Phillip Zimbardo violation of

• experiment continued even when participants wanted to opt out - violated Nuremberg rule 9 • resulted in injury, psychology effects - Nuremberg rule 10 • led to implementation of rules to prevent harmful treatments in future studies

Helskini declaration

• need to publicise research results, including negative/inclusive studies • need to ensure access to interventions that are proven effective addresses: patient / subject rights, type of research, reason for research, funding, consent must be informed

#alltrials campaign

• project advocating that clinical research should adopt the principles of open research • "all trials registered, all trials reported" • all clinical trails should be listed in a registry and their results should always be shared as open data (that some data should be freely available to everyone to use + republish as they wish, without restrictions from copyright, patents or other controlling mechanisms • Thousands of clinical trials have not reported their results; some not registered. • Information on what was done + found in these trials could be lost forever to doctors and researchers leading to bad treatment decisions, missed opportunities for good medicine + trials being repeated. • All trials past and present should be registered + the full methods and the results reported.

Role that ethics committees play in research institutions

•Any educational setting (school, lab, university) that wants to use animals in research / teaching •Bird watching •Taking of samples

importance of ethics in scientific research

•Avoid negative impacts •Meet community standards taxpayer funded