Week 1: Introduction, ethics and value for society
Ethics
Neuropsychological assessment is only meaningful and helpful to a patient if it’s valid and
interpretable. Relationship between brain and behaviour. Clinical neuropsychologists focus on
pathological behaviour.
Important to consider: access and familiarity with technology; environment; privacy;
interpretability – norm scores, Balancing pros and cons. Study evidence of validity of remote
assessment, consider individual patient factors and then make a balanced choice.
Advancement of techniques quite a far way since 1900s. Dualism dominant until 19 th century,
recognition of neurons in 1900s. First imaging method = x-ray (1895, Röntgen). Then
pneumoencephalography (replace a bit of the spinal fluid with oxygen, made tissue more visible, too
painful and risky), Computed (axial) Tomography Scan (CT, x-rays from all angles of the brain.
You can see tissue of brain very well). 18870s ECG, 1895s X-ray, 1930s EEG, 1970s CT-scan, 1980s
MEG, PET. In 1990s, fMRI, TMS, DBS. 2000s genomics, proteomics, DTI, integration of techniques.
Since then: data mining and more advanced methods.
Trade-of for what you want to know and how much you want to know. Is it functional to know
what a particular neuron in the hippocampus does, or do you want to know what this region in the
hippocampus does. Non-invasive Somewhat invasive Invasive methods
EEG PET scan, RCBF Single cell
Questions to consider recordings
- Does the gain in knowledge MEG CT Brain surgery
outweigh the burden to the fMRI Repetitive transcranial DBS
subjects Eye tracking rTMS
- Is there another way to gain this knowledge without bothering these people
- Is it really necessary to do this experiment in a patient population or can we do it in healthy
people first?
Code of ethics required to read. Code of ethics of research in the social and behavioural
sciences involving human participants. Developed by the deans of the social science faculties in the
Netherlands.
- Avoidance of exploitation
- Just distribution of benefits and burden
- Respect for persons have to give informed consent and be allowed to stop at any moment
o Participants are treated as autonomous agents
o Participants with diminished autonomy are entitled to protection
- Respect human dignity
- Scientific validity
- Scientific social and educational relevance
- Respect for rights and specific interests of research participants and/or community/society.
Milgram experiment give electric shock to (actor) someone you don’t see but hear.
Experimenter is in room and pushes participant to continue. Would this be allowed to continue
,nowadays with the code of ethics? What points from the code of ethics does this research violate?
Informed consent is sometimes really difficult.
The WMO Law wet medisch-wetenschappelijk onderzoek. Dutch law regarding scientific
medical research involving people. When doing medical research with people, you have to apply for
this law.
Science and value for society
How can science benefit society?
Validity to make science useful, you’ve to make it valid. Whether that’s with internal validity,
external validity, ecological validity, construct validity, statistical conclusion validity. One validity
might improve the other validity but have. A trade-off with the third validity.
Psychology’s replication crisis, 2015 only 36% of 100 replications of famous studies
replicated results. Percentages of cognitive psychology are a little higher than in social psychology.
Proposed causes (not necessarily evil intentions, but definitely the wrong priorities).
- Publication bias only researches with results get published.
- Publish or perish if you don’t publish enough, you’ll lose your job in certain areas.
Scientists are pressured to publish a lot and not recognise false positives
- Lack of institutional oversight researchers fabricate their results and there wasn’t enough
oversight to catch it, so this went on for years.
- Commercial conflicts of interest
- Inadequate training
- Desire to find a statistically significant result (Bias). we want to believe the hypotheses so
we’re more likely to find results that aren’t there.
- Ambiguity in how best to make these decisions.
What could help?
- More dedicated replications time consuming, fewer rewards, no breakthroughs, harder to
publish
- Improved education on methods and pitfalls
- Pre-registered articles publication no matter the result. They put their hypotheses or
methods out, before they start data collection.
- Data and workflow sharing to create transparency. That way, others can use the same data
to find if they can replicate your study.
- Adversarial collaboration
Research integrity honesty (accurately reporting data), scrupulousness, transparency (writing
methods and results clear enough for others to replicate), independence (staying objective),
responsibility (ethics). If you want to research in the Netherlands, you’ve to adhere to these guidelines.
Threats data fabrication, plagiarism, unethical use of data, position.
1
, Open science movement goal is to make science accessible to anyone. Increase transparency
and quality, increase reproducibility, increase research integrity. Open access publishing, data sharing,
peer-reviewing.
Translational sciences turning observations into effective interventions that improve health of
individuals and public – from diagnostics and therapeutics to medical procedures and behavioural
changes. Broader impact statements. Part of funding application.
Find out what value of research is. Who are the stakeholders, who benefits from research?
Patients, physicians, funding bodies, general public, researchers. You consider what drives your
research question in the design face. Who is interested? Who can collaborate and what can they
contribute to your project?
Recommendations for translation strict reporting guidelines, optimising science
(randomisation, blinding outcome measurements, report exclusions, sample size calculations),
publication and communication (what will I do with the findings?), less focus on impact factor, but
more on actual impact (based on how often an article is cited).
Different levels experimental settings to clinical settings, healthy participants to patients,
college students to broader population, animal literature to humans.
Methods quantitative vs qualitive research. Different results and different validity and
generalisation.
Week 2: large datasets in clinical research
Data sharing
It’s an open science goal. So much data is collected, so why not make it openly accessible?
Many advantages to allow people in different areas (less wealthy areas) to also do research.
Inclusiveness and making it more accessible.
- Reduce needed resources for research
- Pooling datasets combining data of people who do similar research to see if you can find
similar results.
- Allow other questions than your own to be addressed
- Make research more inclusive
- Transparency between colleagues even other areas (journalism/legal) can do research and
check if they get the same results. Legitimise the results.
- Allow more collaboration
- Increase trust in science
Many kinds of data sets questionnaire, task, health, biomarker, brain, etc.
Disadvantages personal investment (why give your hard-earned data away), fears of being
scooped, fear of errors (poorly collected datasets will find their way to the public and conclusions
based on this false data can be made), fear of misuse (open for misinterpretation, making wrong
conclusions), lack of time, privacy concerns (not all data is suitable to be shared, make sure that
2
Ethics
Neuropsychological assessment is only meaningful and helpful to a patient if it’s valid and
interpretable. Relationship between brain and behaviour. Clinical neuropsychologists focus on
pathological behaviour.
Important to consider: access and familiarity with technology; environment; privacy;
interpretability – norm scores, Balancing pros and cons. Study evidence of validity of remote
assessment, consider individual patient factors and then make a balanced choice.
Advancement of techniques quite a far way since 1900s. Dualism dominant until 19 th century,
recognition of neurons in 1900s. First imaging method = x-ray (1895, Röntgen). Then
pneumoencephalography (replace a bit of the spinal fluid with oxygen, made tissue more visible, too
painful and risky), Computed (axial) Tomography Scan (CT, x-rays from all angles of the brain.
You can see tissue of brain very well). 18870s ECG, 1895s X-ray, 1930s EEG, 1970s CT-scan, 1980s
MEG, PET. In 1990s, fMRI, TMS, DBS. 2000s genomics, proteomics, DTI, integration of techniques.
Since then: data mining and more advanced methods.
Trade-of for what you want to know and how much you want to know. Is it functional to know
what a particular neuron in the hippocampus does, or do you want to know what this region in the
hippocampus does. Non-invasive Somewhat invasive Invasive methods
EEG PET scan, RCBF Single cell
Questions to consider recordings
- Does the gain in knowledge MEG CT Brain surgery
outweigh the burden to the fMRI Repetitive transcranial DBS
subjects Eye tracking rTMS
- Is there another way to gain this knowledge without bothering these people
- Is it really necessary to do this experiment in a patient population or can we do it in healthy
people first?
Code of ethics required to read. Code of ethics of research in the social and behavioural
sciences involving human participants. Developed by the deans of the social science faculties in the
Netherlands.
- Avoidance of exploitation
- Just distribution of benefits and burden
- Respect for persons have to give informed consent and be allowed to stop at any moment
o Participants are treated as autonomous agents
o Participants with diminished autonomy are entitled to protection
- Respect human dignity
- Scientific validity
- Scientific social and educational relevance
- Respect for rights and specific interests of research participants and/or community/society.
Milgram experiment give electric shock to (actor) someone you don’t see but hear.
Experimenter is in room and pushes participant to continue. Would this be allowed to continue
,nowadays with the code of ethics? What points from the code of ethics does this research violate?
Informed consent is sometimes really difficult.
The WMO Law wet medisch-wetenschappelijk onderzoek. Dutch law regarding scientific
medical research involving people. When doing medical research with people, you have to apply for
this law.
Science and value for society
How can science benefit society?
Validity to make science useful, you’ve to make it valid. Whether that’s with internal validity,
external validity, ecological validity, construct validity, statistical conclusion validity. One validity
might improve the other validity but have. A trade-off with the third validity.
Psychology’s replication crisis, 2015 only 36% of 100 replications of famous studies
replicated results. Percentages of cognitive psychology are a little higher than in social psychology.
Proposed causes (not necessarily evil intentions, but definitely the wrong priorities).
- Publication bias only researches with results get published.
- Publish or perish if you don’t publish enough, you’ll lose your job in certain areas.
Scientists are pressured to publish a lot and not recognise false positives
- Lack of institutional oversight researchers fabricate their results and there wasn’t enough
oversight to catch it, so this went on for years.
- Commercial conflicts of interest
- Inadequate training
- Desire to find a statistically significant result (Bias). we want to believe the hypotheses so
we’re more likely to find results that aren’t there.
- Ambiguity in how best to make these decisions.
What could help?
- More dedicated replications time consuming, fewer rewards, no breakthroughs, harder to
publish
- Improved education on methods and pitfalls
- Pre-registered articles publication no matter the result. They put their hypotheses or
methods out, before they start data collection.
- Data and workflow sharing to create transparency. That way, others can use the same data
to find if they can replicate your study.
- Adversarial collaboration
Research integrity honesty (accurately reporting data), scrupulousness, transparency (writing
methods and results clear enough for others to replicate), independence (staying objective),
responsibility (ethics). If you want to research in the Netherlands, you’ve to adhere to these guidelines.
Threats data fabrication, plagiarism, unethical use of data, position.
1
, Open science movement goal is to make science accessible to anyone. Increase transparency
and quality, increase reproducibility, increase research integrity. Open access publishing, data sharing,
peer-reviewing.
Translational sciences turning observations into effective interventions that improve health of
individuals and public – from diagnostics and therapeutics to medical procedures and behavioural
changes. Broader impact statements. Part of funding application.
Find out what value of research is. Who are the stakeholders, who benefits from research?
Patients, physicians, funding bodies, general public, researchers. You consider what drives your
research question in the design face. Who is interested? Who can collaborate and what can they
contribute to your project?
Recommendations for translation strict reporting guidelines, optimising science
(randomisation, blinding outcome measurements, report exclusions, sample size calculations),
publication and communication (what will I do with the findings?), less focus on impact factor, but
more on actual impact (based on how often an article is cited).
Different levels experimental settings to clinical settings, healthy participants to patients,
college students to broader population, animal literature to humans.
Methods quantitative vs qualitive research. Different results and different validity and
generalisation.
Week 2: large datasets in clinical research
Data sharing
It’s an open science goal. So much data is collected, so why not make it openly accessible?
Many advantages to allow people in different areas (less wealthy areas) to also do research.
Inclusiveness and making it more accessible.
- Reduce needed resources for research
- Pooling datasets combining data of people who do similar research to see if you can find
similar results.
- Allow other questions than your own to be addressed
- Make research more inclusive
- Transparency between colleagues even other areas (journalism/legal) can do research and
check if they get the same results. Legitimise the results.
- Allow more collaboration
- Increase trust in science
Many kinds of data sets questionnaire, task, health, biomarker, brain, etc.
Disadvantages personal investment (why give your hard-earned data away), fears of being
scooped, fear of errors (poorly collected datasets will find their way to the public and conclusions
based on this false data can be made), fear of misuse (open for misinterpretation, making wrong
conclusions), lack of time, privacy concerns (not all data is suitable to be shared, make sure that
2
