April International Neuroethics Society Webinar: Brain Data Governance and Neuroethics

By Anita Jwa

Image courtesy of Flickr

This piece is the first in a series that will include discussions about the International Neuroethics Society (INS) webinars; these webinars are being hosted monthly throughout 2021 and feature various neuroethics topics.

On April 23, the International Neuroethics Society held a virtual panel discussion on Brain Data Governance and Neurorights. Interdisciplinary panelists – Samir Das (Montreal Neurological Institute, McGill University), Stephan Rainey (Oxford Uehiro Centre for Practical Ethics, University of Oxford), and Fruszina Molnár-Gábor (Heidelberg Academy of Sciences), with Philipp Kellmeyer moderating the discussion – examined the ethical and legal issues in governing the access and use of human brain data collected in various contexts, such as for research or commercial purposes.

Based on his experience in multi-center/cross-border research studies, Samir Das shared challenges in collecting and sharing human brain data, including issues of privacy and security, standardization and interoperability, and building technical infrastructure. In addition to providing several rather straightforward technical solutions, Das urged that an ethical framework that delineates detailed guidance on how to responsibly share data — such as the Ethics and Governance Framework in Canadian Open Neuroscience Platform — could be extremely valuable in addressing these challenges.

Stephen Rainey also argued that recent advances in high-density recording of brain signals could lead to novel insights into brain activity, but such advances also raise critical ethical concerns. The fact that brain signals recorded and processed for one purpose could be reprocessed later for other purposes makes brain data sensitive. In particular regard to brain–computer interface (BCI) technology, data are being used for BCI device customization, which also makes the data personal. Moreover, because data collected from a large number of individual users become part of a BCI device’s operating system, it would be difficult to account for individual data once merged, which could impede the exercise of the right to personal data retrieval or deletion as provided in the European Union’s General Data Protection Regulation (GDPR).

Focusing more specifically on legal issues, Fruszina Molnár-Gábor highlighted the need for specific regulations for different types of data. In terms of brain data, intrusion into privacy could be a more severe threat than is posed in other personal data due to individuals’ lack of control over accessing brain data and making decisions about its use. Losing control over data could also affect other rights, including the right to access, correction, or deletion under the GDPR. Yet, given that data protection and privacy are not absolute rights, Molnár-Gábor emphasized that data security should operate while balancing these rights with other rights and interests — for example, the freedom of research, as well as interests in using data for commercial purposes.

Image courtesy of Flickr

Collectively, the panel considered that, given the highly multidisciplinary nature of brain data governance, the primary challenge lies more in balancing different, conflicting cultures and interests, and agreeing upon common standards, than in implementing technical safeguards. Thus, it would be critical to incorporate different disciplinary insights in developing specific rules for brain data protection. One example would be adopting regulatory instruments such as the Code of Conduct in the GDPR. In addition, the panel proposed other regulatory approaches (e.g., a regulatory protection against the malicious use of brain data). The issue of informed consent was also considered, particularly as it relates to the consumer domain. Panelists examined various formats of consent — such as broad consent, dynamic consent, and consent management through a data trustee or databank — as alternative options to better respect subjects’ autonomy or to better enable more granular consent. The panel also touched upon the need to achieve both technical and legal interoperability across different regulatory regimes to develop an international governance framework, and on the question of whether and how human rights or neuro-rights can be useful in building such frameworks.

For the final third of the session, moderator Philipp Kellmeyer fielded questions from viewers on a range of topics. The closing discussions examined: the fundamental ethical concerns (and potential legal prohibitions) regarding the collection and use of brain data for commercial purposes; considerations for vulnerable populations in governing brain data; protection of brain data after subjects are deceased; and the role of do-it-yourself neurotechnology community or the citizen science/neuro-hacking community and related issues (such as whether the freedom of research and research exemptions could be extended to self-experimentation or proprietary research).

This panel discussion provided a general overview on relevant ethical and regulatory issues around brain data governance and valuable insights on how to better inform subjects and to develop more robust regulatory safeguards for brain data protection. A sound brain data governance framework should aim to maximize the benefits of the collection and processing of brain data for biomedical research while minimizing the potential harms caused by the misuse of brain data. As the panelists discussed, use of brain data in the consumer domain raises different ethical concerns, such as commercialization of brain data, compared to its use in research or clinical contexts, although the line between these uses often blurs. Thus, it would be crucial to contemplate how to differentiate between these uses and to determine an appropriate level of protections for a specific use in developing the governance framework for brain data.

Image courtesy of Pixabay

In addition, one topic that received less attention in the discussion was risks of harm associated with collection and use of brain data. The most salient risks associated with brain data would be unwanted disclosure of health-related or other sensitive information derived from the data, such as the risk of brain disorder (e.g., Alzheimer’s disease) or prediction of personality traits or future behaviors (e.g., criminality). As the panelists noted, significant technological advancement has occurred in the measuring and processing of brain activations with the aid of recent advances in artificial intelligence and machine learning algorithms. Nevertheless, the current state of neurotechnology remains far from reliably decoding mental/emotional state or predicting specific future outcomes. For example, much of the previous work on brain decoding using neuroimaging data suffers from small sample sizes and from high rates of false positives (Poldrack, Huckins, & Varoquaux, 2020; Varoquaux, 2018). Although a regulatory framework for brain data should account for the future prospect of technological advancement, a realistic assessment of state-of-the-art technology would be necessary to prevent unfounded hype, or fear, of the capacity of brain data to read minds or predict specific outcomes, which could lead to undue restrictions on the use of brain data (Choudhury et al., 2014).

It is also important to note that disclosure of brain data and related sensitive information per se would be unlikely to cause harm – it would additionally require the use of the data with intent to endanger the subjects’ interests, as in the cases wherein a subject’s disease biomarker or personality traits derived from brain data are used to damage the subject’s reputation or to discriminate against the subject in employment or health insurance contexts. Therefore, in addition to requiring strong privacy and security measures, regulatory efforts for brain data governance should also be directed at preventing misuse of the data, as was pointed out in the panel discussion. Existing regulations in other comparable contexts, such as the Genetic Information Non-discrimination Act in the United States, would provide a useful reference for developing a regulatory safeguard against the misuse of brain data. This regulatory safeguard would provide the utmost protection for brain data privacy without unnecessarily limiting the use of brain data to understand brain function for the prevention and treatment of brain disorders (Jwa & Poldrack, 2021).

References

1. Choudhury, S., Fishman, J. R., McGowan, M. L., & Juengst, E. T. (2014). Big data, open science and the brain: lessons learned from genomics. Front Hum Neurosci, 8, 239. doi:10.3389/fnhum.2014.00239
   
2. Jwa, A. & Poldrack R. A. (2021). Data Sharing Policies in Neuroimaging Data Repositories, PsyArXiv, doi.org/10.31234/osf.io/cnuy7
   
3. Poldrack, R. A., Huckins, G., & Varoquaux, G. (2020). Establishment of Best Practices for Evidence for Prediction: A Review. JAMA Psychiatry, 77(5), 534-540. doi:10.1001/jamapsychiatry.2019.3671
   
4. Varoquaux, G. (2018). Cross-validation failure: Small sample sizes lead to large error bars. Neuroimage, 180(Pt A), 68-77. doi:10.1016/j.neuroimage.2017.06.061

______________

Anita Jwa is a postdoctoral research fellow at Stanford University Department of Psychology. She graduated from Stanford Law School with a Doctor of Science of Law (JSD) and her primary research interests are ethical, legal, and social implications of advances in neuroscience. She is currently working on a research project on ethical and regulatory issues around neuroimaging data sharing, particularly regarding the risks of subject re-identification using novel software tools and algorithms.

Want to cite this post?
Jwa, A. (2021). April International Neuroethics Society Webinar: Brain Data Governance and Neuroethics. The Neuroethics Blog. Retrieved on , from http://www.theneuroethicsblog.com/2021/06/aprilinternational-neuroethics-society.html