Asilomar survey: researcher perspectives on ethical principles and guidelines for BCI research

Pham, Michelle; Goering, Sara; Sample, Matthew; Huggins, Jane E.; Klein, Eran

doi:10.1080/2326263x.2018.1530010

Cited by 25 publications

(14 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although invasive techniques provide more accurate data than non-invasive techniques, non-invasive modalities are more frequent and appreciated in the research domain. Non-invasive recording techniques for brain activity improve safety and reduce ethical concerns (Burwell et al, 2017 ; Pham et al, 2018 ). Over time, various non-invasive techniques have been used in studies.…”

Section: Introductionmentioning

confidence: 99%

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM

et al. 2022

View full text Add to dashboard Cite

The constantly evolving human–machine interaction and advancement in sociotechnical systems have made it essential to analyze vital human factors such as mental workload, vigilance, fatigue, and stress by monitoring brain states for optimum performance and human safety. Similarly, brain signals have become paramount for rehabilitation and assistive purposes in fields such as brain–computer interface (BCI) and closed-loop neuromodulation for neurological disorders and motor disabilities. The complexity, non-stationary nature, and low signal-to-noise ratio of brain signals pose significant challenges for researchers to design robust and reliable BCI systems to accurately detect meaningful changes in brain states outside the laboratory environment. Different neuroimaging modalities are used in hybrid settings to enhance accuracy, increase control commands, and decrease the time required for brain activity detection. Functional near-infrared spectroscopy (fNIRS) and electroencephalography (EEG) measure the hemodynamic and electrical activity of the brain with a good spatial and temporal resolution, respectively. However, in hybrid settings, where both modalities enhance the output performance of BCI, their data compatibility due to the huge discrepancy between their sampling rate and the number of channels remains a challenge for real-time BCI applications. Traditional methods, such as downsampling and channel selection, result in important information loss while making both modalities compatible. In this study, we present a novel recurrence plot (RP)-based time-distributed convolutional neural network and long short-term memory (CNN-LSTM) algorithm for the integrated classification of fNIRS EEG for hybrid BCI applications. The acquired brain signals are first projected into a non-linear dimension with RPs and fed into the CNN to extract essential features without performing any downsampling. Then, LSTM is used to learn the chronological features and time-dependence relation to detect brain activity. The average accuracies achieved with the proposed model were 78.44% for fNIRS, 86.24% for EEG, and 88.41% for hybrid EEG-fNIRS BCI. Moreover, the maximum accuracies achieved were 85.9, 88.1, and 92.4%, respectively. The results confirm the viability of the RP-based deep-learning algorithm for successful BCI systems.

show abstract

Section: Introductionmentioning

confidence: 99%

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM

et al. 2022

View full text Add to dashboard Cite

show abstract

“…SPECS is intended to be one sort of practice in a set of practices that seeks to support the moral integrity of novel neurotechnology research. Other practices include holding end-user focus groups, which are critical in gaining knowledge of the interests and lived experiences of current and prospective end-users, developing general ethical guidelines for researchers (Pham et al, 2018), and integrating ethicists into research teams (Goering and Klein, 2020). Other ethical engagement methods, like Responsible Research and Innovation (RRI) and Patient and Public involvement (PPI), are worthwhile ways to introduce and integrate the perspectives and concerns of potential end users and representative stakeholders from diverse publics in the process of shaping research agenda and protocols (Fleurence et al, 2013; Jakobson et al, 2019).…”

Section: The Story Of Specsmentioning

confidence: 99%

“… 3. For example, Sara Goering and Rafael Yuste argue for the necessity of ethical guidelines for neurotechnology research given its risks to research participants and broader societal values (Goering and Yuste, 2016). Numerous scholars have responded to the call, seeking to enumerate principles and prescriptions that they regard as helpful for promoting ethically responsible conduct (Greely et al, 2018; Pham et al, 2018). …”

mentioning

confidence: 99%

Fostering the trustworthiness of researchers: SPECS and the role of ethical reflexivity in novel neurotechnology research

Tubig

McCusker

2020

Research Ethics

View full text Add to dashboard Cite

The development of novel neurotechnologies, such as brain-computer interface (BCI) and deep-brain stimulation (DBS), are very promising in improving the welfare and life prospects many people. These include life-changing therapies for medical conditions and enhancements of cognitive, emotional, and moral capacities. Yet there are also numerous moral risks and uncertainties involved in developing novel neurotechnologies. For this reason, the progress of novel neurotechnology research requires that diverse publics place trust in researchers to develop neural interfaces in ways that are overall beneficial to society and responsive to ethical values and concerns. In this article, we argue that researchers and research institutions have a moral responsibility to foster and demonstrate trustworthiness with respect to broader publics whose lives will be affected by their research. Using Annette Baier’s conceptual analysis of trust, which takes competence and good will to be its central components, we propose that practices of ethical reflexivity could play a valuable role in fostering the trustworthiness of individual researchers and research institutions through building and exhibiting their moral competence and good will. By ethical reflexivity, we mean the reflective and discursive activity of articulating, analyzing, and assessing the assumptions and values that might be underlying their ethical actions and projects. Here, we share an ethics dialog tool—called the Scientific Perspectives and Ethics Commitments Survey (or SPECS)—developed by the University of Washington’s Center of Neurotechnology (CNT) Neuroethics Thrust. Ultimately, the aim is to show the promise of ethical reflexivity practices, like SPECS, as a method of enhancing trustworthiness in researchers and their institutions that seek to develop novel neurotechnologies for the overall benefit of society.

show abstract

“…The recent rise in complexity of ML architectures has made it more difficult to explain their outputs. Although there is an overall agreement about the safety and ethical needs for interpreting ML outputs [15,16], there is however no consensus on how this challenge can be addressed. On the one hand, there are advocates for the development of models that are themselves interpretable rather than putting the effort later into making black-box models explainable [17].…”

Section: Introductionmentioning

confidence: 99%

Towards a More Reliable Interpretation of Machine Learning Outputs for Safety-Critical Systems Using Feature Importance Fusion

2021

View full text Add to dashboard Cite

When machine learning supports decision-making in safety-critical systems, it is important to verify and understand the reasons why a particular output is produced. Although feature importance calculation approaches assist in interpretation, there is a lack of consensus regarding how features’ importance is quantified, which makes the explanations offered for the outcomes mostly unreliable. A possible solution to address the lack of agreement is to combine the results from multiple feature importance quantifiers to reduce the variance in estimates and to improve the quality of explanations. Our hypothesis is that this leads to more robust and trustworthy explanations of the contribution of each feature to machine learning predictions. To test this hypothesis, we propose an extensible model-agnostic framework divided in four main parts: (i) traditional data pre-processing and preparation for predictive machine learning models, (ii) predictive machine learning, (iii) feature importance quantification, and (iv) feature importance decision fusion using an ensemble strategy. Our approach is tested on synthetic data, where the ground truth is known. We compare different fusion approaches and their results for both training and test sets. We also investigate how different characteristics within the datasets affect the quality of the feature importance ensembles studied. The results show that, overall, our feature importance ensemble framework produces 15% less feature importance errors compared with existing methods. Additionally, the results reveal that different levels of noise in the datasets do not affect the feature importance ensembles’ ability to accurately quantify feature importance, whereas the feature importance quantification error increases with the number of features and number of orthogonal informative features. We also discuss the implications of our findings on the quality of explanations provided to safety-critical systems.

show abstract

Asilomar survey: researcher perspectives on ethical principles and guidelines for BCI research

Cited by 25 publications

References 40 publications

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM

Fostering the trustworthiness of researchers: SPECS and the role of ethical reflexivity in novel neurotechnology research

Towards a More Reliable Interpretation of Machine Learning Outputs for Safety-Critical Systems Using Feature Importance Fusion

Contact Info

Product

Resources

About