A preliminary survey on the perception of marketability of
Brain-Computer Interfaces (BCI) and initial development of a
repository of BCI companies
F. Nijboer1, B. Z. Allison2, S. Dunne3, D. Plass-Oude Bos1, A. Nijholt1, P. Haselager4
1Human Media Interaction, Faculty of Electrical Engineering, Informatics and Mathematics, University
of Twente, Enschede, the Netherlands
2Graz Technical University, Graz, Austria
3Starlab, Barcelona, Spain
4Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands
femke.nijboer@utwente.nl
Abstract
The marketability of current and future BCI applications may greatly in
uence the decisions of
goverments, the industry and academia. In this paper we rst explored with a survey when respon-
dents (N=145), who were present at the 4th International BCI Meeting, expect that di erent BCI
applications will become commercially available. Second, we surveyed how well existing BCI compa-
nies are known to respondents. Third, we compared the ndings with our own preliminary overview
of the marketability of BCIs and our repository of 28 companies. Respondents were optimistic about
the marketability of BCIs for healthy users and users who need assistive technology, but 72.4 % of the
respondents was unaware that companies already exist which market BCI's. Based on a preliminary
market overview we cautiously suggest that optimism in relation to applications for healthy users
is more appropriate than in relation to BCI-based assistive technology. In future we plan surveys
among a broader range of stakeholders and a more profound analysis of the market.
1 Introduction
Brain-Computer Interface (BCI) research is rapidly increasing, with considerable enthusiasm for applica-
tions for users with and without physical disabilities [1, 2]. A roadmap for BCI research and development
could help address emerging markets and opportunities. The future Brain/Neuronal Computer Inter-
action project (fBNCI; http://future-bnci.org/ aims to develop such a roadmap by drawing on the
expertise of BCI researchers, as well as many other stakeholder groups, including companies, end users,
patient organizations, policy makers, and the general public.
One focus of the roadmap is to evaluate the marketability of current and future BCI applications. Hence,
we surveyed researchers on the marketability of BCIs at the 4th International BCI Meeting, which took
place at the Asilomar conference centre in June 2010. This conference provided an excellent opportunity
to engage many quali ed respondents, although it has to be noted that many other quali ed respondents
(e.g. from the eld of human-computer interaction or ambient intelligence) do not typically attend this
meeting which may bias our ndings (see section 5). Other results from this survey and more demo-
graphical data of the repondents are available elsewhere ([3]) and under review ([4]).
In this paper we focus on three aspects of the survey. First, we explored when respondents expect that
di erent BCI applications will become commercially available (see 3.1). Second, we surveyed how well
existing BCI companies are known to respondents (see 3.2). Third, we compared the ndings with our
own preliminary overview of the marketability of BCIs and our repository of companies (see 3.3).
1
2 Methods
2.1 Respondents
A total of 145 (105 males, 39 females) out of 289 conference attendees responded to the questions about
the marketability of BCIs. Seventy three persons were aged between 18 and 30, sixty-nine persons
between 31 and 55 and two persons were aged between 56 and 70. The sample consisted of experts from
various disciplines (e.g. neuro-, computer or cognitive scientists, electrical engineers, psychologists etc).
One participant did not give demographical data.
2.2 Survey
Participants completed an online survey that included four questions on the expected marketability
of di erent purpose BCIs. In the survey, for convenience, we ordered some BCI types based on their
function: 1) BCIs for healthy users, 2) BCIs as assistive technology, 3) BCI-controlled prostheses and 4)
BCIs as therapy tools.
Speci cally, participants were asked to indicate when they expected to see these types of BCI systems to
become available on the market. Participants could choose from 5 answer options: "never", "between 0-5
years", "between 5-10 years", "more than 10 years" or "it already exists on the market". Participants
who indicated that this type of BCI already exists on the market were asked which group or company
o ers the product.
2.3 Development of a repository of BCI companies
Only companies which o ered BCI products and services were included in the repository. Thus, compa-
nies supplying hardware and software needed for BCI research were not included. The repository was
build through: 1) web searches, 2) written and verbal interviews of experts and 3) iterative postings on
LinkedIn groups ("brain-computer interface group", "neuromarketing" and "BrainGain").
3 Results
3.1 Respondents' expectancies of marketability of BCI
Table 1 presents the respondents' expectancies per BCI class.
BNCI technology for: already
on
market
0-5 yrs 6-10 yrs > 10 yrs never Number
of resps
healthy users 26.6 % 44.8 % 13.8 % 12.4 % 1.4 % 145
users who need AT 16.6 % 42.1 % 33.8 % 7.6 % 0.0 % 145
users who need prostheses 0.7 % 20 % 43.4 % 34.5 % 1.4 % 145
users who need therapy 9.9 % 39.4 % 39.4 % 10.6 % 0.7 % 142
Table 1: Overview of percentage of respondent which indicated when BCI applications will enter (if ever)
the market for healthy users, users who need assistive technology (AT), prostheses and therapy. The last
column shows the number of respondents (resps) who rated the item
3.2 Respondents' knowledge about existing companies
Thirty-two respondents reported which companies they knew that already marketed BCI's for health
users. By far the best known companies are Neurosky (counted 13 times) and Emotiv (counted 14 times).
Other companies that respondents mentioned were: Hitachi, Ambient, OCZ technology, Starwars Science,
Mattel, inc., g.Tec, Brain Actuated Technologies, InteraXon, Zeo, inc. and Interactive Productline.
Twenty-two respondents reported which companies they knew that already marketed BCIs as AT. The
best-known company is g.Tec (counted 16 times). Other groups and companies that respondents listed
were: Ambient, Brain Actuated Technologies, OpenVibe and BCI2000. Nine respondents commented
2
that they knew BCI's as therapy tools already are on the market, but only two gave concrete company
names: Brain master Technologies, inc. and EEGinfo. None of the respondents identi ed a company
that already markets BCI-based prostheses.
3.3 Preliminary overview of companies and marketability
We summarized a preliminary overview of 28 companies related to Brain-Computer Interfacing (see gure
1), which we expect to be more complete at the time of the Graz conference. Currently, neurofeedback
companies are not yet listed in the repository. We invite all readers to comment on this overview
and continue completing and criticizing the overview so we can o er the European Commission the
most accurate summary possible. This overview currently shows that as many as 10 companies market
BNCI's for entertainment and gaming. Six companies o er BNCI as assistive technology, although two
of these companies (Neural Signals, inc and BitBrain) do not yet market products. However, there
is a big di erence in the number of products sold between the two markets. Neurosky, for example,
has sold approximately 1 million Mindsets, whereas g.Tec has sold between 30-40 Intendixes (personal
communication).
Our survey did not address neuromarketing in depth. We counted six companies in this area. BCI
researchers (N=144) have divided opinions on whether neuromarketing technologies do (35.4 %) or do
not count (47.9 %) as BCI systems (Nijboer et al, in review). Nevertheless, neuromarketing developments
in
uence BCIs, and it is predicted that this market will grow tremendously [5, 2]. Finally, two companies
(Brain Fingerprinting inc. and No Lie MRI) provide BCI services for criminal investigation.
BNCI
technologies
health
sector
entertainment
sector
financial
sector
ICT
educational
sector
neuromarketing
serious games
humancomputer
interaction
multimodal
interaction
gaming
art
science
realtime
analysis
neuroeconomics
Interactive
Productline
Existing companies
Zeo, inc.
Neurosky
Emotiv Mattel, inc.
Uncle Milton
PLX
BCI net
Mind
technologies
Neuro-insight
The Mindlab
Advanced
Brain
Monitoring
Sands
technologies
Cortech
solutions
Emsense
safety
security
jurisdiction
BitBrain
technologies
InteraXon.
InteraXon.
nutrition
g.Tec
Brain actuated
technologies
Ambient
wellness
prevention
monitoring
therapy
assitive
technology
addiction
disorders
Neural
Signals, inc.
image
detection
BitBrain
technologies
ambient
intelligence
Brainfingers
Brain Fingerprinting
Laboratories, Inc.
Figure 1: Preliminary overview of existing companies
4 Discussion
The survey results show that respondents were generally optimistic about the near-term marketability of
BCI's for healthy users and user in need of AT. Repondents were slightly less optimistic about BCI for
therapy and expect BCI-controlled neuroprostheses to require more than 10 years to enter the market.
Our preliminary overview shows that companies - more than respondents realize - already market BCIs
for healthy users and users who need AT. However, having a product on the market does not yet mean
that a product is sold successfully. BCI research toward AT has a history of several decennia yet 30-40
products were sold in g.Tec. BCI applications for healthy users are available since less than 5 years and
(at least) more than a million products have been sold. It is remarkable that 72.4 % of the participants
are not aware that BNCI technologies for healthy users are already on the market. Since 10 companies
already exist in this direction we argue that optimism for this type of applications is quite appropriate.
3
In our opinion BNCI for assistive technology is not likely to create large economic value, unless it
becomes available for a broader range of categories of (more numerous) end-users. The societal value of
AT-BCI products is enormous, since they can enable users with disabilities to participate in the society
at interpersonal and professional levels. However, we predict that the research and development of BCI
applications targeted at the general public will more easily mobilize the industry and facilitate the tech
transfer between universities and industry. This will lead to more acceptance of BNCI technology in
the public and possibly to a spillover to the AT industry. In addition, even though the market for BCI-
controlled prosthetics requires a more time-consuming trajectory, ultimately we expect these products
to create large economic and societal value similar to Deep Brain Stimulation, because the number of
potential end-users is relatively high.
The survey reported here contains some
aws. First, since we could not make the entire survey too time-
consuming, some questions were generalized. For instance, we did not specify whether the BCI products
would work with non-invasive or invasive brain signals nor did we consider the usability of di ent BCIs.
However, these speci cations would in
uence the cost, number of available users, training time, time to
market and many other factors. Second, the respondents were mostly people involved primarily in BCI
research and had the time, funds, and enthusiasm to attend a major conference, so the sample may be
positively biased. fBNCI in future will survey a broader range of stakeholders, including more experts
from human-computer interaction and the industry.
5 Conclusion
Although we view these data as a rst crude estimation of the elds' perception of marketability of BCIs,
we cautiously conclude that BCI researchers seem optimistic about how much time it will take before the
rst BCIs for non-medical purposes and AT enter the market. We agree with them that the market for
BNCI applications for healthy users is promising, but we are not convinced that the market for AT-BCI
applications is equally promising simply because the number of end-users is relatively low in comparison.
Moreover, the limited knowledge of respondents about existing companies should raise awareness in the
BCI community that tech transfer is already well established largely without the involvement of the
scientists who consider themselves the core of the BCI eld. A more pro-active role of BCI scientists in
tech transfer would be desirable for high quality product development.
6 Acknowledgements
The authors gratefully acknowledge the support the Future BNCI project (Project number ICT-248320)
and of the BrainGain Smart Mix Program. We thank all respondents and Thorsten Zander, Jan van
Erp, Yann Renard, Fabien Lotte, Ariel Garten, Tom Sullivan and Christoph Guger for valuabe input.
References
[1] A. Nijholt and D. Tan. Brain-computer interfacing for intelligent systems. Ieee Intelligent Systems,
23(3):72{72, 2008.
[2] Brendan Z. Allison. Toward ubiquitous bcis. In Bernhard Graimann, Gert Pfurtscheller, and Brendan
Allison, editors, Brain-Computer Interfaces, The Frontiers Collection, pages 357{387. Springer Berlin
Heidelberg, 2010.
[3] F. Nijboer, J. Clausen, B. Z. Allison, and P. Haselager. Researchers opinions about ethically sound
dissemination of bci research to the public media. International Journal of Bioelectromagnetism, in
press.
[4] F. Nijboer, J. Clausen, B. Z. Allison, and P. Haselager. The asilomar survey: researchers opinions
on ethical issues related to brain-computer interfacing. Neuroethics, in review.
[5] Z. Lynch. The Neuro Revolution: How Brain Science Is Changing Our World. St. Martin's press,
New York, 2009.
4
Brain-Computer Interfaces (BCI) and initial development of a
repository of BCI companies
F. Nijboer1, B. Z. Allison2, S. Dunne3, D. Plass-Oude Bos1, A. Nijholt1, P. Haselager4
1Human Media Interaction, Faculty of Electrical Engineering, Informatics and Mathematics, University
of Twente, Enschede, the Netherlands
2Graz Technical University, Graz, Austria
3Starlab, Barcelona, Spain
4Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands
femke.nijboer@utwente.nl
Abstract
The marketability of current and future BCI applications may greatly in
uence the decisions of
goverments, the industry and academia. In this paper we rst explored with a survey when respon-
dents (N=145), who were present at the 4th International BCI Meeting, expect that di erent BCI
applications will become commercially available. Second, we surveyed how well existing BCI compa-
nies are known to respondents. Third, we compared the ndings with our own preliminary overview
of the marketability of BCIs and our repository of 28 companies. Respondents were optimistic about
the marketability of BCIs for healthy users and users who need assistive technology, but 72.4 % of the
respondents was unaware that companies already exist which market BCI's. Based on a preliminary
market overview we cautiously suggest that optimism in relation to applications for healthy users
is more appropriate than in relation to BCI-based assistive technology. In future we plan surveys
among a broader range of stakeholders and a more profound analysis of the market.
1 Introduction
Brain-Computer Interface (BCI) research is rapidly increasing, with considerable enthusiasm for applica-
tions for users with and without physical disabilities [1, 2]. A roadmap for BCI research and development
could help address emerging markets and opportunities. The future Brain/Neuronal Computer Inter-
action project (fBNCI; http://future-bnci.org/ aims to develop such a roadmap by drawing on the
expertise of BCI researchers, as well as many other stakeholder groups, including companies, end users,
patient organizations, policy makers, and the general public.
One focus of the roadmap is to evaluate the marketability of current and future BCI applications. Hence,
we surveyed researchers on the marketability of BCIs at the 4th International BCI Meeting, which took
place at the Asilomar conference centre in June 2010. This conference provided an excellent opportunity
to engage many quali ed respondents, although it has to be noted that many other quali ed respondents
(e.g. from the eld of human-computer interaction or ambient intelligence) do not typically attend this
meeting which may bias our ndings (see section 5). Other results from this survey and more demo-
graphical data of the repondents are available elsewhere ([3]) and under review ([4]).
In this paper we focus on three aspects of the survey. First, we explored when respondents expect that
di erent BCI applications will become commercially available (see 3.1). Second, we surveyed how well
existing BCI companies are known to respondents (see 3.2). Third, we compared the ndings with our
own preliminary overview of the marketability of BCIs and our repository of companies (see 3.3).
1
2 Methods
2.1 Respondents
A total of 145 (105 males, 39 females) out of 289 conference attendees responded to the questions about
the marketability of BCIs. Seventy three persons were aged between 18 and 30, sixty-nine persons
between 31 and 55 and two persons were aged between 56 and 70. The sample consisted of experts from
various disciplines (e.g. neuro-, computer or cognitive scientists, electrical engineers, psychologists etc).
One participant did not give demographical data.
2.2 Survey
Participants completed an online survey that included four questions on the expected marketability
of di erent purpose BCIs. In the survey, for convenience, we ordered some BCI types based on their
function: 1) BCIs for healthy users, 2) BCIs as assistive technology, 3) BCI-controlled prostheses and 4)
BCIs as therapy tools.
Speci cally, participants were asked to indicate when they expected to see these types of BCI systems to
become available on the market. Participants could choose from 5 answer options: "never", "between 0-5
years", "between 5-10 years", "more than 10 years" or "it already exists on the market". Participants
who indicated that this type of BCI already exists on the market were asked which group or company
o ers the product.
2.3 Development of a repository of BCI companies
Only companies which o ered BCI products and services were included in the repository. Thus, compa-
nies supplying hardware and software needed for BCI research were not included. The repository was
build through: 1) web searches, 2) written and verbal interviews of experts and 3) iterative postings on
LinkedIn groups ("brain-computer interface group", "neuromarketing" and "BrainGain").
3 Results
3.1 Respondents' expectancies of marketability of BCI
Table 1 presents the respondents' expectancies per BCI class.
BNCI technology for: already
on
market
0-5 yrs 6-10 yrs > 10 yrs never Number
of resps
healthy users 26.6 % 44.8 % 13.8 % 12.4 % 1.4 % 145
users who need AT 16.6 % 42.1 % 33.8 % 7.6 % 0.0 % 145
users who need prostheses 0.7 % 20 % 43.4 % 34.5 % 1.4 % 145
users who need therapy 9.9 % 39.4 % 39.4 % 10.6 % 0.7 % 142
Table 1: Overview of percentage of respondent which indicated when BCI applications will enter (if ever)
the market for healthy users, users who need assistive technology (AT), prostheses and therapy. The last
column shows the number of respondents (resps) who rated the item
3.2 Respondents' knowledge about existing companies
Thirty-two respondents reported which companies they knew that already marketed BCI's for health
users. By far the best known companies are Neurosky (counted 13 times) and Emotiv (counted 14 times).
Other companies that respondents mentioned were: Hitachi, Ambient, OCZ technology, Starwars Science,
Mattel, inc., g.Tec, Brain Actuated Technologies, InteraXon, Zeo, inc. and Interactive Productline.
Twenty-two respondents reported which companies they knew that already marketed BCIs as AT. The
best-known company is g.Tec (counted 16 times). Other groups and companies that respondents listed
were: Ambient, Brain Actuated Technologies, OpenVibe and BCI2000. Nine respondents commented
2
that they knew BCI's as therapy tools already are on the market, but only two gave concrete company
names: Brain master Technologies, inc. and EEGinfo. None of the respondents identi ed a company
that already markets BCI-based prostheses.
3.3 Preliminary overview of companies and marketability
We summarized a preliminary overview of 28 companies related to Brain-Computer Interfacing (see gure
1), which we expect to be more complete at the time of the Graz conference. Currently, neurofeedback
companies are not yet listed in the repository. We invite all readers to comment on this overview
and continue completing and criticizing the overview so we can o er the European Commission the
most accurate summary possible. This overview currently shows that as many as 10 companies market
BNCI's for entertainment and gaming. Six companies o er BNCI as assistive technology, although two
of these companies (Neural Signals, inc and BitBrain) do not yet market products. However, there
is a big di erence in the number of products sold between the two markets. Neurosky, for example,
has sold approximately 1 million Mindsets, whereas g.Tec has sold between 30-40 Intendixes (personal
communication).
Our survey did not address neuromarketing in depth. We counted six companies in this area. BCI
researchers (N=144) have divided opinions on whether neuromarketing technologies do (35.4 %) or do
not count (47.9 %) as BCI systems (Nijboer et al, in review). Nevertheless, neuromarketing developments
in
uence BCIs, and it is predicted that this market will grow tremendously [5, 2]. Finally, two companies
(Brain Fingerprinting inc. and No Lie MRI) provide BCI services for criminal investigation.
BNCI
technologies
health
sector
entertainment
sector
financial
sector
ICT
educational
sector
neuromarketing
serious games
humancomputer
interaction
multimodal
interaction
gaming
art
science
realtime
analysis
neuroeconomics
Interactive
Productline
Existing companies
Zeo, inc.
Neurosky
Emotiv Mattel, inc.
Uncle Milton
PLX
BCI net
Mind
technologies
Neuro-insight
The Mindlab
Advanced
Brain
Monitoring
Sands
technologies
Cortech
solutions
Emsense
safety
security
jurisdiction
BitBrain
technologies
InteraXon.
InteraXon.
nutrition
g.Tec
Brain actuated
technologies
Ambient
wellness
prevention
monitoring
therapy
assitive
technology
addiction
disorders
Neural
Signals, inc.
image
detection
BitBrain
technologies
ambient
intelligence
Brainfingers
Brain Fingerprinting
Laboratories, Inc.
Figure 1: Preliminary overview of existing companies
4 Discussion
The survey results show that respondents were generally optimistic about the near-term marketability of
BCI's for healthy users and user in need of AT. Repondents were slightly less optimistic about BCI for
therapy and expect BCI-controlled neuroprostheses to require more than 10 years to enter the market.
Our preliminary overview shows that companies - more than respondents realize - already market BCIs
for healthy users and users who need AT. However, having a product on the market does not yet mean
that a product is sold successfully. BCI research toward AT has a history of several decennia yet 30-40
products were sold in g.Tec. BCI applications for healthy users are available since less than 5 years and
(at least) more than a million products have been sold. It is remarkable that 72.4 % of the participants
are not aware that BNCI technologies for healthy users are already on the market. Since 10 companies
already exist in this direction we argue that optimism for this type of applications is quite appropriate.
3
In our opinion BNCI for assistive technology is not likely to create large economic value, unless it
becomes available for a broader range of categories of (more numerous) end-users. The societal value of
AT-BCI products is enormous, since they can enable users with disabilities to participate in the society
at interpersonal and professional levels. However, we predict that the research and development of BCI
applications targeted at the general public will more easily mobilize the industry and facilitate the tech
transfer between universities and industry. This will lead to more acceptance of BNCI technology in
the public and possibly to a spillover to the AT industry. In addition, even though the market for BCI-
controlled prosthetics requires a more time-consuming trajectory, ultimately we expect these products
to create large economic and societal value similar to Deep Brain Stimulation, because the number of
potential end-users is relatively high.
The survey reported here contains some
aws. First, since we could not make the entire survey too time-
consuming, some questions were generalized. For instance, we did not specify whether the BCI products
would work with non-invasive or invasive brain signals nor did we consider the usability of di ent BCIs.
However, these speci cations would in
uence the cost, number of available users, training time, time to
market and many other factors. Second, the respondents were mostly people involved primarily in BCI
research and had the time, funds, and enthusiasm to attend a major conference, so the sample may be
positively biased. fBNCI in future will survey a broader range of stakeholders, including more experts
from human-computer interaction and the industry.
5 Conclusion
Although we view these data as a rst crude estimation of the elds' perception of marketability of BCIs,
we cautiously conclude that BCI researchers seem optimistic about how much time it will take before the
rst BCIs for non-medical purposes and AT enter the market. We agree with them that the market for
BNCI applications for healthy users is promising, but we are not convinced that the market for AT-BCI
applications is equally promising simply because the number of end-users is relatively low in comparison.
Moreover, the limited knowledge of respondents about existing companies should raise awareness in the
BCI community that tech transfer is already well established largely without the involvement of the
scientists who consider themselves the core of the BCI eld. A more pro-active role of BCI scientists in
tech transfer would be desirable for high quality product development.
6 Acknowledgements
The authors gratefully acknowledge the support the Future BNCI project (Project number ICT-248320)
and of the BrainGain Smart Mix Program. We thank all respondents and Thorsten Zander, Jan van
Erp, Yann Renard, Fabien Lotte, Ariel Garten, Tom Sullivan and Christoph Guger for valuabe input.
References
[1] A. Nijholt and D. Tan. Brain-computer interfacing for intelligent systems. Ieee Intelligent Systems,
23(3):72{72, 2008.
[2] Brendan Z. Allison. Toward ubiquitous bcis. In Bernhard Graimann, Gert Pfurtscheller, and Brendan
Allison, editors, Brain-Computer Interfaces, The Frontiers Collection, pages 357{387. Springer Berlin
Heidelberg, 2010.
[3] F. Nijboer, J. Clausen, B. Z. Allison, and P. Haselager. Researchers opinions about ethically sound
dissemination of bci research to the public media. International Journal of Bioelectromagnetism, in
press.
[4] F. Nijboer, J. Clausen, B. Z. Allison, and P. Haselager. The asilomar survey: researchers opinions
on ethical issues related to brain-computer interfacing. Neuroethics, in review.
[5] Z. Lynch. The Neuro Revolution: How Brain Science Is Changing Our World. St. Martin's press,
New York, 2009.
4
No comments:
Post a Comment