The Growing Use of Biometrics in Education

A biometric is a naturally occurring body feature that can be measured. The National Institute of Standards in Technology (NIST¹) has three definitions:

A measurable physical characteristic or personal behavioral trait used to recognize the identity, or verify the claimed identity, of an applicant. Facial images, fingerprints, and iris scan samples are all examples of biometrics.

Automated recognition of individuals based on their biological and behavioral characteristics.

The science and technology of measuring and statistically analyzing biological data. In information technology, biometrics usually refers to automated technologies for authenticating and verifying human body characteristics such as fingerprints, eye retinas and irises, voice patterns, facial patterns, and hand measurements.

Biometrics have a long history (with fingerprinting becoming standardized in the 19th century, facial recognition in the 60s², typing biometrics in the 80s³, and others in decades since) widespread use and application, including healthcare, business, and government. Their use and application varies significantly from increasing convenience and efficiency in login processes to identifying criminals in public spaces. In each use, they present interesting tradeoffs between privacy, safety, and convenience. Cybersecurity and identification verification is a prerogative of any organization that is working to balance safety and privacy.

More and more technologies like biometrics (leveraging aspects of the human body or body mechanics to verify identity) are working their way into everyday life at work, home, and school. The allure of biometrics is their ubiquitous availability and universal uniqueness. In fact, you might even have used a biometric to unlock your phone and read this by using facial recognition or your fingerprint, a habitual activity you might do 50 or more times a day (each time saving you the time and effort of typing in a password or pin). It’s no surprise that biometrics are becoming more and more common due to the ease of use, low failure rates, high levels of security, and difficulty in mimicking or replicating.

These tools are used already in many corporate settings to safeguard trade secrets, access to facilities, create role based access, and identify unauthorized individuals. In an era where text, audio, and video can all be generated by AI, the use of these in virtual environments alongside physical ones can help to ensure authenticity and security. Even in fully remote workplaces and digital environments, they are often used to ensure the security of sensitive systems such as electronic medical records and financial information.

Use on Campus

The modern campus (both K12 and College) leverages biometrics to ensure safety and security on campus while also providing students seamless and frictionless access to facilities, buildings, events and more. On the one hand: biometrics offer ease of use and verified permissions to students without the hassle of human verification, gating, or lengthy checking processes. On the other, these same detection protocols have turned educational institutions into latent connected surveillance systems.

Whether you’re a student, a teacher, an administrator, or a citizen: verification of your belongingness and access rights and place is happening regularly. You may have experienced an enhanced convenience and frictionless access to the lunch line, have had a proctoring session flagged, or been locked out of your computer or phone. Each use case might employ a different biometric.

Here’s a list of the most widely used biometric validation tools and techniques and how they’ve been introduced on campus:

Facial Recognition

Managing access for authorized individuals (enrolled students, teachers, staff) is obviously a major priority of schools today. Facial recognition for device authentication is already in widespread use for laptops, tablets, and phones (including those provided by the institution).

Already in K12 schools facial recognition systems are used to identify students for use in attendance and for campus safety. On college campuses, facial recognition is being introduced not necessarily for security but to increase student success by measuring attention in classrooms⁴. This nascent application could provide more real time insights into student engagement, attendance, or identify students at risk for stopping out or course failure.

In online proctoring, automated AI systems utilize facial recognition to flag the presence of additional faces on camera. Computer vision also provides proctoring providers with eye movement tracking and other on camera behavior as markers for potential cheating.

Fingerprint, Handprint, Hand Geometry⁵

Perhaps the most useful and widespread in use today, fingerprints, palm prints, and hand geometry provide a growing level of convenience for turning on a computer, phone, or even paying for lunch. You likely are already keying into your phone with a thumb or finger print.

Voice Recognition

Text to speech assistants like Alexa are based on voice recognition and increasingly can be programmed to recognize and differentiate voices. Popular campus lecture capture technologies (like Panopto) utilize voice and speech recognition to identify speakers and correctly attribute captioned text (this is an important feature to ensure accessibility standards).

Again, in online proctoring voice and speech recognition may be utilized to identify multiple speakers within the test takers environment to flag that for further review.

Keystroke Dynamics

How you type on a computer keyboard is a biometric signature. This fact is one of the underlying foundations for our work at Cursive Technology. In cybersecurity, the typing biometric is a great endpoint protection tool (we use it on our computers⁶).

On campuses the most common application is either in the learning management system as a gating feature or during online proctor service onboarding (used to speed up authentication for repeat usage through a quick gating check). In both cases, this is used for identity verification ensuring that the right student is logged in or accessing an assessment/resource.

Researchers also collect this biometric data for its value in learning science research. In the last few decades, this data has been collected only in clinical settings (on computers configured to capture this data in labs). Based on that collected and labeled data, researchers have used and linked keylogging data to measure cognitive load, estimate writing quality, measure proficiency/fluency, measure stress, measure dishonesty, and more.

With Cursive’s tools we enable keystroke data collection for educators and students via both the learning management system and a free browser extension. While authorship verification from session to session is possible and available, teachers and students also benefit from detailed analytics and a growing number of insights based on published research and proprietary research conducted by our team. With new challenges of identifying AI generated content, this data is a rich source of information to verify student writing and effort in the context of their submissions.

Iris Recognition

High-security areas like research labs could use iris scanners for access due to their high accuracy. Campuses with high tech and high security research labs (such as those which house highly dangerous biological agents⁷) are the most likely to incorporate these types of biometrics into their security configurations, though the use cases are likely limited since there are easier ways to process identity verification.

Gait, Vein, Ear, Heartbeat, & Electroencephalogram

Now, there are additional biometrics available but most don’t have applicability to even the most modern campus. This other bucket of biometrics has viability for use, but campuses are unlikely to have them employed in the near term. These biometrics include Vein, Gait, Ear Shape, Heartbeat, and EEG, all of which can be measured.

Gait recognition is being trialed by policy for use in identifying individuals, especially in crowds or buildings. While possible, there is not widespread adoption⁸ and is likely to be met with resistance on campuses outside of its policing and public safety applications.

For body signals like heartbeat and electroencephalogram (EEG) the uses of these could become more interesting with the rise in wearable technologies and advances in human-computer interfaces (HCI). Authenticating using your mind is a wild idea, research into this field is nascent and will be a space to watch in years to come.

Conclusion

The science of measuring body characteristics and functions has created truly amazing applications of technology. Their use in any market attracts justified scrutiny for the companies that make use of and store biometrics. As a result, the sensitive nature of biometric data requires high standards for cybersecurity and privacy. As an end user you should be aware when and if biometric data is being collected, how it is used, and whether you have a claim to it (or the removal of it). Additionally, privacy policies and end user license agreements for companies collecting or using this type of data should always clearly state how they can utilize it (including transferring it, selling it, deleting it, etc.).

While many are used as identification tools, they also can be used to create efficiency, tell us about the learning process, make common interactions more convenient and increase trust. At Cursive, we’re most excited at the prospects of accelerating writing skills development by providing on demand, objective, quantifiable analysis back to writers. We believe that this data could help both teachers and students understand the level, duration, and intensity of effort in the writing process, encourage better writing habits, and build student confidence alongside the human feedback provided by a tutor, teacher, or peer.

In the era of generative AI content, AI agents, and deep fake technology, the use of biometrics is surely going to increase, giving us all opportunities to trade privacy and participation for authenticity, safety, trust, and convenience.