Find Us

123 Main Street
New York, NY 10001

Monday—Friday: 9:00AM–5:00PM
Saturday & Sunday: 11:00AM–3:00PM

Behavioral biometrics: signature analysis | cognitive biometrics

TypingDNA presents the various types of biometrics and explores their use for authentication and beyond. In this blog post, we look into signature analysis and cognitive biometrics.

Signature analysis 

Even since the use of papyrus in ancient times, the way people type has been a fascinating topic. Graphology is looking at the physical characteristics and patterns of handwriting. The first book on handwriting analysis appeared in 1575 and belonged to the renowned scholar, Juan Huarte de San Juan. Still, the British Psychological Society ranks graphology alongside astrology, giving them both “zero validity”. 

Nowadays, graphology has been placed in a secondary role. Instead, the technology of signature analysis is made with special software and is currently used to analyze human signatures and evaluate shapes, contours, and movements used to create a personal trademark. Unlike graphology, the technology is not trying to find any information in the various shapes but rather compares samples to make sure it’s the same person behind. When forged, signatures are made more slowly, thus indicating the fraud. A so-called motion-versus-time is analyzed to make sure signatures are genuine.  

The use cases are multiple, from banking, government, insurance to health-care industries. In most of the cases, the signature analysis was intended to prevent identity theft but has been a modest solution to combat fraud.

The disadvantages are multiple: a person will always have variations in its signature, so any electronic signature will differ from the previous one causing a chain of false positives or false negatives. 

Given the latest innovation in biometric authentication variants, the method can’t be regarded as a reliable anti-fraud solution. 

Cognitive biometrics 

Although more likely to belong to a Matrix movie script, an alternative method to the user authentication, cognitive biometrics has been introduced by Kenneth Revett of the British University in Egypt. The topic is largely discussed in the International Journal of Cognitive Biometrics. This method is based on acquiring information from users by collecting one or more biosignals by presenting users with one or more external stimuli, such as images. There are various techniques to obtain responses. Some include an electroencephalogram (EEG), electrocardiogram (ECG), electrodermal response (EDR), blood pulse volume (BVP). 

Gaze interaction is situated somewhere between cognitive and behavioral biometrics. In the past years, eye trackers or pupillometry have gained popularity among non-conventional authentication methods — the latest innovations in the field show and advancement in analyzing eye movements based on computer interaction. Among the advantages of gaze interaction, we list user-friendliness and reliability due to its, so far impossible way to mimic physiological and psychological response. 

Still, this method can not yet be trusted for stand-alone usage, so in most cases, it is combined with other biometrics such as keystroke dynamics and mouse use analysis.