Physiological and Behavioural Biometrics
Biometrics literally means ‘body measurement’ of any type, but increasingly the term is used in the narrower sense of ‘body measurements and calculations related to human characteristics that can be used to identify and/or group people based on a single or multiple biometric attributes*. These modalities range from ‘physiological’ attributes through to ‘behavioural’ traits and beyond and they operate across the broad spectrum of biometric use cases and applications. In general terms:
Physiological biometrics employ physical, structural, and relatively static attributes of a person such as their fingerprints, the pattern of their iris, contours of their face, or the geometry of veins in their hands. Some modalities are microscopic in nature but still exhibit biological and chemical structures that can be acquired and identified e.g., DNA and odour.
Many physiological biometrics are permanent features and are resistant to change unless they are accidentally or intentionally damaged, degraded or destroyed. Examples, in the case of fingerprints, might include the partial erosion of the friction ridges on the hands of those engaged long term in handling or laying bricks, the loss or amputation of hands/fingers and, in extreme cases, invasive surgery or self-harm to remove or alter the ridge structure.
Behavioural biometrics establish identity by monitoring the distinctive characteristics of movements, gestures, and motor-skills of individuals as they perform a task or series of tasks. This means human movements such as walking (gait analysis) or finger contact with a keyboard (keystrokes) are captured and analysed. Structural physiological factors will be fundamental in determining how such movements are performed by an individual e.g., proportional length of the feet, thigh and shin bones of the leg, the relative size and dexterity of hand/fingers etc. A behavioural biometric system measures how an individual conducts an action or series of actions and then uses this data (user profile) to compare a subsequent performance of the same actions by a person to determine if he/she matches the user profile or is an imposter. The actions in question may be normal, routine functions e.g., interacting with a smartphone. It should be kept in mind that the user profile is not strictly a biometric template, which is a digital representation of a physiological feature, but a record of information, actions and settings associated with the user.
New biometric modalities are currently being developed that challenge even the broadest definition of behavioural biometrics. For example, heartbeats can be used to identify individuals, but they are an involuntary biological process and not an intentional or manipulated action performed by the individual in the same way as, for example, keystrokes. Therefore, behavioural biometrics encompasses a variety of modalities that exhibit both voluntary and involuntary repeated motions and associated rhythmic timings/pressures of body features ranging from signatures, gait, voice, and keystrokes through to eye tracking and heartbeats.
* Some applications have been developed using software and algorithms that are significantly different to standard biometric recognition systems. They are designed primarily to group people rather than identify them individually. These systems may be used to estimate factors such as age, gender or race or evaluate the mood, emotional state, or attentiveness of individuals and/or groups. The highly variable accuracy rates of these types of applications have caused widespread concern and much public debate regarding fundamental human rights, privacy, and the acceptable limits of such technologies in civil society (Refer to the Biometrics Institute Good Practice Framework B.1.1, C.1.1/2/3/4 and B.5.1) and the Biometrics Institute Three Laws of Biometrics – formulate policy then processes and then the technology).