Frequently Asked Questions

Contact us if you've got more questions and we'll do our best to answer.

Mental fatigue is made possible through the use of behavioral biometrics. Behavioral biometrics is the science that seeks to identify and characterize individuals based upon their behavior while carrying out daily tasks. Its underlying assumption is that each individual behaves differently and that these behaviors are unique enough to identify an individual with a satisfactory degree of certainty and to characterize their state through changes in these behaviors. By analyzing user interaction patterns obtained from the use of the mouse, keyboard, or trackpad, and through machine learning techniques, it is possible to establish a baseline and determine, for example, if the user is making more mistakes or performing with less accuracy. This allows Performetric to identify and quantify mental fatigue.
Performetric is the result of an intensive research work in which validation is one of the most important steps. Performetric’s Patent Pending System has been validated and correlated against accepted measures of fatigue and cognitive performance:
  • Automated Neuropsychological Assessment Metric (ANAM) is a library of computer-based assessments of cognitive domains including attention, concentration, reaction time, memory, processing speed, and decision-making. ANAM has been administered nearly two million times in a variety of applications and settings. ANAM provides clinicians and researchers with tests to evaluate changes in an individual’s cognitive status over time;
  • United States Air Force School of Aerospace Medicine (USAFSAM) Mental Fatigue Scale is a golden standard for self-assessment questionnaires for mental fatigue;
  • Electroencephalography (EEG) is a technique that has been used often for the detection of patterns of mental fatigue as well as for the analysis of cognitive overload.
Performetric uses the seven-point USAFSAM Mental Fatigue Scale, created in 1979 by Dr. William F. Storm and Captain (Dr.) Layne P. Perelli of the Crew Performance Branch of the USAF School of Aerospace Medicine, Brooks AFB, San Antonio, Texas, and then used in many field and laboratory tests. The scale items are:
  1. Fully alert. Wide awake. Extremely peppy.
  2. Very lively. Responsive, but not at peak.
  3. Okay. Somewhat fresh.
  4. A little tired. Less than fresh.
  5. Moderately tired. Let down.
  6. Extremely tired. Very difficult to concentrate.
  7. Completely exhausted. Unable to function effectively. Ready to drop.
Levels 1-3 mean that the individual is not tired, while levels 4-5 indicate the the individual is experiencing some fatigue and levels 6-7 that the individual is really tired which could lead to an unsafe or health-damaging scenario.
No. Performetric only uses keystroke dynamics, i.e., how you type, and never knows what you write or even which key you press. Instead, Performetric relies on information such as writing velocity and rhythm or key latency.
Performetric collects data about the user’s interaction patterns with the peripherals (e.g. key latency, writing velocity, movement precision) to characterize interaction performance and quantify fatigue. This may allow to classify, with some degree of certainty, the type of task being carried out (e.g. writing, browsing). Other than that, Performetric does not collect any more specific data about the tasks being carried out.
Performetric only stores historical data describing the mental fatigue of each user (the classified values ranging between 1 and 7), as well as the user’s interaction patterns and profile information. We do not collect or store written text, websites, passwords or any other critical information. For more information about privacy, please check our privacy terms.
Yes you can, although you cannot use more than one computer/machine simultaneously, and each individual computer/machine will go through a learning phase.
Performetric uses a machine learning algorithm to quantify mental fatigue that requires some data from the user in order to know her/his interaction patterns. Specifically, the algorithm requires a minimum of 24h of interaction in order to create a baseline. After this learning phase, the system will continue to collect data in order to adapt to the user.