Emotions and behaviors are intrinsic to human nature. When we delve into understanding an individual’s emotions and reasons, we embark on a journey to comprehend the human psyche better. This journey, however, has its challenges. It raises pertinent questions about the scientific basis of our analyses and the factual evidence supporting them. Some of the questions that often arise include:

  • Does our blood pressure surge when we are engulfed in anger?
  • Does our body language alter when we are anxious?
  • In strategic games like poker, are players conscious of their body language, striving to regulate it to avoid giving away their strategies?
  • Can our feelings of affection for someone be visibly discerned by others?
  • Are the above queries universally accepted, or are they subjects of debate?

The answers to these questions provide insights into the human condition. Analyzing such indicators can offer a deeper understanding of human emotions and reactions. The crux of the matter is not just in the analysis but in interpreting the data derived from it.

Methodology

However, it is essential to differentiate between biomarkers used in medical studies and those used in behavioral analyses. For instance, while incessant blinking might indicate an eye ailment, it does not necessarily rule out other underlying issues. Medical studies primarily focus on identifying biomarkers that signal specific health concerns. Using these biomarkers to gauge stress or emotional states might yield inaccurate results.

On the other hand, non-medical clinical studies aim to understand an individual’s mental state within a particular context. Such studies often reveal reactions that might not be universally understood or accepted. Numerous studies conducted by psychologists, psychiatrists, and other medical professionals have found correlations between specific behaviors—like changes in body language, heart rate, blood pressure, eye movements, and even the frequency of eyelid blinking—and the body’s responses initiated by the central nervous system. While some reactions can be consciously controlled, others are automatic and involuntary.

Consider the example of sudden braking in a vehicle. Even after the immediate danger has passed, the heart continues to race, a manifestation of the body’s “fight or flight” response. If humans were akin to machines, braking would be our sole reaction. Specific reactions, like the facial expression after tasting something bitter, can vary based on age and cultural influences. A child might instantly display a reaction uninfluenced by societal norms, whereas an older individual, molded by cultural expectations, might suppress it.

Such behaviors have been extensively studied using non-medical clinical methodologies. These studies often employ medical equipment to measure physiological responses like heart rate, blood pressure, and body temperature. By integrating this vast array of data, including countless variables, into neural networks, researchers aim to discern correlations between an individual’s state of mind and their reactions, enhancing our understanding of human emotions and behaviors.

The Array of Tests and Experiments

To facilitate advanced machine learning and neural network training, a series of tests and experiments were designed. These were categorized into distinct factors, each scrutinized individually. Through autonomous machine learning and empirical testing, the results were validated. The factors included:

  • Pigmentation Analysis: This detected minute changes in skin tone to identify discoloration due to blood circulation in response to stimuli.
  • Sound Analysis: This assessed changes in sound range, volume, speech rate, and response times.
  • Eyelid and Blink Analysis: This gauged the rate and speed of responses to stimuli.
  • Body Language Analysis: This observed changes in facial and body movements about stimuli, comparing them with concurrent movements like head tilts and leg motions.
  • Thermal Analysis: This monitored temperature fluctuations in various body parts to track blood flow changes in reaction to stimuli.
  • Emotional Analysis: This recognized facial expressions representing specific emotions.
  • Medical Indicators: Using standard equipment, heart rate, blood pressure, and salinity were measured.

Objectives

The primary goal was to automate neural network-based identification and analysis. The process was divided into stages:

  • Identification Phase: Data extraction from images/sounds and detecting patterns.
  • Cataloging Phase: Associating changes with meanings like cognition and emotion.
  • Correlation Phase: Comparing values from different factors obtained during testing.
  • Pattern Phase: Recognizing recurring patterns to infer the subject’s condition.
  • Algorithmic Phase: Formulating computational methods for data processing.
  • Analysis Phase: Weighing the data to produce a comprehensive report.

Stimulation

Subjects were exposed to various stimuli, such as cognitive and emotional questions, images, and sound ranges.

Repetition and Recovery

Breaks were given between and during tests for subjects to recuperate. Tests were repeated to assess consistency over different times and conditions.

Subjects and Cross-Referencing

Initially, 150 individuals from diverse backgrounds participated. They were divided into groups of 30, analyzed separately, and then cross-referenced. Human experts also validated the computerized results.

Experts

The professionals involved included clinical psychologists, medical engineers, neuropsychologists, and polygraph experts.

Data Validation and Accuracy

Data extraction capabilities ranged from 85% to 96%. AI-analyzed prediction accuracy varied between 72% and 93%. The patented CRD model architecture had an accuracy range of 83% to 94%. Ambiguities arose from varying human emotional reactions, which were addressed in the final report.

Ongoing Model Training

Given the myriad of variables influencing human analysis, machine learning is an ongoing process. Data from specific fields will be collected and cross-referenced to refine results for each test.

About Revealense

Revealense is a pioneering deep-tech startup disrupting human behavior analysis based on a patented methodology of advanced psychophysiological stress monitoring, multicultural behavior mapping, and neuropsychology.

Revealense enables the illumination of human business interactions, thereby generating growth, efficiency, diversity, and innovation for firms. Understanding human behavior is at the core of meaningful partnerships, trust, and a culture of innovation because it is people who drive a firm’s success. Revealense provides a comprehensive and in-depth understanding of a person’s mental and psychological state in the context of each case, situation, and event.

We achieve this through ‘Revealense illuminator® insights,’ our deep learning neural network-based responsible AI. It provides an objective and fair evaluation, free of discrimination and bias from communication lapses.

These insights allow accurate decisions that are made every day by various industries, such as HR, banking, insurance, and public safety, helping them to understand their clients and employees better.

Using video analysis of a person’s reaction that decodes its meaning in the context of a given situation, we provide comprehensive insights into human emotions and cognition, as well as advanced stress monitoring for better and more accurate decisions.