Remote eye tracking
Modern eye trackers record eye position and movement based on optical tracking of corneal reflections, thereby allowing the analysis of gaze positions or eye movements on a 2-D screen or in 3-D environments. One of the first eye trackers was built in 1960 and contained a large metal stand for the respondent’s head as well as two full-sized cameras pointing towards each eye.
Since then eye trackers have undergone a technological revolution – modern eye trackers are hardly any larger than shampoo bottles and don’t require the respondent to sit completely still. Besides gaze position, most eye trackers today also monitor pupillary responses and the distance to the screen, providing additional insights into cognitive and emotional processing.
Using eye tracking in human behavior research unlocks several interesting measures, such as attention, interest and arousal. What exactly do we find interesting in videos and pictures? What elicits our attention?
These are only a few of many questions that can be answered by using eye tracking with appropriate software. Drawing heat-maps of accumulated gaze positions overlaid on the original stimulus, reviewing aggregated gaze data and gathering information from customized areas of interests can all help obtain answers to these questions. Besides, pupillary responses can also be used as measure for arousal.
While most peripheral sensors measure body signals that are modulated by cortical processes, direct brain recordings provide direct access to the electric processes underlying human cognition and behavior. Several brain-imaging technologies exist but only few of them qualify for human behavior research. One of the most groundbreaking technologies is electroencephalography (EEG), mainly due to its excellent time resolution, low cost and portability.
EEG electrodes mounted in flexible caps record the electrical voltage distribution across the scalp, which is considered to be generated by neural synchronization inside of cortex, which results in electric fields that are propagated towards the surface. Recent progress in online signal decontamination and feature extraction allows for real-time computation of several core metrics such as cognitive workload or engagement based on the recorded EEG raw data.
While advanced systems allow for customized adjustment and individualization of the transformation matrices from raw data to metrics, several EEG devices even provide pre-calculated metrics without further adjustment, allowing you to jump straight into data collection.
fNIRS (functional Near-Infrared Spectroscopy) records the diffusion of near-infrared light by human skull, scalp and brain tissue, allowing researchers to monitor cerebral blood flow in specified brain regions.
While fNIRS is a relatively new technology, it has already proven to be a very promising tool in human behavior research due to its non-invasive, portable and inexpensive character. fNIRS can, for example, be used to monitor cerebral blood flow in frontal cortex as an indicator of cognitive workload (prefrontal activity) or motivation (prefrontal asymmetry).
In the upcoming months and years the application bandwidth for fNIRS will further grow, allowing you to answer more research questions on human behavior and cognition.
Whenever you would like to accomplish brain imaging with excellent spatial resolution, fMRI (functional Magnetic Resonance Imaging) is the method of choice. An fMRI scanner uses magnetic fields and radio frequencies to measure cognitive processing based on changes in cerebral blood flow in specific regions of the brain.
Further, fMRI can be used to generate structural scans of high spatial precision, representing an accurate and highly precise 3-D rendering of the respondent’s brain. Structural scans obtained with fMRI are particularly useful when combining them with surface EEG recordings. In this case, the spatial precision of the fMRI meets the temporal resolution of the EEG – both combined allow for sub-second reconstructions of generator sources of brain activity associated with cognitive or behavioral processing, which is not possible using each method on its own.
Galvanic skin response (GSR) – also known as electrodermal activity (EDA) or skin conductance (SC) – measures electrical conductance of the skin from body parts that are sensitive to emotional processing or physiological arousal, such as hands or feet.
GSR is quite popular in human behavior research due to its low cost, extremely fast setup time, as well as low intrusion during data acquisition
There are two important factors of facial expressions. One that we all know is that emotional reactions will elicit facial expressions. However, only few people know that facial expressions will elicit emotional reactions. Take a look at a random person and smile and there is 90% chance that the person will smile back at you and feel a little happier in that moment.
Automatic facial expressions analysis is without any doubt one of the most popular technologies within human behavior research since it does not require any sensors being attached to the respondent. Furthermore, facial expression algorithms allow for a more detailed insight into emotional processing as reflected by stimulus-driven or self-elicited emotional reactions such as joy, anger, fear, surprise, confusion etc.
Combining facial expression analysis with technologies that monitor arousal (e.g., pupil dilation, EEG, or GSR) allow for further insights into the intensity of the expressed emotions. All you need to get started is a webcam and the appropriate software.
Electromyography (EMG) records electric activity generated by muscle contractions. Whenever a muscle contracts a burst of electric activity is generated which propagates through adjacent tissue and bone and can be recorded from any neighboring skin areas. Importantly, EMG activity is linearly related to the amount of muscle contraction and the number of contracted muscles.
However, EMG activity (as for instance recorded with the Shimmer EMG sensor) is even measurable when we do not display obvious actions, for example, when we control our body to not perform certain behaviors. This renders EMG an excellent technique to monitor cognitive-behavioral processing in addition to behavioral observations alone.
The electrocardiogram (ECG) is mainly known from hospitals and medical environments, measuring heartbeats and rhythms for the initial diagnosis of cardiac diseases. ECG measures the electrical activity of the heart using electrodes placed in a circular fashion on the chest.
These electrodes register the subtle electrical voltage changes on the skin that arise from the heart muscle. The signal output can then be interpreted as heart rate (units: beats per minute, bpm). Besides the electric ECG signal, optical heart rate (also referred to as photoplethysmogram or PPT) can be obtained by means of a finger or ear clip (for sure you have seen this device in movies or shows). Heart rate, either obtained via ECG or PPT, provides an excellent window into autonomous processes associated with arousal and emotional processing, and due to its ease of use and simple application it is finding its way into human behavior research.
Self-reports constitute the most traditional methodology in human behavior research. They allow respondents to report their feelings, emotions, likings, preferences etc. based on introspection. This is mainly accomplished using questionnaires and surveys.
Many modern human behavior labs employ biometric measures instead of self-reporting, since they believe that they provide more unbiased and quantitative data. However, in an ideal scenario biometric measures are combined with survey data to identify relationships between self-reports and physiology and gain deeper insights into underlying factors driving human cognition and behavior.