Q Sensor 2.0

Emotion data, no strings attached

With the Q Sensor, you're not confined to the lab. The curve-shaped Q Sensor is designed to be worn on the wrist so it is comfortable and unobtrusive to wear all day at work, play, or sleep. This makes it ideal for long-term measurement in clinical and therapeutic research.

Easy and intuitive

Data collection with Q is easy; no messy gels to apply, no wires to tape, and nothing to configure. Simply strap on the Q and it turns on automatically and begins collecting data. Take it off, and it will turn off automatically after two minutes to conserve battery life.

• 24-hour battery life when logging
• 5-hour battery life when streaming
• Bluetooth wireless connectivity
• Storage for 3 months of data
• Downloadable data visualization software (see "Software" tab)
• Data exportable to standard formats

Emotion data, no strings attached

With the Q Sensor, you're not confined to the lab, and with the Q Pod you're not confined to measuring on the wrist either. The size of a box of matches, the Q Pod allows for maximum flexibility, and comes with bands to support measurement on both the palm and wrist. This makes it ideal for short-term measurement.

• 24-hour battery life
• Storage for 3 months of data
• Downloadable data visualization software (see "Software" tab)
• Data exportable to standard formats

Before you buy the Q Sensor lets explore whether it will work for you:

How does measurement on the wrist compare to the fingers?
The traditional place to measure EDA is off the fingertips or palmar surface, but there the sensor interferes with computer use, hand washing, and many other daily activities.  Thus, the palmar surface is not recommended for long-term real-world data measurement outside the laboratory.  The wrist is a relatively new place to measure EDA, and under the majority of tests we’ve seen to date, it gives highly correlated phasic data to the palmar surface.

In one of the most thorough tests done (using the Q’s predecessor device created at MIT, and published in a top peer-reviewed journal) on the wrist (distal forearm) site wearing the new sensor was compared to the traditional fingertip (medial phalanges of index and middle finger) site wearing the traditional FDA-approved sensor.  Groups of people wearing both sensors were subjected to stressor tests of three kinds: (a) physical (13 people), (b) cognitive (15 people), and (c) emotional (13 people). (Poh et al.  “A Wearable Sensor for Unobtrusive, Long-Term Assessment of Electrodermal Activity” IEEE Trans on Biomedical Engineering, Vol. 57, No. 5, May 2010).  Both sensors were worn on the right side of the body (ipsilaterality is important because different circuits in the brain control the right and left EDA – in fact, the right hemisphere controls the right palmar responses, and the left controls the left – at least at the palmar locations; there is some evidence that this ipsilaterality does not hold down on the legs or feet.)

The skin conductance signals from the distal forearms were correlated with those from the ipsilateral fingers.  Based on the median of the correlation coefficients ˜r, correlation between the fingers and distal forearm was very strong (0.76 ≤ ˜r ≤ 0.96) during the baseline and recovery periods for all but the physical and emotional recovery period. There was also a strong correlation between the fingers and distal forearm during the physical (˜r = 0.78) and emotional (˜r =0.72) tasks. During the cognitive task, the correlations were lower, but remained moderately strong (˜r = 0.57).

While no two different positions on a person can be expected to have identical electrodermal responses even using the exact sensor on both sites, the Q sensor readings are highly correlated with the commercial FDA sensor readings under multiple test conditions, and the palmar (finger) and forearm (wrist) sites are also highly correlated during stressor tests made on awake participants.
m

Does EDA tell you something different than measuring heart rate or heart rate variability?
Yes. The heart is innervated by the two main branches of the Autonomic Nervous System:  (1) The Sympathetic Nervous System (SNS), commonly known as the “fight or flight” system, and (2) the Parasympathetic Nervous System (PNS), commonly known as “rest and digest.”  Most organs in the body are innervated by both the SNS and PNS. In order to measure the “excitement” or “arousal” dimension of emotion, the signal of interest is the SNS. The skin is believed by physiologists to be the only organ purely innervated by the SNS, making it the best bodily measure for what “revs your engine” or activates you.  (Note:  while SNS is often called the “fight or flight” system, it activates also with positive excitement and anticipation.)  Scientists currently do not have an accurate way to isolate SNS from measures of the heart.

m

What age groups do you recommend using this device? Is EDA different per age group? Gender? Nationality? Time of the day?

Currently the Q Sensor is being used for various studies that involve people from all age groups including infants. We do not recommend use on newborns or on premature infants where the skin is still developing and may be highly sensitive.

EDA differences have sometimes been reported for different age groups (lower skin conductance for many older people with drier skin, although this is not always the case for every participant) and occasionally for gender differences, although in most of our work we have not seen gender differences.  Remember that most studies in the literature were before the invention of the Q sensor and thus were only for short-term measurement; their results may no longer hold now that large numbers of people are can be measured properly long term. New normative data is needed to see if the old results (from short-term lab measurement) hold up.

EDA has been reported in some studies to show a circadian cycle, in infants and in college students.

m
Can I set alerts when arousal hits a certain level?

The current version of the sensor does not have this feature. However if an individual feels that he/she is highly aroused and wants to record that event they can do so by pressing the button provided on the sensor to mark that event.

m

Affectiva sells a wrist-package and a palm-package – how do I know which one to use when?
While both sites tend to be highly correlated, in some cases the wrist is more sensitive than the palmar surface, meaning it will give higher readings, and in some cases the palm is more sensitive than the wrist, meaning it will give higher readings. Also, the wrist tends to be significantly less obtrusive and have fewer motion artifacts for most activities. However, the palm tends to build up moisture faster when conditions are dry.  Below are some guidelines based on the measurements our team has made over the years.  You can always feel free to try both locations and see what works best.
If you are measuring people when they sleep, the wrist tends to be better than the palm – giving higher readings with more meaningful peaks (these occur frequently during non-REM sleep) and less noise.

Fig.4 Here is a typical night of sleep where the wrist signal (blue) shows stronger      peaks  and less noise than the palmar signal (brown). The gray shaded region is 90    minutes wide, the length of a typical sleep cycle.

If you are measuring people during wake, all-day, then the wrist (or even the lower leg) is better because it is less obtrusive and it will tend to have fewer motion artifacts.  Note that when using dry electrodes on dry skin under dry air conditions (e.g. air conditioning or low humidity air) then it’s important to have an activating exercise shortly after putting the sensors on ("see Tips for activating EDA when you first put sensor on” (below)) else it can take from minutes to hours (depending on dryness and arousal) for the signal to start registering responses.

If you are measuring people during wake, short-term, with dry electrodes in an indoor environment and don’t plan to have participants do anything really activating (perhaps all they are doing is viewing product descriptions and answering questions) then we recommend you measure off the palm. If you’d like to use the wrist under these conditions, then it’s important to do an activating exercise first to build up moisture, then let the person rest 10 minutes or do a calming task (like filling out forms or watching a relaxing video) before starting the task of interest.

m

Once you receive your Q Sensor these tips will help you get up and running:

How tight does the sensor have to be? (What is the optimum fit for the Q sensor?)

To prevent motion artifacts and excessive pressure on the electrodes the sensor should fit snugly around the wrist or palm. The fit should not be so tight that it constricts blood flow or turns your fingers purple.  If the fit feels uncomfortable then wiggle the sensor around to a better position or loosen it until it feels good.  Neither should the fit be loose enough that the sensor is able to rotate around freely on its own, as this will cause motion artifacts to appear in the data. If you see a lot of “spiking” like in the example here then the sensor is either too loose or is placed over a region that is moving a lot (where the sensor can pull away from the skin) and should be moved to a more stable, but still comfortable, position.

Fig.5 Brown=right leg, Blue = left leg, Motion=brown; Gray walking period up/down  stairs is maximum activation (9 minutes)

How can I elicit SCR’s to test my sensor?
You can usually elicit SCR’s by startling a person with a sudden noise, flash of light, or puff of air.  You can also ask them to tell everyone something significant happening in their life. Or you can give them tasks that might elicit cognitive, physical, or emotional effort: lots of mental math, jumping jacks, or something that makes them laugh.

Some psychophysiologists elicit SCR’s in themselves by rapidly sniffing, or by holding their breath (but only very briefly).
m

What is Baseline? How do I establish a baseline? For individuals? Multiple people?
Baseline is generally considered to be the average tonic level of an individual during rest conditions and in the absence of any discrete environmental event/external stimulus.

Standard practice – based on years of studies limited to labs, and usually for a lab session of less than an hour to gather data for scientific research papers – is to have the person wear the EDA sensor for 15 minutes of rest prior to any tasks or measurements. During this period the subject should be asked to sit comfortably, hold still and relax.  If they succeed in truly relaxing, then ideally the signal drops to be low and smooth, and once it does, then the lowest values are usually averaged to provide a baseline.

Fig.3 Activating task, followed by rest to get a baseline. The task was to count backwards from a 4 digit number by 7s as fast as possible and as accurately as possible.

There are several problems with the traditional baseline approach:  Sometimes people cannot relax before or during a lab task – they might have anticipatory arousal, they might have anxiety being in an unfamiliar space not knowing exactly what they are going to be doing, and they might be uneasy being watched.  They might have to use the restroom. They might suddenly remember that they promised a friend they would meet at that time.  They might need more than 15 minutes to relax.  In all these cases, while they may look like they are calm and relaxing on the outside, their levels may be higher than a true baseline.

The Q sensor makes long-term EDA measurement practical, so now you can get a 24/7 baseline and measure what a person’s truly lowest tonic skin conductance levels are over daily life. This solves many of the problems that previously existed with short-term lab-only measurements of EDA baselines. The best way to get a baseline now is to measure long-term, and look for the lowest smooth period of skin conductance where there is no physical movement (where the accelerometer data is smooth), and where the temperature is clearly body temperature.

m

My signal looks flat.  What does it mean?
If it’s flat when you first put the sensor on, then it may be that the electrodes/skin/air are very dry and you need to do something activating for a while to build up some moisture between the skin and the electrodes.

Once you’ve seen SCR’s and you’re sure that everything is working, and the sensor hasn’t moved to a new position, then a flat signal could mean the individual is in a state of rest or is stabile. Stabile is common for some individuals who simply produce fewer SCR’s or who habituate rapidly to repeated presentations of simple stimuli.

Also, it is possible that a person is taking medications, such as anti-cholinergics, or others that can suppress the sympathetic nervous system.  If so, the skin conductance can be very low (below 0.05).  Very dry skin in need of moisturizing can also make it hard to get a good reading. Also, some conditions can be associated with having low skin conductance; for example, many people with ADHD also have very low skin conductance.

Finally, sometimes when you look at your data in the Q software, which auto-scales the vertical dimension, it will look flat because you may have a huge (but invisible) noise spike that is causing the scale to be really large compared to most of the data ("also see How to Easily Adjust Vertical Scaling").

m

Tips for activating EDA when you first put sensor on

Put the sensor on.  If it’s already high (above 1 mS and you see clear skin conductance responses (SCR’s) in the absence of movement (say by having the wearer make multiple “sniffs” or talk about something important to them) then you can skip the steps below. If, however, the level is really low and flat, then it may take a while to show responsivity. Below are some tips that can speed up time to responsivity.

1. Weather-dependent:  Send person outside for 5 mins wearing the sensor.  This can work very quickly if weather is humid or hot.
2. Facility-dependent:  Ask them to go for a brisk 8-10 min walk around the facility, ideally climbing some stairs.  It usually takes a little more than 5 minutes to break a sweat (and see SCRs).
3. If stuck in a room: Tell them they will need to stand up and give a 2-min speech about (something appropriate and personal).  Tell them you’ll be watching it and grading it (or recording it to show somebody important, or something that might matter to them) and they have 3 minutes to prepare starting NOW.  3 minutes later, say, “Time is up” “Please stand up and give a 2-minute speech.  “Go” (A big ticking kitchen timer is ideal.)
4. Another stuck-in-room option:  Give participant a page of mathematics problems to spend 5 minutes solving.  Ask them if they finished high school? College?  (Subtle pressure).  Then tell them to solve as many as possible accurately in 5 minutes, and time them.  (A big ticking kitchen timer is ideal.) “We will score your results and compare it to other (high school/college) graduates.”  Alternatively, give them a task like counting out loud backwards from 3000 by 7’s, as fast as possible and as accurately as possible, and sound a buzzer each time they make an error.

Once you are seeing SCR’s not caused by motion artifacts, then it is probably responding fine.  It is ok if the SCR’s and total values of the EDA are still below 1 mS as long as you are seeing peaks in the data to stimuli that should cause them (like loud startle sounds, etc.) Note that anybody can habituate to the same stimuli presented over and over, so there is no one stimulus that will always work.  Feel free to try things that you think are activating and see what works for an individual situation.

m

I have a lot of SCR’s even when I’m not doing anything – is something wrong?
No, this can be entirely normal.  Some persons are highly reactive with considerable spontaneous generation of SCRs, while others have a relatively steady tonic level of skin conductance without spontaneous SCRs.

m

The Basics:

What is Electrodermal Activity (EDA)?
Electrodermal activity refers to electrical changes measured at the surface of the skin that arise when the skin receives innervating signals from the brain.  For most people, if you experience emotional arousal, increased cognitive workload or physical exertion, your brain sends signals to the skin to increase the level of sweating. You may not feel any sweat on the surface of the skin, but the electrical conductance increases in a measurably significant way as the pores begin to fill below the surface.

m

How is EDA measured?
EDA can be measured in many different ways electrically including skin potential, resistance, conductance, admittance, impedance, and admittance (ref: Electrodermal Activity by Wolfram Boucsein). The Affectiva Q sensor provides a way to capture electrical conductance (inverse of resistance) across the skin. It achieves this by passing a miniscule amount of direct current between two electrodes in contact with the skin.  The units of measurement are microSiemens (μS).

m

Is EDA the same as GSR?
GSR (short for either Galvanic Skin Response, Galvanic Skin Reaction, or Galvanic Skin Reflex) is an old and often misleading term that has been used in reference variety of electrodermal phenomena. GSR can refer to either resistance, conductance, phasic responses, general electrodermal phenomena, or to other things. Over the years, the skin has been shown to be much more complex than a “galvanic element”. Given the potential for confusion the use of the term has declined in favor of more descriptive terms like EDA.
m

Features of EDA Data:

What does a high level mean? A low level?

A level of skin conductance that is higher than an individual’s usual level or baseline reflects situational levels of arousal/activation or responsiveness/attentiveness. The situation could be one that induces stress or interest or excitement or any activity that requires increased level of emotion, cognition, or brain activity.  In contrast, a low level can mean relaxation, and lack of activation, whether good or bad (calm and at peace, or disengaged/bored).
m

Does EDA change with temperature?
EDA can go up whether you are cold or hot. It is not like the old mood rings that merely changed with skin temperature. You can be cold and very stressed, and EDA can go up. You can also be warm and calm, and EDA can drop. If you get really overheated, to the point where you are sweating profusely, then of course it will go up, and probably you aren’t very calm (or comfortable) at that point either. The Q Sensor measures both EDA and skin temperature – making it easy for you to see both signals changing.

m

What are tonic and phasic changes?
Skin conductance measurement is traditionally characterized into two types – tonic and phasic – which can roughly be thought of as “the smooth underlying slowly-changing levels” vs. “the rapidly changing peaks.”

m
Tonic – Tonic skin conductance is generally considered to be the level of skin conductance in the absence of any particular discrete environmental event or external stimuli. This slow-changing level is generally referred to as Skin Conductance Level (SCL). Tonic skin conductance level can slowly vary over time in an individual depending upon his or her psychological state, hydration, skin dryness, and autonomic regulation. Tonic changes in the skin conductance level typically occur in a period of from tens of seconds to minutes.

Fig.1 The two gray shaded areas contain no intentional events while the person was resting. The EDA is smooth and slowly changing, and can be used to estimate tonic SCL. The tonic SCL in each smooth region can be computed as the average of that region.  Here the tonic level is higher after this person exercised.

Phasic – Phasic skin conductance measurements are typically associated with short-term events and occur in the presence of discrete environmental stimuli (sight, sound, smell, cognitive processes that precede an event such as anticipation, decision making, etc).  Phasic changes usually show up as abrupt increases in the skin conductance, or “peaks” in the skin conductance. These peaks are generally referred to as Skin Conductance Responses (SCRs).

m

What is an SCR?
Skin Conductance Responses (SCRs) are abrupt increases in the conductance of the skin. The big SCR that is circled is a nice example of their characteristic form:  they usually have a faster rise time than decay time, unless they occur in rapid succession (like the 3 SCR’s following this big one).

m
m
m
m
m
m
m

Why bother to measure both left and right? Shouldn't the two sides be the same?
Traditionally people have measured only the “non-dominant” hand (left hand for a right-handed person) and most of the EDA results in the literature describe left-handed results.  However, that convention began before neuroscientists showed that a variety of circuits on both sides of the brain are involved in eliciting electrodermal activity, including for example, the left amygdala and the right amygdala, each innervating the left and right sides of the electrodermal response, respectively.  Furthermore, brain imaging has shown that the left and right hemispheres can be differentially activated in some conditions such as depression, and anxiety.  Some researchers have also looked at hemispheric differences in brain imaging related to positive/negative and to approach/withdrawal.  When the brain imaging shows differential activation, we can expect that the skin conductance might also be differentially activated.  Studies can now begin to examine left-right differences on both wrists or both palms using a pair of Q sensors. Initial findings, while in a small group, look consistent with the brain literature, but more studies are needed.
m

Can I measure multiple areas at once?
Yes, you can synchronize multiple sensors to the same computer clock and put the synchronized sensors on as many body locations as you’d like.  (In the Q software, “synchronize” is an option on the top “Q” menu – just click on it when a sensor is plugged into your computer).  The Q software can also display multiple signals on the same time scale or on different time scales.  If you forget to synch a sensor, the Q software can also let you enter a manual start time (double click on the file listed in the upper left window; it will pop up with editable fields.)  Additionally, you can manually slide the signal to where it should be positioned on the time axis.  (Click on the second graph icon in the upper left, the “comparison mode” graph.  Then right click on the signal you want to shift, select “align dataset” and then drag it where you want it.)
m