Psychophysiologists measure electrodermal activity (EDA) in conductance, resistance, and skin potential units. These indices are correlated, but not interchangeable. Measurement procedures, instrumentation, and the biological signals themselves will be described separately.



This unit covers Descriptions of most commonly employed biofeedback modalities: Electrodermal activity (III-A), Sources of artifact (III-B), and Structure and function of the autonomic nervous
system (V-A).
 
Students completing this unit will be able to discuss:

  1. Descriptions of most commonly employed biofeedback modalities: Electrodermal activity
    A. Sensors and sensor placements
    B. Characteristic signals
    C. Signal processing and feedback displays
  2. Sources of artifact
    A. How to identify and correct extraneous biologic activity in recordings
  3. Structure and function of the autonomic nervous system
    A. The effects of commonly employed medications on autonomic activity



A clinician can choose between two exosomatic EDA measures, skin conductance and skin resistance. Which should you use? Psychophysiologists prefer conductance for three reasons. First, conductance more directly reflects the number of active sweat glands and their activity. Second, conductance values are more normally distributed, which aids statistical analysis. Third, conductance increases as arousal and activity increase. This makes more intuitive sense than resistance dropping with increased arousal (Andreassi, 2000).



Skin conductance is monitored by imposing a current across the skin using two or three surface electrodes. Skin conductance does not measure the number of active sweat glands. Instead, skin conductance is an indirect index of sweat gland activity that depends on passing an electric current across the skin. No current, no conductance. This indirect measure is closely related to eccrine sweat gland activity. High levels of sweat gland activity are strongly correlated with high conductance levels (Peek, 1987).

Measurement circuits use a constant voltage or constant current.

A constant voltage system holds the voltage through the skin constant so that current changes with resistance. In contrast, a constant current system holds current constant so that voltage changes with resistance.

The newer constant voltage system is preferred since it adjusts current in response to the number of active sweat glands and directly measures skin conductance (Stern, Ray, & Quigley, 2001). Less current is imposed when there are fewer active sweat glands (Stern, Ray, & Quigley, 2001). This prevents excessive current density from irritating sweat glands and producing a driver artifact that distorts measurement.

Measurement using an AC current is preferred over DC current since this reduces the risk of electrode polarization. AC methods may also be more tolerant of inadequate skin and electrode care. In normal practice, however, skin conductance measurements using AC and DC methods appear comparable. Montagu (1964) found that AC and DC conductance values were strongly correlated (r = +0.99 for skin conductance levels and r = +0.97 for skin conductance responses).



Skin conductance and skin resistance values vary with electrode recording surface; skin potential values do not.

Skin conductance values are expressed in μS (microsiemens) per cm2. Nominal values under standard conditions are 2-100 μS/cm2 (Hassett, 1978). The siemen has replaced the mho as the unit of conductance.

EDA electrodes should be nonpolarizing (not segregate charge) since this reduces conductance values. Biomedical engineers reduce polarization by coating a metal with its own salt. Silver/silver-chloride or zinc/zinc-sulphate electrodes are both used. Silver/silver-chloride electrodes are also used to monitor EEG and EMG.






Electrode recording area should be as large as possible. A larger recording area reduces resistance and spreads current across more sweat glands, reducing sweat gland irritation (Stern, Ray, & Quigley, 2001). Standard electrodes applied to palmar or plantar (sole of foot) sites range from 1.5-2 cm in diameter. Finger electrodes are typically 1 cm or less in diameter (Andreassi, 2000).

Clinicians secure skin conductance electrodes with double-backed adhesive collars (used in EMG), surgical tape, or elastic bands. Velcro is frequently used to attach finger electrodes.



Electrode gel is used to conduct current and should contain a physiologic concentration of saline, which should not exceed 0.05 molar concentration (Hugdahl, 1995). Commercial EEG and EKG gels may distort readings if their saline content approaches saturation (Venables & Christie, 1973). Fowles et al. (1981) recommended mixing Unibase paste with a 3% saline solution for a paste with a 0.05 molar NaCl concentration.



Clinicians attach electrodes to the palmar surface of the hands or fingers or the plantar surface of the foot (sole) where there are the highest density of eccrine sweat glands (about 1,000 glands per square cm). Placement may be bipolar or monopolar (Hugdahl, 1995).

In bipolar recording, both electrodes are placed over active sites. The graphic below, which was generously provided by Peper, Gibney, Tylova, Harvey, and Combatalade (in press) shows palmar and finger placements for SCL sensors. The second phalange may be preferred to the first due to its lower incidence of abrasion or cuts, which can alter the skin's electrical properties.




In monopolar recording, one electrode is placed over an active site and the other over an inactive site with fewer eccrine sweat glands (see below). An additional reference lead may be placed on an inactive site (forearm) on the same side of the body when 60-Hz artifact is a problem (Andreassi, 2000). Following cleaning, a clinician abrades the skin of the inactive site (Stern, Ray, & Quigley, 2001).




Hands are washed before each session to standardize skin condition. SCL significantly drops after washing with soap and water removes surface salt (Peek, 1987).



Skin conductance recordings show tonic (SCL) and phasic (SCR) components. SCRs resemble small waves that ride tidal drifts in SCL (Lykken & Venables, 1971).

Normal skin conductance level values range from 1-30 μS per cm2, while skin conductance response amplitudes range from around 0.01-5 μS (Hugdahl, 1995). SCRs are classified as nonspecific (NS-SCRs) and event related (ER-SCRs). Below is a BioGraph ® Infiniti skin conductance display.





NS-SCRs occur when identifiable stimuli are absent. The typical rate during rest is 1-3 per minute. NS-SCRs must be differentiated from responses due to breathing and movement (Dawson, Schell, & Filion, 1990). These spontaneous fluctuations in skin conductance increase with anxiety and arousal (Hugdahl, 1995).

ER-SCRs are elicited by specific stimuli which may be “novel, unexpected, significant, or aversive” (Dawson, Schell, & Filion, 1990). How do we count ER-SCRs? A criterion of 0.01 μS is typically used to identify an ER-SCR. An ER-SCR is shown below.




                    


An ER-SCR displays several important properties: latency, amplitude, rise time, and half-recovery time. Latency measures the interval between the stimulus and SCR onset, which is 1-3 seconds. Amplitude is the rise in conductance shortly after the stimulus and depends on electrode recording surface (.05-5 μS) (Andreassi, 2000, Schwartz & Andrasik, 2003). Rise time is the interval between SCR onset and peak (1-3 seconds). Finally, half-recovery time (rec t/2) is the interval between SCR peak and 50% recovery of amplitude (2-10 seconds) (Dawson, Schell, & Filion, 1990).



                        


Edelberg (1970) proposed that rapid recovery is associated with attention and goal seeking, while more gradual recovery reflects an individual's emotional state.

What are low and high SCL values? Using 3/8 inch dry electrodes on the volar surface of the fingers, readings below 1 μS are considered low and readings above 10 μS are considered high (Peek, 1987).

Below a BioTrace+ / NeXus-10 screen shows very high resolution (24 bit) skin-conductance feedback with a precision greater than 1/10,000th of a microsiemen.







Skin resistance is measured exosomatically using the procedure outlined for skin conductance. Like SCL and SCR, measurements depend on electrode recording surface. The tonic index is skin resistance level (SRL). Typical values are 10-500 Kohms/cm2 (Hassett, 1978). The phasic component is skin resistance response (SRR). Edelberg (1972) suggested that the criterion for an SRR should be 0.1% of baseline resistance. SRR latency is the same as SCR latency, which is 1-3 seconds (Schwartz & Andrasik, 2003).

An SRR is shown below.




                    




Skin potential is measured endosomatically by placing an active electrode over a site (like the palmar surface of the hand) and a reference electrode over a relatively inactive site (forearm). Skin potential is the voltage difference between sweat glands and internal tissues (Hassett, 1978).

Unlike skin conductance and resistance, measurements do not depend on electrode recording surface. The tonic component is skin potential level (SPL). Typical values are +10 to -70 mV (referenced to the inactive electrode). The phasic component is skin potential response (SPR), which typically includes a negative limb up to 2 mV and a positive limb up to 4 mV (Stern, Ray, & Quigley, 2001).



Skin potential measurement involves the same electrodes and conductive gel used in skin conductance and resistance recording. Recording is monopolar with electrode placement over active and inactive sites. Skin preparation may involve abrasion of the inactive site using tools like a dental burr or needle (Andreassi, 2000). The bias or potential difference between the two electrodes should be kept below 1 mV to prevent confusing gradual electrode drift with a gradual change in skin potential level. A researcher can measure bias before and after data collection by placing the electrode pair in a saline solution and measuring the resulting SPL (Stern, Ray, & Quigley, 2001).



SPR has the same 1-3 second latency as SCR and SRR. The amplitudes of palmar SCRs and SPRs are highly correlated during a single recording session (Wilcott, 1967). The SPR can exhibit a monophasic waveform (negative or positive), biphasic waveform (negative, then positive), or triphasic waveform (negative, positive, then negative). The classic pattern is biphasic. Negative voltage is recorded upward by convention (Stern, Ray, & Quigley, 2001).




                  


Human skin acts like a negatively-charged membrane that is especially permeable to cations (positive ions) (Edelberg, 1971). At rest, SPL mainly reflects the epidermal potential. When SPRs occur, they increase skin potential negativity. This implies that sweat gland potentials contribute more to the SPR than epidermal potentials. Microelectrode studies confirm that the sweat gland pore is more negative than the corneum surface (Edelberg, 1968).
 






The EDA signal is processed by a specialized amplifier. For skin conductance and resistance, an operational amplifier converts current to a proportional DC voltage. Analogous circuitry is used to process the temperature signal.

For skin potential, which is already a voltage, a DC signal amplifier boosts the voltage to usable levels. Following amplification by an operational amplifier, or DC signal amplifier, a low pass filter (2.5 Hz) may be used to reject higher frequency artifacts like 60 Hz.







Four concerns when monitoring EDA are individual differences, movement artifact, skin condition, and room temperature.

Individual differences can significantly affect EDA measurement. Researchers have shown that age, race, lability-stability (Lacey & Lacey, 1958), and stage of the menstrual cycle influence EDA (Stern, Ray, & Quigley, 2001).

Movement artifact results when movement increases electrode contact area (pressing against the skin) or decreases area (lifts electrode off the skin). This problem is less severe with recessed precious metal electrodes prepared with conductive gel. Movement artifact can be reduced by instruction to minimize movement and placing the patient in a stable, comfortable position. Sensors may also be placed on the nondominant hand during tasks that require one hand. When this artifact occurs, it is easily discriminated from valid EDA waveforms by visual inspection of a chart recording (Peek, 1987).

Skin condition affects EDA in several ways. Abrasions or cuts through the highly-resistant epidermis can raise SC readings. Calluses may increase epidermal resistance and lower SC values (Peek, 1987). These problems can be prevented by not placing electrodes over sites that have been abraded, cut, or callused.

SCL values fall sharply after washing hands with soap and water since this removes surface salt. Measurement reliability may be increased by asking the patient to wash immediately before each session (Peek, 1987).

Temperature may lower or raise EDA values. When patients feel cold, SC may be reduced. Warmer than normal rooms may increase SCRs (Venables & Christie, 1980). Excessively high temperatures and humidity may produce thermoregulatory sweating that is unrelated to psychophysiological activity (Peek, 1987). Room temperature and humidity should be regulated across sessions to prevent artifactual values. Time of day, day of week, and season should also be controlled as potentially confounding environmental variables (Stern, Ray, & Quigley, 2001).



You can determine whether a skin conductance display mirrors the your client's electrodermal activity by performing a tracking test, during which you instruct your client that in a few seconds you will clap your hands. Skin conductance should increase within seconds of this warning as seen on the BioGraph ® Infiniti tracking test display below.






A tracking test checks the integrity of the entire signal chain from the skin conductance sensor to the encoder, and the correct software selection of input channels. When used with the ProComp Infiniti system, it ensures that the sensor is intact and snugly inserted into the right encoder input socket, and the correct channel has been chosen for display.











State of the art biofeedback data acquisition systems achieve +/- 5% and +/- 0.2 μS accuracy.




A dummy subject is a circuit board with a fixed resistance that tests the accuracy of an electrodermograph. Two electrodes are snapped on to the circuit board’s posts, and skin conductance or skin resistance are then measured. The electrodermograph should display the value of the resistance.











Now that you have completed this module, describe the method your electrodermograph uses to measure electrodermal activity and the site(s) you use for recording. Summarize the precautions you take to ensure accurate measurements.




Andreassi, J. L. (2000). Psychophysiology: Human behavior and physiological response. Hillsdale, NJ: Lawrence Erlbaum and Associates, Inc.

Cacioppo, J. T., & Tassinary, L. G. (Eds.). (1990). Principles of psychophysiology. New York: Cambridge University Press.

Hassett, J. (1978). A primer of psychophysiology. San Francisco: W. H. Freeman and Company.

Hugdahl, K. (1995). Psychophysiology: The mind-body perspective. Cambridge, MA: Harvard University Press.

Peavey, B. (2003). Effects of drugs on biofeedback. Short course presented at the 34th annual Association for Applied Psychophysiology and Biofeedback convention, Jacksonville, Florida.

Peper, E., Gibney, K. H., Tylova, H., Harvey, R., & Combatalade (in press). Mastery through experience.

M. S. Schwartz, & F. Andrasik (Eds.). (2003). Biofeedback: A practitioner's guide (3rd ed.). New York: The Guilford Press.

Stern, R. M., Ray, W. J., & Quigley, K. S. (2001). Psychophysiological recording (2nd ed.). New York: Oxford University Press.