Repeatability of Measurements of Galvanic Skin Response – A Pilot Study

ABSTRACT

To scrutinize repeatability of measurements of galvanic skin response (GSR) using a complementary therapy device.

GSR diagnosis is closely related to the Chinese system of acupuncture points. These are assumed to show increased electrical conductivity that is super influenced by the (stressed or weak) state of related organs and organ systems including the vagal / simpatico system. GSR techniques apply a constant voltage. For the device used here, a “biophoton therapy device”, a circuit includes a brass electrode that is held in one hand of the test person, and a style electrode for contact with his or her other hand. When the organism is exposed to external influences, skin conductivity may change within split seconds (increase or decrease). Here, it was aimed to transfer information from test substances into the circuit. Paracetamol® was dissolved in water, and both this test probe and control water were put into quartz glass vials. Following a sequence unknown to the tester, the vials were inserted into the input quartz-glass beaker of the device. A decrease in skin conductivity due to the information from Paracetamol® is assumed to be measurable at the acupuncture point “TW-1D-lefthand-pituitary gland”. Repeated measurements with breaks for recovery were carried out. Three sub studies consisted of 5 unblinded and 10 blinded tests each. Depending on the sub study, one to five different tested volunteers and testers were involved. Data were analyzed blind using a two-way analysis of variance with treatment (i.e. test or control) and sub study as independent variables and the frequency of diagnosed indicator drops as the dependent variable. Furthermore, frequency of diagnosed indicator drops was compared to the frequency of non drops both for the test and the control probe in four field tables. In a random distribution, one would expect about 50% in each of the four fields.

A significant main effect for treatment group was observed over all three studies (p<.001). The mean frequency of diagnosed indicator drops was higher for the test probe than it was for control. No main effect was found for sub study (p>.05), indicating that there were no differences between the three sub studies. In addition, no interaction effect between the variables treatment group and sub study was found (p>.05). This means that (a) the significant overall difference between treatment groups holds for each sub study in the same way and (b) the non-difference between sub studies holds both for intoxication and non-treatment. In terms of the pooled results of all three substudies the test probe was correctly detected in 77.5% of cases, with 22.5% false positive results. In reverse, the neutre control was correctly diagnosed in 68.6%, with 31.4% false negative results.

Results suggest that information from substances can influence biological systems and that measurements of electric skin conductivity, as used for diagnostic purposes, are repeatable under standardized and blind conditions.

Keywords: