Table 2 Summary of selected studies – Association between EDA and depression

Barg et al. (Germany, 1996)

− 16 depressed patients (36.1±1.9 years; 50% females), no medication;

− 16 depressed patients (45.8±10.2 years; 50% females) treated with paroxetine;

− 16 depressed patients (45.4±9.9 years; 50% females) treated with imipramine.

SCL, non-specific SCR, SCR

Patients treated with imipramine showed a higher heart rate and lower EDA than the other two groups.

Bernstein et al. (US, 1988)

− 50 schizophrenic patients (34.8±9.5; 30% females);

− 50 depressive patients (32.5±9.7 years; 66% females);

− 50 controls (27.3±7.6 years; 58% females).

SCR recorded during auditory stimuli presentation.

A subsample of each group was told they needed do nothing during tone presentations (habituation series); other subsamples had to press a pedal for each designated target signal, ignoring all nontarget tones.

No differences between hands in SCR for any diagnostic group

Compared to normal controls, schizophrenic patients showed higher nonresponding in the habituation series, but no difference when target tones were presented.

Depressives were equally nonresponsive during the habituation series, but they did not show a decrease in SCR nonresponding to the target tone.

Bernstein et al. (US, 1995)

− 69 schizophrenic patients (34.9±8.5 years; 42% females);

− 45 mood disorder patients (36.6±10.1 years; 62% females);

− 67 normal controls (31.2±10.1 years; 49% females).

SCR recorded during exposure to tones.

Schizophrenic and depressed patients were significantly more often classified as non-responders.

The proportion of non-responders did not vary with neuroleptic or antidepressant medication.

Bob et al. (2011)

− 44 unipolar depressed outpatients, (37.06±8.72);

− 35 healthy controls (35.12±8.10; 57% females).

EDA measured bilaterally.

Absence of differences between EDA in relapse and in remissions as well as between pharmacotherapy response and EDA.

Bonnet et al. (France, 2004)

− 9 non-depressed Chronic Low Back Pain (CLBP) patients, (43 years; 55% females);

− 9 depressed CLBP patients, (46.1 years; 55% females);

− 9 depressed controls free of pain (DCs), (44.1 years; 55% females);

− 9 healthy controls free of pain and depression, (45.3 years; 55% females).

EDA recorded during a 3-min rest period and during a subsequent stimulation period. Stimuli were 3 identical pure tones.

Non-depressed patients presented an increased EDA, especially a higher frequency of non-specific fluctuations, than the 3 other groups.

SCL was lower in the 2 groups of depressed participants than in the healthy control group.

Brankovic (Serbia, 2008)

− 57 patients with major depressive disorder and currently in depressive episode (43.1±10.9 years; 56% females);

− 52 healthy controls (39.8±8.1 years; 60% females).

5 parameters of the SCR recorded when reading emotional eliciting stories.

Initial SCR was larger in controls than depressed patients.

Positive feedback loop was greater in depressed patients than controls.

Negative feedback loop was stronger in depressed patients than controls.

Length of time needed for the magnitude of SCRs to decrease to half of the original response magnitude was longer in depressed patients than controls.

Breyer-Pfaff et al. (1982)

37 patients with primary depressive disorders (22 endogenous and 15 non endogenous) (81% females) treated with amitriptyline.

SCLs, SF, habituation and amplitude of the SCR recorded during the exposure to tones and during active and passive situations.

Measures repeated on the day before treatment began and again after 14 and 28 days of treatment.

Compared to baseline levels, EDA state at day 14 can be considered inhibited and labile (larger phasic and tonic reaction amplitudes).

After 4 weeks of treatment, lability had returned to baseline, but inhibition was unchanged.

No significant correlation between EDA measures and plasma drug levels.

Byrne (Australia, 1975)

− 18 acute depressive inpatients (20-60 years; 50% females):

− 10 “neurotic depressive”;

− 8 “psychotic depressive”;

− 11 normal controls (24-45 years; 45% females).

SCR amplitude, NS.SCR, habituation rate.

Higher mean SCR, higher frequency of NS.SCRs, and slower skin response habituation rata in the neurotic depressive sample. Higher mean SCR amplitude in the control sample.

No significant difference between skin response habituation rates of either the control and neurotic depressive, or control and psychotic depressive samples.

Carney et al. (US, 1981)

− 15 inpatients with primary affective disorder (32±6.9 years; 100% females)

− 15 age-and race-matched controls (31.5±7.5; 100% females)

SCL

Significantly lower SCL in depressed subjects.

No significant correlation between SCL and measures of anxiety or depression for patients with primary affective disorder.

Significant correlations between a measure of depression and SCL in normal controls.

Dawson et al. (US, 1977)

Dawson et al. (1985)

− 20 hospitalized depressed patients (63.8±7.8 years; 80% females);

− 20 non-depressed controls (62.2±7.6 years; 80% females).

SCL and SCR recorded at rest and during a variety of tasks before and after a series of electroconvulsive shock treatments (ECTs)

Depressed patients, compared to non-depressed controls during the pre-ECT test, exhibited lower SCLs, smaller phasic SCRs with longer latencies.

Little EDA changes following ECT.

Overall efficiency of SCL was of 70% and of SCR was 80%.

Donat & McCollough (US, 1983)

− 10 chronically depressed subjects (18-25 years; 100% females);

− 10 controls (18-25 years; 100% females).

SCL

SCL correctly identified 9 of the 10 depressed group subjects and 7 of the 10 control group members under resting conditions.

No difference between depressed subjects and controls under conditions of laboratory stress and of imaginal-role stress.

SCL reliably identified group membership, with lower levels in the depressed group.

Falkenberg et al. (Germany, 2012)

− 16 patients with MDD (37±15 years; 50% females);

− 16 healthy controls (35±14 years; 50% females).

SCR recorded during standardized mood induction using happy and neutral pictures and funny and neutral cartoons.

MDD patients had higher SCRs in the cartoon condition than controls.

Women revealed lower overall SCRs compared to men.

Fraguas et al. (US, 2007)

8 unmedicated patients diagnosed with MDD (35±12 years; 62,5% females) treated with fluoxetine 20 mg per day for 8 weeks.

SCL and SCR measured at basal condition and during four induced emotional states: happy, angry, sad and neutral.

Significant positive correlations between the percentage reduction in depression scores and increases in SCR only during the neutral emotion condition.

Giedke et al. (Germany, 1980)

− 18 patients with primary depression (15 unipolar, 3 bipolar) (46.2±13.1 years; 83% females);

− 27 age-matched healthy controls (45.1±15.4 years; 70.4% females).

GSR/SRR (Skin Resistance Responses).

Response condition: participants were asked to react to the second of two identical tone stimuli.

In the experimental response condition controls increased their number and amplitude of SRRs to tone-stimuli more than patients.

No significant correlation between GSR and both self-rated and physician-rated psychopathology.

Giedke & Heimann (1987)

− 59 drug free patients with primary MDD (47±11 years; 69% females) assigned to a double-blind treatment with amitriptyline or oxaprotiline;

− 30 healthy controls matched for age and sex.

SR and habituation of SR orienting response.

Patients with primary MDD exhibited significantly fewer spontaneous fluctuations of SR and a faster habituation rate of SR orienting response.

SR level did not differ between groups.

Most patient/control differences remained unaffected irrespective of the drug applied.

Greco et al. (Italy, 2014)

− 10 patients affected by bipolar disorder I or II;

− 10 healthy subjects (20-32 years; 50% females).

Tonic and phasic features of EDA measured during an emotional elicitation protocol.

Phasic features well discriminated among depression, mixed state, and euthymia.

Mixed-state shows a strong tonic hypoactivity compared to euthymic state.

Healthy subjects showed no statistical difference on each of the EDA feature patterns.

Hattangadi et al. (Canada, 1968)

- 263 psychiatric inpatients, including 51 depressed patients (27 neurotic and 24 psychotic depression);

- 58 healthy controls.

GSR

Patients responded less than controls to the indifferent stimuli. No significant difference in response frequency between patient categories.

No significant difference in latency, either between normal subjects and patients, or between any two diagnostic categories.

All psychiatric patients in all categories gave lower amplitude responses than the normal group.

No significant difference between medicated and non-medicated patients in any diagnostic group as far as GSR frequency and latency.

Patients receiving antidepressant medication had higher amplitudes than those receiving anti-psychotic medications.

Have et al. (Iceland, 1991)

− 21 patients with Alzheimer dementia (80.6 ± 9.8 years);

− patients with depression (75.9 ± 7 years);

− 19 healthy volunteers (79 ± 11.4 years).

SCL

No difference in SCL among patients' groups.

Very low SCL in the overall sample, likely due to a decline in the number of active sweat glands in old people and diminished sweat production.

Heimann (Germany, 1978)

− 95 depressive and depressive-anxious patients;

− 18 of these 95 treated with amytriptylin for 4 weeks;

− 27 of these 95 retested at a 1-year follow-up.

SRR, SRL (Skin Resistance Level).

Orienting reaction and habituation tested after two minutes of rest, with a series of tones and of flashes.

Assessment of changes in the dimension of activation after 14 and 28 days of treatment with amytriptylin, and 1-year follow up after the first examination during a depressive state.

Patients in the activated and inhibited cluster showed the smallest changes at follow-up, whereas all four patients of the labile-activated cluster change to another group.

Preponderance of cases with lower activation after treatment with amitriptylin.

Iacono et al. (1983)

Iacono & Tuason (1983)

− 26 unipolar depressed patients (37.8±12.7 years; 23% females);

− 24 with bipolar affective disorder (36±12 years; 33% females);

− 46 healthy control patients (35±11.9 years; 17.4% females).

SCL.

The EDA of affective disorder patients was uniformly depressed across all tasks and conditions.

No consistent bilateral asymmetries in EDA were observed.

Iacono et al. (1984)²

− 22 unipolar outpatients in remission (41±12.7 years; 77% females);

− 22 bipolar outpatients in remission (38.5±14.4 years; 68% females);

− 26 controls (41.3±12.2 years; 88% females).

SCLs and SCRs recorded during blowing up a balloon until it burst and during the exposure to tones, and eight familiar sounds.

Compared to the control subjects, the affective disorder patients (especially the unipolar patients) responded significantly less to the balloon task, the highest tones, and the familiar sounds, had lower tonic levels and a larger proportion of them failed to respond to the stimuli.

Several measures of EDA displayed moderately high one-year retest stability.

Ikeda et al. (Japan, 1982)

20 healthy volunteers (23.1 years; 0% females):

− Amitriptyline 30 mg/day, 3 times, for 3 days;

− Nomifensine 100 mg/day, 2 times, for 3 days;

− Placebo 3 times, for 3 days.

SCR recorded before the experiment and 8 hours after the final dose.

In the amitriptyline group, the NS.SCR decreased.

In the nomifensine group, NS.SCR tended to increase.

Kamenskaya & Mikhailova (Russia, 1985)

− 50 manic-depressive patients in the depressive phase prior to treatment (20-45 years), divided into 3 groups according to the nature of the principal affect (anguish, anxiety, or apathy);

− 25 healthy individuals (20-45 years).

SCR in the background and during the presentation of indifferent stimuli (opening the eyes, light flashes) and in a stress situation.

Depressed patients had an increased latency and lower amplitude of the SCR than healthy subjects. The absence of a SCR was observed in 20% of patients, more often with a dominant anxiety affect.

Lader & Wing (UK, 1969)

− 35 inpatients with a primary diagnosis of moderate or severe depression:

− 17 Predominantly agitated (45.8 years; 59% females);

− 13 Predominantly retarded (42.3 years; 69% females);

− 5 Uncomplicated depressives (30.2 years);

− 35 age and sex matched controls.

SCLs, SCRs, NS.SCRs and habituation rate recorded during exposure to tones.

Agitated patients and retarded patients had, respectively, significantly higher and lower mean SCL than controls.

Agitated depressives had the highest conductance levels at the end of the recording session than at the start.

Healthy subjects were the most reactive ones, and habituated significantly more rapidly than did the agitated depressives.

Retarded patients gave a significant lower number of SCRs than agitated patients.

The agitated patients had significantly more and the retarded patients had significantly less fluctuations than the normal subjects.

Lapierre & Butter (Canada, 1978)

− 20 agitated depressed patients (mean age 38 years and 2 months; 75% females);

− 20 retarded depressed patients (mean age 37 years and 8 months; 70% females).

Spontaneous GSR was monitored for 5 minutes. The patient was then given a series of 6 randomized visual and/or auditory stimuli, and the latency period for a EDA response and the magnitude of the response was assessed.

Then: Maprotiline and imipramine administration, double-blind.

Measurement were repeated after 3 hours, 4, 7, 14 and 28 days.

Basal skin resistance was significantly lower for the agitated depressive. The GSR response was consistently greater in the agitated depressives than in the retarded.

Gradual but nonsignificant increase in skin resistance and reduction in the magnitude of the GSR response as treatment progressed in both drug groups. Still the agitated depressives maintained more reactivity on this parameter.

No major difference in GSR was found between imipramine and maprotiline.

Lapierre & Butter (Canada, 1980)

− 20 agitated depressed patients (mean age 38 years; 75% females);

− 20 retarded depressed patients (mean age 38 years; 70% females);

− 20 volunteer controls (mean age 24 years; 75% females).

SCL and SCR recorded prior, during and after visual and auditory non-signal sensory stimulations and a reaction time stimulation.

SCLs significantly higher for both depressed groups.

Depressed patients had longer latency time of onset of an EDA response than in volunteers.

Lemaire et al. (France, 2015)

− 45 bipolar outpatients:

− 15 remitted (41.67±10.19 years; 47% females);

− 15 moderately depressed (43.4±11.65 years; 80% females);

− 15 moderately manic (41.53±8.13 years; 73% females);

− 101 healthy controls (43.86±11.96 years; 49% females).

Maximum SCR amplitude recorded following the presentation of affective and neutral pictures during passive viewing and during experiential suppression.

Negative and positive pictures elicited SCRs of similar maximum amplitudes, greater than those elicited by neutral pictures.

The maximum SCR amplitude was reduced during experiential suppression in all groups but this effect was more pronounced in depressed patients.

Levinson (US, 1991)

− 36 schizophrenic patients (31.4±6.9 years; 36% females);

− 17 patients schizoaffective disorder (30.3±9 years; 59% females);

− 24 inpatients with MDD (psychotic or nonpsychotic), or the mainly affective subtype schizoaffective disorder (36.9±12.4 years; 54% females);

− 25 control subjects (30.6±9 years; 44% females).

SCL, SCR, NS.SCR, SC non-response, habituation score, reaction time recorded during four paradigms:

− a series of tones in a no-task habituation paradigm;

− a similar series of higher tones;

− a series of tones with a button-press (reaction time) task;

− a loud white noise stimulus (without task).

Schizoaffective subjects were more likely to be non-responders, and had lower (faster) mean habituation scores than other groups. Schizophrenic, depressed and controls subjects had similar mean habituation scores and proportions of non-responders.

Depressed subjects showed some evidence of autonomic hyperarousal (higher tonic SC level, trend toward more NS.SCRs).

Lewinsohn et. al (USA,1973)

Two studies on the same sample composed by

− -12 depressed patients;

− -12 psychiatric control patients;

− -12 healthy control subjects.

Autonomic response (skin resistance) to aversive stimulation and adaptation over repeated presentations of the same aversive situation.

In both studies, the depressed group was found to be more responsive to the aversive

stimulus and the overall SCL was highest for the depressed subjects.

Lindsey et al.

(US, 2011)

- 24 currently depressed Seasonal Affective Disorder (SAD) participants (41.58±11.72 years; 92% females);

- 24 demographically-matched controls with no depression (41.83±12.75; 92% females).

SCL, surface facial electromyography and self-reported emotional responses to light- and season-relevant stimuli

Task: 30 digital photographs of 5 scenes presented under 6 conditions (light intensity – 2 conditions; seasonal cues – 3 conditions).

SAD participants displayed more frequent SCR, greater SCR magnitude, and more self-reported depressed mood in response to overcast stimuli and lower SCR magnitude, and less self-reported depressed mood in response to sunny stimuli.

Mardaga & Hansenne (Belgium, 2009)

− 20 subjects with MDD (22-59 years; 55% females);

− 20 healthy controls (24-59 years; 55% females).

SCR recorded following the presentation of neutral, pleasant, and unpleasant pictures.

Pleasant pictures elicited more and larger responses than unpleasant ones in control but not in depressed subjects.

Depressed subjects showed generally faster half-recovery times.

Mestanikova et al. (Slovak Republic, 2015)

− 25 depressed adolescents (14.6± 0.4 years; 52% females);

− 25 age/gender matched healthy controls.

EDA measured in sitting and supine position, at rest.

EDA diminished in the supine resting position in MDD patients compared to healthy controls.

Miquel et al. (Spain, 1999)

− 27 depressed outpatients (4 bipolar disorder in the depressive cycle; 17 MDD; 5 depressive neurosis; 1 subject non-specified depressive disorder) (34.63±10.05 years; 19; 70% females);

− 27 normal subjects (33.44±10.27 years; 63% females).

SCL and SCR recorded during a series of tones.

Depressive patients displayed lower basal SCLs and lower conductance amplitudes to the first stimulus and to stimulus change.

A trend towards habituation was detected in both groups, but neither difference between them, nor differences in habituation speed were found between depressive and healthy subjects.

Mirkin & Coppen (1980

− 13 depressive inpatients (58.6±1.9 years; 72% females).

− 15 normal controls (54.2±1.7 years; 60% females).

SCL and SCR recorded during exposure to tones.

Higher proportion of electrodermal non-responders in patients than controls.

Endogenous depressive patients had significantly lower SCLs than either the non-endogenous patients or controls.

Myslobodsky & Horesh (Israel, 1971)

− 10 endogenous depressive patients (51±6.79 years; 70% females);

− 9 reactive depression patients (43±6.29 years; 89% females);

− 14 normal subjects (48±17.59 years; 57% females).

EDA recorded at rest and during three experimental conditions: a visual-imagery task, a verbal task, a neutral tone habituation sequence.

In endogenous depression EDA was higher on the left hand compared with the right under all the conditions studied.

In reactive depression EDA was higher on the left side during the verbal task and tone habituation sequence and on the right side in the visual task.

Nissen et al. (2010)

− 23 patients with MDD;

35 healthy control matched for sex, age, and IQ.

SCR recorded during the presentation of Conditioned (CS+) and non-Conditioned (CS-) Stimuli.

MDD patients responded stronger to the CS+ than to the CS-.

Noble & Lader (UK, 1971)

− 34 inpatients with a primary diagnosis of depression (12 males, 44.3±13.6 years; 22 females, 35.7±11.4 years);

− Subgroups of 10 agitated and 10 retarded depression patients.

SCLs, SCRs, NS.SCRs recorded at rest and during mental arithmetic, before and after ECT.

The stress of mental arithmetic was associated with increase in skin conductance. The difference between the basal and stress recordings (reactivity) was highly significant. Reactivity was reduced after ECT.

Low skin conductance correlated significantly with severity of depression.

A reduction in NS.SCR correlated with high scores for depressed mood and retardation.

Before ECT the agitated patients had significantly more fluctuations than the retarded group.

No significant overall change in EDA measures after ECT.

O’Kearney & Parry (Australia, 2014)

− 24 depressive episode (39.6±16.98 years; 67% females);

− 24 PTSD (31.7±16.49 years; 50% females);

− 24 healthy controls (35.8±15.73 years; 50% females).

SCR.

PTSD showed higher SCR during trauma recall compared with recall of other events and compared with depressed controls.

No significant difference in reactivity between depressed and non-disordered participants.

Pazderka-Robinson et al. (2004)

− 43 patients with chronic fatigue syndrome (46.3±9.6 years; 100% females);

− 25 depressed patients (35.3±10.5 years; 100% females);

− 44 controls (27.8±9.3; 100% females).

SCL and SCR recorded during an orienting task.

SCLs were markedly lower for the chronic fatigue syndrome group, with no difference between controls and depressives.

Perez-Reyes & Cochrane (US, 1967)

− 108 neurotic depressed inpatients;

− 82 psychotic depressed inpatients;

− 69 healthy volunteers.

SCL and mean number of SCRs per minute.

No difference in initial SCL or SCR frequency, but significant differences in SCR susceptibility among the three groups.

Pruneti et al. (Italy, 2010)

− Outpatients (38.4±9.7 years; 52% females) with the following diagnoses:

− Generalized Anxiety Disorder (GAD, n = 24);

− Major Depression Episode (MDE, n = 14);

− Panic Disorder (PAD, n = 12);

− Obsessive-Compulsive Disorder (OCD, n = 10).

SCR registered in three consecutive phases: baseline (registration at rest), stress presentation, and recovery.

SCR mean values are much higher for GAD and PAD patients than for MDE and OCD.

The amplitude of the SCR was also significantly different among groups.

Rohde et. al (Germany, 2014)

− 43 currently depressed patients (36.34±11.36 years; 60% female);

− 36 controls (32.19±12.26 years; 66% female);

− physiological data of 32 currently depressed participants and 23 controls.

SCR, EMG corrugator activity at baseline and during a Mindful Breathing Exercise.

Mindful episode: the patient mindfully observed his/her breathing.

Drifted episode: the patients had his/her mind wandering, not focused on the task.

SCR: no significant differences were found between currently depressed patients and controls.

In both groups, greater SCRs recorded during drifted episodes, while

weaker SCRs were elicited within mindful episodes.

Rottenberg et al. (2002, USA)

− 71 depressed persons (mean age 33.4 years; 66% females);

− 33 non-depressed healthy controls (mean age 32.3 years; 70% females).

SCL and SCR recorded at baseline and after stimuli (neutral film and cry-inducing film).

During the sad film, non-depressed criers had greater number of SCR than non-depressed non-criers.

Non-depressed non-criers exhibited significant decreases in SCL from the neutral to the sad film.

Depressed criers exhibited significantly smaller increases in SCL and SCR rate to the sad film than did non-depressed criers.

Schneider (US, 1983)

− 10 depressed inpatients (9 with unipolar and 1 with bipolar disorder) (mean age 43.1 years; 0% females);

− 23 schizophrenic inpatients (13 who underwent drug 'washout' and 10 who didn’t) (mean age 45.2 years; 0% females);

− 10 healthy controls (mean age 40.9 years; 0% females).

SCL, SCR and NS.SCR recorded during exposure to tones.

SCLs were higher in the normal and schizophrenic groups than in the other two groups.

The depressives' mean laterality index, unlike that of the controls, reflected lower SCLs and SCRs from the right hand than from the left.

The indices from the other two patient groups did not significantly differ from controls’ ones.

The patient groups generally had a lower proportion of 'responders' than did the health controls group, a lower number of SFs and a lower SF amplitude.

Schneider D. et al. (Germany, 2012)

− 28 in- and out-patients with unipolar MDD (age range 21–54 years; 43% females);

− 28 age- and education-matched control participants.

Galvanic skin conductance (GSC), galvanic skin responses (GSR).

Stimuli: 96 short video-clips conveying one of 4 emotion categories (happy, sad, fear, disgust) or no emotion (neutral).

Patients displayed more GSRs than healthy controls in all emotion categories.

In patients, the number of GSRs did not differ between emotion categories, and neutral video clips evoked less GSRs than happy, sad, fearful and disgusted clips.

Siepmann et al. (Germany, 2001)

12 healthy volunteers (mean age 25 years; 0% females):

− Reboxetine 4 mg, twice a day, for 13 days;

− Placebo, twice a day, for 13 days.

SCL and SCR recorded following a single deep inspiration.

SCR was decreased after multiple dosing with reboxetine.

Siepmann et al. (Germany, 2003)

12 healthy volunteers (mean age 34 years; 0% females):

− Sertraline 50 mg, once a day, for 14 days;

− Placebo, once a day, for 14 days.

SCL and SCR recorded following a single deep inspiration

Sertraline caused a significant reduction of SCL, whereas SCR was not changed.

Siepmann et al. (Germany, 2004b)

12 healthy volunteers (mean age 27 years; 0% females):

− Moclobemide 150 mg, twice a day, for 14 days;

− Placebo, twice a day, for 14 days.

SCL and SCR recorded following a single deep inspiration

SCRs did not change after treatment with moclobemide.

Siepmann et al. (Germany, 2004a)

12 healthy volunteers (25±3 years; 0% females):

− Amitriptyline 25 mg, 3 times daily for 13 days;

− St. John’s wort extract, 3 times daily for 13 days;

− Placebo, 3 times daily for 13 days.

- SCR recorded following a single deep inspiration.

St. John’s wort extract had no effect on SCR.

SCR significantly decreased during treatment with amitriptyline at all time points of measurement.

Sigmon et al. (USA, 2007)

-15 MDD patients, Recurrent, with Seasonal Pattern (MDD-SAD;

80% females);

-15 MDD patients (MDD; 66.7% females);

-15 controls (80% females).

Mean age 38.93 years.

Baseline SCL recordings obtained for a 5-min baseline period and average SCL collected across the baseline period.

SCR at baseline and after presentation of winter and summer scenes were recorded.

MDD-SAD patients exhibited a greater number of SCR and greater SCR amplitude in reaction to the winter scenes, than individuals in the MDD and control groups.

Individuals in the control and MDD groups did not significantly differ on frequency of SCR or strength of SCR amplitude in response to winter scenes. Individuals in the three groups did not significantly differ on SCR amplitude, in reaction to the summer scenes.

Silva et al. (Brazil, 2000)

-29 adult healthy volunteers (86% females) received

- 100 mg Nefazodone (NF) (mean age 25 years);

- 200 mg NF (mean age 26.6 years);

- placebo (mean age 24.9 years).

(Double-blind condition).

Test 1: (1) amplitude of SCR - fluctuations that occurred within 5 s from a sound stimulus; (2) number of spontaneous fluctuations – those occurring be

yond the above mentioned 5-s window; (3) SCL – average level of conductance during the intervals between fluctuations.

Conditioned fear. NF decreased the number of spontaneous fluctuations of skin conductance in a dose-dependent way, although the drug did not affect the amplitude of the SCR to the tone.

Storrie et al. (US, 1981)

− 10 inpatients with Primary Affective Disorder, free from psychoactive medications for at least 3 days prior the initial testing (48±4 years; 0% females);

− 10 healthy subjects without history of depressive dysfunction or psychosis (49±5 years; 0% females).

SCL and SCR recorded during Valsalva maneuvers.

Measures repeated three weeks after initiation of therapy with tricyclic antidepressants or antipsychotic medication.

Laterality hypothesis not confirmed.

Mean SCL and SCR significantly higher in the control group than in depressed patients.

No significant difference in SCL and SCR between treated and untreated patients.

Thorell & D’Elia (Sweden, 1988)³

− 28 in- and out-patients with major depressive episode and dysthymic disorders (42.5; 50% females);

− 59 mentally and somatically healthy subjects.

SCL at the onset of the first stimulus, mean SC fluctuation rate (SCFr) per minute, SCR magnitude (SCRm) to the first stimulus, the SCR rate (SCRr), and the index of SC nonresponding (SCRi).

Significantly higher EDA during remission than during depression according to all 5 EDA variables.

No significant difference in EDA between patients in remission and healthy subjects.

Among suicide attempters, EDA at follow-up did not change significantly, but was not significantly lower than in healthy subjects. However, SCRm and SCRi in suicide attempters were significantly lower at follow-up than in the healthy group.

The extremely hyporesponsive patients, including suicide attempters, when in remission, did not reach the levels of the healthy subjects except for SCFr.

The patient with recurrent major depressive episodes when in remission did not reach the electrodermal responsivity levels found in the healthy subjects.

Thorell et al. (1988, Sweden)

− 59 depressed patients;

− 59 healthy controls.

After 5 min without stimulation,

37 tones were presented. SCL at onset of

the first stimulus; stimulus-unrelated SC fluctuation rate (SCFr) during the whole stimulation session except for 10 s after onset of each stimulus; SC response magnitude (SCRm) to the first stimulus; SCR rate (SCRr); index of SC nonresponding (SCRi).

Cortisol in plasma and in urine were dosed, dexamethasone suppression test (DST) was performed.

No significant correlation between EDA and cortisol in patients’ plasma and urine.

No relationship between EDA and the outcome of DST, neither any simple relationship between EDA and cortisol in plasma and in urine.

Positive correlation between SCL and morning plasma cortisol and negative correlations between EDR and nocturnal urinary cortisol in the suicide attempters.

In the suicide attempters, the already extremely low EDR is additionally suppressed with increased nocturnal urinary cortisol excretion while SCL, and to some extent also EDR, are elevated with increased morning plasma cortisol. This pattern is opposite to that obtained in the healthy subjects and in the non-suicidal patients for SCL and plasma cortisol.

Thorell et al. (Sweden, 1993)

− 50 depressive patients (42.2±13.4 years; 52% females);

− 50 age and gender-matched healthy subjects (42.6±13.5 years; 48% females).

Blood and urine for the measurement of basal hormone levels were collected at 8

a.m., immediately followed by clinical ratings and EDA measurements.

Positive relationships between SCL and basal levels of thyroid hormones in the healthy subjects were absent or reversed in the depressive patients.

No relationships between electrodermal responsivity and thyroid hormones in depressive and healthy subjects.

No significant difference between drug-free patients and patients treated with antidepressant medications regarding the EDA variables.

Toone et al. (UK, 1981)

− 22 schizophrenic patients (mean age 35.8 years; 64% females);

− 11 unipolar depressed patients (mean age 40.6 years; 73% females);

− 4 manic patients (mean age 38.7 years; 50% females);

− 12 anxiety state patients (mean age 30.1 years; 50% females);

− 22 healthy controls (mean age 36.7 in males and 39.3 in females; 36% females).

SCLs, SCRs and SF recorded at rest and during 32 flashes.

No group showed a distinctive pattern of lateral asymmetry; the only significant difference was in the adaptation of SCL during stimulation.

Among depressed males the adaptation of SCL during stimulation on the left exceeded the right; among depressed females the trend was reversed.

The frequency of SF in EDA was greater in the anxiety state and schizophrenic groups and in those patients who reported auditory hallucinations during recording.

Tsai et al. (US, 2003)

− 12 Spanish-speaking latinas with major depressive episode (28.28±7.45; 100% females);

− 10 non-depressed Spanish-speaking latinas (28.28±7.45; 100% females).

SCL during sad and amusing film clips of human and animal content.

Depressed Latinas demonstrated less electrodermal reactivity across all the film clips than non-depressed Latinas.

Ward et al.

(USA, 1983)

− 33 depressed patients (mean age 42.58 years; 36% females);

− 71 healthy controls (mean age 35.59 years; 46% females).

SCL was examined during 15 minutes during the first week of hospitalization.

Healthy and depressed women had significantly lower SCLs than healthy and depressed men; subjects with recurrent depression had significantly lower SCLs than subjects with a first episode depression.

Significant positive correlation between age and SCL in the depressed group but not in the control group.

No significant SCL differences between medicated and unmedicated patients.

No significant SCL differences between endogenous-nonendogenous, situational-non-situational, primary-secondary, or DST suppressor-DST non-suppressor depression.

SCL proved a very sensitive and specific test for distinguishing non-geriatric unipolar depressed adults from physically healthy normal controls.

Ward & Doerr (1986)

− 37 in- and out- patients with major affective disorder (mean age 35.3 years; 59% females);

− 71 controls (control group 1) (mean age 35.5 years; 46% females);

− 334 "stressed" normal controls (control group 2) who were first parents of newborns (mean age 28.7 years; 51% females).

Resting SCL.

Lowest SCLs in the depression group. No difference between the two control groups.

For women, SCL less than 3.0 micronho/cm2 yielded a sensitivity of 95% and a specificity of 91% for Control Group 1. For males, a criterion of SCL less than 4.8 yielded a sensitivity of 93% and a specificity of 89% for Control Group 1.

No significant difference in SCL between drug-free and medicated patients.

SCL was abnormally low in all depression subtypes.

Weckowicz et al (Alberta, 1971)

− 212 depressed inpatients included; data at follow-up available for 170 (mean age 38 years; 69% females).

Galvanic skin responses (GSR) to a noxious stimulus, obtained one minute after the last basal reading.

Patients were tested after admission and before starting therapy. Second assessment after 3 weeks of therapy.

GSR was not a predictor for ECT, it was a near-significant predictor for psychotherapy and for drug therapy.

Williams et al. (1985)

36 primary depressive patients:

− 27 unipolar (37.4±12.5 years; 74% females);

− 9 bipolar (46.2±10.1 years; 55% females).

SCL and SCR recorded for two experimental conditions:

− auditory stimuli during relaxation;

− a task requiring responding to specific tones pressing a foot pedal.

No differences in tonic or phasic EDA in unipolar or bipolar subtype, response to the dexamethasone suppression test, severity of depression, medication status, or gender.

Patients with psychomotor retardation had significantly lower levels of tonic EDA than non-retarded ones.

Zullino et al. (Switzerland, 2015)

− 58 patients randomly assigned to 8-week Venlafaxine treatment (44.9±8 years; 36.7% females) or applied relaxation (45.6±11.3 years; 46.4% females).

Skin conductance.

Patients treated with Venlafaxine had significantly lower EDA than the other group, both at week 4 and 8.