Measures of skin conductance and heart rate in alcoholic men and women during memory performance

Abstract

We examined abnormalities in physiological responses to emotional stimuli associated with long-term chronic alcoholism. Skin conductance responses (SCR) and heart rate (HR) responses were measured in 32 abstinent alcoholic (ALC) and 30 healthy nonalcoholic (NC) men and women undergoing an emotional memory task in an MRI scanner. The task required participants to remember the identity of two emotionally-valenced faces presented at the onset of each trial during functional magnetic resonance imaging (fMRI) scanning. After viewing the faces, participants saw a distractor image (an alcoholic beverage, nonalcoholic beverage, or scrambled image) followed by a single probe face. The task was to decide whether the probe face matched one of the two encoded faces. Skin conductance measurements (before and after the encoded faces, distractor, and probe) were obtained from electrodes on the index and middle fingers on the left hand. HR measurements (beats per minute before and after the encoded faces, distractor, and probe) were obtained by a pulse oximeter placed on the little finger on the left hand. We expected that, relative to NC participants, the ALC participants would show reduced SCR and HR responses to the face stimuli, and that we would identify greater reactivity to the alcoholic beverage stimuli than to the distractor stimuli unrelated to alcohol. While the beverage type did not differentiate the groups, the ALC group did have reduced skin conductance and HR responses to elements of the task, as compared to the NC group.

Keywords: Alcoholism; Emotion; Heart rate; Memory; Psychophysiology; Skin conductance.

Figure 1. This figure shows the flow of the delayed match-to-sample task paradigm.

Rows provide examples of the three facial emotion types (positive, neutral, negative) and the three distractor types (scrambled, nonalcoholic, alcoholic). The rows also represent both matching and non-matching probe faces (match, nonmatch, match). The faces have been obscured in this publication to protect the identity of the individuals who were photographed. The participants saw the entire un-obscured faces.

Figure 2. For the encoded faces, alcoholic participants had fewer skin conductance responses (SCR) than nonalcoholic participants.

Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue), split by gender. The asterisk indicates a significant difference, and error bars represent standard error. No significant interactions were identified. (A) ALC women had significantly fewer SCRs (>0.03 microsiemens (µS)) to the encoded faces than ALC men. (B) SCR amplitudes (µS) to the same stimuli appeared to be similar.

Figure 3. Skin conductance responses (SCR) to distractor cues were reduced for alcoholic participants compared to nonalcoholic controls.

Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue), split by gender. Asterisks indicate significant differences, and error bars represent standard error. No significant interactions were identified. (A) ALC women had significantly fewer SCRs (>0.03 microsiemens (µS)) to the distractor images than ALC men. (B) SCR amplitudes (µS) to the same stimuli revealed a similar pattern.

Figure 4. Skin conductance responses (SCR) to the distractor cues were not significantly related to the type of distractor for alcoholic participants compared to nonalcoholic controls.

Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue). Error bars represent standard error. (A) A significant interaction of Group × Distractor Type was not identified for SCR counts. (B) SCR amplitudes (µS) also did not reveal a significant interaction.

Figure 5. Skin conductance responses (SCR) to probe faces were reduced for alcoholic participants compared to nonalcoholic controls.

Nonmatch SCRs are in blue and match SCRs are in red. Asterisks indicate significant differences between match status, and error bars represent standard error. No significant interactions were identified. (A) Compared to women, men had significantly more SCRs (>0.03 microsiemens (µS)) to the matching probe faces than the nonmatching probe faces. The data are shown for all alcoholic and nonalcoholic participants together. (B) For neutral and positive faces, SCR amplitudes were higher for faces that matched the encoded faces. Data are shown for alcoholic and nonalcoholic men and women together.

Figure 6. Heart rate (HR) tended to decline more in alcoholics in response to the encoded faces than in nonalcoholics.

Bar height indicates beats per minute (BPM). Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue), split by gender. The asterisk indicates a significant difference, and error bars represent standard error. No significant interactions were identified.

Figure 7. The increase in heart rate (HR) to distractor cues depended upon the group and the emotion of the preceding face.

Heart rate (HR) increased by 0.3 beats per minute (BPM) in response to the distractor element, and a significant Group × Gender × Face Valence interaction was identified. Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue). Error bars represent standard error. (A) Shows values for men. (B) Shows values for women.

Figure 8. Heart Rate (HR) responses to the distractor cues were not significantly related to the type of distractor for alcoholic participants compared to nonalcoholic controls.

Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue). Error bars represent standard error. A significant interaction of Group × Distractor Type was not identified for HR counts.

Figure 9. Heart rate (HR) significantly declined but was not significantly related to the group, the type of distractor, or the emotion of the face presented for the probe.

Bar height indicates beats per minute (BPM), with standard error bars. Data are presented for alcoholic participants (ALC, in red) and nonalcoholic participants (NC, in blue), split by gender. No significant interactions were identified.