Skin conductance responses
This protocol is extracted from research article:
The neural circuitry of affect-induced distortions of trust
Sci Adv, Mar 13, 2019; DOI: 10.1126/sciadv.aau3413

SCRs were collected using a PowerLab 4/25T amplifier with a GSR Amp (ML116) unit and a pair of MR-compatible finger electrodes (MLT117F), which were attached to the participants’ left middle and ring finger via dedicated adhesive straps after application of conductance gel. Participants’ hands had been washed using soap without detergents before the experiment. Stable recordings were ensured before starting the main experiment by waiting for signal stabilization during training and stimulation intensity calibration. LabChart (v. 5.5) software was used for recordings, with the recording range set to 40 μS and using initial baseline correction (“subject zeroing”) to subtract the participant’s absolute level of electrodermal activity from all recordings (all specs for devices and electrodes from ADInstruments Inc., Sydney, Australia).

Preprocessing and statistical analyses of SCR data were performed using PsPM (PsychoPhysiological Modelling) (57). Because of technical problems, data from six participants included only one run (of two), and data from one participant were lost. Each participant’s SCR data were downsampled to 10 Hz, low-pass filtered (cutoff frequency, 5 Hz), and subsequently z-transformed. Statistical analysis of the skin conductance data followed the approach commonly used in analyses of fMRI data. Specifically, multiple linear regression was used to estimate SCRs during decisions made in each of the task conditions, that is, during trust and NS control tasks and in the context of threat and no-threat treatment blocks. All events were modeled as Dirac delta functions and convolved using the canonical SCR function together with its temporal derivative (58). We took two precautions to ensure that electrical shocks did not influence estimates of arousal during the decision period: First, we modeled all shock events, entering them as four regressors that reflect the time points of strong or weak shock administration, separately for random (unpredictable) shocks and predictable shocks associated with the block cue. Second, we removed from the regressors of interest all trials during which a shock occurred within an interval from 5 s before the onset of the decision screen until the button press. All these trials were added as two regressors of no interest, reflecting decisions made in the presence of strong and weak shocks. The statistical model therefore included a total of 11 regressors that reflected the onset times of decision screens in trust and NS control trials under expectancy of strong and weak electrical shocks (four regressors of interest during which no shock occurred), cue times indicating the onset of a block, delivery times of tactile stimulation (with four separate regressors including all predictable strong and weak shocks and all unpredictable strong and weak shocks), and the regressors of no interest for decision trials during which a weak or a strong shock occurred. Regressor estimates (β weights) for each condition were then used in follow-up analyses (one-sample t tests contrasting the difference between threatening and nonthreatening conditions and strong versus weak shocks) reported in Results.

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.