Testing a linear time invariant model for skin conductance responses by intraneural recording and stimulation

Abstract

Skin conductance responses (SCR) are increasingly analyzed with model‐based approaches that assume a linear and time‐invariant (LTI) mapping from sudomotor nerve (SN) activity to observed SCR. These LTI assumptions have previously been validated indirectly, by quantifying how much variance in SCR elicited by sensory stimulation is explained under an LTI model. This approach, however, collapses sources of variability in the nervous and effector organ systems. Here, we directly focus on the SN/SCR mapping by harnessing two invasive methods. In an intraneural recording experiment, we simultaneously track SN activity and SCR. This allows assessing the SN/SCR relationship but possibly suffers from interfering activity of non‐SN sympathetic fibers. In an intraneural stimulation experiment under regional anesthesia, such influences are removed. In this stimulation experiment, about 95% of SCR variance is explained under LTI assumptions when stimulation frequency is below 0.6 Hz. At higher frequencies, nonlinearities occur. In the intraneural recording experiment, explained SCR variance is lower, possibly indicating interference from non‐SN fibers, but higher than in our previous indirect tests. We conclude that LTI systems may not only be a useful approximation but in fact a rather accurate description of biophysical reality in the SN/SCR system, under conditions of low baseline activity and sporadic external stimuli. Intraneural stimulation under regional anesthesia is the most sensitive method to address this question.

Impact of a reminder/extinction procedure on threat-conditioned pupil size and skin conductance responses

“…The DCM analysis provides trial by trial estimates of an anticipatory sudomotor burst, modeled as a Gaussian impulse (Bach et al 2010). In line with intraneural recording results (Gerster et al 2018), we fixed the sudomotor burst duration (SD = 0.3 sec) and constrained the (central) latency of the burst between 0 and 2.5 sec. The estimated response amplitudes used in subsequent statistical analysis (CS− and nonreinforced CS+ trials) were z-scored within participants per experimental session (Staib et al 2015).…”

Psychophysiological modeling: Current state and future directions

“…An example where they are opposed is given by individual response functions for SCR. All evidence suggests that the mapping from sudomotor nerve activity to skin conductance depends on subject‐specific anatomical properties, and is variable between persons (Bach, Flandin, Friston, & Dolan, ; Gerster et al, ). Hence, a forward model taking this heterogeneity into account will have a better goodness‐of‐fit than a model assuming a canonical response function across subjects, as we have also shown empirically (Bach et al, ).…”

“…The earliest PsPMs were developed for SCR and explicitly constituted a sequence of two models (Figure a): a neural model that specifies the mapping of the psychological variable onto sudomotor nerve activity (SNA), and a peripheral (effector organ) model that specifies how SNA maps onto measured SCR (Alexander et al, ; Bach et al, ; Lim et al, ). For SCR, this split is useful because the peripheral model can be evaluated on its own by intraneural stimulation and recordings from well accessible peripheral nerves (Gerster et al, ). For some other measures, the peripheral model can be approximated by specific stimuli, for example, one can use luminance changes to elucidate pupil mechanics.…”

“…SCR to pairs of stimuli separated by different intervals (2–9 s) do not depend on the interval, in line with the linearity principle (Bach, Flandin et al, ). A more direct test of the time invariance principle is furnished by intraneural recordings, which show that 60%–75% of SCR variance is explained by a LTI model that takes SN activity as input, although this is still suffering from interfering non‐SN (e.g., vasomotor nerve) activity (Gerster et al, ). A third approach relies on intraneural stimulation while blocking interfering nerve traffic by regional anesthesia.…”

“…In this case, 93%–99% of SCR variance is explained under a LTI model when stimulation frequency is below 0.6 Hz. Above this stimulation frequency, the LTI model cannot be usefully applied due to strong nonlinearities (Gerster et al, ); however, this limitation should be largely irrelevant for most psychological experiments with slower stimulation rates. To summarize, it appears that under suitable conditions, a LTI model is not just an approximation but rather an accurate description of biophysical reality in the SNA/SCR system.…”