Electrocortical and electrodermal responses covary as a function of emotional arousal: A single-trial analysis ANDREAS KEIL, a J. CARSON SMITH, b BETHANY C. WANGELIN, a DEAN SABATINELLI, a MARGARET M. BRADLEY, a and PETER J. LANG a a Center for the Study of Emotion and Attention, University of Florida, Gainesville, Florida, USA b College of Health Sciences, University of Wisconsin–Milwaukee, Milwaukee, Wisconsin, USA Abstract Electrophysiological studies of human visual perception typically involve averaging across trials distributed over time during an experimental session. Using an oscillatory presentation, in which affective or neutral pictures were presented for 6 s, flickering on and off at a rate of 10 Hz, the present study examined single trials of steady-state visual evoked potentials. Moving window averaging and subsequent Fourier analysis at the stimulation frequency yielded spectral amplitude measures of electrocortical activity. Cronbach’s alpha reached values 4.79, across electrodes. Single- trial electrocortical activation was significantly related to the size of the skin conductance response recorded during affective picture viewing. These results suggest that individual trials of steady-state potentials may yield reliable indices of electrocortical activity in visual cortex and that amplitude modulation of these indices varies with emotional engagement. Descriptors: Dense-array EEG, Emotion, Motivation, Arousal, Picture perception, Skin conductance response Hemodynamic imaging and electrophysiological studies have converged to show that activity in visual cortex is facilitated when viewing affective pictures, with the amplitude of neural signals increasing as a function of the emotional intensity rated by par- ticipants (e.g., Bradley et al., 2003). Adding to an increasing body of hemodynamic neuroimaging studies, work capitalizing on event-related potentials (ERPs) has focused on the time course of this modulation. These studies have converged to show late (4 300 ms) ERP enhancements for affectively arousing stimuli, which correlate significantly with hemodynamic parameters (Sabatinelli, Lang, Keil, & Bradley, 2007). There has also been evidence that the steady-state visual evoked potential (ssVEP), which is an oscillatory brain response to rapidly flashing stimuli, shows a similar sensitivity to emotional content (Keil et al., 2003). To achieve sufficient signal-to-noise ratios, the electrophys- iological studies discussed above relied on averaging signals across experimental trials in which different exemplars with sim- ilar emotional content were presented (e.g., erotica, landscapes, mutilation, etc.). Category-based averaging to obtain ERP time series has been a fruitful approach to examining the spatial and temporal electrocortical dynamics associated with emotional picture perception. In addition, covarying cross-category ERPs with averaged data reflecting peripheral physiology has yielded substantial evidence for content-related changes in visual per- ception and physiological reactivity (e.g., Cuthbert, Schupp, Bradley, Birbaumer, & Lang, 2000). For some experimental questions, however, it would be desirable to measure electrocor- tical activation associated with a specific picture in a brief time interval. This might help, for instance, to better understand trial- by-trial changes related to repeated exposure to the same stim- ulus. Alternatively, research in clinical populations may benefit from studying visual responses to single scenes, for example, one representing a high-fear situation. In addition, changes of such specific responses over the course of treatment could be examined using single-trial approaches. The ssVEP is a measure that can produce reliable estimates of electrocortical processes with short epochs of data. It is a con- tinuous brain response elicited by a repetitive visual stimulus that is periodically modulated in intensity at a fixed rate of 6–8 Hz or greater. Phenomenologically, a visual stimulus appears to flicker when using steady-state presentation. Using electroencephalo- graphy, these signals can be recorded at the scalp as an oscillatory waveform that has the same fundamental frequency as the flick- ering stimulus (Regan, 1989). Frequency-domain analyses can then be used to extract the response of interest at the known stimulation frequency, resulting in high signal-to-noise ratios even when the number of trials is limited. The amplitudes of ssVEPs have been shown to be sensitive to cognitive and affective features of experimental tasks, including spatial selective attention (Mu¨ ller, Malinowski, Gruber, & Hillyard, 2003), The authors thank Sabine Heim for comments on an earlier draft of this article. Research was supported by grants from the Deutsche For- schungsgemeinschaft and from the NIMH. Address reprint requests to: Andreas Keil, Ph.D., Department of Psychology and NIMH Center for the Study of Emotion & Attention, University of Florida, PO Box 112766, Gainesville, FL 32611, USA. E-mail: akeil@ufl.edu Psychophysiology, 45 (2008), 516–523. Wiley Periodicals, Inc. Printed in the USA. Copyright r 2008 Society for Psychophysiological Research DOI: 10.1111/j.1469-8986.2008.00667.x 516