Biological Psychology 88 (2011) 65–71 Contents lists available at ScienceDirect Biological Psychology journa l h o me page: www.elsevier.com/locate/biopsycho Prefrontal asymmetry predicts affect, but not beliefs about affect Amanda R.W. Steiner ∗ , James A. Coan University of Virginia, United States a r t i c l e i n f o Article history: Received 25 April 2010 Accepted 23 June 2011 Available online 7 July 2011 Keywords: Emotion Self-report EEG Prefrontal asymmetry Homesickness a b s t r a c t Self-reported emotional feelings are easily biased by situational or identity-related beliefs. Such biases vary as a function of memory. Recent memories draw more on veridical felt affective experience whereas distant memories draw more on situational or identity-related biases. For this study, frontal EEG asym- metry was used to predict feeling- versus belief-based self-reports of freshmen year homesickness in college freshmen and sophomores. Relatively greater right frontal EEG asymmetry predicted greater feeling-based, experiential reports of freshman year homesickness, whereas no associations were found between frontal EEG asymmetry and belief-based, retrospective reports of freshman year homesickness. These results support the status of frontal EEG asymmetry as a measure of affective vulnerability and suggest that links between frontal EEG asymmetry and self-reported affect are detectable to the extent that self-reports capture current emotional feelings and not situational or identity-related beliefs about what one ought to have felt. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Despite concerns about their accuracy, reliability, valid- ity, and utility—about their susceptibility to many forms of bias—psychophysiologists interested in emotional experience remain beholden to subjective self-reports (e.g., Christensen et al., 2003; Gray and Watson, 2007; Levine et al., 2006; Robinson and Clore, 2002b; Watson, 2004; Wilson et al., 2003). This is no less true in the era of social and affective neuroscience, where inferences about emotional experience involving, for example, electroen- cephalography (EEG) or functional magnetic resonance imaging (fMRI), are partially dependent upon (and limited by) experience measures of uncertain reliability and validity (Humrichouse et al., 2007; Watson and Walker, 1996). Some have argued that the problem of self-report bias is so severe that self-report measures are at best irrelevant and at worst misleading (e.g., LeDoux, 1996; Öhman, 1999). By contrast, we feel that sound measurements of subjective experience are central to the successful linking of neural constructs to psychological or expe- riential constructs (cf., Hagemann, 2004). In the coming decades, advances in our ability to infer links between neural functioning and subjective experiences are likely to be critical for the matur- ing disciplines of social and affective neuroscience. The neglect or abandonment of self-report measures would almost certainly ∗ Corresponding author at: Department of Psychology, University of Virginia, P.O. Box 400400, Charlottesville, VA 22904, United States. Tel.: +1 434 982 4750; fax: +1 434 982 4766. E-mail address: arw6z@virginia.edu (A.R.W. Steiner). impede our ability to fully understand emotion at a neural level. To the extent that neural, behavioral, and experiential measures are interdependent, there is a methodological imperative to evaluate the nature of associations between neural activity and self-report measures of emotion. The question is not whether self-report measures of emo- tion are in some broad and objective sense capable of providing an “accurate” representation of emotion-related neural processes. Self-report measures may after all reflect many different forms of emotion-related knowledge, each with its own potential neural sig- nature. For this paper, we are chiefly concerned with distinguishing two sources of information from which an individual may draw in formulating a subjective experience report. The first concerns beliefs about how one ought to respond given one’s sense of identity or situational demands. The second is a kind of “online” conscious readout of ongoing neural or physiological processes. For this paper, we characterize these two forms of self-report as being grounded in belief and feeling, respectively. Of interest to us is whether a common putative psychophysiological measure of personality and affective responding—frontal EEG asymmetry—measures one, the other, or both. 2. Sources of information in self-report: belief versus feeling Robinson and Clore (2002a,b) and others (e.g., Gilbert et al., 1998; Wilson et al., 2003) have argued that the amount of time lapsed between an experience and questions about that experi- ence can lead to important differences in how research participants 0301-0511/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.biopsycho.2011.06.010