A DERIVED TRANSFER OF ELICITING EMOTIONAL FUNCTIONS USING DIFFERENCES AMONG ELECTROENCEPHALOGRAMS AS A DEPENDENT MEASURE MICAH AMD 1 ,DERMOT BARNES-HOLMES 1 , AND JASON IVANOFF 2 1 NATIONAL UNIVERSITY OF IRELAND, MAYNOOTH, IRELAND 2 SAINT MARY’S UNIVERSITY, HALIFAX, CANADA Emotional responses have specific electroencephalographic (EEG) signatures that arise within a few hundred milliseconds post-stimulus onset. In this experiment, EEG measures were employed to assess for transfer of emotional functions across three 3-member equivalence classes in an extension of Dougher, Auguston, Markham, Greenway, & Wulfert’s (1994) seminal work on the transfer of arousal functions. Specifically, 12 human participants were trained in the following match-to-sample performances A1 ¼ B1, A2 ¼ B2, A3 ¼ B3 and B1 ¼ C1, B2 ¼ C2, B3 ¼ C3. After successfully testing for the emergence of symmetry relations (B1 ¼ A1, B2 ¼ A2, B3 ¼ A3 and C1 ¼ B1, C2 ¼ B2, C3 ¼ B3), visual images depicting emotionally positive and emotionally negative content were presented with A1 and A3, respectively, using a mixed stimulus pairing–compounding procedure. A2 was paired with emotionally neutral images. Next, EEG data were recorded as participants were exposed to a forced-choice recognition task with stimuli A1, B1, C1, A2, B2, C2, A3, B3, C3 and three novel stimuli A4, B4 and C4. Results yielded differential EEG effects for stimuli paired directly with emotional versus neutral images. Critically, differential EEG effects were also recorded across the C stimuli that were equivalently related to the A stimulus set. The EEG data coincide with previous reports of emotion-specific EEG effects, indicating that the initial emotional impact of a stimulus may emerge based on direct stimulus pairing and derived stimulus relations. Key words: equivalence classes, emotion, valence, affective neuroscience Many of us are aware of how we seemingly develop and exhibit evaluative responses to- wards individuals and/or objects which we have never previously encountered, based solely upon their appearance (Kates, 1959). From a behavior-analytic perspective, Pavlovian condi- tioning and stimulus generalization processes have often been used to explain how initially neutral stimuli may come to acquire appetitive or aversive properties (Cartwright, 2000; Le Pelley et al., 2010; Watson & Raynor, 1920; Winberg, 2005). For example, if individual A behaves in a manner which is perceived to be aversive leading to the avoidant behavior of individual B, it is possible that another individu- al C, who looks a little like A, can acquire similar aversive properties when encountered initially by B. Functionally, the initial evaluation of any stimulus object, including other persons, in- volves the degree to which perceived likeability (valence) towards the target object changes due to well-established behavioral processes, such as classical conditioning and stimulus generaliza- tion (Domjan, 2005). Here, ‘valence’ simply refers to the degree to which a target stimulus is deemed appetitive (positively valenced) or aver- sive (negatively valenced) (Colombetti, 2005; Fazio, Eiser & Shook, 2004; Tolman, 1932). Another behavioral process by which the valence of a stimulus may be established involves the derived transfer of stimulus functions through equivalence relations (Dougher, Au- gustson, Markham, Greenway, & Wulfert, 1994). This process is said to occur when a specific function established for one stimulus in an equivalence class transfers to other stimuli in the class. Imagine, for example, that a person is taught to match a novel stimulus A with another novel stimulus B and to match B with a third stimulus C and then demonstrates the forma- tion of an equivalence relation (e.g., by match- ing C to A in the absence of differential reinforcement). If A is then used to predict the delivery of mild electric shock, and thus elicits evidence of fear, this fear function may also transfer to the B and C stimuli. Critically, the aversive properties observed for the B and C Research conducted by Micah Amd under the supervision of Jason Ivanoff, Department of Psychology, Saint Mary’s University, Halifax. Manuscript prepared with the assistance of Dermot Barnes- Holmes, Department of Psychology, National University of Ireland, Maynooth. Correspondence concerning this article should be addressed to Micah Amd, Department of Psychology, National University of Ireland Maynooth, Co. Kildare, Ireland (e-mail: kudos.ma@gmail.com). doi: 10.1002/jeab.19 JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR 2013, 99, 318–334 NUMBER 3 (MAY) 318