Contents lists available at ScienceDirect Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr Research report Long-term reliability of the visual EEG Poffenberger paradigm Patrick Friedrich a, ⁎ ,1 , Sebastian Ocklenburg a,1 , Lisa Mochalski a , Caroline Schlüter a , Onur Güntürkün a,b , Erhan Genc a a Institute of Cognitive Neuroscience, Biopsychology, Department of Psychology, Ruhr-University of Bochum, Germany b Stellenbosch Institute for Advanced Study (STIAS), Wallenberg Research Centre at Stellenbosch University, Stellenbosch 7600, South Africa ARTICLE INFO Keywords: Poffenberger paradigm Visual perception Reliability Interhemispheric transfer Corpus callosum ABSTRACT The Poffenberger paradigm is a simple perception task that is used to estimate the speed of information transfer between the two hemispheres, the so-called interhemispheric transfer time (IHTT). Although the original paradigm is a behavioral task, it can be combined with electroencephalography (EEG) to assess the underlying neurophysiological processes during task execution. While older studies have supported the validity of both paradigms for investigating interhemispheric interactions, their long-term reliability has not been assessed systematically before. The present study aims to fill this gap by determining both internal consistency and long- term test-retest reliability of IHTTs produced by using the two different versions of the Poffenberger paradigm in a sample of 26 healthy subjects. The results show high reliability for the EEG Poffenberger paradigm. In contrast, reliability measures for the behavioral Poffenberger paradigm were low. Hence, our results indicate that electrophysiological measures of interhemispheric transfer are more reliable than behavioral measures; the later should be used with caution in research investigating inter-individual differences of neurocognitive measures. 1. Introduction The corpus callosum is the main connection between the two hemispheres and consists of about 200 million axons [1]. It is mainly composed of excitatory glutamatergic fibers, but can serve an inhibitory role due to GABAergic interneurons within the receiving hemisphere [2,3]. It is important for several different cognitive processes, ranging from visual perception [4,5,6,7,8] and motor activity [9,10] to higher cognitive functions such as decision-making [11], working memory [12], learning [13] and language [14]. One particularly important issue in the context of callosal research is to find reliable measurements of callosal functions. This issue is currently of special interest because of concerns about the replicability of neuroscientific and psychological studies [15]. Historically, a commonly used method to investigate the function of the corpus callosum is the classical Poffenberger paradigm [16,17]. In this task, visual stimuli (e.g. flashing white circles) are presented either in the left or the right visual half field. Participants have to react by pressing a button with either the ipsilateral or contralateral hand. Trials in which stimulus and reacting hand are on the same side are called “uncrossed”, whereas trials in which stimulus and reacting hand are on opposite sides are called “crossed”. In the uncrossed condition, neural correlates of perception and motor re- sponse are located within the same hemisphere. In the crossed condition, however, the perception is primarily located in one hemi- sphere while the other hemisphere accomplishes the motor output. A comparison between the reaction times (RT) of uncrossed and crossed conditions has shown that for uncrossed trials the RTs are on average 3 milliseconds (ms) faster than for crossed trials [18]. Subtracting the RTs in the uncrossed condition from the RTs in the crossed condition results in the so-called “crossed-uncrossed” difference (CUD). Since this measure is estimated by RTs and thus behavior, one can also name it behavioral CUD (bCUD). This difference measure is thought to reflect the additional processing time of the crossed condition, in which the perceptual information has to transfer from one hemisphere to the other in order to trigger the motor response. Hence, the bCUD is interpreted as an estimate for interhemispheric transfer time (IHTT), which in turn should be associated with structural properties of commissural fiber bundles. Indeed, an association between structural variability of white matter fibers and bCUD has been suggested for the corpus callosum: For example, a smaller bCUD is associated with bigger callosal size ([19]: r= -0.50, p < 0.05), as well as higher fractional anisotropy (Schulte et al., 2005: r = -0.54, p < 0.05), a measure of microstructural integrity that reflects an efficient white matter architecture [20]. Nevertheless, other studies did not find an association between bCUD http://dx.doi.org/10.1016/j.bbr.2017.05.019 Received 7 December 2016; Received in revised form 24 March 2017; Accepted 10 May 2017 ⁎ Corresponding author at: Abteilung Biopsychologie, Institut für Kognitive Neurowissenschaft, Fakultät für Psychologie, Ruhr-Universität Bochum, Universitätsstraße 150, 44780 Bochum, Germany. 1 These authors contributed equally to this work. E-mail address: patrick.friedrich@rub.de (P. Friedrich). Behavioural Brain Research 330 (2017) 85–91 Available online 12 May 2017 0166-4328/ © 2017 Elsevier B.V. All rights reserved. MARK