Affective versus cognitive responses to musical chords: An ERP and behavioral study.

Abstract

When people listen to music, various brain processes are involved, including those that assess the rhythmic, tonal, timbral, and structural properties of the music-which we will call cognitive processing-as well as other processes, such as affective processing. Cognitive processing can, for example, allow us to implicitly notice when a note is out of key or out of rhythm by building a musical context out of the incoming sound stream. By affective processing, on the other hand, we refer to those brain processes that allow us to perceive or feel emotion in response to music, and to feel enjoyment of music or even find it beautiful. Only few studies on the processing of musical sounds have differentiated the psychological mechanisms underlying cognitive and affective processes (Gagnon & Peretz, 2000), as have some studies of other types of aesthetic stimuli, such as pictures (Mandler & Shebo, 1983). Even less is known about the brain correlates of those processes and their chronometric succession (e.g., see reviews by Brattico, Bogert, & Jacobsen, 2013; Jacobsen, 2006).Our current understanding of the information processing of music in the brain is mostly based on cognitive responses to musical sound patterns and chords, such as discrimination of sound feature changes or detection of tonality violations (for reviews, see the books by Koelsch [2012] and Peretz & Zatorre [2003]). A useful method to study the chronometric ordering of stages in auditory processing is the event-related potential (ERP), a neurophysiological signal elicited in response to a sensory or behavioral event, which comprises several components that are supposed to originate from distinct neuronal populations, (Luck, 2014). The ERP is measured by averaging time-accurate electroencephalogram (EEG) recordings over many trials of stimuli and across tens of participants. ERP studies have evidenced the early sensory encoding of sound features in the primary auditory cortex, as indexed by the N1 component, which peaks centrally at around 100 ms after stimulus onset (Naatanen & Picton, 1987). The subsequent stage of predictive coding and error detection (the error between predictions based on the regularities in the auditory stimulation and the incoming sound) is indexed by the mismatch negativity (MMN) component of the ERP, which is elicited by a feature change in a repeating stimulus sequence and peaks frontally around 150 to 200 ms after stimulus onset (Friston, 2005; Naatanen, Gaillard, & Mantysalo, 1978; Picton, Alain, Otten, Ritter, & Achim, 2000). The change can be simple, such as a sequence of identical tones being interrupted by a tone of a different frequency (Sams, Paavilainen, Alho, & Naatanen, 1985), or it can be more abstract, such as the violation of basic rules in tone patterns or combinations (e.g., the contrast between descending and ascending intervals: Saarinen, Paavilainen, Schoger, Tervaniemi, & Naatanen, 1992; or major and minor chords: Virtala et al., 2011).Cognitive processing of more complex rules involving an integration of sounds over longer time spans has been associated with the early right anterior negativity (ERAN), as well as the later P300. Similarly to the MMN, the ERAN peaks frontally around 200 ms after the onset of an unexpected musical sound; however, whereas the MMN is based on regularities between presently occurring successive sounds, the ERAN is a marker of unusual chord successions that depends on representations of harmonic regularities from long-term memory and can be caused by longdistance music-syntactic dependencies (Koelsch, Gunter, Friederici, & Schroger, 2000; Koelsch & Jentschke, 2010). A later ERP component related to chord processing is the P300, an endogenous, attentive response related to judgment of a stimulus (Squires, Squires, & Hillyard, 1975). The P300 peaks parietally 220 to 500 ms after the onset of a target tone in a discrimination task (Polich, 2007), of an incongruous tone presented at the end of a melody (e. …