Cognitive Neuroscience and Neuropsychology

The neural basis of auditory perception can be investigated non-invasively by recording event-related brain potentials (ERPs) during auditory stimulation.1 ERPs can be conceptualized as a sequence of components each with its own set of neural generators and reflecting different psychological functions.2 The ERP component termed the mismatch negativity (MMN) reflects a pre-attentive auditory deviance detection process based on auditory sensory memory traces.3–5 MMN is typically elicited by infrequent discriminable changes in sound parameters such as frequency, intensity, or duration in otherwise homogeneous series of a repetitive standard stimulus. MMN relies on a transient store of auditory sensory information for detecting deviation from invariant features of the acoustic environment (such as repetition of a stimulus).5,6 This transient auditory store is probably identical to what experimental psychology terms as auditory sensory memory (or echoic memory).3,7 Because MMN is elicited even when subjects disregard the auditory stimulation (e.g. while playing a visual computer game or performing a difficult visual discrimination task) MMN can be used as a pre-attentive index of auditory sensory memory traces.8,9 Most MMN studies have employed pure tones to probe auditory sensory memory traces. For such simple sounds, acoustic parameters and perceived qualities are not easily distinguishable. By employing complex missing-fundamental tones10 that allowed one to separate the effects of spectral and virtual pitch11 on the MMN, it was shown recently that the memory traces involved in the MMN process encoded these sounds in terms of their perceived (virtual) pitch rather than in terms of the spectral pitch of their components.12 Missing-fundamental tones are perceived as having a pitch identical to that of their fundamental frequency, even though the fundamental itself is not present in the frequency spectrum of these complex tones.10 A subsequent study revealed that resolving the missing-fundamental pitch requires relatively long acoustic samples.13 This was demonstrated by the elicitation of MMN by 500 ms missing-fundamental tones but not at a stimulus duration of 150 ms. Moreover, the subjects’ performance in discriminating the same complex tones indicated that the these sounds could only be discriminated at the longer stimulus duration.13 The goal of the present study was to determine whether or not MMN is elicited by infrequent changes in sound timbre with no corresponding change in pitch.14 Timbre denotes the perceived auditory quality describing sound ‘colour’. Timbre can be divided to three components: the temporal component (duration of the constant-amplitude part of the sound, rise and fall times), the spectral Cognitive Neuroscience and Neuropsychology