Omission mismatch negativity of speech sounds reveals a functionally impaired temporal window of integration in schizophrenia
Introduction
Humans can detect changes in sounds automatically. This phenomenon first reported by Näätänen et al. (1992) who measured an event-related potential named mismatch negativity (MMN). The MMN is elicited by an infrequent change in homogenous repetitive stimuli, even in the absence of attention. The major MMN sources are the auditory cortex, including Heschl’s and superior temporal gyrus (Hari et al., 1984, Scherg et al., 1989). The MMN has been measured to evaluate the mechanisms underpinning mental disease in terms of cognitive function. It is generated by sensing changes relative to auditory sensory memory (Näätänen et al., 1978). Thus, auditory sensory memory could play a role by transiently keeping the previous sound-image for comparison with each incoming sound. The existence of a transient memory naturally suggests that successive acoustic information should be divided into a unitary image with a given time contrast due to its memory size. Such a unitary event storage was considered to be related to the brain mechanism, the so-called temporal window of integration (TWI), which integrates the closely presented sounds into single information units.
Several studies on the TWI have been reported to date. Tervaniemi et al. (1994) reported that the omitted second tone elicited the MMN only when the offset-onset interval between the paired tones was 40 or 140 ms, but not when this interval was 240 or 340 ms. This finding supported the temporal integration hypothesis that sensory memory traces integrate auditory information concerning closely spaced stimuli into a unitary event. Yabe et al., 1997, Yabe et al., 1998 conducted experiments to determine the exact TWI duration using a paradigm in which a click sound was used as a standard, and its omission was deviant while various stimulus onset asynchrony (SOA) in different blocks. Their results suggested that the duration of the temporal window was 160–170 ms, as a definite MMN was elicited with a constant SOA ranging from 70 to 160 ms, whereas the MMN was not elicited with SOA > 170 ms. This finding indicated that the memory trace of sound-image should be represented as time-functional information from 0 to 170 ms. Additionally, Yabe et al., 2001, Yabe et al., 2004 assessed the effects of omission deviants in complex tone sound stimuli corresponding to the TWI on the MMN, and reported that the amplitude decreased and the latency was prolonged, as omissions were located in late parts of the stimuli. They also reported that the difference in the relative latency to omission indicated the accuracy of the tracing function in sensory memory. Furthermore, discriminative sensitivity was not uniform over the TWI duration because their MMN responses declined nonlinearly toward the temporal window terminal. Moreover, this scientific team (2001) also elicited the MMN with the same peak latency but with a different peak amplitude using two types of deviant stimuli differed from each other only in the period after the silent segment occurrence. The mismatch process obtained by their paradigm was triggered at the moment of change but still lasted after detection of deviation. This finding suggested that standard and deviant stimuli were treated as a unitary event within a TWI. Especially, the TWI integrates neighboring sounds and/or silent into a single percept.
A recent research in healthy control participants using omission in complex speech sounds suggested that the MMN latency was prolonged and the amplitude decreased, when omissions were located in late parts of stimuli (Hoshino et al. 2020, unpublished data, in submission). This implies that the automatic auditory discrimination function in the TWI is most robustly strengthened at approximately 50 ms, and discriminative capacity decreases toward the latter half of the TWI. In psychiatry, the attenuation of MMN amplitude in patients with schizophrenia has been reported in many studies, and is suggestive of dysfunction in the preattentive processing of auditory information (Shelley et al., 1991). In particular, the duration MMN (dMMN) has a more significant decrease in amplitude (Todd et al., 2008). Moreover, the dMMN abnormality has a greater detection sensitivity for schizophrenia compared to the frequency and intensity MMN because of the potential functional impairment of the TWI in patients with schizophrenia. Especially, impairments in the dMMN may be enhanced owing to greater deficits in the latter half of the TWI. Further, after studying the sound effects of the MMN in patients with schizophrenia, Kasai et al. demonstrated a language-related dysfunction at the early stage of auditory processing of speech sounds by comparing the MMN of tone and speech sounds (Kasai et al., 2002, Kasai et al., 2003). Therefore, their results suggested that speech sounds were more appropriate in investigating the characteristics of schizophrenia pathophysiology compared to the tone sounds.
Based on these findings, we considered that a paradigm corresponding to the TWI, composed of speech sound, might identify the functional changes in auditory sensory memory associated with the TWI in patients with schizophrenia. We hypothesized that the MMN corresponding to the latter part of the TWI may be different in such patients compared with healthy controls, when assessed by measuring the MMN to omitted segments in complex speech sounds. Therefore, the present study aimed to identify the functional changes in auditory sensory memory associated with the TWI in patients with schizophrenia by investigating the effects of omission in complex speech stimuli on the MMN.
Section snippets
Participants
Our participants were 17 male patients with schizophrenia and 15 age- and sex-matched healthy control individuals (Table 1). All patients were diagnosed with schizophrenia at Fukushima Medical University according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition. All patients received antipsychotic medications. Clinical symptoms were assessed using the Positive and Negative Syndrome Scale and Global Assessment of Functioning. The ethical committee of Fukushima Medical
Results
The grand average difference waveforms for each deviant stimulus in both groups are presented in Fig. 2 by using the average of T5/T6, as a reference. The median relative latency and amplitudes of both groups are presented in Fig. 3. We observed significant differences between the relative median latencies to omitted segments at different times over the TWI. In patients with schizophrenia, significant differences were observed at Fz and Cz when stimulating each ear and the left ear,
Discussion
In this study, we investigated the MMN effects on latency and amplitude using complex speech stimuli with omissions over the course of the TWI. The MMN latency was shorter when either ear was stimulated by omissions at D2, D3 and D4 as a common tendency in both groups. In addition, the latency tended to be prolonged for both ears for omissions at D5 and D6. In the schizophrenia group, these tendencies were significantly different. In addition, the latency corresponding to D1 was prolonged when
Conclusions
We examined the effects of partial temporal omission on MMN in patients with schizophrenia using a speech sound stimulus paradigm corresponding to the TWI. Regarding the MMN pattern of intercomparison in each group, the MMN latency was shorter when either ear was stimulated by omissions at the middle part of TWI, presenting a common tendency in both groups. In addition, the latency tended to be prolonged for both ears for omissions at the late part of TWI. However, in the schizophrenia group,
Author contributions
Conception and design of the study: YM, HY. Acquisition of EEG: YM, HH, KK. Data analysis: YM, HH. Drafting the manuscript and figures: YM. All authors reviewed the manuscript and approved it for submission.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We appreciate the cooperation and time commitment of all physicians that enrolled in this work.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (29)
- et al.
Prediction of psychosis by mismatch negativity
Biol Psychiatry
(2011) - et al.
Responses of the primary auditory cortex to pitch changes in a sequence of tone pips: neuromagnetic recordings in man
Neurosci Lett
(1984) - et al.
Impairment of early cortical processing in schizophrenia: an event-related potential confirmation study
Biol Psychiatry
(1993) - et al.
Neuromagnetic correlates of impaired automatic categorical perception of speech sounds in schizophrenia
Schizophr Res
(2003) - et al.
Effects of olanzapine on auditory P300 and mismatch negativity (MMN) in schizophrenia spectrum disorders
Prog Neuropsychopharmacol Biol Psychiatry
(2005) - et al.
Hemispheric lateralization in an analysis of speech sounds. Left hemisphere dominance replicated in Japanese subjects
Brain Res Cogn Brain Res
(2000) - et al.
The neural basis of sublexical speech and corresponding nonspeech processing: a combined EEG-MEG study
Brain Lang
(2014) - et al.
Early selective-attention effect on evoked potential reinterpreted
Acta Psychol (Amst)
(1978) - et al.
Right hemisphere dominance of different mismatch negativities
Electroencephalogr Clin Neurophysiol
(1991) - et al.
Mismatch negativity: an index of a preattentive processing deficit in schizophrenia
Biol Psychiatry
(1991)