| | Mismatch Negativity (MMN) evoked by sound duration contrasts: An unexpected major effect of deviance direction on amplitudesAccepted 3 October 2008. Abstract ObjectiveVerify and explore unexpected results suggesting an effect of deviance direction (shorter or longer deviants) on the amplitude of MMNs evoked by sound duration contrasts. MethodsMMNs were recorded using the oddball paradigm on ten adults. Four standard stimulus durations (100, 150, 200 and 250 ms) were used and deviants were 50% shorter or longer. Behavioral data (hit rates, d′, and reaction times) were collected after the electrophysiological sessions. ResultsMMNs were larger for short than for long deviants. There was no effect on MMN latencies. Hit rates and d′ data were almost at ceiling level for all conditions even for the longest standard – long deviant combination in which the MMN was abolished. ConclusionsWe argue that the deviance direction effect on MMN amplitudes can be explained by the delay between the moment of deviance detection and the end of the deviance quantification process. SignificanceA major effect of deviance direction on amplitudes was confirmed. This effect, which was confined to electrophysiological data, is to be taken into account when using duration contrasts to probe the processing of temporal information. 1. Introduction  Contrary to vision, in which a still picture frozen in time may convey a lot of information, audition processes stimuli that intrinsically unfold in time. The temporal course of auditory stimuli conveys important information in speech (Rosen, 1992) and music (Drake and Botte, 1993). Deficits in encoding temporal aspects of auditory information are thought to be involved in several types of speech or speech-related disorders (e.g. auditory neuropathy; Zeng et al., 2005; dyslexia; Wright et al., 1997). The processing of auditory temporal information by the brain may be examined, with a high temporal resolution, by recording Event-Related Potentials (ERPs). Among auditory ERPs, the MMN is considered to be an attention-independent process indexing the automatic detection of deviance of a rarely occurring stimulus with respect to frequent, standard background stimuli (for reviews, see Näätänen, 1992, Lang et al., 1995, Ritter et al., 1995, Näätänen and Alho, 1997, Schröger, 1997, Picton et al., 2000a, Näätänen et al., 2007). There are numerous studies suggesting that the MMN is the outcome of an automatic comparison process between a new, deviant stimulus and the memory trace formed by the sensory representation of the standard stimulus within short-term memory (Novak et al., 1990, Näätänen, 1990, Näätänen, 1992, Cowan et al., 1993, Cowan, 1995, Näätänen and Alho, 1995). The MMN can be elicited by many types of acoustic deviance (e.g. frequency contrast: Alho et al., 1990; intensity contrast: Näätänen et al., 1987; spatial localisation contrast: Paavilainen et al., 1989; duration contrast: Näätänen et al., 1989b) as well as by contrasts in abstract features of complex stimuli such as in music (e.g. Tervaniemi et al., 1993), speech (e.g. Kraus et al., 1992), or even spatial (Colin et al., 2002b; Stekelenburg et al., 2004) and phonetic (Colin et al., 2002a, Colin et al., 2004) illusory percepts. The MMN peak amplitude is usually reported to be positively correlated with contrast size (e.g. Sams et al., 1985; Näätänen et al., 1989a; Pihko et al., 1995; Amenedo and Escera, 2000; Jaramillo et al., 2000; Pakarinen et al., 2007), which, as attested by behavioral experiments, is itself positively correlated with the easiness of deviance detection (e.g. Sams et al., 1985; Pakarinen et al., 2007). MMN peak latency, on the contrary, decreases as contrast size and easiness of deviance detection increase (e.g. Sams et al., 1985; Näätänen et al., 1989a; Pihko et al., 1995), although this may not be the case for every dimension (see for example results from Pakarinen et al., 2007 showing similar MMN latencies across contrast sizes for intensity and duration contrasts). The present report stems from observations made during a study of MMN parameters across a wide range of duration contrasts. The data accumulated up to now strongly suggested that there was an unanticipated systematic amplitude difference between MMNs evoked by shorter vs longer deviants. The data analysis reported here was performed in order to verify the serendipitous finding of an effect of deviance direction on the MMN parameters, an issue that has been hardly addressed in the literature. Studies on the effect of duration deviance direction on MMN amplitude are controversial. In most of the studies examining the MMN elicited by duration contrasts, duration decrements were used (e.g. Kaukoranta et al., 1989; Näätänen et al., 1989b; Pihko et al., 1995; Joutsiemi et al., 1998; Tervaniemi et al., 1999; Kasai et al., 2001; Escera et al., 2002; Grimm et al., 2004, Grimm et al., 2006; Ilvonen et al., 2004; Näätänen et al., 2004; Takegata et al., 2004; Grimm and Schröger, 2005; Sysoeva et al., 2006). Only Michie et al. (2000) and Todd et al., 2000, Todd et al., 2003 used longer duration deviants in their studies on temporal discrimination in schizophrenia because it had been shown that a long deviant condition better discriminates between patients with a diagnosis of schizophrenia and healthy controls (Shelley et al., 1991). Only a few studies directly compared the effects of duration increments and decrements. Working with spectrally rich tones, Amenedo and Escera (2000) used two standard durations (50/100ms) and shorter or longer deviant tones differing from the standards by 10%, 20%, 30%, 40% or 60%. The MMN amplitude was found to be independent of the direction of the change and of standard duration. The authors concluded that brain detection of these duration deviance did not depend on the absolute physical length of the two sounds but on their relative duration. Similar results were obtained with pure tones by Näätänen et al. (1989b) using 25–50 and 100–200ms contrasts and by Shelley et al. (1991) using a 50–100ms contrast. On the other hand, Catts et al. (1995) as well as Jaramillo et al. (1999) found that long deviant pure tones evoked larger MMN amplitudes than short deviant ones. However, the same authors failed to obtain such effect with white-noise bursts (Jaramillo et al., 2000) and even found the reverse pattern for vowels (Jaramillo et al., 1999). Finally, in a very recent study, Takegata et al. (2008) compared duration decrements and increments using 200/120 and 400/240ms standard/deviant contrasts. The first contrast yielded similar MMNs for short and long deviants, whereas for the second contrast, an MMN was found for short deviants only. This study analysed the latencies and amplitudes of the MMNs evoked by duration contrasts based on four standard durations, each of which being associated with a short and a long deviant of identical relative deviance magnitude. For each duration of the standard stimulus, the short and long deviants durations differed from the standard ones by a fixed proportion (50%). The null hypothesis was that, because of the use of a single relative duration contrast (corresponding to a constant Weber fraction), the MMNs latencies and amplitudes would be identical across conditions. Deviance magnitude took the form of a fixed proportion added to or subtracted from the standard duration rather than a fixed absolute amount because it was assumed that it is more difficult to detect a fixed duration deviance among long duration standard stimuli than among short ones. Considering e.g. a fixed duration deviance of 50ms, a 150ms deviant in a 200ms standard condition would likely be more easily detected than a 250ms deviant in a 300ms standard condition, whereas with a fixed proportion of 50% both a 100ms deviant in a 200ms standard condition and a 150ms deviant in a 300ms standard condition should equally well be detected. As the size of the contrast is known to impact on MMN amplitude at least (e.g. Sams et al., 1985; Näätänen et al., 1989a; Pihko et al., 1995), it was important that deviance size be of equal proportion in each condition, to avoid a confounding effect on this parameter. The choice of 50% as the proportion of the standard duration by which the deviants differed from the standards stemmed from published studies suggesting that a minimal ratio of 20% is required to elicit an MMN, but that ratios of 40% and 60% yield clear-cut MMNs for both increments and decrements in duration (e.g. Amenedo and Escera, 2000; see also Näätänen et al., 1993 for similar findings with several proportions of Inter-Stimulus-Interval deviance). Moreover, Näätänen et al. (1989b) showed that with a standard/deviant ratio of 50%, the MMN amplitude did not vary across standard durations ranging from 25ms to 400ms. By selecting a 50% ratio allegedly giving rise to invariant MMN amplitudes across standard durations, contrast values of this study were kept well above the near threshold values associated with diminished amplitudes (Sams et al., 1985; Näätänen et al., 1989a; Amenedo and Escera, 2000; Jaramillo et al., 2000; Pakarinen et al., 2007) and increased latencies (Sams et al., 1985; Näätänen et al., 1989a) which could have introduced a confounding variable. Stimuli were spectrally rich tones in order to enhance MMN amplitudes (Tervaniemi et al., 2000a, Tervaniemi et al., 2000b, Zion-Golumbic et al., 2007), the four standard durations were 100, 150, 200 and 250ms allowing comparison of these data with those reported in the literature (e.g. Amenedo and Escera, 2000; Catts et al., 1995; Grimm et al., 2004; Jaramillo et al., 2000; Kaukoranta et al., 1989; Näätänen et al., 1989b; Pihko et al., 1995; Shelley et al., 1991; Sysoeva et al., 2006; Takegata et al., 2004). The same stimuli were also used in psychophysical measurements in order to correlate the MMN results with behavioral discrimination performances across all conditions of duration contrasts. 2. Methods 2.1. Subjects Ten right-handed subjects (7 women) aged 18–52 years (mean 26.5) participated in the experiment as paid volunteers. All were in good health and had normal auditory function (see details in the Procedure section). They gave their informed consent to participate after the details of the procedure had been explained to them. The experimental protocol has been approved by the ethical committee of the Brugmann Hospital, where the neurophysiological recordings took place. 2.3. Procedure The stimuli were delivered using Tucker-Davis Technologies (TDT) hardware (System II). They were binaurally presented with Etymotic earphones (model ER-3A) connected through a 25cm long silicon tubing ending into a hollowed foam cylinder inserted into the entrance of the ear canal. Before the electrophysiological experiment, the auditory threshold of each subject was measured for each stimulus duration using TDT software (PsychoSig 3.12). Thresholds were measured using an adaptive two-alternative forced choice paradigm with a one-up, two-down rule. This corresponds to tracking the 70.7% point on the psychometric function (Levitt, 1971). Subjects were provided with a visual display of each presentation interval and a feedback about the correctness of each answer. A maximum of 60 trials or 12 reversals, whichever was first reached, were delivered. The signal level was initially modified by 5dB, then by 2dB after the first three reversals. Thresholds were computed by averaging the signal levels across the last eight reversal points. During the electrophysiological experiment, subjects were comfortably seated in an armchair fitted with a back headrest. They were instructed to ignore the auditory stimuli while reading a self-selected book, so as to minimize the risk of contamining the data by attention-related ERP components such as N2b and P300 (Näätänen, 1991; Picton et al., 2000b). Stimuli were generated by TDT hardware (System II) and software (SigGen 3.51 and SigPlay 3.3) that controlled for their duration and intensity. The TDT was linked to an InstEP stimulation system (software ver. 3.3) that controlled for the presentation order and timing of the stimuli. For each subject, all standard and deviant stimuli were presented 60dB above the individual auditory threshold computed as the mean threshold of the ten duration stimuli. In order to minimize the risk of MMN habituation reported to occur when sessions are too long (McGee et al., 2001), stimuli were presented in short sequences of about 12 minutes. In each short sequence, 500 standard stimuli and 60 deviant stimuli of each kind (short and long) were randomly delivered with a Stimulus Onset Asynchrony (SOA) of 1200ms. A complete MMN data bank for each standard duration was made up of 10 short sequences, in order to include 600 deviant stimuli of each kind (short and long) before data analysis. Moreover, for each type of deviant, one series of 600 “deviant alone” trials was presented in addition to the 10 short sequences. These “deviant alone” series consisted in single tones presented at the same SOA as in the oddball sequences. The presentation order of the short sequences was pseudo-randomized so that the same sequence was never presented more than once in immediate succession. The deviant alone sequences were randomly interspersed among the oddball sequences. Subjects were free to take any type of break they wished between sequences. 2.5. MMN computation and measurement Data averaging was performed with an InstEP system (software ver 3.3). Recording epochs with a total duration of 1024ms were averaged separately for the deviant alone and the deviant in oddball sequences stimuli. The pre-stimulus baseline was 332ms. Rejection of artefacted sweeps from the final averages was performed on the basis of a ±100μV criterion for all channels. Averaged waveforms were converted into ASCII format and exported to a spreadsheet program for further analysis and plotting. The MMN was computed as the differential waveform obtained by subtracting the potential evoked by deviant stimuli presented alone from the one evoked by deviant stimuli presented in a sequence of standards (Kraus et al., 1995). In order to maximize signal-to-noise ratio and take the entire MMN equivalent dipole into consideration, we re-referenced the waveforms recorded at Fz to the average of the mastoids (Näätänen et al., 2004; Pettigrew et al., 2004a, Pettigrew et al., 2004b; Pakarinen et al., 2007; Pablos Martin et al., 2007; Takegata et al., 2008). For each deviant-standard contrast, the presence of an MMN was objectively ascertained from the re-referenced waveforms by a method proposed by Kraus et al. (1993). This method relies on the computation of t-tests, comparing the point-to-point amplitudes of the waveforms evoked by the standard and deviant stimuli (or by the deviant alone and deviant in oddball sequences stimuli in the present case), in order to determine the latency period over which the grand averages were significantly (p<.05) different from zero. For each contrast, t-tests were computed within a 300ms time period symmetrically centered on the visually identified MMN latency peak on the grand average waveforms. Only putative MMN peaks falling within the normal MMN latency range (80–220ms) were considered (Sable et al., 2003). Peak latency and amplitude values were gathered, on Fz differential waveform, for each subject and each contrast. Peak amplitude was computed as the difference between the mean amplitude of the pre-stimulus baseline and the most negative peak occurring during the period of significant t-tests. For illustration purposes, the baseline was centered on zero. The MMN latency, was determined as the peak latency of the maximal MMN peak amplitude computed from the theoretical point in time at which deviance could be detected (i.e. from the end of the deviant stimulus for the short deviant contrasts and from the end of the standard stimulus for the long deviant contrasts). 2.6. Psychophysical experiment In order to avoid carry-over effects of attention that could spoil the inattentive condition requested for MMN recordings, uncontaminated by N2b and P300 components (Näätänen, 1991; Picton et al., 2000b), the behavioral discrimination data were collected after the electrophysiological session. The stimuli were the same as in the electrophysiological experiment and were delivered in the same way. Only one oddball sequence of 620 trials was presented for each condition. In order to avoid tiredness effects, a short break was proposed after the first half of each sequence, i.e. after 30 deviants of each kind had been presented. The subjects were informed that they would hear shorter and longer deviant sounds among constant duration standard ones. Their task was to press a key when they perceived a shorter tone and another key when they perceived a longer one. Reaction times were collected and, to avoid biases due to intrinsic sound duration, they were not computed from sound onset but from the moment at which deviance occurred. For each subject and each standard-deviant contrast, discrimination performances were assessed by computing d′ values according to the signal detection theory (Green and Swets, 1966). In this study, perfect detection (100% of hits and 0% of false alarms) should give a d′ value of 4.98 after correction for ceiling and floor effects, whereas chance responses (50% of hits and 50% of false alarms) should give a d′ value of 0. 2.7. Statistical analysis Two-way ANOVAs were conducted to assess whether standard duration and deviance direction affected MMN latencies and amplitudes and behavioral hit rates and reaction times. 3. Results 3.1. Auditory thresholds The auditory thresholds measured in each subject for each of the ten standard and deviant stimulus durations showed a trend to lower thresholds with longer durations, but a one-way ANOVA indicated that this trend was not significant (F<1). Moreover, individual thresholds varied in a much more erratic manner with respect to duration, precluding a correction for the stimulus duration effect in a consistent way for every subject. We therefore used the mean individual auditory threshold to determine the individual auditory intensity at which all stimuli were presented. 3.2. MMN assessment Fig. 1, Fig. 2, Fig. 3, Fig. 4 illustrate differential waveforms grand averaged across all subjects and evoked by the short and long deviants against each of the four possible standard durations. The visual analysis of Fig. 1, Fig. 2, Fig. 3, Fig. 4 shows that short deviants elicited clear-cut MMNs inverted in polarity between scalp convexity and mastoids across all standard durations, whereas long deviants evoked a questionable MMN for the 200ms standard and probably none for the 250ms standard. As shown in Fig. 5, objective t-tests analysis applied on the re-referenced enhanced waveforms revealed the presence of a significant MMN for all conditions but the long deviant in the context of the longest standard duration. 3.3. Statistical analysis of the MMN parameters Fig. 6 illustrates the distributions of the MMN mean latencies and amplitudes, measured on the re-referenced waveforms in the eight experimental conditions. Two-way ANOVAs were carried out on MMN peak latencies and amplitudes with standard duration and deviance direction as independent variables. As no statistically significant MMN was observed for the long deviant in the context of the long standard condition, these ANOVAs were performed for the 100, 150 and 200ms standard conditions only. According to the null hypothesis, there was no effect of standard duration (F<1) or of deviance direction (F(1,9)=2.61, p>.05) on MMN latencies. The interaction between both factors failed to reach significance too (F<1). MMN peak amplitudes were significantly larger for short than for long deviants (F(1,9)=14.61, p<.01; difference: 1.72μV). Although the visual inspection of the amplitude distributions (see Fig. 6) suggested an interaction between standard duration and deviance direction (the amplitude differences between short and long deviants gradually increasing across standard durations), neither standard duration (F<1), nor the interaction (F(2,18)=3.13, p=.07) between the latter and deviance direction reached significance. In order to include the short deviants in the context of the long standard contrast (250 vs 125ms) in the statistical analysis, we computed a one-way ANOVA on MMN peak amplitudes with standard duration (four levels), in the context of short deviants only, as independent variable. This analysis confirmed, for short deviants, the absence of significant effect of standard duration (F(3,27)=1.24, p>.05). 3.4. Psychophysical experiment Results (hit rates, reaction times and d′) are displayed, for each contrast, in Table 1. | | |  | | 100ms standard | 150ms standard | 200ms standard | 250ms standard | |
|---|
| | 50ms deviant | 150ms deviant | 75ms deviant | 225ms deviant | 100ms deviant | 300ms deviant | 125ms deviant | 375ms deviant | |
|---|
| | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
|---|
| Hit rates % | 88.83 | 6.29 | 76.00 | 9.98 | 88.5 | 6.30 | 82.83 | 9.33 | 92.83 | 3.40 | 89.17 | 6.65 | 95.33 | 3.39 | 88.17 | 8.41 | | | Reaction times | 846.22 | 88.84 | 864.60 | 96.14 | 870.10 | 77.27 | 823.30 | 74.37 | 865.60 | 65.57 | 804.10 | 93.10 | 839.20 | 61.90 | 767.70 | 116.25 | | | d′ | 4.21 | 0.60 | 3.69 | 0.59 | 4.23 | 0.69 | 3.97 | 0.70 | 4.38 | 0.43 | 4.23 | 0.59 | 4.66 | 0.42 | 4.18 | 0.61 | | | | |
Two-way ANOVAs were performed on hit rates, d′ and reaction times with standard duration (four levels) and deviance direction (short vs long) as independent variables. Hit rates exhibited a significant effect of standard duration (F(3,27)=4.07, p<.05). Post hoc tests (HSD Tukey) revealed that the only statistically significant differences were between the 100ms standard condition and the 200ms standard condition (p<.05) and between the 100ms standard condition and the 250ms standard condition (p<.05), hit rates being, respectively, 8.6% and 9.3% higher in the 200 and 250ms conditions than in the 100ms condition. There was no deviance direction effect (F(1,9)=3.52, p>.05) and the interaction between deviance direction and standard duration was not significant (F(3,27)=2.61, p>.05). Similarly, for d′ there was neither deviance direction effect (F(1,9)=3.79, p>.05), nor interaction between deviance direction and standard duration (F(3,27)=1,76, p>.05) but an effect of standard duration (F(3,27)=3.45, p<.05). Post hoc tests (HSD Tukey) showed that d′ values were significantly higher (p<.05) in the 250ms standard condition (mean d′=4.41) than in the 100ms standard condition (mean d′=3.95). As regards reaction times, the effect of standard duration was significant (F(3,27)=5.30, p<.01), the longest standard duration (250ms) giving rise to shorter reaction times than the 100ms (p<.01; difference: 52ms) and 150ms (p<.05; difference: 44ms) standard durations. There was no effect of deviance direction (F(1,9)=4.69, p=.058) but the interaction between both factors also reached significance (F(3,27)=13.05, p<.0001). This interaction is graphically illustrated in Fig. 7. Post hoc tests revealed that reaction times were shorter for long deviants than for short deviants at the three longest standard durations (p<.01 for 150ms standards and p<.001 for 200 and 250ms standards). There was no effect of standard duration for short deviants, whereas, for long deviants, reaction times significantly decreased between the 100 and 150ms standard duration conditions (p<.05) and between the 150 and 250ms standard duration conditions (p<.001). Other significant differences bearing on pairs of reaction times measured at combinations of different standard durations and different deviance directions resulted from this pattern of interaction but are not reported since they do not bring any contribution to the interpretation of the data. 4. Discussion 4.1. Auditory thresholds as a function of stimulus duration It is well known that when the duration of an acoustic stimulus falls below several hundred ms, its threshold raises in a way proportional to the inverse of its duration, a phenomenon called temporal integration (see Gerken et al., 1990 for a review). In this study, although a trend toward lower thresholds for longer durations was clearly present, it could not be statistically substantiated, probably because of the combined effect of a small number of subjects and of a high intra individual variability. The latter mechanisms precluded any adjustment of individual stimulus levels as a function of the respective durations in order to control the relative level of the stimulus accurately with respect to its threshold, the so-called Sensation Level (SL). 4.2. Effect of duration contrast parameters on MMN amplitudes We found that 50% shorter deviants gave rise to significantly larger MMNs than 50% longer deviants. This finding was unexpected. In many studies investigating both short and long deviants, the MMN peak amplitude was found to be independent of the direction of the change, whatever the type of sounds used: pure tones (Näätänen et al., 1989b), spectrally rich tones (Amenedo and Escera, 2000), or white-noise bursts (Jaramillo et al., 2000). Only a few studies reported a deviance direction effect on MMN amplitudes. Whereas Catts et al. (1995) and Jaramillo et al. (1999), using pure tones, found larger MMNs with long deviants than with short deviants, when using vowels, Jaramillo et al. (1999) reported the opposite pattern: a larger MMN with short deviants than with long ones. The latter authors explained the effects of stimulus type on deviance direction as reflecting different processing levels involved in MMN generation according to the nature of the stimuli, leading to different consequences on deviance direction effects. This study conducted only on spectrally rich sounds did not address this issue, which certainly deserves further systematic studies. Based on a review of the literature, Takegata et al. (2008) suggested the existence of a temporal threshold for MMN elicitation, the duration MMN being reduced in amplitude or even abolished when the longer sound exceeds the threshold duration of 400ms. Using vowels, sinusoids, music chords and white noise, they compared an above-threshold (240 vs 400ms contrast) and a below-threshold (120ms vs 200ms contrast) condition. They found that in the below-threshold condition, the MMN was not sensitive to deviance direction but was abolished for duration increment in the above-threshold condition. It should be noted that in this latter study and all previously listed studies, one or at best two standard duration conditions were examined. The present study, using four different standard durations, offered the possibility to study the impact of deviance direction on the MMN peak amplitude in a much more systematic way. The present finding of a detrimental effect of long deviants on MMN amplitude for every sound duration tested shows that an asymmetry between MMNs evoked by short and long deviants is already present for stimulus durations currently used for research and clinical purposes. The inspection of Fig. 5, Fig. 6 even suggests that the asymmetry between short and long deviants may not appear once any threshold is reached, but appears progressively. The finding of a larger MMN with short deviants might have been accounted for in terms of a double deviance, a phenomenon known to enhance the MMN (e.g. Takegata et al., 1999). Indeed, for short sounds (<200ms), perceived loudness correlates with sound duration. Consequently, for standard or deviant durations shorter than 200ms, the two stimuli differ both in terms of duration and loudness. Published studies on the effect of sound duration on loudness provide highly variable data except for a general agreement that, at a given intensity, loudness increases with duration (Moore, 1982). The results are highly dependent on the measurement method and on the frequency content of the stimuli. Without a proven method valid for our stimulus structure, it was impossible to compute the theoretical impact of duration changes on loudness. Nevertheless, we reject this hypothesis because if it explained the deviance direction effect, this effect should diminish with longer standard durations, which is certainly not the case as ascertained by ANOVA, the visual analysis of Fig. 6 even suggesting that the amplitude difference between short and long deviants increases with stimulus duration. Moreover, Todd and Michie (2000) compared the MMNs elicited by duration increments (125ms deviants among 50ms standards) in tones matched in intensity or in tones matched in loudness and found no contamination of the duration MMN by perceived loudness effects. Another potential explanation rests on the concept of the Temporal Window of Integration (TWI). According to this hypothesis, if the temporal distance between sound onset and offset exceeds the duration of the integration window (initiated by sound onset and lasting about 200ms after which a unitary representation of the stimulus starts to fade away), then the relation between onset and offset is not represented in such a way that the pre-attentive deviance–detection system can register a change in this relation (Grimm and Schröger, 2005). Fig. 5 can indeed be interpreted as showing a severe reduction of MMN amplitude when one of the two sounds compared is longer than 250ms (with even an abolition above 300ms). However, we argue that an effect of TWI is not likely to account for the present results. Indeed, there is no reason why a TWI effect would have not impacted behavioral results that remained almost at ceiling level for all standard durations. Moreover, in other studies conducted with longer standard durations, clear-cut MMNs evoked by short deviants were found for standard durations up to 800ms (Näätänen et al., 2004). The present deviance direction effect therefore seems specifically related to MMN evoked by long deviants and not to be paralleled by behavioral results since hit rates were maximal and reaction times minimal for the condition in which the MMN was abolished. Measuring MMN to frequency-modulated deviants, Grimm and Schröger (2005) showed that the later the deviation occurred within the sound, the smaller the MMN amplitude was, suggesting that the temporal distance between stimulus onset and deviance onset is relevant for MMN elicitation. As, for all standard duration conditions, the temporal distance between sound and deviance onsets was always shorter for short deviants than for long deviants, such a temporal distance effect may, at least partially, explain the present asymmetry between short and long deviants. There is however another possible contribution for the asymmetry between MMNs evoked by short and long deviants. One major difference between the two conditions is that in the case of short deviants, deviance detection and deviance quantification are both achieved at the same time (i.e. at the end of the deviant stimulus), whereas, for long deviants, deviance detection is triggered at the end of the standard stimulus, but deviance quantification cannot be completed before the end of the deviant stimulus. Since the MMN parameters are obviously determined both by the moment of deviance detection (MMN latency dependent on the moment of deviance occurrence within the stimulus) and by the magnitude of deviance (MMN latency and amplitude dependent on deviance amount), an optimal situation arises when all the information is present at the time of deviance detection. This is not the case with long deviants, for which deviance quantification is not completed before the end of the deviant stimulus. For long deviants, there is therefore a delay between deviance detection and completion of the entire mismatch process that may lead to poorer MMNs when the temporal overlap between the two processes decreases at longer deviant durations. While this paper was under revision, Takegata et al. (2008) published a paper also suggesting a lack of temporal overlap between deviance detection and quantification to explain the asymmetry they observed between temporal decrement and increment MMNs. This latter hypothesis also contributes to shed light on the detrimental standard duration effect bearing on long deviants: indeed, as clearly shown in the upper row of Fig. 5, the MMNs evoked by short deviants are very similar and well synchronized whereas the MMNs evoked by long deviants are progressively desynchronized, up to the point of MMN abolition, as standard duration increases. As shown in the right part of Fig. 6, although it is not statistically substantiated (p for interaction=.07), there is a trend for amplitude differences between short and long deviants to increase progressively across standard durations. In the present paradigm, the longer the standard is, the longer the long deviant. Consequently, as the duration difference between the standard and the long deviant increases, the time taken by the quantification process also increases leading to poorer MMNs. Let us acknowledge however that the effect of temporal distance between deviance detection and quantification does not account for the observation that the MMN elicited by duration decrements is dramatically reduced for very long standard durations. For example, with short deviants and standard durations of 800ms and longer, Näätänen et al. (2004) observed a significant diminution of the MMN peak amplitudes. Similarly, 40% duration decrements did not elicit an MMN in the context of 667ms (Grimm et al., 2006) or 1000ms standard sounds (Grimm et al., 2004), at least in ignore conditions. This effect may nevertheless contribute to a reduction of the MMN amplitude evoked by long deviants in the context of short standards. 4.3. Psychophysical experiment As regards hit rates and d′, we found that subjects were more effective for detecting deviants in the longest standard duration relative to the shorter one, without any effect of deviance direction. A potential explanation to account for this observation is that longer sounds contain more information than short sounds, therefore facilitating the behavioral discrimination process. This effect of standard duration was in striking contrast with the one found for the MMN amplitude, the MMN being even abolished for the longest standard duration, at least for the long deviant. Moreover, we found a deviance direction effect on the MMN amplitudes (long deviants giving rise to smaller amplitudes than short deviants) but not on hit rates. Such a discrepancy between behavioral data (which were almost at ceiling level for each standard duration) and MMN data was also reported by Grimm et al. (2004), and by Grimm and Schröger (2005) who observed a severe reduction of the decrement MMN with very long duration standards (1000ms) while behavioral hit rates, in similar conditions, were above 90%. Taken together, those data suggest that the finding of a poorer MMN with long deviants and the absence of MMN for the long deviant in the context of the longest standard is not related to a deficit in the discrimination process, but is an exclusively electrophysiological phenomenon. In the context of the present paradigm, we propose that the MMN reduction was caused by the combined effects of a late onset of deviance within the stimuli and a lack of temporal overlap between deviance detection and deviance quantification. As regards reaction times, we found that subjects reacted faster as standard duration increased, especially in the context of long deviants. Again, we failed to observe any correlation between this psychophysical effect and MMN parameters. However, the standard duration effect on reaction times is probably not related to discrimination abilities indexed by the MMN but to a classical motor preparation effect. It is well known that reaction times decrease as the time interval between a warning signal (here, sound onset) and the subsequent target (here, deviance detection) increases (e.g. Bertelson and Tisseyre, 1968). 5. Conclusion We found a major effect of deviance direction on MMN amplitude. Since it has no psychophysical counterpart, we propose that this phenomenon is of a purely electrophysiological nature and may be accounted for in terms of degree of synchronization of MMN generating neurons. For short deviants, deviance detection and quantification both occur at the same time, giving rise, whatever standard duration (within the values used in this study), to well-defined MMNs with latencies accurately indexing the moment of deviance detection. For long deviants, deviance quantification occurs later than deviance detection and this difference increases with standard durations. Consequently, MMN generating neurons are less and less well synchronized as standard durations increase, giving rise to poorer and poorer MMNs. In order to unravel the mechanisms explaining the deviance direction effect on MMN amplitudes (effect that may even be dependent on the stimulus type), further studies using a greater number of standard durations (again with short and long deviants) are needed. In the meantime, these results are sufficiently compelling to raise a caveat when using the duration MMN to probe the processing of temporal information for research and clinical (namely, schizophrenia and autism, which are one of the several currently recognized applications of the MMN) purposes. A better understanding of the neurophysiological processes underlying deviance direction effects and of the different stimuli parameters that are likely to modulate such effects are needed in order to prevent spurious findings of abnormal MMNs following an inappropriate selection of stimulus parameters. On the other hand, a careful selection of the stimuli that elicit less well-defined MMNs may enable clinicians to make the MMN more sensitive to slight anomalies of the central nervous system. For example, the less well-defined MMN for duration increments than for duration decrements may be at the root of group differences between schizophrenics and healthy controls observed for duration increments only (Catts et al., 1995; Shelley et al., 1991). We should also take into consideration that deviance direction effects may affect other types of temporal intervals than duration. For example, the literature contains some MMN data bearing on the comparison of short and long temporal deviance created by manipulating the Inter-Stimulus-Interval (Ford and Hillyard, 1981; Sable et al., 2003). Although one has to remain cautious because Inter-Stimulus-Interval (ISI) manipulations can imply the supramodal perception of rhythm, which is known to be characterized by perceptual and electrophysiological non-linearities (Repp, 1992; Desain and Honing, 2003; Pablos Martin et al., 2007), some published data comparing the effect of shorter and longer deviant ISI suggested a similar asymmetry expressed at the level of evoked responses (Ford and Hillyard, 1981; Sable et al., 2003). Acknowledgement This work was financially supported by the “Fyssen Foundation” to Cécile Colin, by the “Belgian Kid’s Fund” and the “David and Alice Van Buuren Foundation” to Emily Markessis, by Grant EC9014000007 from Université Libre de Bruxelles to Monique Radeau, by the “Emile Defay Funds” and the “Brugmann Foundation” to Paul Deltenre. We thank Roger Gheldof and Patrick Laudin for their engineering help as well as Christian Labarre for his help in revising the English text. References Alho et al., 1990. 1.Alho K, Sainio K, Sajaniemi N, Reinikainen K, Näätänen R. Event-related brain potential of human newborns to pitch changes of an acoustic stimulus. Electroenceph Clin Neurophysiol. 1990;77:151–155. MEDLINE Amenedo and Escera, 2000. 2.Amenedo E, Escera C. The accuracy of sound duration representation in the human brain determines the accuracy of behavioral perception. Eur J Neurosci. 2000;12:2570–2574.
CrossRef
Bertelson and Tisseyre, 1968. 3.Bertelson P, Tisseyre F. The time-course of preparation with regular and irregular foreperiods. Q J Exp Psychol. 1968;20:297–300. Catts et al., 1995. 4.Catts SV, Shelley AM, Ward PB, Liebert B, McConaghy N, Andrews S, et al. Brain potential evidence for an auditory sensory memory deficit in schizophrenia. Am J Psychiatry. 1995;152:213–219. Colin et al., 2002a. 5.Colin C, Radeau M, Soquet A, Colin F, Deltenre P. Mismatch negativity evoked by the McGurk–MacDonald effect: evidence for a phonological representation within the auditory sensory short term memory. Clin Neurophysiol. 2002;113:495–506. Abstract | Full Text |
Full-Text PDF (210 KB)
|
CrossRef
Colin et al., 2002b. 6.Colin C, Radeau M, Soquet A, Dachy B, Deltenre P. Electrophysiology of spatial scene analysis: the Mismatch Negativity (MMN) is sensitive to the ventriloquism illusion. Clin Neurophysiol. 2002;113:507–518. Abstract | Full Text |
Full-Text PDF (205 KB)
|
CrossRef
Colin et al., 2004. 7.Colin C, Radeau M, Soquet A, Deltenre P. Generalization of the generation of an MMN by illusory McGurk percepts: voiceless consonants. Clin Neurophysiol. 2004;115:1989–2000. Abstract | Full Text |
Full-Text PDF (275 KB)
|
CrossRef
Cowan et al., 1993. 8.Cowan N, Winkler I, Teder W, Näätänen R. Memory prerequisites of mismatch negativity in the auditory Event-Related Potential (ERP). J Exp Psychol Learn. 1993;19:909–921. Cowan, 1995. 9.Cowan N. Attention and memory. An integrated framework. Oxford: Oxford University Press; 1995;. Desain and Honing, 2003. 10.Desain P, Honing H. The formation of rhythmic categories and metric priming. Perception. 2003;32:341–365. MEDLINE |
CrossRef
Drake and Botte, 1993. 11.Drake C, Botte MC. Tempo sensitivity in auditory sequences: evidence for a multiple-look model. Percept Psychophys. 1993;54:277–286. MEDLINE Escera et al., 2002. 12.Escera C, Corral MJ, Yago E. An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes. Cognitive Brain Res. 2002;14:325–332. Ford and Hillyard, 1981. 13.Ford JM, Hillyard SA. Event-related potentials (ERPs) to interruptions of a steady rhythm. Psychophysiology. 1981;18:322–330. MEDLINE |
CrossRef
Gerken et al., 1990. 14.Gerken GM, Bhat VKH, Hutchison-Clutter M. Auditory temporal integration and the power function model. J Acoust Soc Am. 1990;88:767–778. MEDLINE |
CrossRef
Grimm et al., 2004. 15.Grimm S, Widmann A, Schröger E. Differential processing of duration changes within short and long sounds in humans. Neurosci Lett. 2004;356:83–86. MEDLINE |
CrossRef
Grimm and Schröger, 2005. 16.Grimm S, Schröger E. Pre-attentive and attentive processing of temporal and frequency characteristics within long sounds. Cognitive Brain Res. 2005;25:711–721. Grimm et al., 2006. 17.Grimm S, Roeber U, Trujillo-Barreto NJ, Schröger E. Mechanisms for detecting auditory temporal and spectral deviations operate over similar time windows but are divided differently between the two hemispheres. NeuroImage. 2006;32:275–282. MEDLINE |
CrossRef
Green and Swets, 1966. 18.Green DM, Swets JA. Signal detection theory and psychophysics. New-York, NJ: Wiley; 1966;. Ilvonen et al., 2004. 19.Ilvonen T, Kujala T, Kozou H, Kiesiläinen A, Salonen O, Alku P, et al. The processing of speech and non-speech sounds in aphasic patients as reflected by the mismatch negativity (MMN). Neurosci Lett. 2004;366:235–240. MEDLINE |
CrossRef
Jaramillo et al., 1999. 20.Jaramillo M, Alku P, Paavilainen P. An event-related potential (ERP) study of duration changes in speech and non-speech sounds. Neuroreport. 1999;10:3301–3305. Jaramillo et al., 2000. 21.Jaramillo M, Paavilainen P, Näätänen R. Mismatch negativity and behavioural discrimination in humans as a function of the magnitude of change in sound duration. Neurosci Lett. 2000;290:101–104. MEDLINE |
CrossRef
Joutsiemi et al., 1998. 22.Joutsiemi SL, Ilvonen T, Sinkkonen J, Huotilainen M, Tervaniemi M, Lehtokoski A, et al. The mismatch negativity for duration decrement of auditory stimuli in healthy subjects. Electroenceph Clin Neurophysiol. 1998;108:154–159. MEDLINE Kasai et al., 2001. 23.Kasai K, Yamada H, Kamio S, Kakagome K, Iwanami A, Fukuda M, et al. Brain lateralization for mismatch response to across- and within-category change of vowels. Neuroreport. 2001;12:2467–2471. MEDLINE |
CrossRef
Kaukoranta et al., 1989. 24.Kaukoranta E, Sams M, Hari R, Hamalainen M, Näätänen R. Reactions of human auditory cortex to a change in tone duration. Hear Res. 1989;41:15–22. MEDLINE |
CrossRef
Kraus et al., 1992. 25.Kraus N, McGee T, Sharma A, Carrell T, Nicol T. Mismatch negativity event-related potential is elicited by speech stimuli. Ear Hear. 1992;13:158–164. MEDLINE Kraus et al., 1993. 26.Kraus N, McGee T, Micco A, Sharma A, Carrell T, Nicol T. Mismatch Negativity in school-age children to speech stimuli that are just perceptibly different. Electroenceph Clin Neurophysiol. 1993;88:123–130. MEDLINE Kraus et al., 1995. 27.Kraus N, McGee T, Carrell TD, Sharma A. Neurophysiologic bases of speech discrimination. Ear Hear. 1995;16:19–37. MEDLINE Lang et al., 1995. 28.Lang AH, Eerola O, Korpilahti P, Holopainen I, Salo S, Aaltonen O. Practical issues in the clinical application of mismatch negativity. Ear Hear. 1995;16:117–129. Levitt, 1971. 29.Levitt H. Transformed up–down methods in psychoacoustics. J Acoust Soc Am. 1971;49:467–477.
CrossRef
McGee et al., 2001. 30.McGee TJ, King C, Tremblay K, Nicol TG, Cunningham J, Kraus N. Long-term habituation of the speech-elicited mismatch negativity. Psychophysiology. 2001;38:653–658. MEDLINE |
CrossRef
Michie et al., 2000. 31.Michie PT, Budd TW, Todd J, Rock D, Wichmann H, Box J, et al. Duration and frequency mismatch negativity in schizophrenia. Clin Neurophysiol. 2000;111:1054–1065. Abstract | Full Text |
Full-Text PDF (218 KB)
|
CrossRef
Moore, 1982. 32.Moore BCJ. An introduction to the psychology of hearing. London: Academic Press; 1982;. Näätänen et al., 1987. 33.Näätänen R, Paavilainen P, Alho K, Reinikainen K, Sams M. The mismatch negativity to intensity changes in an auditory stimulus sequence. Electroenceph Clin Neurophysiol. 1987;(Suppl. 40):125–131. Näätänen et al., 1989a. 34.Näätänen R, Paavilainen P, Alho K, Reinikainen K, Sams M. Do event-related potentials reveal the mechanism of the auditory sensory memory in the human brain ?. Neurosci Lett. 1989;98:217–221. MEDLINE |
CrossRef
Näätänen et al., 1989b. 35.Näätänen R, Paavilainen P, Reinikainen K. Do event-related potentials to infrequent decrements in duration of auditory stimuli demonstrate a memory trace in man?. Neurosci Lett. 1989;107:347–352. MEDLINE |
CrossRef
Näätänen, 1990. 36.Näätänen R. The role of attention in auditory information processing as revealed by event-related potentials and other brain measures of cognitive function. Behav Brain Sci. 1990;13:201–288. Näätänen, 1991. 37.Näätänen R. Mismatch negativity (MMN) outside strong attentional focus: a commentary on Woldorff et al.. Psychophysiology. 1991;28:478–484. MEDLINE |
CrossRef
Näätänen, 1992. 38.Näätänen R. Attention and brain function. Hillsdale, NJ: Lawrence Erlbaum; 1992;. Näätänen et al., 1993. 39.Näätänen R, Jiang D, Lavikainen J, Reinikainen K, Paavilainen P. Event-related potentials reveal a memory trace for temporal features. Neuroreport. 1993;5:310–312. MEDLINE Näätänen and Alho, 1995. 40.Näätänen R, Alho K. Mismatch negativity – a unique measure of sensory processing in audition. Int J Neurosci. 1995;80:317–337. MEDLINE |
CrossRef
Näätänen and Alho, 1997. 41.Näätänen R, Alho K. Mismatch negativity – the measure for central sound representation accuracy. Audiol Neurootol. 1997;2:341–353. MEDLINE |
CrossRef
Näätänen et al., 2004. 42.Näätänen R, Syssoeva O, Takegata R. Automatic time perception in the human brain for intervals ranging from milliseconds to seconds. Psychophysiology. 2004;41:660–663. MEDLINE |
CrossRef
Näätänen et al., 2007. 43.Näätänen R, Paavilainen P, Rinne T, Alho K. The mismatch negativity (MMN) in basic research of central auditory processing: a review. Clin Neurophysiol. 2007;118:2544–2590. Abstract | Full Text |
Full-Text PDF (1003 KB)
|
CrossRef
Novak et al., 1990. 44.Novak GP, Ritter W, Vaughan HG, Wizniter ML. Differentiation of negative event-related potentials in an auditory discrimination task. Electroenceph Clin Neurophysiol. 1990;75:255–275. MEDLINE |
CrossRef
Paavilainen et al., 1989. 45.Paavilainen P, Karlsson ML, Reinikainen K, Näätänen R. Mismatch Negativity to change in spatial location of an auditory stimulus. Electroenceph Clin Neurophysiol. 1989;73:129–141. MEDLINE |
CrossRef
Pablos Martin et al., 2007. 46.PablosMartin X, Deltenre P, Hoonhorst I, Markessis E, Rossion B, Colin C. Perceptual biases for rhythm: the Mismatch Negativity latency indexes the privileged status of binary vs non-binary interval ratios. Clin Neurophysiol. 2007;118:2709–2715. Abstract | Full Text |
Full-Text PDF (212 KB)
|
CrossRef
Pakarinen et al., 2007. 47.Pakarinen S, Takegata R, Rinne T, Huotilainen M, Näätänen R. Measurement of extensive auditory discrimination profiles using the mismatch negativity (MMN) of the auditory event-related potential (ERP). Clin Neurophysiol. 2007;118:177–185. Abstract | Full Text |
Full-Text PDF (299 KB)
|
CrossRef
Pettigrew et al., 2004a. 48.Pettigrew CM, Murdoch BE, Kei J, Chenery HJ, Sockalingam R, Ponton CW, et al. Processing of English words with fine acoustic contrasts and simple tones: a mismatch negativity study. J Am Acad Audiol. 2004;15:47–66. MEDLINE |
CrossRef
Pettigrew et al., 2004b. 49.Pettigrew CM, Murdoch BE, Ponton CW, Kei J, Chenery HJ, Alku P. Subtitled videos and mismatch negativity (MMN) investigations of spoken word processing. J Am Acad Audiol. 2004;15:469–485. MEDLINE |
CrossRef
Picton et al., 2000a. 50.Picton TW, Alain C, Otten L, Ritter W, Achim A. Mismatch Negativity: different water in the same river. Audiol Neurootol. 2000;5:111–139. MEDLINE |
CrossRef
Picton et al., 2000b. 51.Picton TW, Bentin S, Berg P, Donchin E, Hillyard SA, Johnson R, et al. Guidelines for using human event-related potentials to study cognition: recording standards and publication criteria. Psychophysiology. 2000;37:127–152. MEDLINE |
CrossRef
Pihko et al., 1995. 52.Pihko E, Leppäsaari T, Lyytinen H. Brain reacts to occasional changes in duration of elements in a continuous sound. Neuroreport. 1995;6:1215–1218. MEDLINE Repp, 1992. 53.Repp BH. Probing the cognitive representation of musical time: structural constraints on the perception of timing perturbations. Cognition. 1992;44:241–281. MEDLINE |
CrossRef
Ritter et al., 1995. 54.Ritter W, Deacon D, Gomes H, Javitt DC, Vaughan HG. The mismatch negativity of event-related potentials as a probe of transient auditory memory: a review. Ear Hear. 1995;16:52–67. MEDLINE Rosen, 1992. 55.Rosen S. Temporal information in speech: acoustic, auditory and linguistic aspects. Philos Trans R Soc Lond B Biol Sci. 1992;336:367–373. MEDLINE |
CrossRef
Sable et al., 2003. 56.Sable JJ, Gratton G, Fabiani M. Sound presentation rate is represented logarithmically in human cortex. Eur J Neurosci. 2003;17:2492–2496.
CrossRef
Sams et al., 1985. 57.Sams M, Paavilainen P, Alho K, Näätänen R. Auditory frequency discrimination and event-related potentials. Electroenceph Clin Neurophysiol. 1985;62:437–448. MEDLINE Schröger, 1997. 58.Schröger E. On the detection of auditory deviants: a pre-attentive activation model. Psychophysiology. 1997;34:245–257. MEDLINE |
CrossRef
Shelley et al., 1991. 59.Shelley AM, Ward PB, Catts SV, Michie PT, Andrews S, McConaghy N. Mismatch negativity: an index of a preattentive processing deficit in schizophrenia. Biol Psychiatry. 1991;30:1059–1062.
Full-Text PDF (414 KB)
|
CrossRef
Stekelenburg et al., 2004. 60.Stekelenburg JJ, Vroomen J, de Gelder B. Illusory sound shifts induced by the ventriloquist illusion evoke the mismatch negativity. Neurosci Lett. 2004;357:163–166. MEDLINE |
CrossRef
Sysoeva et al., 2006. 61.Sysoeva O, Takegata R, Näätänen R. Pre-attentive representation of sound duration in the human brain. Psychophysiology. 2006;43:272–276. MEDLINE |
CrossRef
Takegata et al., 1999. 62.Takegata R, Paavilainen P, Näätänen R, Winkler I. Independent processing of changes in auditory single features and feature conjunctions in humans as indexed by the mismatch negativity (MMN). Neurosci Lett. 1999;266:109–112. MEDLINE |
CrossRef
Takegata et al., 2004. 63.Takegata R, Nakagawa S, Tonoike M, Näätänen R. Hemispheric processing of duration changes in speech and non-speech sounds. Neuroreport. 2004;15:1683–1686. MEDLINE |
CrossRef
Takegata et al., 2008. 64.Takegata R, Tervaniemi M, Alku P, Ylinen S, Näätänen R. Parameter-specific modulation of the mismatch negativity to duration decrement and increment: evidence for asymmetric processes. Clin Neurophysiol. 2008;119:1515–1523. Abstract | Full Text |
Full-Text PDF (353 KB)
|
CrossRef
Tervaniemi et al., 1993. 65.Tervaniemi M, Alho K, Paavilainen P, Sams M, Näätänen R. Absolute pitch and event-related brain potentials. Music Percept. 1993;10:305–316. Tervaniemi et al., 1999. 66.Tervaniemi M, Lehtokoski A, Sinkkonen J. Test–retest reliability of mismatch negativity for duration, frequency and intensity changes. Clin Neurophysiol. 1999;110:1388–1393. Abstract | Full Text |
Full-Text PDF (231 KB)
|
CrossRef
Tervaniemi et al., 2000a. 67.Tervaniemi M, Ilvonen T, Sinkkonen J, Kujala A, Alho K, Huotilainen M, et al. Harmonic partials facilitate pitch discrimination in humans: electrophysiological and behavioral evidence. Neurosci Lett. 2000;279:29–32. MEDLINE |
CrossRef
Tervaniemi et al., 2000b. 68.Tervaniemi M, Schröger E, Saher M, Näätänen R. Effects of spectral complexity and sound duration on automatic complex-sound pitch processing in humans – a mismatch negativity study. Neurosci Lett. 2000;290:66–70. MEDLINE |
CrossRef
Todd and Michie, 2000. 69.Todd J, Michie PT. Do perceived loudness cues contribute to duration mismatch negativity (MMN)?. Neuroreport. 2000;11:3771–3774. MEDLINE Todd et al., 2000. 70.Todd J, Michie PT, Budd TW, Rock D, Jablensky AV. Auditory sensory memory in schizophrenia: inadequate trace formation?. Psychiatry Res. 2000;96:99–115. Abstract | Full Text |
Full-Text PDF (317 KB)
|
CrossRef
Todd et al., 2003. 71.Todd J, Michie PT, Jablensky AV. Association between reduced duration mismatch negativity (MMN) and raised temporal discrimination thresholds in schizophrenia. Clin Neurophysiol. 2003;114:2061–2070. Abstract | Full Text |
Full-Text PDF (328 KB)
|
CrossRef
Wright et al., 1997. 72.Wright BA, Lombardino LJ, King WM, Puranik CS, Leonard AM, Merzenich MM. Deficits in auditory temporal and spectral resolution in language-impaired children. Nature. 1997;387:176–178. MEDLINE |
CrossRef
Zeng et al., 2005. 73.Zeng FG, Kong YY, Michalewski HJ, Starr A. Perceptual consequences of disrupted auditory nerve activity. J Neurophysiol. 2005;93:3050–3063. MEDLINE |
CrossRef
Zion-Golumbic et al., 2007. 74.Zion-Golumbic E, Deouell LY, Whalen DH, Bentin S. Representation of harmonic frequencies in auditory memory: a mismatch negativity study. Psychophysiology. 2007;44:671–679.
CrossRef
a Unité de Recherches en Neurosciences Cognitives, Université Libre de Bruxelles (U.L.B.), Belgium b FNRS, Bruxelles, Belgium c Faculté de Médecine, Université Libre de Bruxelles (U.L.B.), Belgium d Unité de Neurophysiologie, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain (U.C.L.), Bruxelles, Belgium e UMR 7593 CNRS-Paris VI, CHU Pitié-Salpétrière, Paris, France f Laboratoire de Psychologie Expérimentale, Université Libre de Bruxelles (U.L.B.), Belgium g Laboratoire de Neurophysiologie Clinique, Hôpital Brugmann, Bruxelles, Belgium  Corresponding author. Address: Clinique de Neurophysiologie, Bat. EM (rotonde–1) Hôspital Brugmann, 4, Place Van Gehuchten, B-1020 Brussels Belgium. Tel.: +32 2 477 22 86; fax: +32 2 477 24 56.
PII: S1388-2457(08)01005-5 doi:10.1016/j.clinph.2008.10.002 © 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT01005-5/journalimage?img=PIIS1388245708010055.gr1.lrg.gif&fig=fig1&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)
01005-5/journalimage?img=PIIS1388245708010055.gr2.lrg.gif&fig=fig2&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)
01005-5/journalimage?img=PIIS1388245708010055.gr3.lrg.gif&fig=fig3&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)
01005-5/journalimage?img=PIIS1388245708010055.gr4.lrg.gif&fig=fig4&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)
01005-5/journalimage?img=PIIS1388245708010055.gr5.lrg.gif&fig=fig5&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)
01005-5/journalimage?img=PIIS1388245708010055.gr6.lrg.gif&fig=fig6&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)
01005-5/journalimage?img=PIIS1388245708010055.gr7.lrg.gif&fig=fig7&kwhquery=&issn=1388-2457&ishighres=false&allhighres=false&free=yes','journalimage');)