Auditory sensory memory in 2-year-old children: an event-related potential study

Size: px

Start display at page:

Download "Auditory sensory memory in 2-year-old children: an event-related potential study"

Juniper Crawford
5 years ago
Views:

1 LEARNING AND MEMORY Auditory sensory memory in -year-old children: an event-related potential study Elisabeth Glass, te achse and Waldemar von uchodoletz Department of Child and Adolescent Psychiatry, Psychosomatic and Psychotherapy, Ludwig-Maximilians-University, Munich, Germany Correspondence to Elisabeth Glass, Department of Child and Adolescent Psychiatry, Psychosomatic and Psychotherapy, Ludwig-Maximilians-University, Waltherstr, D- 87 Munich,Germany Tel: ; fax: ; Lisa.Glass@lrz.uni-muenchen.de Received8 January 8; accepted January 8 Auditory sensory memory is assumed to play an important role in cognitive development, but little is known about it in young children. The aim of this study was to estimate the duration of auditorysensorymemoryin-year-oldchildren.werecordedthe mismatch negativity in response to tone stimuli presented with di erent interstimulus intervals. Our ndings suggest that in -year-old children the memory representation of the standard tone remains in the sensory memory store for at least s but for less than s. Recording the mismatch negativity with stimuli presented at various interstimulus intervals seems to be a useful method for studying the relationship between auditory sensory memory and normal and disturbed cognitive development. NeuroReport 9:569^57 c 8 Wolters Kluwer Health Lippincott Williams & Wilkins. Keywords: auditory sensory memory, children, duration of memory trace, interstimulus interval, mismatch negativity Introduction Auditory short-term memory is generally assumed to be crucial for cognitive development. The adequate functioning of short-term memory has, for example, been postulated to be a precondition for language acquisition [], and deficits in auditory short-term memory are thought to cause language and other developmental disorders []. The knowledge about short-term memory in young children is, however, limited, as it is usually determined by means of behavioural methods. Those demand focussed attention, motivation, and understanding of tasks. uch methods are inappropriate in young children or children with cognitive impairments. The aim of this study was to examine auditory sensory memory in -year-old children using a method that is independent of attention and that does not require a behavioural response. An objective indicator of auditory memory that is suitable to use with young children is the mismatch negativity (MMN), an event-related potential (ERP), which is associated with a particular facet of auditory short-term memory, the auditory sensory memory []. The MMN reflects an automatic, preattentive memorybased comparison process []. It is conventionally elicited by an infrequently occurring divergent auditory stimulus (deviant) in a stream of identical sounds (standard). The frequently occurring standard induces a neural representation in the sensory store, and the MMN is the result of a violation of anticipated information detected by comparison of the deviant with the memory trace of the standard. An MMN can be recorded only if the memory trace of the standard has not yet decayed. On the basis of the assumption that the lifetime of memory traces in the auditory sensory memory store can be examined by varying the interstimulus interval (II) [4], MMN measures have been used to study memory processes in both basic and clinical research [5 7]. MMN studies investigating auditory sensory memory in adults have shown that the MMN cannot be elicited reliably if the II is longer than approximately s [8 ]. Cheour et al. [] examined newborns and found an MMN in response to frequency changes with an II of 7 ms but not with an II of 4 ms. In a different study, they showed that in school-age children an MMN could be elicited with 4 ms []. Gomes et al. [] studied three groups of healthy children (6 7, 8, years) and a group of adults. In all groups an MMN was present if the II was ms, but for 8 ms an MMN was evoked only in the -year-old and -year-old children and in the adults. In this field, however, there is a lack of systematic studies, particularly in children during the crucial phase of cognitive development. Information regarding auditory sensory memory is of importance for understanding the relationship between auditory short-term memory and normal and disturbed cognitive functioning. Earlier studies have provided evidence that the duration of sensory memory lies somewhere between.7 and 8 s in young children, however, this information is too imprecise for practical use. The aim of this study was to determine the duration of auditory sensory memory in -year-old children. More precise knowledge about this facet of short-term memory in young children could allow an early detection of memory deficits in children with cognitive impairments, which may in turn enable an early and specific intervention. Methods Participants The participants were 7 normally developing -year-old children. The MMN was elicited in three conditions with different IIs. By dividing the children into two groups, only c Wolters Kluwer Health Lippincott Williams & Wilkins Vol 9 No 5 6 March 8 569

2 GLA ETAL. one experimental session was required for each child. In group ( children) IIs of 5 and ms and in group (5 children) IIs of 5 and ms were used. Of the 7 children, 6 were girls and boys (group : and ; group : 6 and 9, respectively). The participants were aged between 5 and 7 months (group : mean 5.4 months, D.6; group : mean 5.8 months, D.7). On the basis of the parent report, there was no evidence for developmental or general medical disorders in any of the children. The children had normal scores on screening instruments for cognitive abilities and language development. The children s hearing was normal as assessed with an otoacoustic emission screening test. Of the 7 mothers, 6 (7%) had completed at least years of school. The general design of the study was approved by the ethics commission of the university medical faculty. The parents of all children gave their written informed consent. timuli and procedure Tones of (standard) and Hz (deviant) were used as stimuli (duration: ms; rise and fall time: ms). These frequencies were chosen because they are easily discriminable and had been used in earlier auditory sensory memory studies in children and adults [5,]. Measurement duration was shortened by using a timesaving paradigm [4]. timuli were grouped in trains of four tones, separated from each other by a fixed interval of 5 ms within the train. The trains began with either a standard or a deviant stimulus, all nonleading stimuli were standards only (Fig. ). An II of 5, or ms was used between trains, depending on the particular experimental condition. During later data analysis, only ERP responses to tones that occurred as the first tone in a train were averaged, ensuring that the number and the relative position of averaged standards and deviants were comparable. In a given II condition, 6 tones were presented, 4 standards and deviants (4%). The presentation of tones was divided into four blocks of 4 stimuli each, with trains arranged in pseudo-randomized order within the blocks, and with a break of 4 s between blocks. Two II conditions were presented per child, a short one of 5 ms and a longer one of either ms (total duration: 5 min) or ms (total duration: 4 min) within the same session and in balanced order. During the EEG recording, the child was seated in an upright child s seat and was either shown a silent video or, if necessary, was kept occupied with quiet play. The child s mother sat in the same room and completed questionnaires. timuli were presented through a loudspeaker that was located in front of the child at a distance of. m, resulting in a sound pressure level of 74 db. D tandard stimuli D Deviant stimuli tandard stimuli used for averaging Interstimulus interval 5 ms Interstimulus interval 5// ms Fig. Mismatch negativity paradigm for probing the duration of auditory sensory memory. Event-related potential recording The electroencephalogram (EEG) was recorded using Ag/AgCl sintered electrodes attached to an elastic electrode cap (Easy Cap, Herrsching, Germany). Electrodes were placed according to the International ystem. The horizontal electro-oculogram was recorded from electrodes placed at the outer canthus of each eye. Most of the children did not accept an electrode underneath their eye, and therefore only Fp was used for elimination of vertical eye artefacts. The EEG electrodes were referenced to the right mastoid during the recording and rereferenced off-line against linked mastoids. Data acquisition was carried out using a BrainAmp system (Brain Products, Gilching, Germany). The online bandpass filter was set to.6 and Hz (sampling rate: 5 Hz; impedances at the beginning of measurements: o5 ko). Off-line processing included high-pass (.8 Hz) and low-pass ( Hz) filtering, artefact rejection (exclusion of responses with EEG activity at fronto-central electrodes 48 mv, Fp activity4 mv, horizontal electro-oculogram 45 mv), segmentation ( to + 6 ms after stimulus onset), baseline correction ( to ms) and averaging. After exclusion of participants with fewer than 8 remaining responses for standards or deviants, deviant responses (D: ; range: 8 7) and standard responses (D: ; range: 8 7) were averaged per participant and II condition. No significant difference was observed in the number of deviant or standard responses between the II conditions (t-test). Data analysis The data for the fronto-central electrodes (F, Fz, F4, C, Cz and C4) were extracted for further analysis. The MMN was obtained by subtracting standard from deviant evoked responses for each II condition, and was measured as the mean amplitude across a time window from 4 to 5 ms after stimulus onset. This time window was chosen based on tests for significant differences (Po.5) between standard and deviant responses (one-sample t-tests) [5], covering all intervals of significant differences in any of the three II conditions for the six fronto-central electrodes (intervals are listed in Table ). tatistical analysis of the ERP data was performed using analysis of variance. The six electrodes were divided into three topographic categories of a within-subject factor side (left, centre, right). Between-subject factors II or group were used for comparison of II conditions between the groups of participants (i.e. groups and ). For withingroup comparisons a repeated-measures design with II as the within-subject factor was used. Greenhouse Geisser corrections were applied when appropriate (e values are reported). Mismatch negativity comparability between both samples of the short interstimulus interval condition In the 5-ms II condition, prominent and similar MMN responses were obtained for both groups of participants (Fig. ). No statistically significant differences in mean MMN amplitudes were found between these samples: no main effect of group [F(,5)¼.7, P¼.79] and no interaction of group with side [F(,7)¼.8, e¼.994, P¼.44]. We, therefore, regard both samples as being comparable. 57 Vol9 No5 6March8

3 AUDITORY ENORY MEMORY Results Mismatch negativity in the interstimulus interval conditions 5, and ms Figure shows the grand average of MMN waveforms for the three II conditions, 5, and ms. For the 5-ms condition the grand average of the combined groups (groups and ) is shown. A distinct MMN was found for II¼5 ms and II¼ ms, whereas no negativity was elicited in the -ms II condition (Table and Fig. ). To test the statistical significance of differences among the II conditions, two analyses of variance of the mean MMN amplitudes were performed, one for the comparison of the 5-ms and -ms conditions and one for the -ms and -ms conditions. The comparison of the 5-ms and -ms II conditions for group yielded no significant effect of II [F(,)¼.6, P¼.84] and no interaction of II with side [F(,4)¼.49, e¼.9, P¼.688]. In contrast, a significant difference was found between the -ms and -ms conditions: there was a main effect of II [F(,5)¼5.6, P¼.7], but no II side interaction [F(,7)¼.98, e¼.896, P¼.5]. As the 5-ms II condition was examined in both groups of participants, and to cross check the above results, we also compared the 5-ms condition of the second group with the -ms condition of group. As expected, this also Table igni cant di erences between standard and deviant responses Range (ms) II (ms) N F Fz F4 C Cz C ^5 4^48 4^44 5^96 ^ ^ ^4 8^4 8^4 ^ ^6 ^ 5 ^ ^ ^ ^ ^ ^ II, interstimulus interval. Running t-tests; Po.5 at Z4 consecutive data points. yielded no significant effect of II [F(,5)¼.49, P¼.6] and no interaction of II with side [F(,7)¼.96, e¼.955, P¼.94]. Plots of the mean amplitudes for the different II conditions highlight the similarity of mean MMN amplitudes in the 5-ms and -ms conditions and the clear differences between mean amplitudes in the -ms and -ms conditions (Fig. 4). Discussion It is generally thought that evaluation of the MMN with different IIs probes the duration of auditory sensory memory. Our investigation of the MMN with different IIs in -year-old children showed a prominent MMN in the 5- ms and -ms II conditions, but not in the condition with an extended II of ms. The existence of a distinct MMN for the two shorter intervals suggests that the children were able to differentiate between the standard and deviant tone stimuli. As there were no significant differences between MMN amplitudes elicited with stimulus conditions of 5 and ms, it can be assumed that the memory trace of the standard is available for comparison with deviant stimuli for at least ms. The absence of an MMN in the -ms condition suggests that the neural representation of the preceding standards reaches a limit within s, after which it is no longer accessible for the comparison process underlying the generation of MMN. Previous studies that investigated the duration of auditory sensory memory by means of MMN examined children of either younger or older ages. Cheour et al. [] reported a duration of auditory sensory memory between 7 and 4 ms in newborns, and Gomes et al. [] found that memory traces remained for longer than 8 s in children older than years but decayed within 8 s in younger school-age children. These findings are in line with our results, which suggest that in -year-old children auditory F Fz [ms] [ms] F4 [ms] C Cz C4 [ms] [ms] [ms] Fig. Mismatch negativity of groups and for the 5 -ms interstimulus interval condition (thick line: group ; thin line: group ; reference: linked mastoids). Vol9 No5 6March8 57

4 GLA ETAL. F Fz F4 [ms] [ms] [ms] C Cz C4 [ms] [ms] [ms] Fig. Mismatch negativity for interstimulus interval conditions of 5, and ms (grey areas indicate time intervals for which mean amplitudes were calculated; thick line: 5 ms; thin line: ms; dashed line: ms; reference: linked mastoids). 5 Central Left Right memory traces remain in the sensory memory store for at least s and decay within s. The concept of using the MMN in determining the duration of sensory memory is, however, controversial [4,8]. Memory processes may be task and function specific, and therefore, it should be taken into account that the temporal limit of sensory memory found with this paradigm may not have general validity. It can be hypothesized, however, that during the longer time interval between the stimuli the sensory memory trace becomes inaccessible for the preattentive detection of auditory changes reflected by the MMN. Conclusion In connection with findings from earlier studies, our results provide evidence for a developmental increase in the duration of auditory sensory memory. As the MMN can be elicited without confounding by focussed attention, motivation or understanding of tasks and can be recorded II [ms] Fig. 4 Mean Mismatch negativity amplitudes for di erent interstimulus interval (II) conditions in left (dashed line), central (solid line) and right (dotted line) fronto-central regions. in young children, MMN examination with variable IIs could be a powerful tool for studying the relationship between auditory short-term memory and cognitive development at an early stage, and for early identification of short-term memory deficits in children with specific developmental disorders. Acknowledgements The authors thank Monika Hage and Elisabeth Greiner for their assistance in collecting the ERP data. We are also particularly grateful to all the children and their parents for participating in this study. Parts of the results have been generated in the context of the doctoral thesis of E. Glass at the medical faculty of the Ludwig-Maximilians University of Munich (in preparation). This research was supported by the Friedrich-Baur-tiftung, Grant 57/6 awarded to teffi achse. References. Gathercole E, Baddeley AD. Working memory and language. Hove, East ussex: Psychology Press;.. Montgomery JW. Working memory and comprehension in children with specific language impairment: what we know so far. J Commun Disord ; 6:.. Näätänen R. Mismatch negativity: clinical research and possible applications. Int J Psychophysiol ; 48: Kujala T, Tervaniemi M, chroger E. The mismatch negativity in cognitive and clinical neuroscience: theoretical and methodological considerations. Biol Psychol 7; 74: Barry JG, Hardiman MJ, Line E, White KB, Yasin I, Bishop DV. Duration of auditory sensory memory in parents of children with LI: a mismatch negativity study. Brain Lang 7; (Epub ahead of print), doi:.6/ j.bandl Ceponienè R, Hukki J, Cheour M, Haapanen M-L, Ranta R, Näätänen R. Cortical auditory dysfunction in children with oral clefts: relation with cleft type. Clin Neurophysiol 999; : Cowan N, Winkler I, Teder W, Näätänen R. Memory prerequisites of mismatch negativity in the auditory event-related potential (ERP). JExp Psychol Learn Mem Cogn 99; 9: Vol 9 No 5 6 March 8

5 AUDITORY ENORY MEMORY 8. Winkler I, chroger E, Cowan N. The role of large-scale memory organization in the mismatch negativity event-related brain potential. J Cogn Neurosci ; : ams M, Hari R, Rif J, Knuutila J. The human auditory sensory memory trace persists about s: neuromagnetic evidence. J Cogn Neurosci 99; 5:6 7.. Böttcher-Gandor C, Ullsperger P. Mismatch negativity in event-related potentials to auditory stimuli as a function of varying interstimulus interval. Psychophysiology 99; 9: Cheour M, Ceponienè R, Leppänen P, Alho K, Kujala T, Renlund M, et al. The auditory sensory memory trace decays rapidly in newborns. cand J Psychol ; 4: 9.. Cheour M, Haapanen M-L, Hukki J, Ceponienè R, Kurjen-luoma, Alho K, et al. The first neurophysiological evidence for cognitive brain dysfunctions in children with CATCH. Neuroreport 997; 8: Gomes H, ussman E, Ritter W, Kurtzberg D, Cowan N, Vaughan HG Jr. Electrophysiological evidence of developmental changes in the duration of auditory sensory memory. Dev Psychol 999; 5: Grau C, Escera C, Yago E, Polo MD. Mismatch negativity and auditory sensory memory evaluation: a new faster paradigm. NeuroReport 998; 9: chroeger E. Measurement and interpretation of the mismatch negativity. Behav Res Methods, Instrum Comput 998; : 45. Vol9 No5 6March8 57

Development of auditory sensory memory from 2 to 6 years: an MMN study

Journal of Neural Transmission, 2008; Online First DOI 10.1007/s00702-008-0088-6 Die Originalarbeit ist verfügbar unter: www.springerlink.com Development of auditory sensory memory from 2 to 6 years: an