Scaffolding vocal development: maternal responsiveness to ...
1
v Vocalization type per min v Maternal responsiveness to type of vocalization Alicja Radkowska (a) , Iris Nomikou (b) , Giuseppe Leonardi (c) , Katharina Rohfling (c) & Joanna Rączaszek-Leonardi (a) Results – repeated measures ANOVA Hypotheses v In comparison to non-speech-related vocalization, infant’s amount of speech-related vocalization will increase with age. v Mothers will respond more to speech-related vocalizations regardless of infants’ age (in comparison to non-speech-related ones). v Mothers reactions to speech-related vocalizations will be quicker regardless of infants’ age (in comparison to non-speech-related ones). Acknowledgements: This work was supported by the by the NCN-DFG collaborative Beethoven project EASE (2014/15/G/HS1/04536) and by Cluster of Excellence Cognitive Interaction Technology “CITEC” (EXC 277) at Bielefeld University funded by the German Research Foundation (DFG) and Warsaw University Foundation. We thank the participating families. Visit 1 3 m Visit 2 4 m Visit 3 5 m Visit 4 6 m Visit 5 7 m Visit 6 8 m German longitudinal study of parent – infant interactions Study design (a) University of Warsaw, (b) University of Portsmouth, (c) Paderborn University Results – Cross-Recurrence Analysis: Diagonal Profiles Scaffolding vocal development: maternal responsiveness to infant speechlike vocalizations at three, six and eight months Participants v Vocalization type: speech-related (SL) per minute, non-speech-related (NS) per minute v Mother responsiveness to vocalization type within 2 sec window – proportion: responses to SL/SL vocalizations; proportion: responses to NS/NS vocalizations; v Mean time of mother response (MRT) to vocalization type: MRT to SL, MRT to NS v ELFRA 2 – Overall productive vocabulary at 24 months reported by parents via a report Measures v 16 German dyads (9 boys, 7 girls) v Age: 1 st visit M = 3,44; SD = 0,18 ; Age 4th visit M = 6,24; SD = 0,26; Age 6th visit M = 8,06, SD = 0,14 Results – Cross-Recurrence Analysis: Diagonal Profiles Introduction Parental responsiveness is an important feature of parent-infant interactions for infant’s development and learning, including language. Responsiveness is also a part of a social feedback loop between parents and children (Warlamount, Richards, Gilkerson &Oller, 2014) and might have consequences for their later language development (Tamis- LeMonda, Bornstein & Baumwell, 2001; Wu & Gros-Louis, 2014). Yet, most studies concerned with responsiveness to infants’ vocalizations have investigated infants from 8 months on. Parents are prompt to contingently respond to speech-related vocalizations of 8-48 months old children (Warlamount et al., 2014). In the present work we checked if parents promote spontaneous, speech- related sounds by timed responses even earlier in development. We aimed to investigate longitudinally differences in parental responsiveness to early speech-related and non-speech-related infants’ vocalizations when infants were 3, 6 and 8 months old. Furthermore, the temporal characteristic of responses to those two types of vocalizations was explored. • trend for main effect of type F(1, 14) = 4,18; p = 0,06; η 2 = 0,23 Ø SL vocalizations responded to more than NS • main effect of type F(1, 13) = 4,9; p < 0,05; η 2 = 0,27 Ø MRT shorter for SL than NS vocalizations • n.s. v Mean reaction times to vocalization type Whole group Groups: scoring high (left) or low (right) in ELFRA2 Overall productive vocabulary v For 3 month olds: the ‘peak’ for SL in diagonal profile for whole group is present in both, high and low ELFRA2 groups. v For 6 month olds: the ‘peak’ for SL in the profile for high ELFRA2 group (albeit small) which is not present in low ELFRA2 group; v For 8 month olds: the ‘peak’ for SL observable in profile for whole group is present in profile for high ELFRA2 group, but not in low ELFRA2. Mothers of infants in low ELFRA2 group seemed to not distinguish their reactions between SL and NS vocalizations. Sample cross-recurrence plot Conclusions v In the 3-8 months age window infants tended to produce more speech-related vocalizations with age in comparison to non-speech-related, but differences didn’t reach significance. v Mothers showed higher responsiveness to speech-related vocalizations in comparison to non-speech-related regardless of infants’ age. v Duration of mean reaction times of maternal responses to speech-related vocalizations was shorter than MRT to non-speech-related vocalizations, regardless of the age of the infant. Taken together, these results suggest that mothers early in development start to support emerging vocal communicational skills by selective and timed responses. v Diagonal profiles from CRQA might be a robust method for characterization of temporal dynamics of maternal responses to infant vocalizations. v Early differences in maternal responses to different types of infants’ vocalizations may contribute to trajectories of development and later language outcomes. Analysis v repeated measures ANOVA v Cross-recurrence quantification analysis (Dale, Warlamount & Richardson, 2011) • the level of recurrence rate (% REC) was extracted from recurrence plots • plotted diagonal-wise recurrence profiles (%REC as a function of the lag in a window of ± 5 seconds)
Transcript of Scaffolding vocal development: maternal responsiveness to ...
v Vocalizationtypepermin vMaternalresponsivenesstotypeofvocalization
AlicjaRadkowska(a),Iris Nomikou (b),GiuseppeLeonardi (c),KatharinaRohfling (c) &JoannaRączaszek-Leonardi (a)
Results– repeatedmeasuresANOVA
Hypothesesv Incomparisontonon-speech-relatedvocalization,infant’samountof
speech-relatedvocalizationwillincreasewithage.vMotherswillrespondmoretospeech-relatedvocalizationsregardless
ofinfants’age(incomparisontonon-speech-relatedones).vMothersreactionstospeech-relatedvocalizationswillbequicker
regardlessofinfants’age(incomparisontonon-speech-relatedones).
Acknowledgements:ThisworkwassupportedbythebytheNCN-DFGcollaborativeBeethovenprojectEASE(2014/15/G/HS1/04536)andbyClusterofExcellenceCognitiveInteractionTechnology“CITEC”(EXC277)atBielefeldUniversityfundedbytheGermanResearchFoundation(DFG)andWarsawUniversityFoundation.Wethank theparticipating families.
Visit 13 m
Visit 24 m
Visit 35 m
Visit 46 m
Visit 57 m
Visit 68 m
Germanlongitudinalstudyofparent– infantinteractionsStudydesign
(a) UniversityofWarsaw,(b)UniversityofPortsmouth,(c) PaderbornUniversity
Results– Cross-RecurrenceAnalysis:DiagonalProfiles
Scaffolding vocal development: maternal responsiveness to infant speechlike vocalizations at three, six and eight months
Participantsv Vocalizationtype:speech-related(SL)per
minute,non-speech-related(NS)perminutevMotherresponsivenesstovocalizationtype
within2secwindow– proportion:responsestoSL/SLvocalizations;proportion:responsestoNS/NSvocalizations;
vMeantimeofmotherresponse(MRT)tovocalizationtype:MRTtoSL,MRTtoNS
v ELFRA2– Overallproductivevocabularyat24monthsreportedbyparentsviaareport
Measuresv 16Germandyads(9boys,7girls)v Age:1st visitM=3,44;SD=0,18;Age4thvisitM=6,24;SD=0,26;Age6thvisit M=8,06,SD=0,14
Results– Cross-RecurrenceAnalysis:DiagonalProfiles
IntroductionParental responsiveness is an important feature of parent-infantinteractions for infant’s development and learning, including language.Responsiveness is also a part of a social feedback loop between parentsand children (Warlamount, Richards, Gilkerson &Oller, 2014) and mighthave consequences for their later language development (Tamis-LeMonda, Bornstein & Baumwell, 2001; Wu & Gros-Louis, 2014). Yet,most studies concerned with responsiveness to infants’ vocalizationshave investigated infants from 8 months on.
Parents are prompt to contingently respond to speech-relatedvocalizations of 8-48 months old children (Warlamount et al., 2014). Inthe present work we checked if parents promote spontaneous, speech-related sounds by timed responses even earlier in development.
We aimed to investigate longitudinally differences in parentalresponsiveness to early speech-related and non-speech-related infants’vocalizations when infants were 3, 6 and 8 months old. Furthermore, thetemporal characteristic of responses to those two types of vocalizationswas explored.
• trendformaineffectoftypeF(1,14)=4,18;p=0,06;η2 =0,23Ø SLvocalizationsrespondedto
morethanNS
• maineffectoftypeF(1,13)=4,9;p<0,05;η2 =0,27Ø MRTshorter forSLthan NS
vocalizations
• n.s.
vMeanreactiontimestovocalizationtype
Whole groupGroups:scoring high(left)or low (right)inELFRA2
Overall productive vocabulary
v For3month olds:the‘peak’forSLindiagonalprofileforwhole group is present inboth,highandlow ELFRA2groups.
v For6month olds:the‘peak’forSLintheprofileforhighELFRA2group (albeit small)which is notpresent inlow ELFRA2group;
v For8month olds:the‘peak’forSLobservable inprofileforwhole group is present inprofileforhighELFRA2group,butnotinlow ELFRA2.Mothers ofinfants inlow ELFRA2group seemed tonotdistinguish their reactions between SLandNSvocalizations.
Sample cross-recurrence plot
Conclusionsv In the 3-8 months age window infants tended to produce more speech-related
vocalizations with age in comparison to non-speech-related, but differences didn’treach significance.
vMothers showed higher responsiveness to speech-related vocalizations incomparison to non-speech-related regardless of infants’ age.
v Duration of mean reaction times of maternal responses to speech-relatedvocalizations was shorter than MRT to non-speech-related vocalizations, regardlessof the age of the infant. Taken together, these results suggest that mothers early indevelopment start to support emerging vocal communicational skills by selectiveand timed responses.
v Diagonal profiles from CRQA might be a robust method for characterization oftemporal dynamics of maternal responses to infant vocalizations.
v Early differences in maternal responses to different types of infants’ vocalizationsmay contribute to trajectories of development and later language outcomes.
Analysisv repeatedmeasuresANOVAv Cross-recurrencequantification
analysis(Dale,Warlamount &Richardson,2011)
• thelevelofrecurrencerate(%REC)wasextractedfromrecurrenceplots
• plotteddiagonal-wiserecurrenceprofiles(%RECasafunctionofthelaginawindowof± 5seconds)
