Speech and language abilities underpin much of human communication, and enhance, augment, and focus our cognitive, creative and social skills.
In the Speech and Language group, we conduct research with infants, children and adults both within and between languages, focussing on areas such as:
Genetic and environmental influences in speech perception, early word learning and language and literacy development
Native and non-native language speech perception and the role of language background in second language (L2) learning
First and second language acquisition, bi- and multilingualism
The perception of vowels, consonants, lexical tones and prosody in speech
More information on specific research themes can be found below.
Participate in Speech and Language Research
The Speech and Language research group at the MARCS Institute run experiments that require monolingual and bilingual adult participants. These experiments include speech perception and production, language acquisition and literacy, as well as studies that involve and brain-imaging.
If you are interested in participating in research with the Speech and Language research group, the first step is to register your interest. The information you provide us in the registration form will be stored on a confidential database that will only be accessible by members of the research team for recruiting purposes.
If you are interested in participating in our research, please register your interest (opens in a new window).
Even before birth, infants attend to and perceive subtle differences in human speech sounds and speech prosody. Speech perception develops by a process of attunement: the accretion of attention to sounds that are not present in the ambient language.
The Speech and Language program at The MARCS Institute is home to two major theoretical speech perception formulations. Professor Catherine Best's (1995) Perceptual Assimilation Model (PAM) sets out how perception of phonetic details and formation of language-specific phonological categories affect word learning and vocabulary development, first and second language (L1 & L2) learning, and the development and maintenance of bilingualism.
Professor Anne Cutler's theory (see, e.g., Cutler, 'Native Listening', 2012) explains the interplay of universal and language-specific factors, such as rhythmic structure and phonotactic constraints, in the recognition of spoken language. It also explains one of the most difficult tasks in understanding a foreign language — segmenting the acoustically continuous speech stream into psycholinguistically distinct words.
These theories, along with Professor Denis Burnham's heuristic framework of Language Specific Speech Perception (Burnham, 1986; Burnham, Tyler & Horlyck, 2002), afford the development of a rich array of empirical investigations in which infants, children and adults are tested for their perception of speech segments (consonants, vowels, and lexical tones) that are relevant or irrelevant in their language. This approach, along with within- and between- language and accent studies, allows conclusions to be drawn about how speech perception develops in different linguistic environments.
Language acquisition and literacy
From the age of about six months to nine months, infants begin to attend selectively to those sounds and patterns that are specific to the language(s) they hear around them. In turn, this early language specific speech perception bootstraps ongoing language acquisition such as word segmentation (Kooijman, Hagoort & Cutler, 2009), word recognition and vocabulary learning (Best, Tyler, Gooding, Orlando & Quann, 2009) and even later reading ability (Burnham, 2003; Horlyck, Reid & Burnham, in press).
In the Speech and Language program, we take the investigation of speech perception beyond the infancy period in order to elucidate processes of speech perception, word learning, language acquisition, and the development of literacy.
The production of speech, unlike its perception, is not intact at birth and involves the complex juxtaposition of various motor movements that are not fully mastered until a child is four or five years old.
The ambient speech environment assists in the acquisition of speech, but while this process is seemingly easy, the inability of L2 learners beyond puberty to master the native accent speaks to the specificity of the learnt motor control. This results in distinctly different sound repertoires in different languages, and a wide variety of dialects and accents within the one language.
Both are studied in the Speech and Language program in order to elucidate the processes of speech production development, articulatory control, and the perception-action loop.
Special speech registers
In a series of studies, Dr. Christine Kitamura and her colleagues have investigated parents' infant-directed speech (IDS) and found that there are three separable components (attentional, affective, and linguistic, notably vowel hyperarticulation) of IDS (REFS).
The findings have discovered that IDS changes in nature over the first year of life (Kitamura & Burnham, 2003; Kitamura, Thanavisuth, Luksaneeyanawin & Burnham, 2002), with these changes matching infants' speech preferences (Kitamura & Lam, 2009; Kitamura & Notley, 2009), and that IDS to hearing-impaired infants or to infants with degraded speech input is not hyperarticulated (Lam & Kitamura, 2010; 2012).
The general conclusion is that parental speech input is under direct control of the cognitive and sensory abilities of the infants. In spin-off studies, it has been found that there is hyperarticulation of tones in Cantonese IDS (Xu & Burnham, 2010); and vowel hyperarticulation in foreigner- (Uther, Knoll & Burnham, 2009), and computer- (Burnham, Joeffry & Rice, 2010), but not pet-directed (Burnham, Kitamura & Vollmer-Conna, 2002) speech. This further defines the functions and determinants of these special speech registers.