The link between sequence and language processing in the adolescent brain

This work addresses the link between auditory skill and language proficiency. The ability to hear is essential to normal language acquisition: what is addressed here is whether particular skills in the analysis of auditory patterns are needed for language development. Many studies have examined the relevance of analysis of simple sound patterns in single sounds or pairs of sounds to language skill. We are interested in whether the analysis of more-complex sound sequences is of particular relevance to language skill. This has face value given that listeners are required to chop up the speech stream into the correct speech units before those units can be properly understood. We propose the hypothesis that sequence analysis based on a number of different cues is critical to speech understanding: sequences of pitch like melodies, sequences of time-intervals like rhythms and sequences of timbre or 'tone colour'.

Sequence analysis requires high-level analysis in the auditory brain including the auditory cortex and continues to develop over adolescence. We propose to study the link between auditory analysis and language skill in early and late adolescence to examine the hypothesis based on our previous work (in early adolescence) that different types of sequence analysis are important at different stages of adolescence. To take rhythm as an example, in early adolescents there is a relationship between the analysis of simple regular rhythms (like that of a metronome) and language whilst in older subjects the relationship is with more abstract rhythms like when subjects are required to 'find a beat'.

The work is important because many subjects fail to acquire language normally by listening to it. Evidence suggests that this process goes wrong in dyslexia, defined as a disorder of reading and spelling that is not explained by low intelligence or lack of opportunity. Dyslexia occurs in 5% of the population and causes lifelong problems to the individual and great costs to society. If we prove that there are critical links between auditory sequence analysis and language skill this would suggest new behavioural interventions based on auditory sequence training to help language. This would also suggest further investigation of the parts of the brain responsible for sequence analysis in order to understand developmental language disorder.

We propose to carry out work at a local school with which we have developed a relationship over the last five years. We would test different levels of sequence complexity in the pitch, timing and timbre domains in two large cohorts (>200 each) at the ages of twelve and sixteen. The skill of these subjects in auditory analysis would be related to a battery of standardised language assessment measures. We would specifically test the hypothesised relationship between sequence analysis and language skill in these two groups and hypothesised differences in the relationship between sequence analysis and language at different ages. We also wish to examine underlying bases for sequence analysis like tone memory.

Further, we propose to investigate possible brain bases for the relationship between sequence analysis and language skill that builds upon pilot work in which we have demonstrated that there is a correlation between brain structure in the parietal lobe of the brain and both skills. We can do this non-invasively using magnetic resonance imaging on random groups chosen from the younger and older cohorts at the Newcastle Magnetic Resonance Centre (NMR) of Newcastle University .

The outcome of this work would be an understanding of the relationship between sound-sequence analysis and language skill and its brain basis, as a target for future training or intervention strategies.

The relevance of auditory processing to language skills is controversial. The need for acoustic analysis in early language acquisition is a self-evident truth: the controversy concerns what specific aspects of speech over what times scales are particularly relevant to language acquisition and further development , about which knowledge is limited. In this study we propose to carry out a systematic assessment of the perception of pitch, time intervals and timbre over different levels of sequence complexity from single sounds to sequences. The work tests the central hypothesis that there is a critical relationship between sound-sequence analysis and language skill, and examines whether this relationship changes over adolescence, when there are rapid changes in the auditory system. Auditory testing would be carried out on two cohorts aged 12 and 16 (>200 in each) using custom software and hardware in a classroom setting. Language measures would be carried out using a standardised battery of tasks. Analysis would seek correlation between overall auditory and language skill and, specifically, correlation between auditory sequence tasks and language skill. The analysis of two cohorts allows us to test the hypothesis that different types of sequence analysis correlate with language skill at different ages. Additionally we would carry out voxel based morphometry to test the hypothesis that the intraparietal sulcus is a critical substrate for both auditory-sequence skill and language skill, and that the substrate for sequence and language analysis changes over adolescence.

Immediate beneficiaries from this research would include any educational or health professionals who wished to understand the relevance of complex sound analysis to language skill. Currently the work we carry out is based on group inference and will allow a general understanding of the process. The relationship between sound-sequence analysis skill and language skill was mentioned in discussions to which Griffiths contributed at the House of Lords about musical education. This work highlights auditory bases for links between musical and language analysis that interest policy makers and teachers.

The work represents a programme in which new sequence measures are evaluated in an effort to develop assessment tools that allow single-subject inference and the definition of individuals with disorders. I see child and adult patients in my Central Auditory Clinic in London (run with a NeuroOtologist) and adult patients in the Newcastle Cognitive Clinic: a number of these have language impairment that is not explained by abnormal cochlear function or conventional neurological disorder. A realistic intermediate aim for this work is to develop single measures or combinations of measures that might better characterise what is wrong with such patients, and such measures would be welcomed by audiologists, speech and language therapists, neurootologists, and educational psychologists. I have a lot of contact with audiologists, especially in USA, many of whom are very keen for suggestions for a 'standard test battery'. It is likely that language disorders are so heterogeneous that this is not possible, but in the proposal we seek to develop a systematic approach that does not champion particular tests but might suggest dimensions of auditory analysis that are critical to language skill. We will also systematically test higher auditory cognition, especially auditory attention and working memory for sounds, that might be relevant to any links between auditory perceptual and language skills.

The most important pathway to impact for work of this kind is the potential development of training strategies that might improve language using auditory tasks. Although such training was introduced for all children in one US State more than ten years ago, trials of therapy have met with mixed success. But in our view successful therapy is still at best a possible intermediate impact of this work. An immediate critical need is a systematic understanding of the fundamental relationship between auditory analysis and language in normal and abnormal development that we focus on here.