Ylva Østby

In this thesis, the developmental brain structure and development of memory functions are investigated, both separately and conjointly. First, the different developmental trajectories of cortical and subcortical brain structures are described. Then, the neurobiological foundations of two instances of memory are investigated in the context of development. The aim of the three studies presented here is to discover relationships between brain maturation, in terms of grey and white matter volumes, cortical thickness and microstructural properties of white matter pathways, and verbal working memory and visuo-spatial long-term memory. The body of literature on brain development is currently growing fast, but unresolved issues still remain. Among these is the controversy regarding hippocampal development: does its volume increase or decrease during late childhood and adolescence?

A first step towards understanding the biological foundations of memory development includes a detailed mapping of brain maturation, both in terms of cortical, sub-cortical and white matter development. While a general understanding of structural brain maturation is beginning to achieve foothold in the field of neuroscience, less is known about the relationships between brain maturational events and cognitive development. This is particularly so when it comes to memory development. Thus, although the current thesis can only handle a limited selection of remembering as performed by children and adolescents, this approach constitutes a step towards bridging the gap between biological and cognitive accounts of memory development.

In the first study, volumetric analyses showed great heterogeneity of brain maturation from 8 to 30 years, with cerebral cortex and subcortical structures differing in their declining trajectories towards adulthood. A result that stood out from the rest of the subcortical developmental paths was slight a developmental increase in volumes of medial temporal lobe structures, the amygdala and the hippocampus, of which the hippocampus is particularly important for long-term memory.

On the behavioral level, in the second study, developmental differences where also found for different aspects of long-term memory, as measured by the Rey-Osterrieth Complex Figure Test. While recall performance showed a steady increase from 8-19 years, the relative retention of the material between the 30 minutes recall condition and the 1 week recall condition, remained steady at about 83 % throughout the age-span. These differences mapped onto the brain developmental differences, as recall performance was related to cortical thickness in left orbitofrontal cortex, while retention was related to hippocampal volume.

Another way that diversity in brain maturational processes proved to be important in memory development was shown in the third study, where the two processes of cortical thinning and increasing white matter pathway organization where jointly put to the test of predicting working memory performance in development. In this study, theoretically based pre-selected regions within a fronto-parietal network of the left hemisphere, and the connecting pathway of the left superior longitudinal fasciculus (SLF), were related to performance on a digit span task, showing that each of a number of regions of the parietal and lateral prefrontal cortex, as well as the SLF, were almost equally important correlates of performance of both a simple storage span task (Digit Span Forwards) and a complex span task (Digit Span Backwards).

Developmentally, the contribution of each of cortical and white matter tract variables seemed to have different impact at different time periods of development, with white matter tract properties being predictive of performance in early adolescence, followed by cortical thickness being the most important correlate of working memory in late adolescence.

Taken together, the three studies underscore the importance of considering multiple facets of brain maturation in studying memory development, both for short-term and long-term memory. Not all brain developmental paths are created equal, and their relationships with various memory processes, like working memory span, long-term memory recall and –retention, are diverse and complex. Considerations of methodological constraints include, among others, the limitations inherent in a cross-sectional design, the problems with keeping the memory variables the same across age groups, and issues concerning the MRI-derived brain variables. Also, the theoretical dilemma of assigning a causal direction to the relationships between brain variables and memory performance is discussed.

Publisert 9. nov. 2011 09:56