European Commission logo
polski polski
CORDIS - Wyniki badań wspieranych przez UE
CORDIS
Zawartość zarchiwizowana w dniu 2024-06-18

Tuning Attention during Language Learning

Final Report Summary - TUNINGLANG (Tuning Attention during Language Learning)

Speech is a very complex stimulation with rapid acoustic transitions building slower variations and following regularities in the way these sounds are combined. How is our cognitive system able to extract the richness of vocabulary and rules characterizing a language from this complex stimulation? For other functions such as memory or visual perception, the attention system optimizes learning simplifying the input by focusing in those elements that are relevant for the task or goal at hand. In the particular case of language, the inherently time dependent flow of speech makes temporal orienting of attention (TO) a likely candidate to be engaged and help in the learning process. This project was interested in understanding how temporal orienting of attention is differently tuned to assist in the extraction of words and rule dependencies in speech. The project allowed us to observe the tight relationship between attention and language in the process of learning, in developmental, anatomical and functional terms.
Developmentally, we could observe that infants start being able to control when to attend with no influence of external cues in the second year of life and this ability improves in the following years when executive functions allow they to flexibility shift their attention. It is the interaction between TO abilities and executive functions that was found to relate to rule learning, in infants, children and adults. The work performed in adults allowed us to discover that, in addition, these TO requiring the executive system engage the left dorsal fronto-parietal system and correspond to a second stage in the learning process that allows individuals to create an attentional focus based on the a priori knowledge learned. This a priori is built from the preceding initial stage where statistical computations extract predictions between elements in the dependencies and recruit a more ventral left fronto-parietal system. Functionally these networks are complemented by the basal ganglia that appears to have a role in the updating of the expectations progressively through learning, acting in the interaction between the dorsal and ventral fronto-parietal systems. Interestingly this ventral system overlaps between the attention and language systems at the anatomical level and may be a network enabling the link between attention and language in the left hemisphere. In contrast, word learning required also temporal tuning of attention through different processes. Distinct parts of the language networks and their connections were involved for this. In this case, the connectivity between frontal and temporal regions at the functional and structural levels allows a synchronization with the speech input that has consequences for optimal word learning abilities.