PhD Project 6
Sharpening sensory predictions by linguistic primes
Studies within the visual domain have shown that top-down prediction facilitates perception by sharpening the response to expected percepts in early visual cortex. Here, we aim to investigate whether this also holds for cross-modal priming between language and vision. We will use words to prime visual stimuli and measure magnitude and sharpness of responses in primary and high-order visual areas. We will study how the prediction for specific visual features, e.g. face gender characteristics, depends on vocabulary. We will further manipulate the predictions by varying the frequency of congruent prime-target pairs, and by using primes with different levels of specificity.
This project proposes novel approaches (from the modern perspective on perception as afforded by predictive inference) to explore the classical question on the interaction between perception and language. The main hypothesis in this project is based on very recent advances in the research on sensory perception. This study therefore will bring novel ideas into the discussion on language-perception interaction.
The behavioural experiments on the effects of language on face gender perception have been supplemented with a computational model, with the purpose of establishing the effect of language on different stages of perceptual decision-making. A series of additional behavioural experiments was designed and conducted, focusing on the effects of gender-associated words on face perception. The results of these experiments are in preparation for publication. To investigate the timing of priming effect and the level of linguo-perceptual interactions, a MEG experiment has been designed and approved by the local neuroimaging community (DCCN). The data acquisition is planned for 2017. A pilot/behavioural experiment is being carried out.