PhD Oral Exam - Bruno Richard, Psychology

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

The design and function of the human visual system is thought to have been shaped by the environment and tasks humans have performed throughout evolution and experience. Thus, it is important to establish the association between the properties of natural scenes and the responses of the visual system to these features at a behaviourally relevant level (e.g., psychophysically). A relevant property of natural scenes - which human observers are sensitive to - is the slope of the orientation averaged Fourier amplitude spectrum (α). It characterizes the decrease in amplitude as a function of spatial frequency (1/fα), and has an average value of 1.0 (on logarithmic axes) in natural scenes. The process of detecting a change in α is thought to stem from the activity of one or more spatial frequency channels, but their exact contribution remains to be defined. Thus, the overarching goal of this dissertation was to assess the contribution of spatial frequency channels to the detection, and discrimination, of a change in α. We first set out to find psychophysical evidence that α discrimination thresholds were dependent on the spatial frequency content of noise stimuli with 1/f amplitude spectra. Our results showed that α discrimination is regulated by low spatial frequency contrast. Thus, we attempted to modulate the activity of spatial frequency channels with trans-cranial Direct Current Stimulation (tDCS) and change α discrimination thresholds. But, while capable of modulating contrast detection, we found tDCS to be an ill-suited technique to modulate multiple channel activity under suprathreshold conditions (i.e., broadband noise discrimination). Consequently, we turned to a classification image paradigm to understand how different spatial frequencies contribute to the identification of a change in α. We found the identification of α and the identification of a change in α (e.g., discrimination) are completed according to different processes. That is, identification is specific to α, while discrimination is not; observers identified a change in αs according to increases in luminance contrast in the same low spatial frequency bands. Taken together, these findings demonstrate that the discrimination of α, regardless of its value, is a low spatial frequency dependent process.