Scottish Vision Group Meeting,Isle of Skye, Scotland,5–7 April 2019

摘要

Round Table Discussion: Virtual reality and the future of vision research Paul Hibbard ~(1), Rafa? Mantiuk ~(2) and Gizem Rufo ~(3) ~(1)University of Essex ~(2)University of Cambridge ~(3)Facebook Reality Labs Abstract Recent advances in consumer-grade virtual reality (VR) equipment have opened up new research topics for vision science. In this discussion session, panellists from a wide range of backgrounds (psychophysics, computer graphics, industry research) will together provide a diverse perspective on VR both as an experimental tool and as a topic of study. Each panellist will give a ten-minute introduction on their work with VR, and these presentations will be followed by a 30-minute round-table discussion session with participation from the audience. Keynote Lecture: Integrative active vision Iain D. Gilchrist University of Bristol Abstract Vision involves looking and seeing: both are required to support visually guided behaviour. The process of looking involves the generation of saccadic eye movements to point the high-resolution fovea to relevant locations in the environment. The brain network that supports the generation of saccadic eye movements has been studied in detail for over fifty years and we have a detailed understanding of both its anatomy and neurophysiology. What is less clear is how the saccadic network interacts with other brain systems. In this talk I will review examples of our research that attempt to address this issue. I will explore interactions between the saccadic system and face processing, respond timing, salience and choice. Together these studies describe a rich and complex pattern of interactions that allow the brain to generate integrated visually guided behaviour. Attentional enhancement of relevant features precedes the suppression of irrelevant features even when distractors are cued Plamen A. Antonov, Ramakrishna Chakravarthi and S?ren K. Andersen University of Aberdeen Abstract Selective attention may allocate limited visual processing resources mainly by enhancing attended or suppressing unattended information. In most experiments, attentional cues indicate to-be-attended rather than to-be-ignored stimuli, thereby potentially biasing selectivity towards enhancement. Here we compared cued shifts of feature-selective attention between conditions in which attended stimuli were cued with conditions in which unattended stimuli were cued. Two superimposed (red or blue), flickering, random dot kinematograms (RDKs) that elicited steady-state visual evoked potentials (SSVEPs) were presented. Auditory cues instructed participants to either attend the relevant (e.g. “attend red”) or ignore (e.g. “ignore blue”) the irrelevant RDK on a trial-by-trial basis and to detect brief coherent motion events in the attended RDK while ignoring such events in the unattended RDK. In ‘attend’ trials, enhancement of SSVEPs elicited by attended stimuli preceded the suppression of unattended stimuli, confirming previous findings. In ‘ignore’ trials, hit rates were lower and stabilized ～560 ms later post-cue than in ‘attend’ trials. Enhancement still preceded suppression, but importantly, these processes were also shifted back by ～630 ms. We interpret this as the result of semantically translating “ignore” to “attend” cues instead of using them to directly suppress irrelevant stimuli (e.g. “ignore blue” is substituted by “attend red”). This interpretation is also supported by ERPs elicited by the auditory cues. Overall, our findings suggest that distractor suppression is not under direct voluntary control. Gaze contingencies in joint attention as social reward that drives motor learning Malgorzata Kasprzyk and Bert Timmermans University of Aberdeen Abstract We investigate whether the supposedly rewarding nature of joint attention can drive saccadic motor learning similarly to monetary rewards. Interactionists suggest that social interaction plays a crucial role in development and learning of social skills because seeing others reacting to our actions constitutes an intrinsic reward (Schilbach et?al., 2013). Recent studies demonstrated that some forms of social interaction (e.g. joint attention) activate similar brain areas to rewards such as money supporting the notion of this rewarding nature of social interaction (Pfeiffer et?al., 2014). However, it is unclear whether social interaction per se can drive learning similarly to those rewards (Schilbach et?al., 2010). The current study is an eye-tracking paradigm based on a Milstein and Dorris (2007) study, and requires participants to interact with an anthropomorphic avatar or a cylinder and then conduct a saccade towards a left or right peripheral target. Fast saccades are rewarded by the avatar following participant’s gaze (social reward) or the increase of accumulated gains displayed on the cylinder (monetary reward) with one side being more rewarded then the other (80% vs 20%). We expect that both social and monetary rewards will facilitate