2017年第二届曲阜视觉科学会议知识库

PURPOSE: The current study aimed to investigate the underlying mechanism of how perceived duration of a visual target is modulated by perceptual as well as contextual magnitude induced by Arabic numbers.
METHODS: The current study combined context manipulation with duration judgement. In the learning phase, participants were trained to categorize 340, 400, and 460ms intervals as short, medium, and long durations, respectively. The subsequent testing phase included small magnitude context (10, 20, 30, 10, 20, 30), large magnitude context (70, 80, 90, 70, 80, 90) and mixed magnitude context (10, 20, 30, 70, 80, 90). The participants first rated whether each of the six numbers in the contextual sequence was ‘‘smaller’’ or ‘‘larger’’ than 50. Then, a small (10 or 20) or large (80 or 90) target number was presented visually with a varying duration (340~460ms). Participants were required to categorize each target as a short, medium, or long duration. 
RESULTS: The perceived durations increased with physical durations significantly. The NTA effects, i.e., the over-/under- estimation of the durations of perceptual large/small numbers, were observed only during small, but not large or mixed magnitude context. More critically, the interaction of numerical context and target magnitude was significant. Large targets preceded by small magnitude context were rated as lasting longer, relative to large magnitude context. Whereas, small targets preceded by large magnitude context were rated as lasting longer, relative to small magnitude context.
CONCLUSIONS: The current results extend our understanding to the mechanism of NTA. The magnitude information existing in perceptual as well as contextual space can modulate perceived duration of a visual target and lead to a repetition suppression effect, suggesting inseparability of a hypothetic magnitude continuum across the border between physical present and contextual past. The repetition suppression effect is likely due to a reduction of neural activity of numerical representation, consistent with the neural amplitude hypothesis.