Human visual perception is contingent on a visual system that can efficiently bridge the gap between perceived and physical reality, all while balancing the physical and metabolic constraints on processing. Early visual cortex has been identified as the foundation for this bridge, where modulatory signals, like attention and emotion, facilitate and bias processing towards the most relevant visual information for the task at hand. However, there remains an incomplete understanding of how visual processing is shaped by attention to time and features, and with emotional state. To address these gaps in knowledge, the experiments outlined in this dissertation use psychophysics, fMRI, and computational modeling to explore the mechanisms that subserve changes in perception due to attention- and emotion-based modulation of early visual cortex in humans.
First, we examined the mechanisms that subserve temporal attention – the allocation of attention to a specific moment in time (Chapter 2). To do so, we manipulated temporal attention in an equivalent noise approach, measuring fine orientation discrimination thresholds in visual noise (n = 12). Under a normalization framework, results revealed that temporal attention selectively enhanced perceptual thresholds for a target signal across increasing noise levels. Moreover, a control experiment (n = 6) confirmed that temporal attention did not reduce uncertainty in target detection, but instead selectively enhanced the appearance of target information.
Next, we examined how attention alters spatial frequency processing – the encoding of coarse and fine details in an image (Chapter 3). Previous studies in non-human primates have shown that attention modulates spatial frequency processing, but this remains untested in human early visual cortex. To do so, we leveraged an fMRI population spatial frequency tuning (pSFT) mapping approach while manipulating attention to spatial frequency (n = 8). Consistent with previous studies, attention altered spatial frequency processing across the visual field, but with respect to the attended spatial frequency and population receptive field eccentricity. Specifically, voxels with more peripheral receptive fields experienced an increase in preferred SF and a decrease in tuning bandwidth, while more foveal populations experienced a decrease in preferred SF and an increase in tuning bandwidth. These results support the spatial resolution hypothesis, suggesting that internal goals and required spatial resolution together bias visual processing.
Lastly, we used the same fMRI pSFT mapping approach to test how emotional arousal alters spatial frequency processing in early human visual cortex (Chapter 4). Previous studies have found behavioral deficits in high spatial frequency processing in fearful contexts, where priority is given to lower spatial frequencies. However, there is a lack of human evidence supporting this change at the visuocortical population level. In our experiment (n = 7), we manipulated affective arousal with emotional sounds while concurrently mapping pSFT and recording physiological measures of arousal (pupillometry and skin conductance). Despite ambiguous physiological measures of arousal, we found a significant and non-monotonic effect of sound on pSFT that varied as a function of spatial frequency preference – a pattern of results that mirrors how attention modulates spatial frequency processing.
Together, these findings speak to the malleability of the human visual system in selectively altering visual processing with respect to attentional and emotional cues. Chapter 1 provides an important insight into how attention to time selectively enhances visual target signals in increasingly noisy environments. Additionally, Chapters 2 and 3 are the first to provide quantitative measures for how attention and arousal alter spatial frequency processing in early visuocortical populations, contributing to a deeper understanding of the adaptability of the human visual system to internal and environmental demands.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/47961 |
| Date | 25 January 2024 |
| Creators | Ramirez, Luis Daniel |
| Contributors | Ling, Sam |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
| Rights | Attribution 4.0 International, http://creativecommons.org/licenses/by/4.0/ |
Page generated in 0.0019 seconds