In this set of studies the modulation of feedback-based cognitive skill learning was investigated by modulating a probabilistic classification learning (PCL) task to be either emotional or neutral. In the current task, based on the weather prediction task, cue cards were presented on the screen and subjects were asked to predict what they would come across while walking in the woods, in the emotional condition a snake/spider or in the neutral condition a flower/mushroom. Chapter 1 is a review of the animal and human literature of multiple memory systems, amygdala modulation of multiple memory systems, and sleep-dependent procedural memory consolidation.Chapter 2 examined how emotional arousal affected performance, strategy use, and sympathetic nervous system activation in our manipulated PCL task. Subjects highly fearful of the outcomes in the emotional condition showed overall greater skin conductance responses compared to the other groups, as well as retardation in initial cue-outcome acquisition. Individuals who were not fearful of the outcome stimuli used more complex (optimal) strategies after a 24-hr period of memory consolidation relative to the other groups, reflecting greater implicit knowledge of the probabilistic task structure.The purpose of the experiment in Chapter 3 was to examine consolidation-based stabilization and enhancement in an emotional cognitive skill task. There was no effect of sleep on retention or savings on percent correct or strategy use in both the emotional and neutral PCL task. These results conform to recent evidence that probabilistic learning does not show sleep-dependent performance enhancements.Chapter 4 investigated the neural correlates of emotional PCL with functional magnetic resonance imaging. There was greater amygdala and striatal activity in the emotional versus neutral group on Day 1. There was also increased activity in the striatum on Day 2, suggesting an early and lasting bias of emotion on procedural learning. Additionally, there were differences in neural recruitment by subjects using complex versus simple implicit strategies.The findings from this series of experiments have implications for the assessment of psychopathologies that show dysfunction in affective and striatal areas, such as obsessive-compulsive disorder and Tourette's syndrome, and for the development, eventually, of optimal therapies. / Dissertation
| Identifer | oai:union.ndltd.org:DUKE/oai:dukespace.lib.duke.edu:10161/448 |
| Date | 13 December 2007 |
| Creators | Thomas, Laura Anderson |
| Contributors | LaBar, Kevin S |
| Source Sets | Duke University |
| Language | en_US |
| Detected Language | English |
| Type | Dissertation |
| Format | 1362432 bytes, application/pdf |
Page generated in 0.002 seconds