Theses and Dissertations
This collection includes both ASU Theses and Dissertations, submitted by graduate students, and the Barrett, Honors College theses submitted by undergraduate students.
Displaying 1 - 2 of 2
Filtering by
- Creators: Adelson, P David
- Creators: Glenberg, Arthur
Description
Epilepsy is a group of disorders that cause seizures in approximately 2.2 million people in the United States. Over 30% of these patients have epilepsies that do not respond to treatment with anti-epileptic drugs. For this population, focal resection surgery could offer long-term seizure freedom. Surgery candidates undergo a myriad of tests and monitoring to determine where and when seizures occur. The “gold standard” method for focus identification involves the placement of electrocorticography (ECoG) grids in the sub-dural space, followed by continual monitoring and visual inspection of the patient’s cortical activity. This process, however, is highly subjective and uses dated technology. Multiple studies were performed to investigate how the evaluation process could benefit from an algorithmic adjust using current ECoG technology, and how the use of new microECoG technology could further improve the process.
Computational algorithms can quickly and objectively find signal characteristics that may not be detectable with visual inspection, but many assume the data are stationary and/or linear, which biological data are not. An empirical mode decomposition (EMD) based algorithm was developed to detect potential seizures and tested on data collected from eight patients undergoing monitoring for focal resection surgery. EMD does not require linearity or stationarity and is data driven. The results suggest that a biological data driven algorithm could serve as a useful tool to objectively identify changes in cortical activity associated with seizures.
Next, the use of microECoG technology was investigated. Though both ECoG and microECoG grids are composed of electrodes resting on the surface of the cortex, changing the diameter of the electrodes creates non-trivial changes in the physics of the electrode-tissue interface that need to be accounted for. Experimenting with different recording configurations showed that proper grounding, referencing, and amplification are critical to obtain high quality neural signals from microECoG grids.
Finally, the relationship between data collected from the cortical surface with micro and macro electrodes was studied. Simultaneous recordings of the two electrode types showed differences in power spectra that suggest the inclusion of activity, possibly from deep structures, by macroelectrodes that is not accessible by microelectrodes.
Computational algorithms can quickly and objectively find signal characteristics that may not be detectable with visual inspection, but many assume the data are stationary and/or linear, which biological data are not. An empirical mode decomposition (EMD) based algorithm was developed to detect potential seizures and tested on data collected from eight patients undergoing monitoring for focal resection surgery. EMD does not require linearity or stationarity and is data driven. The results suggest that a biological data driven algorithm could serve as a useful tool to objectively identify changes in cortical activity associated with seizures.
Next, the use of microECoG technology was investigated. Though both ECoG and microECoG grids are composed of electrodes resting on the surface of the cortex, changing the diameter of the electrodes creates non-trivial changes in the physics of the electrode-tissue interface that need to be accounted for. Experimenting with different recording configurations showed that proper grounding, referencing, and amplification are critical to obtain high quality neural signals from microECoG grids.
Finally, the relationship between data collected from the cortical surface with micro and macro electrodes was studied. Simultaneous recordings of the two electrode types showed differences in power spectra that suggest the inclusion of activity, possibly from deep structures, by macroelectrodes that is not accessible by microelectrodes.
ContributorsAshmont, Kari Rich (Author) / Greger, Bradley (Thesis advisor) / Helms Tillery, Stephen (Committee member) / Buneo, Christopher (Committee member) / Adelson, P David (Committee member) / Dudek, F Edward (Committee member) / Arizona State University (Publisher)
Created2015
Description
ABSTRACT
Learning a novel motor pattern through imitation of the skilled performance of an expert has been shown to result in better learning outcomes relative to observational or physical practice. The aim of the present project was to examine if the advantages of imitational practice could be further augmented through a supplementary technique derived from my previous research. This research has provided converging behavioral evidence that dyads engaged in joint action in a familiar task requiring spatial and temporal synchrony end up developing an extended overlap in their body representations, termed a joint body schema (JBS). The present research examined if inducing a JBS between a trainer and a novice trainee, prior to having the dyad engage in imitation practice on a novel motor pattern would enhance both of the training process and its outcomes.
Participants either worked with their trainer on a familiar joint task to develop the JBS (Joint condition) or performed a solo equivalent of the task while being watched by their trainer (Solo condition). Participants In both groups then engaged in blocks of alternating imitation practice and free production of a novel manual motor pattern, while their motor output was recorded. Analyses indicated that the Joint participants outperformed the Solo participants in the ability to synchronize the spatial and temporal components of their imitation movements with the trainer’s pattern-modeling movements. The same group showed superior performance when attempting to freely produce the pattern. These results carry significant theoretical and translational potentials for the fields of motor learning and rehabilitation.
Learning a novel motor pattern through imitation of the skilled performance of an expert has been shown to result in better learning outcomes relative to observational or physical practice. The aim of the present project was to examine if the advantages of imitational practice could be further augmented through a supplementary technique derived from my previous research. This research has provided converging behavioral evidence that dyads engaged in joint action in a familiar task requiring spatial and temporal synchrony end up developing an extended overlap in their body representations, termed a joint body schema (JBS). The present research examined if inducing a JBS between a trainer and a novice trainee, prior to having the dyad engage in imitation practice on a novel motor pattern would enhance both of the training process and its outcomes.
Participants either worked with their trainer on a familiar joint task to develop the JBS (Joint condition) or performed a solo equivalent of the task while being watched by their trainer (Solo condition). Participants In both groups then engaged in blocks of alternating imitation practice and free production of a novel manual motor pattern, while their motor output was recorded. Analyses indicated that the Joint participants outperformed the Solo participants in the ability to synchronize the spatial and temporal components of their imitation movements with the trainer’s pattern-modeling movements. The same group showed superior performance when attempting to freely produce the pattern. These results carry significant theoretical and translational potentials for the fields of motor learning and rehabilitation.
ContributorsSoliman, Tamer (Author) / Glenberg, Arthur (Thesis advisor) / Helms Tillery, Stephen (Committee member) / McBeath, Michael (Committee member) / Amazeen, Eric (Committee member) / Arizona State University (Publisher)
Created2015