Contemporary DRAM systems have maintained impressive scaling by managing a careful balance between performance, power, and storage density. In achieving these goals, a significant sacrifice has been made in DRAM's operational complexity. To realize good performance, systems must properly manage the significant number of structural and timing restrictions of the DRAM devices. DRAM's efficient use is further complicated in many-core systems where the memory interface has to be shared among multiple cores/threads competing for memory bandwidth. In computer architecture, caches have primarily been viewed as a means to hide memory latency from the CPU. Cache policies have focused on anticipating the CPU's data needs, and are mostly oblivious to the main memory. This work demonstrates that the era of many-core architectures has created new main memory bottlenecks, and mandates a new approach: coordination of cache policy with main memory characteristics. Using the cache for memory optimization purposes dramatically expands the memory controller's visibility of processor behavior, at low implementation overhead. Through memory-centric modification of existing policies, such as scheduled writebacks, this work demonstrates that performance-limiting effects of highly-threaded architectures combined with complex DRAM operation can be overcome. This work shows that an awareness of the physical main memory layout and by focusing on writes, both read and write average latency can be shortened, memory power reduced, and overall system performance improved. The use of the "Page-Mode" feature of DRAM devices can mitigate many DRAM constraints. Current open-page policies attempt to garner the highest level of page hits. In an effort to achieve this, such greedy schemes map sequential address sequences to a single DRAM resource. This non-uniform resource usage pattern introduces high levels of conflict when multiple workloads in a many-core system map to the same set of resources. This work presents a scheme that provides a careful balance between the benefits (increased performance and decreased power), and the detractors (unfairness) of page-mode accesses. In the proposed Minimalist approach, the system targets "just enough" page-mode accesses to garner page-mode benefits, avoiding system unfairness. This is accomplished with the use of a fair memory hashing scheme to control the maximum number of page mode hits. High density memory is becoming ever more important as many execution streams are consolidated onto single chip many-core processors. DRAM is ubiquitous as a main memory technology, but while DRAM's per-chip density and frequency continue to scale, the time required to refresh its dynamic cells has grown at an alarming rate. This work shows how currently-employed methods to schedule refresh operations are ineffective in mitigating the significant performance degradation caused by longer refresh times. Current approaches are deficient -- they do not effectively exploit the flexibility of DRAMs to postpone refresh operations. This work proposes dynamically reconfigurable predictive mechanisms that exploit the full dynamic range allowed in the industry standard DRAM memory specifications. The proposed mechanisms are shown to mitigate much of the penalties seen with dense DRAM devices. In summary this work presents a significant improvement in the ability to exploit the capabilities of high density, high frequency, DRAM devices in a many-core environment. This is accomplished though coordination of previously disparate system components, exploiting integration of such components into highly integrated system designs. / text

The purpose of this study was to further explore previously observed differences in phonological processing between adults who do and do not stutter through a list recall task. Three types of lists of words were generated according to their associations with a lure word: phonological, semantic, and hybrid. For the experimental task, participants were instructed to listen to recordings of lists of 12 words, 4 of each type, and immediately recall them in any order. We looked at recall accuracy and rate of production of each list's associated lure word. For recall accuracy, phonological lists were lowest, hybrid lists were in the middle, and semantic lists were highest. For production of the critical lure, phonological lists were the lowest, semantic lists were in the middle, and hybrid lists were highest. The pattern was the same for recall accuracy and critical lure production for both talker groups; however, the adults who stutter had lower means for each condition in both cases. The results provide further evidence that there are systematic and significant differences in the phonological working memory efficiency of AWS and AWNS, which may contribute to fluency differences. / text

published_or_final_version / Medicine / Doctoral / Doctor of Philosophy

Reduced autobiographical memory specificity is considered to be a vulnerability factor for depression and is a significant predictor of the course of depression. Previous studies have shown that different natures of rumination (abstract versus concrete) have different effects on autobiographical memory specificity. Knowing more about the relationship among depression, rumination, and autobiographical memory specificity is important for understanding the cognitive biases in depression. This study explored whether rumination of different valences had different effects on autobiographical memory specificity in participants with major depressive disorder. A 2 (group: MDD, control) x 2 (rumination: positive, negative) x 2 (time: pre, post) mixed design was used. Fifty-two currently depressed people and 52 nonpsychiatric controls completed this experiment. They did the Autobiographical Memory Test and the mood ratings before and after either the positive or the negative rumination task. In the rumination task, they were requested to focus their attention on some specific thoughts about themselves. Results showed a significant group (depressed, control) x time (pre, post) interaction effect for the number of specific memories. This was a result of a significant decrease in specific memories retrieved after negative rumination in the depressed group, but not after positive rumination. No significant result was found among the nonpsychiatric controls. These findings seem to be related to the inhibitory deficit of depressed people in keeping task-irrelevant negative materials from the working memory. Therefore, the working memory capacity is lowered and fewer specific memories are retrieved. / published_or_final_version / Clinical Psychology / Doctoral / Doctor of Psychology

Not available / text

Not available / text

Not available / text

Procedural memory consolidation has been shown to enhance a variety of perceptual and motor skills during sleep. Only recently has this effect been investigated in trained musicians performing music. I tested the extent to which a music performance skill benefited from sleep-based consolidation overnight and whether this process may be inhibited when musicians learn two melodies in juxtaposition. 60 experienced musicians, all nonpianists, learned to perform either one or two 13-note piano melodies during evening training sessions. The musicians practiced each melody with their nondominant hand by repeating it from beginning to end during 12 30-second practice blocks alternating with 30-second rest intervals. All participants were retested on the target melody the following morning in three 30-second retest blocks alternating with 30-second rest intervals. Participants who learned only one melody in the evening showed overnight gains in the number of correct key presses per block (CKP/B) in the target melody at retest. Participants who learned the target melody and an additional melody at training showed no overnight gains in CKP/B in the target melody. Participants who learned both melodies and then immediately were retested on the target melody at training showed overnight gains in CKP/B in the morning retest of the target melody--gains similar to those observed among the participants who learned only the target melody at training; this group showed no decrement in the performance of the target melody in the retest at the end of training, which indicates that there were no immediate interference effects apparent in the target melody after having learned the second melody. These results show that experienced learners performing a familiar type of task, and one that includes auditory processing demands, benefit from overnight consolidation of procedural memories. These benefits may be inhibited, however, when musicians learn similar, competing tasks in juxtaposition. / text

This study explored the relationship between working memory (WM) deficits and reading difficulties using secondary data analysis on data collected from 63 English speaking students in two urban school divisions in Saskatchewan participating in a larger SSHRC funded study (Marche, McIntyre, Claypool, 2013). First, this study addressed whether the WM profiles of individuals with reading difficulties were different from those of individuals without reading difficulties. The results showed that individuals with reading difficulties scored lower than individuals with average reading ability on measures of verbal short-term memory (STM), verbal WM, and visuospatial WM. Second, this study looked at the differential effects of computer-based WM training on the WM profiles of children with and without reading difficulties. The results showed that after WM training, there was a difference between the visuospatial STM scores of individuals with and without difficulties, when reading ability was determined by the combination of a decoding and comprehension task. Furthermore, a difference was also noted between the visuospatial WM scores of individuals with and without word decoding difficulties, and the visuospatial STM, verbal WM, and visuospatial WM scores of individuals with and without reading comprehension difficulties. Additionally, the verbal STM scores of individuals with reading comprehension difficulties were marginally different than the scores of individuals without. No differences were found between individuals who did not participate in WM training. The limitations of the study, as well as the implications for practice and future research, are discussed.

Not available / text