MA, Kent State University, 2012, College of Arts and Sciences / Department of Psychological Sciences

A wealth of research has highlighted that retrieval practice promotes subsequent memory, particularly when the retrieval attempt is successful. Recently, several papers have demonstrated that the amount of successful retrieval practice dramatically influences final test performance. For example, Vaughn and Rawson (2011) had participants learn Lithuanian-English word pairs by retrieving them from one to five times (i.e., learning to varying criterion levels) during practice. Despite retrieval practice always occurring in the forward direction during practice (A - ?), performance increased as a function of initial criterion level both on final forward (A - ?) and backward (? – B) cued recall tests. Importantly, the performance gain across criterion levels appeared asymmetric, as the gains were much larger in the forward versus backward cued recall direction. However, one potential explanation for the observed asymmetry in criterion level gains is that the materials strongly favored forward cued recall, as retrieving Lithuanian versus English is inherently more difficult for native English speakers. The present experiments utilized English-English pairs to more appropriately investigate whether criterion level gains are symmetric or asymmetric. According to the associative symmetry hypothesis (ASH), enhancing memory in a forward direction equally enhances memory in the backward direction. According to the independent association hypothesis (IAH), enhancing associative memory in a forward direction does not enhance associative memory in the backward direction. Although recent review articles support ASH, we present recall and recognition data which highlight asymmetric associative gains (supporting IAH) in the context of criterion learning.

Committee: Katherine Rawson PhD (Committee Chair); William Merriman PhD (Committee Member); John Dunlosky PhD (Committee Member); Karin Coifman PhD (Committee Member) Subjects: Cognitive Psychology

Doctor of Philosophy, The Ohio State University, 2014, Psychology

A powerful theoretical framework for exploring recognition memory is the global matching framework, in which a cue's memory strength reflects the similarity of the retrieval cues being matched against all of the contents of memory simultaneously. Contributions at retrieval can be categorized as matches and mismatches to the item and context cues, including the self match (match on item and context), item noise (match on context, mismatch on item), context noise (match on item, mismatch on context), and background noise (mismatch on item and context). I present a model that directly parameterizes the matches and mismatches to the item and context cues, which enables estimation of the magnitude of each interference contribution (item noise, context noise, and background noise). The model was fit within a hierarchical Bayesian framework to ten recognition memory datasets that employ manipulations of strength, list length, list strength, word frequency, study-test delay, and stimulus class in item and associative recognition. Estimates of the model parameters revealed at most a small contribution of item noise that varies by stimulus class, with virtually no item noise for single words and scenes. Despite the unpopularity of background noise in recognition memory models, background noise estimates dominated at retrieval across nearly all stimulus classes with the exception of high frequency words, which exhibited equivalent levels of context noise and background noise. These parameter estimates suggest that the majority of interference in recognition memory stems from experiences acquired prior to the learning episode.

Committee: Per Sederberg PhD (Advisor); Roger Ratcliff PhD (Committee Member); Jay Myung PhD (Committee Member) Subjects: Psychology

Doctor of Philosophy (PhD), Ohio University, 2006, Electrical Engineering & Computer Science (Engineering and Technology)

This dissertation is focused on the development of system level architectures and models of dynamically self-reconfigurable systems for machine intelligence. This research is significant for building brain-like intelligent systems. Although the development of deep submicron very large scale integration (VLSI) system, nanotechnology and bioinformatics facilitate building such intelligent systems, yet it is very challenging to study how these kinds of complex, reconfigurable systems can self-develop their connectivity structures, accumulate knowledge, make associations and predictions, dynamically interact with environment, and self-control to accomplish desired tasks. A new framework of “learning-memory-prediction” for machine intelligence is proposed in this research, and it serves as the foundation for building intelligent systems through learning in dynamic value systems, memorizing in self-organizing networks, and predicting in hierarchical structures. These systems are characterized by on-line data driven learning, distributed structure of processing components with local and sparse interconnections, dynamic reconfigurability, self-organization, and active interaction with environment. Learning is the fundamental element for biologically intelligent systems. The proposed online value system is able to learn and dynamically estimate the value of any multi-dimensional data set, and such value system can be used in reinforcement learning. Feedback mechanism is introduced in the self-organizing learning system to allow the machine to be able to memorize information in its distributed processing elements and make associations. After the information is learned and stored in the associative memory, a biologically-inspired anticipation-based temporal sequence learning architecture is proposed. All systems proposed in this research are hardware-oriented. A novel computing paradigm that can achieve low power consumption for designing large scale, high density intelligent (open full item for complete abstract)

Committee: Janusz Starzyk (Advisor) Subjects:

Doctor of Philosophy (PhD), Wright State University, 2023, Human Factors and Industrial/Organizational Psychology PhD

Instead of characterizing transfer from short-term memory to long-term memory as the relocation of information from one structural system to another, I propose a theory that conceives of transfer as the learning processes that act on and transform the representations of the information itself. Dynamic Associative Memory posits that recently encoded memories are supported by active maintenance and the relevance of the current context. Over time, the current context becomes less relevant; therefore, the brain must learn contextually invariant associations between memories so that they may support themselves. I instantiated my theory in the ACT-R cognitive architecture and created a new module to automate and fully integrate attentional refreshing into the architecture. The DAM module extends ACT-R's spreading activation to allow activation to be shared among related items in declarative memory. It implements a novel associative learning process based on causal inference that stochastically generates new memory traces for associations between items proportionate to the causal power of one item to predict the other. I also developed another module to provide ACT-R models with a principled method for updating temporal context, and I proposed similarity functions for quantifying the contextually invariant relatedness of hierarchical relationships and the contextually mediated relatedness of features. I ran three simulation studies, systematically manipulating cognitive load, encoding instructions, and the repetition and semantic content of the to-be-remembered items, to investigate the fitness and predictions of the new model. Recall of elaborated words was better than unelaborated words, which were recalled better than non-words. Recall of lists composed of items with less semantic content benefited more from repetition. The model failed to reproduce the benchmark cognitive load effect in immediate recall, but the effect returned in delayed recall, suggesting that (open full item for complete abstract)

Committee: Ion Juvina Ph.D. (Committee Chair); Joseph Houpt Ph.D. (Committee Member); Glenn Gunzelmann Ph.D. (Committee Member); Valerie Shalin Ph.D. (Committee Member); Herbert Colle Ph.D. (Committee Member) Subjects: Cognitive Psychology

Master of Science in Computer Engineering, University of Dayton, 2020, Electrical and Computer Engineering

Implementing cognitive algorithms on robots is one potential direction to realize autonomous artificial agents. There is an effort to push robotics and artificial intelligence into many aspects of daily life. An important step in this process is leveraging concepts known to work from human cognitive features on computer systems to improve the performance of robotic systems. Spiking Neural Networks (SNNs) allow these computational models to be instantiated in a low size, weight, and power (SWaP) form factor due to the biological efficiencies they approximate. This paper shows an associative memory in the form of an SNN, an application of the associative memory, and some performance benchmarking. The model is created using a neural network simulator and run on a low SWaP CPU and Intel's neuromorphic processor, Loihi, an artificial intelligence accelerator highly optimized for spiking neural algorithms. In addition, the model is employed on a mobile robotic platform that explores the real world and uses online learning to make associations. When the model was run on Loihi the overall power usage decreased as well as the run time of the simulation as compared to the low SWaP CPU proving beneficial to implement the neuromorphic hardware.

Committee: Tarek Taha Dr. (Advisor); Eric Balster Dr. (Committee Member); Trevor Bhil Dr. (Committee Member) Subjects: Computer Engineering; Neurobiology

Master of Science, Miami University, 2020, Biology

The ability to learn about the surrounding environment is advantageous for many arthropods when searching for mates, avoiding predators, or foraging. It has been demonstrated that arachnids are capable of both simple and complex forms of learning within these situations. Simple forms of learning, such as classical conditioning have been well explored in arachnids, but studies on more complex learning, such as contextual associations, are still needed. Here I present data on learning by the wolf spider Pardosa milvina (Araneae, Lycosidae) in connection with foraging. Spiders underwent training to learn either a simple or complex association. During training subjects were exposed to either unpalatable (quinine-coated), palatable (sucrose-coated), or neutral (water-coated) prey within either a maple or peppermint scented environment. After training, P. milvina was tested in a novel two-choice maze where individuals selected to travel either towards peppermint or maple cues to investigate potential preferences resulting from learning. Some significant effects of treatment and time on spider attack behaviors were found during the training period, though these factors yielded no significant effects on behavior during maze testing. Overall, some behavioral trends of complex groups during training and simple groups during testing align with expected results of learning these association types.

Committee: Ann Rypstra (Advisor); Alan Cady (Committee Member); Nancy Solomon (Committee Member) Subjects: Animals; Behavioral Sciences; Biology

Doctor of Philosophy (PhD), Ohio University, 2019, Electrical Engineering & Computer Science (Engineering and Technology)

Artificial intelligence (AI) is being widely applied to various practical problems, and researchers are working to address numerous issues facing the field. The organizational structure and learning mechanism of the memory is one such issue. A cognitive agent builds a representation of its environment and remembers its experiences to interpret its inputs and implements its goals through its actions. By doing so it demonstrates its intelligence (if any), and it is its learning mechanism, value system and sensory motor coordination that makes all this possible. Memory in a cognitive agent stores its knowledge, knowledge gained over a life-time of experiences in a specific environment. That is, memory includes the “facts”, the relationships between them, and the mechanism used to learn, recognize, and recall based on the agent's interaction with the world/environment. It remembers events that the agent experienced reflecting important actions and observations. It motivates the agent to do anything by providing assessment of the state of the environment and its own state. It allows it to plan and anticipate. And finally, it allows the agent to reflect on itself as an independent being. Hence, memory is critical for intelligence, for it is the memory that determines a cognitive agent's abilities and learning skills. Research has shown that while memory in humans can be classified into different types, based on factors such as their longevity and cognitive mechanisms used to create and retrieve them, they all are achieved using a similar underlying structure. The focus of this dissertation was on using this principle, i.e. different memories created using the same underlying structure, to implement memory for cognitive agents using a biologically plausible model of neuron. This work was an attempt to demonstrate the feasibility of implementing self-organizing memory structures capable of performing the various memory related tasks necessary for a cognitive agent using a c (open full item for complete abstract)

Committee: Wojciech Jadwisienczak (Advisor) Subjects: Electrical Engineering

PHD, Kent State University, 2015, College of Arts and Sciences / Department of Biological Sciences

Psychological illnesses such as anxiety disorder, depression, and posttraumatic stress disorder share a common element: patients often engage in rumination, which is characterized by intrusive thoughts and persistent memory of negative events. Indeed, a hallmark of the stress response is the enhanced capability for learning in emotionally salient situations, especially those involving aversive stimuli. A rodent model of chronic mild stress (CMS) was used in this series of experiments to examine effects on associative learning and behavior, mediated through stimulation of central beta-adrenergic receptors (β-ARs). Animals exposed to CMS develop behaviors associated with depression (e.g. decreased exploration, social withdrawal) and fear (freezing in a pertinent context) when observed in a place with environmental cues associated with the stress paradigm; when observed in a place without such cues, no such manifestation occurs. Interestingly, we also found that prolonged exposure to stress decreases behavioral inhibition, as stressed rodents placed in a passive avoidance apparatus show decreased retention latencies 24h post training, relative to home cage controls. We also investigate the role of β-ARs, where we found that administration of the beta blocker propranolol attenuates or blocks the development of these depressive behaviors and also prevents the stress-enhanced learning shown in a contextual fear conditioning task; however, β-AR involvement is more complex in passive avoidance. A final study suggests that chronic exposure to stress may sensitize β-adrenergic receptor signaling, as we observed an increase in amygdaloid adenylate cyclase activity in stressed subjects, which was blocked with the use of propranolol; however, further results were inconclusive. Overall, we conclude that stress-induced anxiety and depressive behaviors initially develop as a consequence of the enhanced learning that is a part of the stress response, and that the effect is mediated (open full item for complete abstract)

Committee: John Johnson PhD (Advisor); Heather Caldwell PhD (Committee Member); Eric Mintz PhD (Committee Member); Aaron Jasnow PhD (Committee Member); David Riccio PhD (Other) Subjects: Biology; Neurobiology; Neurosciences; Physiological Psychology; Physiology

PhD, University of Cincinnati, 2009, Arts and Sciences : Psychology

Children with traumatic brain injury (TBI) often experience memory deficits, although the nature of these deficits, their functional outcome, and possible recovery are not well understood. The present fMRI study examined the neural activation patterns in a group of young children who sustained mild to moderate TBI in early childhood (n = 8), and a group of healthy control children (n = 14) during a verbal paired associate learning task that promoted the use of two mnemonic strategies differing in efficacy. The children with TBI demonstrated intact memory performance and were able to successfully utilize the mnemonic strategies. However, the TBI group also demonstrated altered brain activation patterns during the task compared to the control children. These findings suggest that early childhood TBI may alter activation within the network of brain regions that support associative memory even in children who show good behavioral performance, and these changes likely persist for years after the injury.

Committee: C.Y. Peter Chiu PhD (Committee Chair); Paula K. Shear PhD (Committee Member); Shari L. Wade PhD (Committee Member); M. Douglas Ris PhD (Committee Member) Subjects: Psychology

Doctor of Philosophy, Miami University, 2008, Psychology

Rabbit eyeblink classical conditioning (EBCC) is a task widely used to understand the neurobiological correlates of associative learning and memory. The hippocampus has proven to be an important structure in acquiring the association between the conditioning stimuli and the development of conditioned responses. This study used a brain-computer interface to trigger conditioning trials in the presence (T+) or absence (T-) of a frequency component of the hippocampal field potential historically related to sensory processing and attention, termed theta. The presence of theta has been shown to facilitate learning in this task as well as accelerate learning related unit responses in the hippocampus, but the precise nature of its beneficial effect in hippocampal neurophysiology has yet to be determined. In this study, tetrodes were lowered into the dorsal hippocampus of four groups of rabbits including those who received paired or unpaired conditioning stimuli in both T+ and T- theta conditions. Specialized signal processing software compared the extracted data streams from each wire of the tetrode to separate the waveforms into single neuron responses. Each sorted neuron was then categorized according to its firing properties as pyramidal cells or one of the known types of interneurons that exist in the hippocampus. The individual units were analyzed for their relation to the pretrial period as well as for how they responded to the conditioning stimuli. Interneurons were highly correlated with hippocampal state used to trigger the trials, supporting the existence of interneurons that systematically vary with the ongoing theta activity. Interneuron responses to the conditioning stimuli were generally dependent on hippocampal state, demonstrating excitatory responses in T+ groups and suppression in T- groups. Pyramidal cells that demonstrated suppression to the conditioning stimuli were also more common in T- groups, while excitatory pyramidal cells were more related to ass (open full item for complete abstract)

Committee: Stephen Berry PhD (Advisor); Allan Pantle PhD (Committee Member); Dragana Ivkovich Claflin PhD (Committee Member); Kathleen Killian PhD (Committee Member) Subjects: Behaviorial Sciences; Physiological Psychology; Psychobiology; Psychology