Добірка наукової літератури з теми "Associative recognition memory; state-trace analysis"

Studies have shown associative-memory decline in aging. While the literature is inconclusive regarding the source of the deficit, some researchers argue that it is caused by impaired encoding and maintenance processes in working-memory (WM). Successful retrieval of a stimulus depends on its sequential presentation in the learning list: stimuli at the beginning or the end of the learning list benefit from higher retrieval probability. These effects are known as “primacy” and “recency” effects, respectively. In the case of the primacy-effect, stimuli at early list positions benefit from extensive rehearsal that results in enhanced consolidation and trace in long-term memory (LTM). In the case of the recency-effect, target stimuli at later serial positions are still maintained in WM and can therefore be effortlessly retrieved. Considering these effects could shed light on the involvement of WM in associative-binding. Both behavioral and neuroimaging researchers have studied associative-decline in aging. However, no work has explicitly tested age differences in memory for items versus associations as a function of stimuli serial position (SSP). In the current study, 22 younger and 22 older adults were recruited to participate in a study aimed to test the separate and joint effects of both SSP and aging on memory-recognition of items and associations. In the task used, retrieval was manipulated for SSP (beginning/middle/end of the list) and item/associations recognition modes. We hypothesized that greater associative-decline will be observed in older adults, specifically for recently presented material. The results showed that both groups presented a significant associative-deficit at the recency positions; this decrease was additive and did not correspond to the expected interaction effect. Further analysis showed that the source of associative-memory decline for stimuli at recency position in older adults resulted from an increase in false-alarm (FA) rates. These results support the involvement of WM-binding impairment in aging.

Testing the state of associative memory using the paired associates method revealed a significant decrease in the effectiveness of associative memorization in veterans of the Anti-Terrorist Operation (ATO) / Joint Forces Operation (JFO) with traumatic brain injuries, as evidenced by a larger number of errors made by them and a longer duration of their passing the test compared to the subjects of the control group. In the examined control group, the formation of associative memory was provided by a consistent system of top-down and bottom-up control of memorizing associative pairs of words and forming their images in both right and left hemispheres, while in veterans of the ATO/JFO with traumatic brain injuries in the formation of associative memory was more involved the left hemisphere than the right one, which indicated the overwhelming contribution of the semantic analysis of words to the formation of a connection between them. After the presentation of the first word of an associative pair, the retrieval of the second word in the control group provided a balanced neural network in the right and left hemispheres. At the same time, increased activity in the left hemisphere and the corresponding neural network in the low-frequency range were found in the group with traumatic brain injuries, which indicated a greater involvement of verbal semantic analysis of words when reproducing the second word. At the same time, the lack of synchronization in the beta-range in the central-parietal zone may indicate a deterioration of the postretrieval processes that ensure the processes of inhibition after recalling the corresponding words. Thus, in the control group, when memorizing associative pairs of words, coordinated processes of perception of individual words, their recognition, semantic analysis and the formation of semantic connections between them and their connections with the corresponding images were found, which is a reflection of coordinated inter-hemispheric information processes while subjects with traumatic brain injuries predominantly relied on the phonetic and semantic aspects of associative word pairs.

Convolutional neural networks (CNNs) have powerful representation learning capabilities by automatically learning and extracting features directly from inputs. In classification applications, CNN models are typically composed of: convolutional layers, pooling layers, and fully connected (FC) layer(s). In a chain-based deep neural network, the FC layers contain most of the parameters of the network, which affects memory occupancy and computational complexity. For many real-world problems, speeding up inference time is an important matter because of the hardware design implications. To deal with this problem, we propose the replacement of the FC layers with a Hopfield neural network (HNN). The proposed architecture combines both a CNN and an HNN: A pretrained CNN model is used for feature extraction, followed by an HNN, which is considered as an associative memory that saves all features created by the CNN. Then, to deal with the limitation of the storage capacity of the HNN, the proposed work uses multiple HNNs. To optimize this step, the knapsack problem formulation is proposed, and a genetic algorithm (GA) is used solve it. According to the results obtained on the Noisy MNIST Dataset, our work outperformed the state-of-the-art algorithms.

The integrity of the hippocampus and surrounding medial-temporal cortices is critical for episodic memory, with the hippocampus being posited to support relational or configural associative learning. The present event-related functional magnetic resonance imaging (fMRI) study investigated the role of specific medial-temporal lobe structures in learning during relational and item-based processing, as well as the extent to which these structures are engaged during item-based maintenance of stimuli in working memory. fMRI indexed involvement of the hippocampus and underlying cortical regions during performance of two verbal encoding conditions, one that required item-based maintenance of word triplets in working memory and the other that entailed the formation of inter-item associations across the words in each triplet. Sixteen subjects were scanned using a rapid event-related fMRI design while they encountered the item-based and relational processing trials. To examine the correlation between fMRI signal in medial-temporal structures during learning and the subject's subsequent ability to remember the stimuli (a measure of effective memory formation), subjects were administered a yes-no recognition memory test following completion of the encoding scans. Results revealed that the hippocampus proper was engaged during both relational and item-based processing, with relational processing resulting in a greater hippocampal response. By contrast, entorhinal and parahippocampal gyri were differentially engaged during item-based processing, providing strong evidence for a functional neuroanatomic distinction between hippocampal and parahippocampal structures. Analysis of the neural correlates of subsequent memory revealed that activation in the bilateral hippocampus was reliably correlated with behavioral measures of effective memory formation only for those stimuli that were encoded in a relational manner. Taken together, these data provide evidence that the hippocampus, while engaged during item-based working memory maintenance, differentially subserves the relational binding of items into an integrated memory trace so that the experience can be later remembered.

In this paper, a tracking and capturing method for abnormal behavior in marketing system is designed. The main steps of the method include: (1) Trace generation based on function injection. This method selects NOP, HLT and other instructions of untrusted processes to fill the memory area, and injects the test. The function and control process execute the function to obtain Trace; (2) Trace & capture-based sandbox interception system function analysis method, first establish a finite state automaton model describing the instruction address translation of the untrusted process call system function, according to the state of the automaton. The conversion determines whether the system function calling process has interception behavior, and further identifies the system function of the sandbox interception according to the position of the state machine conversion instruction in the automatic machine. This paper designs and implements the prototype system, called TC-analyser, and solves the key problems of the function injection timing selection, Trace intermediate representation and other implementation processes. Finally, choose Chromium and Adobe Reader to test the TC-analyser’s interception recognition capability and compare it with Hooks hark. The experimental results show that the TC-analyser has good interception and recognition capabilities.

Emotion is known to influence multiple aspects of memory formation, including the initial encoding of the memory trace and its consolidation over time. However, the neural mechanisms whereby emotion impacts memory encoding remain largely unexplored. The present study used a levels-of-processing manipulation to characterize the impact of emotion on encoding with and without the influence of elaborative processes. Participants viewed emotionally negative, neutral, and positive scenes under two conditions: a shallow condition focused on the perceptual features of the scenes and a deep condition that queried their semantic meaning. Recognition memory was tested 2 days later. Results showed that emotional memory enhancements were greatest in the shallow condition. fMRI analyses revealed that the right amygdala predicted subsequent emotional memory in the shallow more than deep condition, whereas the right ventrolateral PFC demonstrated the reverse pattern. Furthermore, the association of these regions with the hippocampus was modulated by valence: the amygdala–hippocampal link was strongest for negative stimuli, whereas the prefrontal–hippocampal link was strongest for positive stimuli. Taken together, these results suggest two distinct activation patterns underlying emotional memory formation: an amygdala component that promotes memory during shallow encoding, especially for negative information, and a prefrontal component that provides extra benefits during deep encoding, especially for positive information.

The main purpose of this paper is to focus on the design of Viterbi Decoder (VD) with low power, which is significant for receiver section of data communication applications such as Radar, Satellite, Telephone and Automatic speech recognition. The Viterbi decoder algorithm consists of three most important blocks – Branch Metric Unit (BMU), Add Compare and Select (ACS) Unit and Survivor Memory Unit (SMU). BMU computes the metrics between the input and output state transitions. ACS unit include the Path Metric Unit (PMU), which computes the metrics with the sequence to a next state of a path and selects the lower metric value as a survivor path. SMU stores the data bits which utilizes the trace back method to fetch the likelihood path from the current state to a previous state. An ACS unit is an essential block for VD. The basic recursive ACSU design consists of Ripple Carry Adder (RCA), Comparator and a Selector block, which consume more area, power and operates with high junction temperature. To overcome these drawbacks, a modified ACSU design is implemented with recursive cancellation technique. ACS unit is modified by including a trace back mechanism to obtain a low latency and high speed in VD. It is designed with low complexity multiplexers, adders, logical AND gate and comparator block. This breaking recursive ACSU design utilizes less power, high throughput, low latency and also operates at low temperature. This analysis and simulation process are accomplished using Vivado Design Suite.

Background and Purpose. Recent studies suggest that abnormal structure and function in the brain network were related to cognitive and emotional impairment in hyperthyroid patients (HPs). The association between altered voxel-mirrored homotopic connectivity (VMHC) and neuropsychological impairment in HPs remains unclear. This study is aimed at investigating the association between the disrupted functional coordination and psychological dysfunction in hyperthyroidism. Method. Thirty-three hyperthyroid patients and thirty-three matched healthy controls (HCs) were recruited, and they received resting-state functional magnetic resonance imaging (fMRI) scans and neuropsychological evaluation. The VMHC value was computed to reveal the functional coordination between homotopic regions in both groups. The neurobehavioral relevancy method was employed to explore the relationship between the altered VMHC and emotional, cognition measures. Further receiver operating characteristic (ROC) curve analysis was adopted to examine the power of changed regional VMHC in classifying the patients with hyperthyroidism. Results. Compared with the HCs, the HPs exhibited significantly declined VMHC values in the bilateral medial frontal gyrus (MeFG). The interhemispheric asynchrony in the MeFG was positively correlated with Z scores of episodic memory. The ROC analysis further determined that abnormal VMHC in the MeFG could efficiently distinguish the HPs from the HCs (area under the curve AUC=0.808, P<0.001). Conclusion. The altered interhemispheric coordination in the hub of the default mode network may implicated in the modulation of episodic memory in HPs patients and the distinct feature of the interhemispheric asynchrony may be treated as a potential target for the early recognition and intervention for the HPs with cognitive impairments. Clinical Trial Registration. This is a study of the neurological basis for dysfunction of mood and cognition in hyperthyroid patients: a resting-state fMRI study (registration number: ChiCTR-OOC-16008607).

The aim of this study is to explore the pattern of change in multiple measures of cognitive abilities in a sample of oldest-old adults, comparing two different time metrics (chronological age and time to death) and therefore examining both underlying conceptual assumptions (age-related change and terminal decline). Moreover, the association with individual characteristics as sex, education, and dementia diagnosis was also examined. Measures of cognitive status (Mini-Mental State Examination and the Swedish Clock Test) and tests of crystallized (knowledge and synonyms), memory (verbal memory, nonverbal long-term memory, recognition and correspondence, and short-term memory), and visuospatial ability were included. The sample consisted of 671 older Swedish adult participants of the OCTO Twin Study. Linear mixed models with random coefficients were used to analyse change patterns and BIC indexes were used to compare models. Results showed that the time to death model was the best option in analyses of change in all the cognitive measures considered (except for the Information Test). A significant cognitive decline over time was found for all variables. Individuals diagnosed with dementia had lower scores at the study entrance and a faster decline. More educated individuals performed better in all the measures of cognition at study entry than those with poorer education, but no differences were found in the rate of change. Differences were found in age, sex, or time to death at baseline across the different measures. These results support the terminal decline hypothesis when compared to models assuming that cognitive changes are driven by normative aging processes.

The aim of this thesis is to investigate competing explanations of the processes underlying associative recognition. Like recognition memory for individual items, associative recognition is currently understood through two different classes of model. The first is the single-process model class which holds that associative recognition decisions are based on a continuum of associative memory strength. The second is the dual-process model class, which holds that associative recognition decisions are based on two sources of information, called familiarity and recollection. Familiarity is conceptualised as a fast-acting, context-free „feeling of knowing‟, while recollection is said to be a slower, more conscious process allowing for the recall of detail and context. Familiarity may play a role in associative recognition through a mechanism called unitisation, whereby two distinct stimuli are bound into a single individual memory trace. State-trace analysis is a method to determine the number of latent variables or processes that contribute to performance on a set of tasks, under mild assumptions. A critical diagnostic feature is the dimensionality of the state-trace plot – a plot of performance on one dependent variable against the other. If associative recognition depends on a single latent variable then manipulation of experimental factors affecting memory should result in a unidimensional state-trace plot. If associative recognition depends on two or more latent variables which are differentially affected by the experimental factors then a bidimensional state-trace will result. State-trace analysis therefore provides a method of discriminating a class of single-process models from a class of dual-process models. State-trace analysis was applied to associative recognition in four experiments. Each experiment utilised two independent variables that previous research had suggested could differentially affect familiarity and recollection. Experiment 1 investigated associative recognition of word pairs by manipulating attention and study presentation frequency. Experiments 2 investigated associative recognition of word pairs under conditions designed to encourage unitisation by pairing an encoding-based unitisation manipulation with a working memory load manipulation. Experiment 3 manipulated the same unitisation instructions as well as varying study time. Experiment 4 examined the effect of unitisation using pairs of faces and manipulated visual similarity and study time. State-trace analysis of the four experiments consistently revealed unidimensional state-trace plots. Using a recently developed monotonic regression statistical test, unidimensionality could not be rejected at either aggregate or individual participant level. Therefore, no evidence was found for the differential activation of familiarity and recollection in associative recognition. The results of this thesis are therefore consistent with a single-process account of associative recognition. These results also pose a challenge to dual-process models to identify alternate experimental manipulations that reveal the involvement of different component processes such as recollection and unitized familiarity.
Thesis (Ph.D.) -- University of Adelaide, School of Psychology, 2014

Research Doctorate - Doctor of Philosophy (PhD)
Research in cognitive science and neuroscience has the shared goal of understanding how cognitive and neural representations and processes mediate the observed relationships between stimuli and responses in different experimental paradigms. The almost ubiquitous basis for inferences about the number of processes or latent dimensions involved, is the observation of a dissociation; an interaction due to an unequal or opposite effect of one independent variable on the levels of another independent variable. However, it has been clearly shown that dissociations do not provide strong evidence for the need of an extra dimension. In this thesis, which is a collection of published and submitted papers, we describe an extension of the dissociation methodology – state-trace analysis (Bamber, 1979) – that does provide a rigorous basis for this inference. In the first section, an informal introduction to state-trace analysis is provided. We also develop Bayesian methods suitable for quantifying state-trace evidence in favour of a one-dimensional or multi-dimensional explanation, as well as for refining state-trace experiments. In the second section of this thesis, an application of state-trace analysis is presented that examines the question of whether human face recognition is special in the sense that faces can be encoded in terms of a dimension or dimensions additional to those available to most other objects. Over a series of experiments, and using the new methods developed in Section One, we confirm that the encoding of unfamiliar faces is special and discuss the need to extend this type of analysis to other psychological phenomena.