A-53 Exploring the Relationship between Spoken Language and Verbal Short-term Memory Assessment Tasks

Abstract Objective Examine whether performance on spoken language assessment measures may be associated with performance at different phases of verbal learning and recall tasks. Method The assessment records of 222 American Veterans with diverse neuropsychiatric conditions were analyzed using Exploratory Factor Analyses. There were no exclusion criteria. All participants completed the Visual Naming (VisNam), Sentence Repetition (SenRep), Controlled Word Association (COWA), and Token Tests of the Multilingual Aphasia Examination (MAE), and Benton Serial Digit Learning Test – 8 Digits (SDL8). Individual assessment instruments were factored using Principal Component Analyses (PCA). A three-factor solution of the SDL-8 was co-factored with the spoken language components of the MAE to identify common sources of variance. Results A three-factor solution of the SDL8 separated trials into three overlapping factors consisting of early (SDL8_Early), middle (SDL8_Middle), and late (SDL8_Late) trials. Co-factoring the three new scales with the verbal components of the MAE produced a five-factor model explaining 84.563% of the shared variance: 1) SDL8_Early loaded with SenRep, 2) SDL8_Middle loaded with SenRep, 3) SDL8_Late loaded with Token, 4) SDL8_Late loaded with COWA, and 5) VisNam alone formed the fifth factor. Conclusions The results suggest that rote repetition is largely associated with early trials and slightly associated with middle trials, while late trials are largely associated with auditory comprehension and slightly associated with verbal fluency. This may be indicative of a shift in use of spoken language abilities to accommodate increasing levels of complexity in presented verbal short-term memory tasks and thus reflective of a change on learning strategy to optimize performance.

Download Full-text

A-65 Verbal Abilities May Influence Performance in Assessment of Verbal Short-Term Memory Tasks

Archives of Clinical Neuropsychology ◽

10.1093/arclin/acz034.65 ◽

2019 ◽

Vol 34 (6) ◽

pp. 925-925

Author(s):

A Guerra ◽

J Moses ◽

J Rivera ◽

M Davis ◽

K Hakinson

Keyword(s):

Verbal Memory ◽

Verbal Fluency ◽

Short Term Memory ◽

Word Association ◽

Memory Task ◽

Short Term ◽

Factor Solution ◽

Term Memory ◽

Verbal Abilities ◽

Three Factor

Abstract Objective Examine whether verbal abilities may help explain the learning strategies people employ when completing a short-term verbal memory task. Methods The assessment records of 296 American Veterans with diverse neuropsychiatric conditions were analyzed using Exploratory Factor Analyses. There were no exclusion criteria. All participants completed the Benton Serial Digit Learning Test – 9 Digits (SDL-9) and Visual Naming (VisNam), Sentence Repetition (SenRep), Controlled Word Association (COWA), and Token Tests of the Multilingual Aphasia Examination (MAE). Individual assessment instruments were factored using Principal Component Analyses (PCA). A three-factor solution of the SDL-9 was co-factored with the verbal components of the MAE to identify common sources of variance. Results A three-factor solution of the SDL-9 separated trials into three overlapping factors consisting of early (SDL-9_Early), middle (SDL-9_Middle), and late (SDL-9_Late) trials. Co-factoring the three new scales with the verbal components of the MAE produced a four-factor model explaining 67.85% of the shared variance: 1) SenRep loaded with SDL-9_Early, 2) COWAT loaded with SDL-9_Middle and SDL-9_Late, 3) Token loaded with SDL-9_Late, and 4) Vis Nam loaded with SDL-9_Late. Conclusions The results suggest that individuals may engage verbal abilities differently as they progress from simpler to more difficult verbal short-term memory tasks. It appears performance in early trials is mostly associated with rote repetition and performance on middle trials is mostly associated with verbal fluency, while performance on the late trials is associated with a combination of verbal fluency, auditory comprehension, and conceptual organization/naming. This may therefore indicate a shift in learning strategy to meet increased cognitive demands.

Download Full-text

Cognitive impairment associated with drug use in old age people

European Psychiatry ◽

10.1016/j.eurpsy.2017.01.1765 ◽

2017 ◽

Vol 41 (S1) ◽

pp. s876-s876

Author(s):

E. Ros-Cucurull ◽

C. Cardona-Rubira ◽

E. García-Raboso ◽

R.F. Palma-Álvarez ◽

L. Grau-López ◽

...

Keyword(s):

Cognitive Impairment ◽

Drug Use ◽

Short Term Memory ◽

Verbal Learning ◽

Drug Misuse ◽

Verbal Information ◽

Short Term ◽

Term Memory ◽

Old Adults ◽

Visual Construction

IntroductionSubstance use disorder is a growing phenomenon among old adults. It is usually significantly undervalued, misidentified, under diagnosed and poorly treated. It has been related to cognitive impairment but there are few studies focused on the elderly.AimTo evaluate the relationship between drug use and cognitive impairment in old adults.MethodsWe conducted a prospective study (basal and 6 month follow up) in 67 patients over 65 years old seeking for treatment for drug misuse (alcohol and prescription drugs, mainly benzodiacepines) in addiction and dual diagnosis unit in Barcelona. A specific protocol was performed to evaluate attention, executive function, working memory, learning capacity, fonetic and visual fluency, decision-making, visual construction and cognitive flexibility (FCT, CPT-II, N-BACK, COWAT FAS, TAP, SDMT, IGT, CVLT, TOL, RFFT, STROOP). Patients were compared with a control group (healthy non drug users) with same characteristics (gender, age range and education status). The protocol consisted in two separated sessions of 90 minutes each one performed by a neuropsychologist.ResultsResults obtained suggested that patients under drug misuse had worse scores in fluency, visual construction, memory and attention compared with controls. After 6 month treatment and achieving abstinence patients improve in cognitive skills as verbal learning, short-term memory and free recall of verbal information. Cognitive impairment profile changes depending on the substance abused (alcohol or benzodiacepines).ConclusionsDrug use can produce deleterious effects in old adults. However, those who achieve abstinence may improve some cognitive functioning as verbal learning, short-term memory and free recall of verbal information.Disclosure of interestThe authors have not supplied their declaration of competing interest.

Download Full-text

Video-Based Sentiment Analysis with hvnLBP-TOP Feature and bi-LSTM

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v33i01.33019963 ◽

2019 ◽

Vol 33 ◽

pp. 9963-9964 ◽

Cited By ~ 3

Author(s):

Haoran Li ◽

Hua Xu

Keyword(s):

Sentiment Analysis ◽

Extraction Method ◽

Short Term Memory ◽

Recognition Accuracy ◽

Principal Component ◽

Short Term ◽

Term Memory ◽

Feature Extraction Method ◽

Long Short Term Memory ◽

New Feature

In this paper, we propose a new feature extraction method called hvnLBP-TOP for video-based sentiment analysis. Furthermore, we use principal component analysis (PCA) and bidirectional long short term memory (bi-LSTM) for dimensionality reduction and classification. We achieved an average recognition accuracy of 71.1% on the MOUD dataset and 63.9% on the CMU-MOSI dataset.

Download Full-text

Prediction of Wind Turbine-Grid Interaction Based on a Principal Component Analysis-Long Short Term Memory Model

Energies ◽

10.3390/en11113221 ◽

2018 ◽

Vol 11 (11) ◽

pp. 3221 ◽

Cited By ~ 9

Author(s):

Yining Wang ◽

Da Xie ◽

Xitian Wang ◽

Yu Zhang

Keyword(s):

Principal Component Analysis ◽

Wind Turbine ◽

Short Term Memory ◽

Principal Component ◽

Wind Farms ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory ◽

Lstm Network ◽

Oscillation Frequencies

The interaction between the gird and wind farms has significant impact on the power grid, therefore prediction of the interaction between gird and wind farms is of great significance. In this paper, a wind turbine-gird interaction prediction model based on long short term memory (LSTM) network under the TensorFlow framework is presented. First, the multivariate time series was screened by principal component analysis (PCA) to reduce the data dimensionality. Secondly, the LSTM network is used to model the nonlinear relationship between the selected sequence of wind turbine network interactions and the actual output sequence of the wind farms, it is proved that it has higher accuracy and applicability by comparison with single LSTM model, Autoregressive Integrated Moving Average (ARIMA) model and Back Propagation Neural Network (BPNN) model, the Mean Absolute Percentage Error (MAPE) is 0.617%, 0.703%, 1.397% and 3.127%, respectively. Finally, the Prony algorithm was used to analyze the predicted data of the wind turbine-grid interactions. Based on the actual data, it is found that the oscillation frequencies of the predicted data from PCA-LSTM model are basically the same as the oscillation frequencies of the actual data, thus the feasibility of the model proposed for analyzing interaction between grid and wind turbines is verified.

Download Full-text

Stock Price Prediction Based on Principal Component Analysis and Long Short-Term Memory Network

Advances in Applied Mathematics ◽

10.12677/aam.2020.911225 ◽

2020 ◽

Vol 09 (11) ◽

pp. 1954-1960

Author(s):

甲刘

Keyword(s):

Principal Component Analysis ◽

Stock Price ◽

Short Term Memory ◽

Principal Component ◽

Short Term ◽

Term Memory ◽

Stock Price Prediction ◽

Price Prediction ◽

Memory Network ◽

Long Short Term Memory

Download Full-text

Remaining Useful Life Prediction of Airplane Engine Based on PCA–BLSTM

Sensors ◽

10.3390/s20164537 ◽

2020 ◽

Vol 20 (16) ◽

pp. 4537

Author(s):

Shixin Ji ◽

Xuehao Han ◽

Yichun Hou ◽

Yong Song ◽

Qingfu Du

Keyword(s):

Principal Component Analysis ◽

Hybrid Model ◽

Short Term Memory ◽

Principal Component ◽

Remaining Useful Life ◽

Short Term ◽

Term Memory ◽

Airplane Engine ◽

Useful Life ◽

Long Short Term Memory

The accurate prediction of airplane engine failure can provide a reasonable decision basis for airplane engine maintenance, effectively reducing maintenance costs and reducing the incidence of failure. According to the characteristics of the monitoring data of airplane engine sensors, this work proposed a remaining useful life (RUL) prediction model based on principal component analysis and bidirectional long short-term memory. Principal component analysis is used for feature extraction to remove useless information and noise. After this, bidirectional long short-term memory is used to learn the relationship between the state monitoring data and remaining useful life. This work includes data preprocessing, the construction of a hybrid model, the use of the NASA’s Commercial Aerodynamic System Simulation (C-MAPSS) data set for training and testing, and the comparison of results with those of support vector regression, long short-term memory and bidirectional long short-term memory models. The hybrid model shows better prediction accuracy and performance, which can provide a basis for formulating a reasonable airplane engine health management plan.

Download Full-text

Artificial Neural Networks in Motion Analysis—Applications of Unsupervised and Heuristic Feature Selection Techniques

Sensors ◽

10.3390/s20164581 ◽

2020 ◽

Vol 20 (16) ◽

pp. 4581

Author(s):

Marion Mundt ◽

Arnd Koeppe ◽

Franz Bamer ◽

Sina David ◽

Bernd Markert

Keyword(s):

Neural Networks ◽

Principal Component Analysis ◽

Motion Analysis ◽

Lower Limb ◽

Short Term Memory ◽

Principal Component ◽

Joint Angles ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory

The use of machine learning to estimate joint angles from inertial sensors is a promising approach to in-field motion analysis. In this context, the simplification of the measurements by using a small number of sensors is of great interest. Neural networks have the opportunity to estimate joint angles from a sparse dataset, which enables the reduction of sensors necessary for the determination of all three-dimensional lower limb joint angles. Additionally, the dimensions of the problem can be simplified using principal component analysis. Training a long short-term memory neural network on the prediction of 3D lower limb joint angles based on inertial data showed that three sensors placed on the pelvis and both shanks are sufficient. The application of principal component analysis to the data of five sensors did not reveal improved results. The use of longer motion sequences compared to time-normalised gait cycles seems to be advantageous for the prediction accuracy, which bridges the gap to real-time applications of long short-term memory neural networks in the future.

Download Full-text