and Chinesespeaking adults (n = 32) solved single-digit multiplication problems. In one condition, problems were presented as visual digits (e.g., 8 x 9). In the other condition, problems were presented as auditory number words in the participant's first language (e.g., /eit/ /taimz//nain/). Chinese-speaking adults made proportionately more operand-intrusion errors (e.g., 4 x 8 = 24) than English-speaking adults. Both groups made more operand-intrusion errors with auditoi T than with visual presentation. These findings are similar to those found when participants solve problems presented as visual number words (e.g., eight x nine), suggesting that in both cases the activation of phonological codes interferes with processing.

Accurate runoff prediction can provide a reliable decision-making basis for flood and drought disaster prevention and scientific allocation of water resources. Selecting appropriate predictors is an effective way to improve the accuracy of runoff prediction. However, the runoff process is influenced by numerous local and global hydrometeorological factors, and there is still no universal approach about the selection of suitable predictors from these factors. To address this problem, we proposed a runoff prediction model by combining machine learning (ML) and feature importance analysis (FIA-ML). Specifically, take the monthly runoff prediction of Yingluoxia, China as an example, the FIA-ML model uses mutual information (MI) and feature importance ranking method based on random forest (RF) to screen suitable predictors, from 130 global climate factors and several local hydrometeorological information, as the input of ML models, namely the hybrid kernel support vector machine (HKSVM), extreme learning machine (ELM), generalized regression neural network (GRNN), and multiple linear regression (MLR). An improved particle swarm optimization (IPSO) is used to estimate model parameters of ML. The results indicated that the performance of the FIA-ML is better than widely-used long short-term memory neural network (LSTM) and seasonal autoregressive integrated moving average (SARIMA). Particularly, the Nash-Sutcliffe Efficiency coefficients of the FIA-ML models with HKSVM and ELM were both greater than 0.9. More importantly, the FIA-ML models can explicitly explain which physical factors have significant impacts on runoff, thus strengthening the physical meaning of the runoff prediction model.

Normalized difference vegetation index(NDVI) is the most commonly used factor to re ect vegetation growth status, and improving the prediction accuracy of NDVI is of great signi cance to the development of regional ecology. In this study, a new NDVI forecasting model based on the combination of time series decomposition(TSD), convolutional neural network (CNN) and long short-term memory (LSTM) was proposed. In order to verify the performance of TSD-CNN-LSTM model and explore the response of NDVI to climatic factors, two forecasting models of temperature and precipitation based on its own historical information and four NDVI forecasting models that based on Temperature, precipitation and its own historical information were established. The results show that TSD-CNN-LSTM model based on its own historical information had the best forecasting performance, with the RMSE, NSE, MAE to be 0.4996, 0.9981, 0.4169 for temperature, 5.6941, 0.9822, 3.9855 for precipitation and 0.0573, 0.9617, 0.0447 for NDVI, respectively. Meanwhile, the NDVI forecasting models based on climatic factors show that the model based on the combination of temperature and precipitation has the better effect than that based on single temperature and single precipitation. Combined with the results of correlation analysis it can be inferred that NDVI changes are most signi cantly in uenced by the combination of temperature and precipitation, followed by temperature, and least in uenced by precipitation. The above ndings can provide a meaningful reference and guidance for the study of vegetation growth with climate changes.

Normalized difference vegetation index(NDVI) is the most commonly used factor to reflect vegetation growth status, and improving the prediction accuracy of NDVI is of great significance to the development of regional ecology. In this study, a new NDVI forecasting model based on the combination of time series decomposition(TSD), convolutional neural network (CNN) and long short-term memory (LSTM) was proposed. In order to verify the performance of TSD-CNN-LSTM model and explore the response of NDVI to climatic factors, two forecasting models of temperature and precipitation based on its own historical information and four NDVI forecasting models that based on Temperature, precipitation and its own historical information were established. The results show that TSD-CNN-LSTM model based on its own historical information had the best forecasting performance, with the RMSE, NSE, MAE to be 0.4996, 0.9981, 0.4169 for temperature, 5.6941, 0.9822, 3.9855 for precipitation and 0.0573, 0.9617, 0.0447 for NDVI, respectively. Meanwhile, the NDVI forecasting models based on climatic factors show that the model based on the combination of temperature and precipitation has the better effect than that based on single temperature and single precipitation. Combined with the results of correlation analysis it can be inferred that NDVI changes are most significantly influenced by the combination of temperature and precipitation, followed by temperature, and least influenced by precipitation. The above findings can provide a meaningful reference and guidance for the study of vegetation growth with climate changes.