Viewing electronic screens for a long time can cause visual fatigue, even induce eye problems such as refractive errors, dry eyes, etc. The relationship between screen display and visual fatigue has received much attention. This article aims to study the optimal display color temperature and color mode under two common ambient illuminations—normal office lighting (450 lux) and dark environment (3 lux). In a 2 × 2×3 experimental design, 36 participants evaluated the effects of two display color modes (light mode and dark mode) and three screen color temperatures (2800 K, 4500 K, and 6500 K) on visual fatigue under two ambient illuminations. We used eye movement tracking technology to collect blink rate and pupil diameters as objective indicators and used the Richter scale to collect subjective visual fatigue and preference score as subjective indicators. Results showed that at 450 lux, lower visual fatigue was found in the dark mode, 4500 K and 6500 K. At 3 lux, the visual fatigue was lower in the dark mode and 4500 K color temperature. Whether at daytime or night, the 2800 K color temperature resulted in the highest visual fatigue. Our findings complement the research on screen display and visual fatigue. They have practical implications for reducing visual fatigue caused by prolonged viewing of electronic screens.

Natural language models have been widely used for their impressive performance in various tasks, while their poor robustness also puts critical applications at high risk. These models are vulnerable to adversarial examples, which contain imperceptible noise that leads the model to wrong predictions. To ensure such malicious examples are imperceptible to humans, various word-level attack methods have been proposed. Previous works on word-level attacks attempt to generate adversarial examples by substituting words in sentences. They utilize different candidate substitution selection methods and substitution strategies to improve attack effectiveness and the quality of generated examples. However, previous works are all unigram-based attack methods, which ignore the connection between words. The unigram nature of these methods downgrades fluency, increases grammatical errors, and biases the semantics of adversarial examples, making adversarial examples easier to be detected by humans. In this paper, to improve the quality of textual adversarial examples and makes the adversarial example more imperceptible to human, we propose a black-box word-level attack method called Dynamic N-Gram Based Attack (DyGram). DyGram tokenizes the entire sentence into multiple n-gram units, rather than individual words as in previous works, and substitutes words in a sentence in descending order of n-gram unit importance. Extensive experiments demonstrate that DyGram achieves higher attack success rates than previous attack methods and improves the quality of generated adversarial examples in terms of the number of perturbed words, perplexity, grammatical correctness, and semantic similarity.

Difficulty balance in the game level design is particularly important for user experience (UX). This study uses cognitive control and flow theory as a starting point, and based on the functional near-infrared spectroscopy technology, this research obtained oxygenated hemoglobin concentration (ΔHbO2) signals from three brain areas in the prefrontal cortex (PFC), namely the dorsolateral prefrontal cortex (DLPFC), the ventral prefrontal cortex (VLPFC), and the frontal area (FPA). Combined with the subjective flow state scale to explore the reaction difference between the video game players (VGPs) and the non-video game players (NVGPs) during the entire period of playing video games and the moment of game difficulty changes. The results show that VGPs and NVGPs have significant differences between flow state and HbO2 signals in different brain regions of the PFC. The players’ flow state is affected by the game difficulty. The HbO2 signals in the players’ DLPFC and FPA have a linear dependence relation with the flow state. This research proves that VGPs and NVGPs have different UX with game difficulty changes from the cognitive perspective. The results provide a reference for the future game level design about game difficulty balance to achieve a better UX in the game.