Oral Presentation

Visuospatial cognitive ability, also known as spatial intelligence, is defined as the skill to generate, retain, search, and transform well-structured images, which is considered as a core element of STEM career fields. Although mathematics and geometry education has been implemented in Taiwan for many years, the existing teaching methods only focus on the learning of independent concepts. Thus, limits the efficiency in improving students’ spatial intelligence. To break through the current education framework, we developed a virtual reality visuospatial training software. The software will focus on the training of the basic components of spatial intelligence via a game-based design. We recruited 89 elementary school students in 5th grade to validate our software on both traditional computer and virtual reality devices to discern the difference between platforms. Elementary school students need to go through a month of learning and training, and to compare the differences between the pre and post training. It will be verified that virtual reality visuospatial cognitive training can increase the students' spatial intelligence. After the intervention, we found that both elementary school students' performance in training program and their spatial intelligence has significant improvement, and the performance in training is significantly related to spatial intelligence, proofing that this virtual reality visuospatial cognitive ability training program is beneficial to elementary students' development of spatial intelligence.

Emotion recognition and empathy are interconnected processes critical to social understanding and interactions. Developmental studies on emotion recognition are abundant; however, the relationship between empathy and emotion recognition remains unclear. In this study, we employed multiple tasks to explore emotion recognition in children and adults and its association with EQ and motor empathy. Fifty-eight children (33 5-6-year-olds, 25 7- 9-year-olds) and 61 adults (24 young adults, 37 mothers) joined this study. Each participant received an EQ questionnaire and the Dynamic Emotion Recognition task, where participants were to identify four basic emotions (happy, sad, fearful, angry) from neutral to fully expressed state, and the Motor Empathy task, where participants' facial muscle activity when watching videos was recorded. For the dynamic emotion recognition, children’s accuracy for “fearful,” “angry,” and “sad” expressions was significantly lower than that of the adults. For motor empathy, the 7 to 9-year-old children exhibited the highest facial muscle activity, whereas the young adults showed the lowest engagement. Importantly, individual EQ scores positively correlated with the motor empathy index in adults (but not children). In conclusion, the present study echoes the previous literature that the development of basic emotion recognition is heterogeneous; identifying negative emotions (i.e., anger, fear, sad) is still difficult for children aged 5-9 and continues to develop in late childhood. Our results also demonstrated that adults with higher EQ show stronger motor empathy. This is a new finding of this study and deserves further exploration.

Research has consistently shown that athletes develop cognitive advantages specific to the demands of their sports. This study explores how the type of sport—team or individual—influences athletes' collaborative decision-making abilities, particularly assessing whether team sports athletes display superior group decision-making abilities compared to their individual sports counterparts. We recruited 40 collegiate athletes, half from team sports like soccer and volleyball and half from individual sports such as swimming. They were grouped into dyads by sport type to perform visual search tasks both cooperatively and individually. We found no significant differences in accuracy and response times across the groups. However, workload capacity analyses from Systems Factorial Technology revealed that team sports athletes had distinct advantages in information integration. This advantage was particularly evident as these athletes made more accurate and quicker decisions within the earlier response time interval compared to the predicted performance when they performed the task alone. The results suggest that the dynamic and interactive nature of team sports may enhance athletes' ability to handle complex, fast-paced decision-making scenarios. These findings underscore the potential of sports-specific cognitive training to develop decision characteristics, particularly in team sports environments.

With advancements in technology, analyzing emotional expression in drawings has become more precise, leveraging digital data compared to traditional methodology. This study explores how cultural differences in drawing styles and color usage across emotions are influenced by the Big Five personality traits. We used OpenCV and Python software to analyze participants' self-drawings of four emotions: anger, happiness, sadness, and fear. Our sample includes a total of 235 cases, with Taiwan representing 42.6% and Thailand 57.4%. There were 105 females (44.7%) and 130 males (55.3%). The mean age was 21.91 years (SD = 1.52). For the Big Five traits, the distribution is as follows: extraversion (10.6%), agreeableness (42.6%), conscientiousness (17.0%), neuroticism (9.4%), and openness (20.4%). Countries showed significant differences across all emotion categories (p<0.001) and in certain color usage and drawing styles (p<0.001). Logistic regression analysis revealed that individuals from Taiwan were more likely to use abstract styles for all emotions significantly (p<0.001). The Big Five traits had a minor effect on drawing styles. For anger, those from Taiwan (p<0.001) and those with high conscientiousness were more likely to use abstract styles (p=0.017). Conscientious individuals were less likely to use abstract styles for sadness (p=0.04), while agreeable individuals were less likely to use narrative styles for sad expressions in drawings (p=0.044). In conclusion, technology enables precise analysis of emotional expressions in drawings. Differences in styles and color usage are influenced by country and personality traits, suggesting cultural and personality factors significantly impact artistic expression.

Schizophrenia is characterized by profound disruptions in social functioning, yet the underlying cognitive mechanisms remain unclear. Specifically, it is unclear whether the deficits in multisensory integration (MSI) observed in these patients are restricted to basic sensory processes or also impair the integration of social information. This study aims to discern whether MSI deficits in schizophrenia affect both self-related and non-self-related information processing equally, or if they are particularly detrimental to tasks involving social information processing. To address this issue, we employed an audio-visual detection version of both non-self-related (e.g., perceptual judgments) and self-related tasks (e.g., preference judgments), concurrently recording event-related potentials (ERPs), and applied System Factorial Technology (SFT). The study consisted of 48 chronic patients with schizophrenia who met the DSM-5 diagnostic criteria. Our findings showed that faster response times in multisensory (AV) condition than unisensory (A or V) conditions in both non-self-related and self-related tasks. However, patients displayed lower information processing efficiency in self-related tasks. ERP analysis showed that auditory N100 amplitudes were larger in the AV condition than in the unisensory (A + V) condition for both tasks. For visual processing, the AV condition elicited a larger visual P100 amplitude in non-self-related tasks compared to the A + V condition, with no significant differences in self-related tasks. This pattern suggests a potential deficit in the early visual processing of social information among patients with schizophrenia. Our findings suggest that while schizophrenia may not affect the speed of sensory processing, it does impair the efficiency of integrating self-related information. The study highlights the need for further exploration into how deficits in multisensory integration extend beyond basic sensory processing to affect social cognition in schizophrenia.

The self-prioritization effect refers to the phenomenon where people perceive external stimuli more accurately when the stimuli are related to themselves rather than to others. We used this effect as a tool to investigate the mechanism of empathy. According to the self-other merging hypothesis, empathy expands the self-representation to include others, which should reduce the self-prioritization effect. Conversely, the self-other distinction hypothesis suggests that empathy heightens the differences between self and others, thus increasing the self-prioritization effect. To differentiate between these hypotheses, we designed a pattern discrimination task along with EEG recordings and analyzed the data using signal detection theory. Different performance indices supported different hypotheses. The sensitivity index (d') and the attention index of the P3 event-related potential component indicated a higher self-prioritization effect in the high-empathy group compared to the low-empathy group, supporting the self-other distinction hypothesis at the perceptual and attentional levels. However, the response bias index and reaction time revealed an opposite pattern, supporting the self-other merging hypothesis at the response level. Our results suggest that empathy operates differently across various levels of self-representation. 

Despite the increasing integration of automated systems in decision-making, the impact of the credibility and reliability of these systems on human decisions remains underexplored. This study aims to address this gap by investigating how the credibility and reliability of automated information influence decision-making performance in tasks of varying difficulty. Participants engaged in a perceptual categorization task with automated aids of varying credibility and reliability while performing difficult and easy tasks. Utilizing the single-target self-terminating (STST) rule from System Factorial Technology, decision efficiency was measured by comparing performance with automated assistance against a baseline model without aid. Results revealed a robust validity effect, characterized by discrepancies in response times and accuracy between valid and invalid automated cues, particularly when participants used high-reliability automation. This effect was amplified by high- credibility information in difficult tasks. While credibility enhanced the validity effect, suggesting that decision-makers were more influenced by credible aids, it did not affect the decision efficiency as measured by the STST capacity. Instead, reliability emerged as the critical factor enhancing decision efficiency, especially in difficult tasks. These findings highlight the nuanced roles of reliability and credibility in aided decision-making, underscoring the importance of ensuring reliable automation in perceptual decision-making. They suggest that while credibility enhances the perceived value of information, reliability primarily drives decision efficiency. This research contributes to a deeper understanding of how automated systems can be optimized to support human decision-making, emphasizing the need to balance both credibility and reliability in the design of such systems.

Developmental reading difficulties may be associated with visual-spatial deficits. Poor readers often make mirror errors cannot discriminate between symmetrical patterns, such as d and b as different letters. Here we investigated the effect of symmetry on visual word form perception in developmental dyslexic (DD) and chronological aged (CA)-matched typical juvenile (aged 12-17 years). We employed real-, non-, jiagu- and scrambled characters in hanzi with symmetric or asymmetric compositions as stimuli. In the behavioral experiment, observers had to judge whether the two stimuli presented on both sides of the central fixation were identical. Both groups responded more accurately in the symmetric compositions than the asymmetric ones. This symmetry effect was found for all character types except for the real-characters. In a block-design functional magnetic resonance imaging (fMRI) experiment, we measured blood oxygenation level dependent (BOLD) activations when observers were asked to match the presented character with that of the previous trial. The visual word form area (VWFA) in the left fusiform gyrus showed significantly differential cortical activations to real- versus scrambled characters for the CA group but not for the DD group. The CA group showed stronger BOLD activations in the lingual and fusiform areas while processing non-characters versus real-characters compared to the DD group. Critically, the DD group recruited the occipitotemporal areas to a greater extent for symmetry analysis in visual word forms. Taken together, the intact symmetry processing suggests that dyslexic readers may fail to break the mirror generalization during visual word form recognition.

In the previous literature, within-person recognition has often been neglected. Jenkin et al. (2011) developed a unique paradigm to study within-person recognition; however, most research has examined only the Caucasian population. Moreover, it is unclear how individual differences correlate with standardized face-processing tests in diverse populations. This study investigates within-person face recognition in the context of Pakistani and Taiwanese adults and explores the correlations among within-person recognition, facial discrimination, and standardized face perception tests. To date, 22 Taiwanese adults (mean age=22.5±3.2) and 12 Pakistani adults who live in Taiwan (mean age=28.6±3.0) participated in this study. Each participant received two conditions of Face Identity Sorting (to sort 40 photos of two celebrities by identity), a Face Discrimination Task, a Cambridge Face Memory Test (CFMT), and a Twenty-Items Prosopagnosia Index (PI-20) questionnaire. Results showed that in the Face Identity Sorting, both groups sorted fewer piles for familiar faces while more piles for unfamiliar faces (Pakistani: familiar mean = 2.6±0.9, unfamiliar mean = 5.9±1.2, Taiwanese: familiar mean = 2.7±0.9, unfamiliar mean = 8.1±3.0. The two groups did not differ in the Face Discrimination Task (Taiwanese: 17.0 ±2.8, Pakistani: 17.9±2.0), CFMT (Taiwanese: 67.9±8.4, Pakistani: 69.4±10.4), and PI-20 (Taiwanese: 48.6±12.3, Pakistani: 43.2±8.3). Correlational analyses revealed that individual PI-20 scores positively correlated with misidentification errors in Taiwanese participants. In conclusion, our study is among the first to investigate within-person recognition in the context of Pakistani and Taiwanese adults, demonstrating a robust effect of familiarity on within-person recognition.

Perceptual disambiguation is crucial for human evolution as it allows for adapting to ambiguous threats with limited exposure. Previous studies showed how prior information affects recognizing ambiguous stimuli consciously, but it remains unclear whether prior experience can be automatically or unconsciously applied to disambiguate stimuli. Here we used fMRI to monitor brain activity during the Mooney Images Paradigm under discontinuous flash suppression. Findings replicated effects observed consciously and revealed neural disambiguation even in unconscious conditions, particularly in occipital and temporal regions like V1, V2, FG, IT, and MT. This finding suggests the brain automatically applies prior experiences even without conscious perception, indicating a disparity between neural and conscious recognition.

Face and object perception play an important role in our daily life. It is hypothesized that face processing involves a stronger degree of holistic processing—the tendency to perceive face parts as indivisible wholes—compared to object processing. This study aims to investigate how human process faces and objects, with a particular focus on the organization of their mental representations using a theory-driven tool, Systems Factorial Technology. Participants were recruited to perform a redundant-target version of a composite task, where they were required to make a two-alternative forced-choice classification decision (face or object) about a stimulus. The stimuli consisted of either the upper or lower parts of a face, an object, a 70/30 morph of a face and an object, or a 30/70 morph of a face and an object. Results revealed that participants were more accurate and faster in processing faces than objects. Additionally, 7 out of 9 participants tended to process faces coactively, while 6 out of 9 participants adopted parallel exhaustive processing for objects. Furthermore, participants were more efficient in processing faces compared to objects, indicating a stronger degree of holistic processing for faces. In conclusion, this study provides strong evidence supporting the notion that faces are processed holistically whereas objects are processed in an analytic fashion.