Abstract

In the present study, two experiments were conducted to examine how neighbor words affect Korean visual word recognition. Previous studies have shown that word frequency has a crucial role in word recognition. In addition, some researchers argue that neighbor words that are orthographically or phonologically similar to a target word also affect word recognition. In Korean, neighbor words can be defined as a group of words that share the first syllable with the target word. The type frequency of the syllabic neighbor words refers to the number of neighbor words and the token frequency refers to the accumulated word frequency of the neighbor words. Although previous studies on Korean visual word recognition have shown that the word frequency effect emerges, there are few studies on effects of the type or the token frequency using a factorial design. To this end, we conducted a lexical decision task, in which the type frequency was manipulated in Experiment 1 and the token frequency was manipulated in Experiment 2. The results showed that neither the type nor the token frequency affect response times of the lexical decision task. The results suggest the necessity to further discuss the nature and the characteristics on the effect of syllabic neighbor words in Korean visual word recognition.

Abstract

This study investigated action monitoring in female college students with high trait anxiety using event-related potentials (EPRs) and Simon task. Based on the scores of State-Trait Anxiety Inventory Y version trait scale and Welsh Anxiety Scale, high trait anxiety (n=24) and low trait anxiety (n=23) groups were determined. The Simon task consists of two conditions-congruent and incongruent conditions. The locations of stimulus and response were the same in the congruent condition, whereas the locations of stimulus and response were different in the incongruent condition. Participants were instructed to press one of four buttons assigned to the location of a stimulus. High and low trait anxiety groups showed comparable behavioral performance. In terms of ERPs, high trait anxiety group exhibited significantly larger amplitude of error-related negativity (ERN) and correct response negativity (CRN) on Cz electrode site than low trait anxiety group. And there were negative correlations between ERN/CRN amplitudes on Cz and trait anxiety level. The results of this study indicate that female college students with high trait anxiety have a different action monitoring pattern from that of low anxiety group and ERN/CRN amplitudes could be used as trait markers of anxiety disorder.

Abstract

P300 concealed information test (P300 CIT) assumes that all stimuli except the target stimuli show the same P300 amplitudes in innocent participants. However, according to the P300 theory, the assumption of the equality of stimuli can be easily violated, thus false positive rate of P300 CIT can be increased. A series of studies were conducted to confirm this possibility. In study 1, we identified how much the assumption of the equality of stimuli was violated in single-probe protocol. This assumption was invalidated in about 30% of the participants in the study using a mock crime, but it was almost met in the study using autobiographical information. In study 2, we conducted an experimental study to evaluate the assumption of equality of stimuli in multiple-probe protocol. It turned out that the assumption was not met in about 25% of participants. In study 3, a Monte Carlo study was conducted to estimate the false positive rates by the level of non-equality of stimuli (0, 1, 2, 3, 4, and 5), the number of probes (1, 3, and 6), the number of trials (30, 48, and 66), and the significance levels (.10 and .05). When the level of non-equality of stimuli was set to 0, the false positive rates were properly controlled regardless of the number of probes, the number of trials, and the significance levels. However, the false positive rates was increased in all conditions when the level of non-equality of stimuli was 1 or higher. The false-positive rates got higher when the level of non-equality of stimuli got higher, the number of probes got smaller, and the number of trials got higher. When the level of non-equality of stimuli was 2, the false positive rates ranged from 12% to 25% at the significance level .10. While the level of non-equality of stimuli was 5, it fell within the ranged of 18% to 38%. In the discussion, we proposed methods to control false positive rates in the presence of non-equality of stimuli.

Abstract

Abstract

Our brain tends to associate stimulus features across the senses in a non-random manner. For example, people show consistency in labelling a rounded shape ‘maluma’/‘bouba’ and a spiky shape ‘takete’/‘kiki’. Previous studies have attributed this phenomenon to the correspondence between sound and shape, but without controlling for other potential factors (i.e., linguistic/orthographical factors). The present study examines the role of acoustic aspect per se by manipulating articulatory gestures to generate synthetic speech sounds not confined to a specific language. Participants were asked to choose either a rounded or spiky shape to indicate the shape that better matched each synthetic speech sound. The results demonstrate that shape choice was systematically mapped on to the dimensions manipulated to generate the sounds. These results indicate that acoustic features indeed drive the association between sound and visual shape.

Abstract

The purpose of the current study was to investigate whether phonological information of the visually presented word is activated before lexical access. To do so, the present study compared ERPs of the irregular words and regular words. Since N320 component has been known as an ERP component reflecting grapheme to phoneme conversion, we would expect different N320 waves between irregular and regular words if phonological information is activated before lexical access. As expected, the results showed larger N320 for irregular words than regular words. The present study supported the idea that phonological information is activated from the sublexical level during visual word recognition processing.

Abstract

Approximate Number Sense (ANS) is an intrinsic intuition and innate cognitive function. It helps children to roughly determine if a quantity is larger or smaller than another. In the current study, we examined relative roles of ANS and intelligence on early arithmetic development in children aged 4-6. Our purpose was 1) to examine whether children aged 4-6 can solve non-symbolic multiplication problems and 2) to examine effects of ANS on early arithmetic development after controlling out effects of intellectual ability. Twenty one young children (4-6 years old) performed non-symbolic multiplication tasks (i..e, multiplication 2 and 4) and a ANS task. Results showed that children as young as 4 years old can solve both non-symbolic multiplication problems above chance level. Also, we found that the efficiency of ANS measured by ANS RTs played a critical role for non-symbolic multiplication 2 task performance even after controlling out effects of IQ. Overall, we conclude that preschoolers as young as 4 years old who don't have prior learning experience of multiplication problems can solve non-symbolic multiplication problems, which might be explained by children's ANS functions.

Abstract

The present study examined whether a reverse distance effect (RDE) is consistently observed for ordinality judgment using numbers, Korean letters and the alphabet. RDE refers to a phenomenon in which better performance is observed for judgment on stimuli that are closer to each other. We examined whether performance on these tasks are correlated with academic achievement in math, Korean and English domains. Indeed, RDEs were observed from all three tasks. This result reveals that the order of numbers and letters are similarly processed and is consistent with the results of previous studies reporting RDE. Performance was better for order judgment of numbers compared to Korean letters, and for Korean letters compared to the alphabet. This reveals that ordinality judgment of letters are less efficient compared to numbers and that ordinality judgment in the native language is more efficient compared to a foreign language. Linear regression analysis revealed that ordinality judgment performance using numbers and Korean letters significantly predicted math achievement. All three ordinality judgment performance predicted achievement in Korean and English domains. These results suggest that ordinality judgment of numbers and letters is related to achievement in not only math but also language. The present study is the first to examine the relationship between ordinality judgment performance and language achievement. We hereby propose that ordinal representations may be more domain-general than previously conceived, going beyond their presupposed numerical nature.

Abstract

Nonword trials are a necessary part of the lexical decision task and their composition can affect the responses to experimental words in the task. To investigate the role of nonwords in visual recognition of Korean words, the nonword type and its proportion in a list were manipulated. The results showed that the frequency effects of bisyllabic Sino-Korean (SK) words were smaller when they appeared among nonwords made of SK syllables than among nonwords with non-SK syllables or non-used syllables. The effect of nonword type was also stronger for low-frequency words. In addition, the proportion of nonwords included in a list influenced lexical decisions to words. The smaller the proportion of nonwords in a list, the faster decisions were made to words. The proportion of nonwords did not interact with word frequency. The present results call for more attention to nonwords in Hangul word recognition.