• eye movement;
  • NEO;
  • non-verbal communication;
  • psychological status;
  • visual information


  1. Top of page
  2. Abstract

Aims:  This paper examines the relationship between observational behavior and the observers' result assumptions, using a contained diagram that includes significant non-verbal information, such as gestures. The ability of care workers to assess a patient's mental status on the basis of non-verbal information would be considered an important skill necessary to understanding patient condition.

Methods:  One hundred and eleven subjects were asked to take a test exploring two types of psychological status, and their eye movements were monitored during testing. Path analysis was used to examine the relationships among eye movements, individual personalities, and test results.

Results:  In stimulus 1, the neuroticism of the personality was determined. In stimulus 2, openness was determined in the range of eye movement. There is a relationship between the total length of eye fixation time and the answer time at the eye fixation point. Although there was no significant influence found in stimulus 1, the openness still tended to influence the eye fixation point.

Conclusions:  The eye fixation point increased among those people with high openness scores. It was theorized that they attempted to obtain information from the stimulus being provided.

THE ABILITY TO interpret non-verbal information, including facial expressions and gestures, to determine someone's state of mind is a very important skill in mental health-care settings. It is often difficult for mentally ill people to verbally express their condition. Therefore, professionals must provide patients with an environment that allows patients to more easily express their condition; professionals must also gain insight into different conditions and desires through observation. The observation act has two stages. In the first stage, perception comes from the sensory feeling receptor. Alternatively, the recognition in the second stage involves processing the perceived information and deriving meaning from it.

With regard to the first phase, the sensory receptor refers to the five human senses: sight, touch, smell, hearing, and taste. People always communicate with each other, not only with words, but also by sending and receiving information through the five senses.1 Specifically, when two people communicate, words compose only 35% of their exchange. That is, 65% of human communication takes place through non-verbal information, such as affect, gestures, speech pace, and other behavior.2

This paper therefore focused on the visual information that becomes the center of perception for non-verbal information. A large and growing body of literature has reported on people's psychological status using visual information. Such research has produced a comprehensive system, the Facial Action Coding System (FACS). Using this system to analyze facial anatomy, a theory was constructed about the mechanisms of the human visage and the emotions that arise when that visage is observed to visibly change.3 Although studies on emotional expression have previously distinguished facial categories, such as pleasure, sadness, surprise, rage, hate, contempt, and fear, recent research has tended to focus on facial research using FACS. Despite this, Russell claimed that the face provides only determinative information and does not offer information about underlying emotions.4,5 Hence, it is difficult to recognize all information by merely looking at a person's face. It is more difficult to understand psychological conditions due to a multitude of factors that includes facial expression and other relevant information. Thus, it would be of interest to investigate not only the role of facial expression but also the process of capturing complicated non-verbal visual information to intuit psychological status. To accomplish this, a diagram containing postural information as well as a personal body zone could be used to clarify the process of recognition in the activities of daily life. Additionally, it would seem that further investigation is needed to clarify the observer's tendency to absorb visual information and assess psychological status, although each person's unique way of capturing visual information is a variable, the results of which are just beginning to be discovered. It is possible that the unique individuality of a person may be reflected in a particular behavior that surfaces during absorption of visual information. The aim of the present paper was to examine the relationship between eye movement and the accompanying physiological index, as well as the correlation between eye movement and personality. In this way, we recognize and clarify those factors at play in capturing the visual information used to assess psychological status.


  1. Top of page
  2. Abstract

Study design

Eye movement

When measuring the acquisition of information resulting from visual stimulus, the trajectory of eyesight depends upon the movement of both the eyes and the head.

The study was designed to measure only saccadic eye movement, to allow evaluation of parameters (eye stop point, eye fixation time, eye movement speed, and eye movement range). The examinees looked forward only, without moving their heads. This method excluded compensatory eye movement. As a consequence, image stimulus is produced. This also allowed the exclusion of convergent eye movement. Moreover, it is possible to exclude smooth pursuit eye movement by using a still image. In this way, only saccadic eye movement is measured when using a still image as a stimulus presentation. This makes it possible to examine the eye stop point as the fixation point. The decision was made to use monochromatic images because visual stimulus of strong colors can affect eye movements.


Several studies to develop a personality inventory have been carried out on the basis of a wide range of character theories. Category classifications such as those found in the DSM-IV are used as clinical views. Within quantitative experimental work, dimension classifications, for instance the Tridimensional Personality and Character Inventory6 and the NEO Personality Inventory (NEO-PI)7 are used.

Alternatively, the NEO-PI, produced by Costa and McCrae,7 was developed from a five-factor model as a structured measure. The instrument is able to collect a word (the characteristic word) that shows a difference among individuals, analyze the factors, and classify them. Therefore, the Japanese version of the inventory is the most appropriate for seeing how action relates in other personality inventories because five factors arise from an action characteristic description.8–10

Assessing psychological status

Upon initially seeing someone, the viewer first wonders if he or she knows that person. The viewer will next assess the psychological status of the person he or she sees by reading facial expression. This ability is important for identifying individuals and examining faces in social situations. Studies in this field have examined non-verbal communication ability. Studies by Ekman and Friesen, and Benner focused on emotional expression.3,11 Stern et al. studied the blink.12 Thus far, we know there is both a universal element and a cultural element involved in anticipating someone's psychological status. Van Geert systematized the generalizations of empirical and technical.13 Ekman et al. demonstrated generalizations in the Facial Affect Program, which indicates that the perception of emotions and their anticipation are inborn.14 Those studies, however, analyzed only one parameter of non-verbal communications. It is also necessary to examine anticipation of the psychological status in more natural settings.

The decision was therefore made in the present study to use a framework exploring the relationships between a set of eye movements, emotions and their recognition, memory, and how individual personality is influenced by the former factors in the process of assessing psychological status. The present study (Fig. 1) focused on the relationships between (i) the pattern of perception demonstrated by eye movement; (ii) the individual personality; and (iii) the result of anticipating psychological status (whether or not the assessment is correct).


Figure 1. Flow chart of the supposition process.

Download figure to PowerPoint



Prospective participants were first–fourth-year undergraduate university students studying to be health-care workers. These participants were confirmed to have no ocular disease and to be capable of the required eye movements. People found in the structural interviews to be mentally ill were excluded from the study. The number of participants who met all criteria was 43 in the Department of Acupuncture and Moxibustion, 31 in the Department of Judo Seifuku, and 37 in the School of Nursing Science. A total of 111 individuals participated in this study, which was conducted between 4 and 11 July 2007.

Measurement of the personality

Inspection form

The Revised NEO-PI (NEO-PI-R) developed by Costa and McCrae7 was used to measure individual personality. The inventory can comprehensively measure personality using five factors: neuroticism (N), extraversion (E), openness (O), agreeability (A), and conscientiousness (C).8,15 The instrument has 240 questions with a 5-point Likert scale.

N specifically measures anxiety, hostility, depression, and impulsiveness. It is generally understood that people with a high level of N have negative feelings, have difficulty controlling their anger, and tend to be unstable. E measures whether a person is active, an excitement-seeker, and gregarious. People with a high level of E tend to like excitement and are social and active. O measures aesthetics, fantasy, and ideas. People with high levels of O tend to be curious and open to experiences. A measures trust in others, altruism, and tenderness. People with high levels of A are likely to be kind and eager to help others. C measures self-discipline and the drives for achievement and order. People who exhibit a high level of C tend to have a strong will and to be serious about accomplishing their objectives.

After informed consent, all participants were asked to complete the NEO-PI-R.

Experimental environment

The environment in which the experiments were carried out was lit with 30 lux. One chair and a Panasonic LCD monitor (TH50PHD8K; Panasonic, Tokyo, Japan) were placed within the environment at a 230-cm distance from each other. The monitor that displayed stimuli for the objects was 80 cm wide × 60 cm high. The monitor was placed against the subjects ±7.43° of vertical and ±9.87° of level. The subjects were asked to sit in the chair with their heads facing straight toward the wall (Fig. 2).


Figure 2. Experiment environment.

Download figure to PowerPoint

Visual stimuli

A set of 36 sheets of photographs was used as the visual stimulus.16 This set of photographs, How to Expand Your SIQ (Social Intelligence Quotient), aims to increase emotional literacy on the basis of non-verbal information and social intelligence. The photographs were of either one or more people in actual daily living situations. Each sheet includes two or three questions that explore the psychological status of the person or persons pictured. There are also answer keys for the questions. Rates of correct answers obtained from 2400 people have been reported.

All people represented in the photographs, however, are European. It is generally understood that cultural differences could affect participant response, therefore it was not appropriate to use these photographs. For this reason, it was decided to limit use to two photographs depicting only two people as the stimuli presentation (Figs 3,4). The selected photographs are simple and easy to understand, and the answer rates from a pilot study of 50 Japanese participants rated as highly as the original study.


Figure 3. Stimulus 1 (Question: These two women are working at the same workplace. Which is the boss? Right, left or doesn't understand.).

Download figure to PowerPoint


Figure 4. Stimulus 2 (Question: These two? Husband and wife in 2 after marriage, husband and wife of the put on or doesn't understand.).

Download figure to PowerPoint

Eye movement recording

The TalkEye II (processor: TKK 2940a, cover type tracking detector: TKK 2940g), which is manufactured by Takei Scientific Instruments (Niigata, Japan), was used.

Standard physiological data

Among the standard physiological data obtained for measurement were viewings of two diagram titles shown one after the other. First, a question was shown. The participant was asked to sign when they remembered the question and then a diagram was shown. The first segment ended when the participant answered. This was repeated. Whether the participant was able to judge correctly the psychological status (the correct answer) and how long it took for the participant to answer were both examined. The answer time was measured from the data of the eye movement. Also, eye movements were measured by the TalkEye II while examinees viewed the stimuli(Fig. 5).


Figure 5. Experiment procedure.

Download figure to PowerPoint

Visual information

The receipt of visual information was calculated using certain eye movement parameters: eye stop point, eye fixation time, eye movement speed, eye movement range, difference between the number of blinks and the pupil diameter, and standard physiological data. TalkEye II version 2.4.1 (Takei Scientific Instruments) was used to analyze eye movements and make calculations.

Eye movement speed

The movement speed is defined as the speed of movement of the viewpoint. The speed of the eye movement represents the average movement speed while the stimulus is being shown to the participant and is measured as movement speed.

Eye movement range

To evaluate the eye movement range, the means and standard deviation of X and Y axes against the eyes' angle were calculated. The range of the ellipse, the diameter of which was a standard deviation in the X and the Y coordinates, comprised the eye movement range, which refers to the wide area of view.


Fixation refers to the act of seeing someone and trying to cull information from the viewing. Even while watching closely, however, the eyes still move in micro-saccade increments to maintain resolution. Therefore, in the present study it was decided that <10°/s movements within the range of the micro-saccade could be measured as if the eyes had stopped on the identical point. If >100 ms had elapsed, then the point was established as the stop point. The total eye fixation time was considered the eye stop if it was >100 ms in the eye stop point.


The questions are how the eyes moved while the participant was answering the questions and to what extent did individual personality influence correct answers and the level of eye movement. To answer these questions, path analysis was used to detect the rate of the data appropriation to the causality model. We used a method of maximum likelihood and set an estimation method for standardization that assessed value while still seeking a correction index.

The χ2 test, comparative fit index (CFI), and root mean square error of approximation (RMSEA) were used to evaluate the model. SPSS version 11.0 (SPSS, Chicago, IL, USA) and AMOS version 5.0 (AMOS, Chicago, IL, USA) were used for data processing.

Ethics considerations

This study was approved by the Ethics Committee at the facilities where the experiment was implemented. The nature and course of the study were explained to the participants both orally and in writing. The participants were also instructed that they could withdraw at any time and that the information obtained in the study would not be used outside of the study. Only those people who agreed to do so participated in the study. Any study information that could identify individuals if obtained was recorded in symbols. The contrast table that listed individual participants' names was strictly guarded at the laboratory under lock and key. Participant names and other identifying information were scrambled. This method maintained the safekeeping of the conduct contrast table data, and the locker in which it was kept required unlatching with a flash memory device with a security lock function. The data will be destroyed within 6 months after release of the study.


  1. Top of page
  2. Abstract

Overview of the examinees

The present subjects were participants who could complete both the NEO-PI-R and the eye movement measurements. With regard to the examinees, 99 individuals were shown stimulus 1, and 100 individuals were shown stimulus 2, each out of a total of 111 individuals. The demographics for stimulus 1 were 52 men and 47 women, with an average age of 21.1 ± 3.4 years. The breakdown by department was 33.3% (33 individuals) from the School of Nursing Science, 30.3% (30 individuals) from Judo Seifuku, and 36.4% (36 individuals) from Acupuncture and Moxibustion.

The demographics for stimulus 2 were 51 men, 49 women, with an average age of 21 ± 3.4 years. The breakdown by department was: 34.0% (34 individuals) from the School of Nursing Science, 28.0% (28 individuals) from Judo Seifuku, and 38.0% (38 individuals) from Acupuncture and Moxibustion (Table 1).

Table 1.  Subject characteristics
DepartmentStimulusSexYear of undergraduate university of courseTotal
Male (%)Female (%)First (%)Second (%)Third (%)Fourth (%)
  1. Stimulus 1, n = 99; stimulus 2, n = 100.

Acupuncture and Moxibustion Department119 (19.2)17 (17.2)1 (1.0)4 (4.0)18 (18.2)13 (13.1)36 (36.4)
219 (19.0)19 (19.0)2 (2.0)5 (5.0)18 (18.0)13 (13.0)38 (38.0)
Judo Seifuku Department123 (23.2)7 (7.1)3 (3.0)5 (5.1)21 (21.2)1 (1.0)30 (30.3)
221 (21.0)7 (7.0)3 (3.0)5 (5.0)19 (19.0)1 (1.0)28 (28.0)
School of Nursing Science110 (10.1)23 (23.2)10 (10.1)23 (23.2)0033 (33.3)
211 (11)23 (23.0)10 (10.0)24 (24.0)0034 (34.0)


The average ± SD of each personality in stimulus 1 was N, 115.0 ± 23.1; E, 108.8 ± 19.2; O, 119.3 ± 16.4; A, 107.1 ± 17.4; and C, 96.5 ± 21.7. There was no apparent skew within the five NEO-PI-R factors.

The results of stimulus 2 were: N, 115.2 ± 22.9; E, 109.3 ± 19.1; O, 119.1 ± 17.0; A, 107.3 ± 17.4; and C, 96.9 ± 21.4. There was no apparent skew within the five NEO-PI-R factors (Table 2).

Table 2.  NEO Personality Inventory–Revised scores
 nNeuroticism (N) mean ± SDExtraversion (E) mean ± SDOpenness(O) mean ± SDAgreeableness (A) mean ± SDConscientiousness (C) mean ± SD
Stimulus 199115.3 ± 23.1108.8 ± 19.2119.3 ± 16.4107.1 ± 17.496.5 ± 21.7
Stimulus 2100115.2 ± 22.9109.3 ± 19.1119.1 ± 17.0107.3 ± 17.496.9 ± 21.4

Measurements during stimuli presentation

Reply results

For stimulus 1 there were 21 incorrect answers (21.2%) and 78 correct answers (78.8%). For stimulus 2 there were 37 incorrect answers (37.0%) and 63 correct answers (63.0%; Table 3).

Table 3.  Supposition result and physiological index of the stimulus presentation
 nCorrect answer n (%)Wrong answer n (%)Answer time (s) Mean ± SDPupil area (dot) Mean ± SD
Stimulus 19978 (78.8)21 (21.2)6.06 ± 4.59953.20 ± 542.02
Stimulus 210063 (63.0)37 (37.0)6.52 ± 3.961133.97 ± 586.27
Answer time

The average ± SD for the answer in each stimulus was 6.06 ± 4.59 s (n = 99) in stimulus 1 and 6.52 ± 3.96 s (n = 100) in stimulus 2 (Table 3).

Physiological index

The pupil area change in response to the stimulus was 953.20 ± 542.02 dot (n = 99) during stimulus 1 and 1133.97 ± 586.27 dot (n = 100) during stimulus 2. No comparison was conducted due to the number of blinks. The answer times were found to be below the average blink interval (Table 3).

Eye movement

Eye movement range

The average ± SD eye movement range was 11.67 ± 12.17 (n = 99) during stimulus 1 and 12.18 ± 16.13 (n = 100) during stimulus 2s (Table 4).

Table 4.  Eye movement results
 nEye movement range (deg2) Mean ± SDEye movement speed (°/s) Mean ± SDEye fixation point (point) Mean ± SDTotal eye fixation time (ms) Mean ± SDAverage eye fixation time (ms) Mean ± SD
Stimulus 19911.67 ± 12.1743.21 ± 32.064.44 ± 6.43735.00 ± 1290.18144.99 ± 38.96
Stimulus 210012.18 ± 16.1344.15 ± 45.484.83 ± 6.35783.48 ± 1132.95102.55 ± 73.23
Eye movement speed

The average ± SD eye movement speed was 43.21 ± 32.06°/s (n = 99) during stimulus 1 and 44.15 ± 45.48°/s (n = 100) during stimulus 2 (Table 4).


The average ± SD eye fixation point of stimulus 1 was 4.44 ± 6.43 points (n = 99) and that for stimulus 2 was 4.83 ± 6.35 points (n = 100; Table 4). The average ± SD total eye fixation time for each closely watched stimulus were 735.00 ± 1290.18 ms (n = 99) during stimulus 1 and 783.48 ± 1132.95 ms (n = 100) during stimulus 2 (Table 4). The average ± SD eye fixation time for one eye stop point in each stimulus was 144.99 ± 38.96 ms (n = 99) during stimulus 1 and 102.55 ± 73.23 ms (n = 100) during stimulus 2 (Table 4).

Personality and eye movement and answer to stimulus presentation

Path analysis showed that stimulus 1 did not trigger the E, A, or C personality factors for this model because of the low path coefficient with the answer time and eye movement. Moreover, the eye movement was not applied to the model within the average eye fixation time or the correct answer and answer time because the average eye fixation time did not have any correlation. It should be the positive result of the χ2 of 8.037 (d.f. = 13), with P = 0.841, CFI = 1.000, and RMSEA < 0.001 after using the standardization assessed value and improving the model's fit. For N in the personality, some level of influence was found in the eye movement range (path coefficient = −0.21). For the eye fixation point, correlation was found between the total length of eye fixation time (path coefficient = 0.96) and the answer time (path coefficient = 0.48). No significant difference was determined for O, but the possibility of influencing the eye fixation point (path coefficient = 0.14) was noted. No factor was found to have significant difference correlating to the correct answer, but there was a tendency for the eye fixation point (path coefficient = 0.07), answer time (path coefficient = −0.18), and eye movement range (path coefficient = 0.12) to be related (Fig. 6a; Table 5). After the path analysis for stimulus 2, the path coefficient for the answer time and the eye movements in the C, N, and E personality factors were not found to be applicable to the model due to their low values. Eye movements were found not to apply to the model regarding average eye fixation time and the correct answer and answer time because no relationship was found. It became the positive result of the χ2 = 12.809 (d.f. = 13), with P = 0.463, CFI = 1.000, and RMSEA < 0.001 after using the standardization assessed value and improving the model's fit. For the personality O factor, influence in the eye fixation point was established (path coefficient = 0.23). For the eye fixation point, the influence of the total length of eye fixation time (path coefficient = 0.98) was found between the answer time (path coefficient = 0.54) and the eye movement range (path coefficient = −0.18; Fig. 6b; Table 6).


Figure 6. Relation between the personality characteristic and the eye movement and the stimulus presentation supposition result for (a) stimulus 1 (χ2 = 8.037; d.f. = 13; P = 0.841; comparative fit index [CFI] = 1.000; root mean square error of approximation [RMSEA] < 0.001) and (b) stimulus 2 (χ2 = 12.809; d.f. = 13; P = 0.463; CFI = 1.000; RMSEA < 0.001). *P = 0.05; **P = 0.01; ***P = 0.001.

Download figure to PowerPoint

Table 5.  Standardization general effect: Stimulus 1
 OpennessNeuroticismEye fixation timeAnswer time
Eye fixation point0.142<0.001<0.001<0.001
Answer time0.068<0.0010.477<0.001
Total eye fixation time0.137<0.0010.963<0.001
Correct answer−0.005−0.026−0.036−0.177
Eye movement range−0.026−0.208−0.182<−0.001
Table 6.  Standardization general effect: Stimulus 2
 OpennessAgreeablenessEye fixation pointEye movement rangeAnswer time
Eye fixation point0.228<0.001<0.001<0.001<0.001
Eye movement range−0.0420.065−0.185<0.001<0.001
Answer time0.1170.0100.5130.161<0.001
Total eye fixation time0.224<0.0010.982<0.001<0.001
Correct answer−0.006−0.001−0.026−0.022−0.138


  1. Top of page
  2. Abstract

Stimulus presentation

The diagram by Archer was used as the stimulus presentation.16 The diagram is designed to show awareness of non-verbal complex facial communications. This study adopted the diagram after observing the high percentage agreement of correct answers in the pilot study in an effort to exclude the influence of cultural difference in the study results. In general, the aim is to use a test that has been created in the same culture in which the examinees live in order to measure non-verbal communication ability. Many studies, however, are mostly descriptive and few were qualitative studies with Japanese people. That is, there have been few experimental studies in Japan. Therefore, measuring tools in this field have not been developed for Japanese subjects.

Personality factors, eye movement, and stimulus presentation supposition

In the present study we clarified how personality influences the absorption of visual information that leads to correct psychological status assessments. The personality factors E and C were not found to relate to the model in the two segments of stimulus. The factor that influenced eye movement most in the two stimuli was found to be O. Individuals with a high O factor are curious about both the internal and external realms and tend to be open to experiences.

According to a study using a Necker Cube diagram by Ellis and Stark,17 the eye stop point tends to focus at the top of the cube reversed when it perceives that a diagram is reversed. There are many eye stop points when a person is receiving many keys that connect to the correct answer from the diagram. The individual with high O is supposed to have more clues.

The factor that influenced eye movement range in stimulus 1 was N and in stimulus 2 it was A. Perhaps one of the factors causing this difference might be the contents difference of stimulus 1 and stimulus 2.

In stimulus 1, N was found to influence eye movement range in the diagram where two people are in the workplace and are asking a job question. Therefore, in such a situation, reacting to the hierarchy requires delicacy and adds to psychological distress when N is high. Next, stimulus 2 involved a diagram asking about the closeness of the relationship between two people. Examinees viewing this diagram exhibited A in relation to their eye movement range. Because a psychologically healthy person sympathetically and altruistically assists others when A is high, it is possible to attempt to assess the closeness of two people by examining the information from a wider range of eye movement.

Eye movement and stimulus presentation supposition

To gather information from the stimulus presentation, the examinees' eyes must stop and recognize information. As a consequence, a judgment is made. To examine the direct and indirect effects of the judgment, we analyzed the relationships between the eye fixation point, each piece of eye movement data, answer time, and correct answer. The relationships between the eye fixation point, the total length of eye fixation time, the eye movement range, and the answer time showed that the eye fixation time increases when there are many eye fixation points in both stimulus 1 and 2. The range of eye movement narrows when the stop time increases. Individuals with a tendency to stop their eyes at a fixation point and linger, result in lapses of time without answer.

Regarding the relationships between eye fixation point, answer time, and correct answer, it was common both in stimulus 1 and 2 that the relationships were very weak. Regarding answer times, however, there was a correlation between the correct answer and the incorrect answer with eye fixation points; a positive relationship was found with the correct answer. As a result, it was theorized that the judgment was made quickly in the visual process and that the eye stop points increased to more efficiently gather information, even after assessing the correct psychological status.


The purpose of the present study was to clarify the relationships between the personality, eye movements, a physiological index, and recognition of appropriate connections within a diagram to assess the factor by which study participants captured visual information and made assessments about psychological status.

O was found to influence eye fixation point. Nonetheless, the eye fixation point's influence upon the correct answer was found to be very weak. The relationships among the factors were also weak. These results, however, could provide a model for the role of personality and eye movements in making suppositions about psychological status. It was theorized that the judgment was made quickly in the visual process and that the eye stop points increased to more efficiently gather information, even after assessing the correct psychological status.

Further investigation using increased stimulus presentations and sample numbers is needed, however. This is because the personality factors that influence range of eye movement necessarily change when the diagram and the stimulus presentation questions are different.


  1. Top of page
  2. Abstract
  • 1
    Vargas M. Louder Than Words: An Introduction to Nonverbal Communication. Iowa State Press, Ames, 1986.
  • 2
    Birdwhistell R. Kinesics and Context: Essays on Body Motion Communication. University of Pennsylvania Press, Philadelphia, 1970.
  • 3
    Ekman P, Friesen W. Facial Action Coding System: A Technique for the Measurement of Facial Movement: Investigator's Guide 2 Parts. Consulting Psychologists Press, Palo Alto, 1978.
  • 4
    Russell JA. Is there universal recognition of emotion from facial expression? A review of the cross-cultural studies. Psychol. Bull. 1994; 115: 102141.
  • 5
    Russell JA. Facial expressions of emotion: What lies beyond minimal universality? Psychol. Bull. 1995; 118: 379391.
  • 6
    Cloninger CR. A systematic method for clinical description and classification of personality variants. A proposal. Arch. Gen. Psychiatry 1987; 44: 573588.
  • 7
    Costa P, McCrae R. Neo PI-R Professional Manual. Psychological Assessment Resources, Odessa, FL, 1992.
  • 8
    Shimonaka J, Nakazato K, Gondou K, Takayama M. Construction and factorial validity of the Japanese NEO-PI-R. Jpn. J. Pers. 1998; 6: 138147.
  • 9
    Tsuji H, Fujishima Y, Natsuno Y, Mukoyama Y. The personality trait theory and the five factor model: The characteristic concept, the structure, and the measurement. Jpn. Psychol. Rev. 1997; 40: 239259.
  • 10
    Murakami Y, Murakami C. The dominant measurement composition of the five-factor personality test. Jpn. J. Pers. 1997; 6: 2939.
  • 11
    Benner P. From novice to expert: Excellence and power in clinical nursing practice. Am. J. Nurs. 1984; 84: 1480.
  • 12
    Stern JA, Bremer DA, McClure J. Analysis of eye movements and blinks during reading: Effects of valium. Psychopharmacologia 1974; 40: 171175.
  • 13
    Van Geert P. Green, red and happiness: Towards a framework for understanding emotion universals. Cult. Psychol. 1995; 1: 259.
  • 14
    Ekman P, Friesen WV, O'Sullivan M et al. Universals and cultural differences in the judgments of facial expressions of emotion. J. Pers. Soc. Psychol. 1987; 53: 712717.
  • 15
    Shimonaka J, Nakazato K, Gondou K, Takayama M. NEO-PI-R. NEO-FFI Manual for the Japanese Version. Tokyo Shinri, Tokyo, 1999.
  • 16
    Archer D. How to Expand Your SIQ (Social Intelligence Quotient). M Evans & Co, New York, 1980.
  • 17
    Ellis SR, Stark L. Eye movements during the viewing of Necker cubes. Perception 1978; 7: 575581.