Additional Factors in Personal Space1
(Key words : personal space, vertical dimension, effect of color)
The purpose of this study was to examine the effects of vertical dimension ( sitting or standing) on personal space, and to obtain information on the use of color boards. The latter was concerned with considering the utility of colored communication board while interacting with a physically disabled person of linguistic impediment. A sitting position occupied less personal space compared to a standing one. Brown-colored boards encouraged students into the narrowest personal space and gave them a feeling of "settling down". This suggested that color had a specific meaning in communication through colored boards.
The term personal space was coined by Sommer (1959) who referred to the invisibly-bounded area by which we separate ourselves from others -areas of different size, depending on circumstances (Wolfgang, 1997). This issue is age old and still novel areas have many for further research. Since Tanaka (1973) who investigated the anisotropy of personal space, approach distance and approached distance were examined using experimental
(Tanaka, 1973 ; Shibuya, 1985 ; Suzuki, 1988) or projective methods (Shibuya, 1976 ; Shibuya, 1985, 1989 ; Kodama and Shindo, 1995).
With regard to the definition of personal space, Hayduck (1978) postulated a circular cylinder above waist and a frustum of a cone below the waist. No have researchers paid attention to the vertical point of view in personal space. We sometimes observed young persons sitting on the floor in close distance with each other compared to standing position in daily life situation. We attempted a consideration of this vertical point of view in Experiment I.
1 I wish to express my gratitude to Ms M. Oka, Ms Y. Kamada, and Ms N. Kohchi for their help in carrying out the experiments and analyzing the data.
Kiyofumi lcHIKA w A
In Experiment II, we dealt with the effects of color-boards on the personal space. We tried to obtain some preliminary information on communication board. Some physically disable people can communicate only with the aids of special communication boards printed with a rating scale, alphabets or letters, in a face-to-face situation. Our preliminary experi- ment suggested that some kinds of colors put someone's mind at ease while facing other through this board. On the other hand, a clear crystal board might make people conscious of glances of people on the opposite side of the board 2•
In an early stage in Japan, Tanaka (1973) reported the. anisotropy of personal space.
Tanaka ( 1973) insisted that an increasing degree of angle from the front side corresponded with a decreasing personal space. Wada (1999) dealt with four orthogonal directions. Wada ( 1999) showed that forward direction had significantly wider personal space compared to backward, left, and right sides. Ikeuchi and Murosaki (2000) indicated that forward and backward sides took significantly wider than left and right sides, and no difference was observed between forward and backward.
As suggested by Hayduck (1978), the effects of vertical dimension on personal space have not been focused on until now. When we observed young persons sitting on a floor, they usually gathered in close distance with each other. We called this "Unko-Suwari", i. e.
sitting crouched down on one's legs.
One of factors to affect the anisotropy of personal space would be eye contact. Suzuki ( 1988) did not obtain the anisotropy of personal space and he considered the reason being the possibility of eye contact for all directions. In Tanaka (1973) and Wada (1999), eye contact was possible only in the forward direction. On the other hand, Ikeuchi et al., (2000) found that the anisotropy of personal space had no relationship to eye contact in all directions.
This suggested that eye contact was not a necessarily conclusive factor for the anisotrophic structure of personal space. One of the factors conceived to affect the anisotrophy of personal space was a kind of personality or individual factor. As for the effect of anxiety on the personal space, Kodama and Shindo (1995) reported that high anxiety took wider distance from others than people with less anxiety. On the other hand, Shibuya ( 1989) considered that people with high anxiety could avoid their anxiety by moving closing to others.
We examined the effect of vertical dimension (standing or sitting position) for direction and anxiety on personal space.
· 2 Many thanks to the suggestion by K. Nakamura, associate professor, Kagawa University in this regard.
To explore the effect of vertical factors (sitting pos1t10n, standing position) with anisotropy (forward, backward, left, right), two within subject's variables of experimental design was adopted.
Fifty-six female undergraduate students of Kagawa University participated as subjects.
They were all unknown to the other 6 female confederates who were also undergraduate students of the same university. All of the students participated as subjects with their corrected eyesight wearing glasses or contact lenses.
Experimental situation :
Two white-tape measures of 2 cm width were attached on the floor in the experimental room (roughly 7mX 7m), and crossed each other at right angles in the center of the room.
The length of the tape was 3. 3 m for both forwards and backwards on both sides of the subject, and 3. 1 m for her left and right sides.
The confederates were trained in order to have a constant speed of gradual approach to the subject, and to be as unchanging as possible for their facial expression to the subject, and so forth.
Instruction for standing position :
The subjects were instructed to stand at the orthogonal point in a natural upright posture, and to make a signal with a motion of their hand at the beginning of their "feeling of embarrassment" to their approaching partner (confederate). After this signal by the subject, fine adjustments regarding the distance between the partner (that is, confederate and the subject) were allowed to the subject. The subject and the confederate were faced with each other maintaining their facial expressions as natural as possible. When the subjects faced the approaching confederates from behind them, the subjects were instructed to look towards the feet of the confederates who were instructed to approach slowly looking towards the subject's waist. When the confederates approached from the direction of left or right to the subjects whose eye direction was forwards, they moved sideways without any eye contact maintained forward eye direction. This meant that both of the pairs' visual line was the same direction to forward direction.
Kiyofumi lcHIKA w A
Instruction to sitting position :
The subjects were instructed to sit with their legs bent beneath them (sitting Japanese
"Seiza" style) using a special small chair without legs. The subjects faced the front at the intersection of axes that run at right angles to one another by means of two white-tapes in the experimental room. The confederate approached the subject slowly while keeping her knees on the floor. The subjects responded by raising their hand to indicate the beginning to recognized "embarrassed feeling" to their partner (confederate). The fine arrangement regarding this response was permitted just after the raising of the sign. When the direction was backwards, the subjects looked towards the knees of the confederates moving back their trunk of the body. While the confederates approached the subjects from their left or right sides, the confederates looked around the front side. This meant that both of the partners (subjects and the confederates) faced forwards. The experimenter stood on the left or right side of the subjects according to a counter-balanced consideration in order to avoid effects of the existence of experimenter.
(a) Check for the relationship of the subjects with the confederates: This was used in order to discard the subjects who were acquainted to their partner (confederates) with each other.
(b) State Anxiety: Twenty items with a 6 point rating scale were used (Shimizu and Imae, 1981).
(c) Trait Anxiety : This was measured by 20 items with a 6 point rating scale adopted from Shimizu et al. , (1981).
(d) Rating of first impressions of the partner: Twenty-three items were used to get the first impression of the subjects to the partner (confederates) with a 7 point rating scale (ex- cerpted from Wada, 1999).
(e) Interpersonal Attraction: Seven items with a 7 point rating scale were prepared to mea- sure interpersonal attraction to the partner (extracted from Miyoshi, 1988).
One of the subjects was discarded because of an incomplete rating scale. Consequently, we analyzed 55 subjects. To ensure the homogeneity of the confederates, a three way analysis of variance -confederate ( 6 persons) X position (standing, sitting) X direction (forward, backward, left, right) - was accomplished. The main effect of confederate and also interactions did not attain significance. As a result, we obtained a large degree of homogeneity among the confederates in this experiment.
Effects of position and direction on personal space :
An ANOV A of position (sitting position, standing position) X direction (forward, backward, left, right) was computed. Both of these two variables were within factors among subjects. The significant main effect of position was obtained, F(l, 54) =90. 52, P
<.001 (see Table 1).
The main effect of direction attained significance, F(3, 162) =42. 07, P
make clear the anisotropic structure, multiple comparisons were accomplished among the di- rections using Ryan's method.
Table 1 Means and SDs of personal space in each condition
"--- Position StandinE position Sitting )OSition
~Mean SD Mean SD
Forward 125.61 51.90 99.61 45.66
Backward 88.87 45.38 80.45 41.21
Left 88.20 40.84 66.29 38.49
Right 87.50 40.72 67.90 39.22
Total 97.55 44.71 78.56 41.15
Consequently, we obtained significant differences between the personal space for forward and backward (t (162) =7. 66, P
<.05), left (t (162) =9. 70, P
<.05), and right (t (162)
=9. 57, P
<.05). There was no significant evidence between left and right, and between backward and right or left.
A significant interaction of position with direction, F (3, 162) = 5. 53, P
<.0012, was attained. The simple main effects were found in sitting rather than the standing position for all directions: forward, F (1,216) =58. 02, P
< .0001; backward, F (1,216) =6. 08, P
<.01; left, F (1,216) =41. 20, P
<.0001; and right, F (1, 216) =32. 97, P
(see Figure 1) .
These results meant that the personal space in sitting position showed a narrower space than the standing one for all directions. The anisotropic structure of personal space was indicated in the shape of jutting out forward.
Standing position Sitting position
Figure 1. Interaction of vertical position with direction
Effects of personality on personal space:
We utilized the Trait-State Anxiety Inventory (STAI) for one of personality variables.
After calculating each mean score for these scales, we categorized high, middle and low groups by the aid of mean
+ .05 SD formula. We categorized three groups for each of the two anxiety scales as follows: for trait anxiety, high (n = 19), middle (n = 22), and low (n
=14), for state anxiety scale, high (n=l5), middle (n=23), and low (n=l7). An ANOVA of personality (high, middle, low) X position (standing, sitting) was computed to examine the effects of anxiety on personal space. The result of this analysis of variance indicated that the main effect of trait anxiety came up short with just slightly significant level, F (2, 52) =3. 15, P
<.051. Ryan's multiple comparison tests meant that significantly narrower personal space in the high anxiety group was observed than with the middle (P
<.018) or low anxiety group (P
<.052; see Figure 2). This suggested that regardless of the standing or sitting position, high anxiety persons showed a strong trend of having narrow personal space. This suggested that persons of high anxiety might reconcile their anxiety with approaching closely to other persons. The interaction of trait anxiety group with position and direction did not attained significance.
As for state anxiety, no significant results were obtained in the main effects and at any interactions.
100 - - - - 95
85 80 75 70
65 - - - -
High Low Middle
Figure 2. Effect of anxiety on personal space
Effects of interpersonal attraction on personal space:
On the basis of mean
+ .5 SD, the attraction for the partner was categorized into high (n=20), middle (n=22), and low (n=l3). A three way ANOVA of attraction (3) X position (2) X direction (4) revealed no significant differences in the main effect and interactions of attraction, except for main effects of position, direction and its interaction mentioned already.
In this experiment, attraction to the partner was not a significant variable for personal space.
We examined the effects of vertical position on the anisotropic structure, which operationally took four directions. The sitting position had a relation with narrow personal space compared to the standing position, regardless of the direction of anisotropic structure.
Hayduck (1978) proposed a hypothesis of personal space considering a viewpoint of vertical dimension. Hayduck (1978) showed a standing figure with a circular cylinder above waist height and a frustum of a cone below the waist. The truncated cone meant the narrowest region of personal space around the ankle region.
We showed that the sitting position occupied significantly narrower personal space compared with the standing position, regardless of the four directions. With regard to the anisotropism of personal space in standing positions, forward took significantly broader personal space than backwards, left, and right directions. On the other hand, among backwards, left, and right directions, we did not obtain significant differences. As for the sitting position, significant differences were obtained between every direction, except for
between left and right. These results for anisotropism, were not consistent with Wada (1999), indicating significant results shown in the following inequality sign: forward
left and right.
Regarding effects of personality on personal space, high-scored subjects for the trait anxiety indicated a strong trend (P
< .051) of narrow personal space both in standing and sitting positions. Kodama et al. , (1995) reported that high-scored students in the trait anxiety showed a wide personal space compared to the low ones, using STAI (State-Trait Anxiety Inventory). On the other hand, Shibuya (1989) showed the opposite results using MAS (Manifest Anxiety Scale). Shibuya (1989) explained that the high anxiety people wanted to gather with people and that this caused narrow personal space. However, we should be reminded that these two studies took projective method to measure the personal space. Direct measuring of personal space in our study obtained a result in accordance with Shibuya (1989).
We observed no relationships of attraction to the partner with personal space. The mean attraction score combining 7 items took a score range from 2. 64 to 6. 08 on the 7 point rating scale. However, Mean and SD were 5. 15 and . 67 respectively. This meant that most of subjects expressed high attraction to the partner and narrow score range. The reason of this result might be due to this skewed distribution.
To obtain multivariate relationships with personal space, we accomplished multivariate analyses. Criterion variable was the personal space of the averaged one for all directions and the explanatory variables were selected one item from each of the state anxiety, first impression to the partner, and the interpersonal attraction scale.
In the situation of standing position, following multiple regression equation was obtained:
= + .181 X (unpleasantness) - . 313 X (unconfident) - . 103 X (reli- able). F (3,216) = 11. 26, P
<.0001, R = . 368, R2 = . 135. As for the explanation variables, unpleasantness was extracted from state anxiety scale, "unconfident" was selected from the item of the first impression for the partner (confederate), and "reliable" was an excerpt from one of items in the interpersonal attraction for the partner.
In the sitting situation, we obtained the following multiple regression equation: Personal space
= + .192 X (unrelaxed) - . 273 X (unconfident) - . 155 X (reliable). Here, F (3,216) = 12. 52, P
<.0001, R = . 385, R2 = . 148.
In the standing situation, getting an impression of "unconfident" (P
<.0001) and reli- ance (P
< .110) towards the partner caused narrow personal space. On the other hand, subjects' unpleasant feelings in the state anxiety (P
<.005) caused subjects stand apart from their partner. In the sitting condition, causing narrow personal space was explained by "un- confident" (P
< .0001) and also reliant impressions (P
< .014) to the partner. Taking wide personal space was explained by unrelaxed feelings (P
<.002) in the situation.
In Experiment I, subjects responded to the approached distance by the partner (confederates). The subjects could adjust the distance between the partner and themselves in accordance with their degree of feeling comfortable. We can usually avoid the space between oneself and the other while maintaining uncomfortable distances in daily life. However, some sort of physically handicapped persons cannot adjust this space in any situation. In Experiment II, we assumed a situation of communication with the help of see-through communication board. We can communicate with the help of a kind of board to .some people with speech impediment. With the aid of the communication board, we can understand the response by way of confirming the turning direction of one's gaze on the board. In this case, the communication board would be printed with a kind of scale, figure and so forth.
Little attention and no research have been given to the point of effects of colors on the communication board. As a preliminary study, we examined whether colors caused different effects on the personal space. The purpose of Experiment II was to obtain exploratory information with regard to communicational situation through colored crystal-clear boards, using students. What kinds of colors would be useful while we attempted to interview with the situation utilizing crystal clear board?
The personal space through the colored crystal-clear boards was measured within subject condition. The color boards used in this study were clear, orange, sky-blue, brown, lemon, and smoky-grey. These were acrylic crystal-clear color boards with 32 cm (length) X 18 cm
( width) X 2 mm (thickness). These colors used were decided by preliminary experiments.
The fifty-six female subjects ( one was discarded as described in Experiment I) and the 6 female confederates were the same as those who had participated in E?(periment I. The experimental situation was same as Experiment I, except for utilizing color boards which were held up by the subjects in front of their faces.
The following instructions to the subjects and the confederates were prepared: "Please stand holding a crystal-clear colored board set around 10 cm in front of your face and facing the front of the orthogonal intersection of two measuring tapes." This distance of 10 cm apart from face was decided by preliminary experiment. To the partner (confederate): "Please stand facing to the subject on the terminal point of the measuring tape. Please slowly approach the
Kiyofumi lcHIKA WA
subject moving along the measuring tape. " The subjects were instructed to look at their partner through the colored crystal-clear board, and to make a signal with a motion of their hand at the beginning of their embarrassment towards their partner. Just after this signal, the subjects were allowed to adjust the point of distance between the confederate and themselves.
During the experiment, the subjects and the confederates were instructed to keep their facial expressions as natural as possible.
Color preference was measured through 5 items on a 7 point rating scale. The personal- ity scale of anxiety was utilized from Experiment I.
We analyzed the 55 subjects mentioned in Experiment I. A two-way ANOVA of confederates (6 students) X colors (clear, orange, sky-blue, brown, lemon, smoky-grey) was accomplished to confirm the homogeneity of the confederates. The main effect of con- federate was not obtained and an interaction was weak trend, F (25, 245) = 1. 34, P
These results confirmed the homogeneity of the confederates.
Effects of co/ors on personal space:
An one-way ANOV A ( within subject) for personal space through 6 color-boards in ad- dition to the result without the board obtained in the Experiment I, was computed with the use of a mean for each condition. The personal space through boards (M = 116. 66, SD = 49. 80) attained significance, F (5,270) =2. 29, P
<.046, compared to the condition without board (M = 125. 61, SD = 52. 38).
(cm) 123 122 121 120 119 118 117 116 115 114 113 112
Clear Sky-Blue Orange Brown
Figure 3. Effect of colors on personal space
Another one-way analysis of variance was calculated to clarify the effects of six colors on the personal space. The main effect of colors attained significance, F (5,270) =2. 29, P
<.046. Multiple comparisons were accomplished by means of Fisher's method (see Figure 3) . The lemon-colored board occupied a significantly broader personal space compared to orange (P
<.018), sky-blue (P
<.037), brown (P
<.002), and smoky-gray (P
Preference of color board:
An ANOV A was accomplished to inquire into the differences among color preference.
The significant difference among colors was obtained, F (4, 216) = 14. 27, P
Fisher's multiple comparison attained the following significant differences: sky-blue was pre- ferred than brown (P
<.0001), lemon (P
<.0001) and smoky-grey (P
<.0001); orange was preferred over brown (P
<.014), lemon (P
<.0004) and smoky-grey (P
brown was preferred over smoky-grey (P
< .008) .
(Preference) 5.8 5.6 5.4 5.2 5 4.8 4.6 4.4 4.2 4 3.8
Clear Sky-Blue Smoky-Grey
Figure 4. Preference for colors
Figure 4 shows these results. The degree of personal space through crystal-clear colored boards did not correspond with the preference of the color. It signifies that the preference for color was independent of the personal space or taking distance from the other person.
Experiment II inquired into the effects of colors on the personal space as a preliminary attempt. Some people might be able to communicate with the help of special communication board in face to face procedures. We assumed a situation that would enable us in face to face situations to get information through a crystal-clear board depicted with rating scale, alphabets or letters. Some kinds of physically handicapped interviewees could only convey their
expressions through such a tool as the clear board mentioned above.
One of the evocative results indicated that using a board regardless any kinds of color took narrow personal space than one without such a board. This suggested that facing others without a board made people nervous to make the eye contact, compared with using such a board regardless of the color. The anxiety caused by eye contact would be softened by the existence of a board between people.
The narrowest personal space was observed for the brown crystal-clear board. On the other hand, the widest personal distance was indicated in the situation of lemon colored board.
Accordingly, we accomplished an analysis of variance for the degree of settling down through the color boards. The significant main effect of the degree of settling down was obtained, F (5, 270)
=15. 20, P
<.0001 (see Figure 5). Multiple comparisons by Fisher's method attained the following significant differences regarding the degree of "settling down": brown
<.0001), orange (P
<.0001), sky-blue (P
<.042) and lemon (P
4.4 4.2 4
3.8 3.6 3.4 3.2
Clear Sky-Blue Lemon
Orange Brown Smoky-Grey
Figure 5. Degee of settling down for each color
It should be suggested, from what has been described above, that a kind of color board (for example, brown) would be useful aid to communicate or get information in face to face situations. It would be worthwhile examining the actual and practical applications more closely in the foreseeable future. To obtain more detailed information regarding communica- tion with boards, further studies would be taken for aged people, handicapped people and so forth, other than students.
Aono (2003) reviewed personal space with a focus on gender concerns. In both of Experiment I and Experiment II, we dealt with only female subjects and female pairs (i. e. , confederates). Gathering male subjects and male confederates was not easy matter, however, gender differences were still one of non-negligible variables that affected personal space.
Aono, A. (2003). Gender Differences in Interpersonal Distance: From the View Point of Oppression Hypothesis. The Japanese Journal of Experimental Social Psychology. 42, 201-218. (In Japanese with English title and English abstract)
Hyduck, L. A. (1978) . Personal Space: An Evaluative and Orienting Overview, Psychological Bulletin, 85, 117-134.
Ikeuchi, M., and Murosaki, M. (2000). A study of Personal Space. Graduation thesis of Kagawa University. (Unpublished) ( *)
Kodama, M., and Y. Shindo, (1995). The Effect of Trait Anxiety on Personal Space. Waseda Studies in Human Sciences, 8, 15-24. (In Japanese with English abstract)
Miyoshi, K. (1988). Effects of Similarity of Need for Uniqueness on Interpersonal Attraction.
Graduation thesis of Kagawa University. (Unpublished) ( *)
Shibuya, S. (1976). A Fundamental Study of Social Space. The Japanese Journal of Psychology, 47, 119-128. (In Japanese with English title and English abstract)
Shibuya, S. (1985). A study of the shape of personal space. Bulletin of Yamanashi Medical Univer- sity, 2, 41-49. (In Japanese with English title and English abstract)
Shibuya, S. (1989). Relationship between MAPS and MAS: -Analysis in Relation to Personal Space-. Bulletin of Yamanashi Medical University, 6, 29-33. (In Japanese with English title and English abstract)
Shimizu, H., and K. Imae, (1981). State-Trait Anxiety Inventory for Japanese. Japanese Journal of Educational Psychology, 1981, XXIX, 348 - 353. (In Japanese) ( *)
Sommer, R. (1959). Studies in personal space. Sociometry, 22, 247-260.
Suzuki, M. (1988). A Basic Study of Personal Space. Waseda Studies in Human Sciences, 1, 23 - 29. (In Japanese with English title and English abstract)
Tanaka, M. 0973). The Anisotropic Structure of Personal Space. Japanese Journal of Educational Psychology, XXI, 223- 232. (In Japanese with English title and English abstract)
Wada, Y. (1999). Personal Space in dyads. Graduation thesis of Kagawa University. (Unpub- lished) ( *)
Wolfgang, A. (Ed.) (1997). Nonverbal Behavior-Perspectives, applications, lntercultural Insight.
Seattle: Hogrefe & Huber Publishers.
( *) Translated into English by the author.