تفاوت های جنسیتی در تصویرسازی

The influence of gender on several measures of mental imagery was assessed using performance tests and imagery questionnaires. A group of 202 female and male university undergraduates were administered three performance tests, and three imagery questionnaires. The performance tests were the Measure of the Ability to Form Spatial Mental Imagery (MASMI), the Spatial Scale of the Primary Mental Abilities (PMA), the Mental Rotation Test (MRT), and the Measure of the Ability to Rotate Mental Images (MARMI). The questionnaires were: The Object-Spatial Imagery and Verbal Questionnaire (OSIVQ), the Vividness of Visual Imagery Questionnaire-2 (VVIQ-2), and the Betts’ Questionnaire Upon Mental Imagery (Betts’ QMI). Men obtained higher scores than women on the performance tests. No significant gender differences were observed on the imagery questionnaires. Women obtained higher scores than men on the Object scale of the Object-Spatial Imagery and Verbal Questionnaire. The results are discussed and new lines of research are proposed.

Numerous studies have assessed gender differences in mental imaging ability (see Richardson, 1994, for a review), but the findings remain inconsistent primarily due to the wide array of types of imagery (Campos, 1998a, Horowitz, 1983 and Richardson, 1977), and the disparity between tests that are intended to measure the same mental imaging ability yet fail to show correlations from one test to another (Campos, 1998b, Campos, 2009 and Campos, 2012). Thus, any reference concerning gender differences and imaging ability must take into account the type of imagery and the measurement instrument used (Campos, Pérez-Fabello, & Gómez-Juncal, 2004). The most extensively used instruments for assessing an individual’s imagery ability are imagery questionnaires, and performance tests. The most widely used questionnaire is the Vividness of Visual Imagery Questionnaire (VVIQ; Marks, 1973 and McKelvie, 1995). No correlations or nonsignificant correlations have been found between imagery questionnaires and performance tests (Burton and Fogarty, 2003, Campos, 2009, Campos, 2012 and Ernest, 1977). Ernest (1983) used three measures of spatial-imagery: the Space Relations (Bennett, Seashore, & Wesman, 1947), the Minnesota Paper Form Board (Likert & Quasha, 1941), and Flags (Thurstone & Jeffrey, 1956), and three self-assessments of imagery: the Betts’ Questionnaire Upon Mental Imagery (Sheehan, 1967), the Test of Visual Imagery Control (Richardson, 1969), and the Individual Differences Questionnaire (Paivio, 1971). Correlations between performance tests and questionnaires ranged between .00 and −.14. Correlations between performance tests ranged from .51 p < .001 to .66 p < .001, and correlations between questionnaires ranged from .28 p < .001 to .83 p < .001. Burton and Fogarty (2003) correlated the Card Rotation Test ( Ekstrom, French, Harman, & Dermen, 1976), the Cube Comparison Test ( Ekstrom et al., 1976), and the Spatial Relations Test ( Thurstone & Thurstone, 1962/2002) with the Vividness of Visual Imagery Questionnaire (VVIQ, Marks, 1973), and the Betts’ Questionnaire Upon Mental Imagery ( Sheehan, 1967), and found that correlations ranged from .12 to .17. Correlations between performance tests ranged from .58 p < .01 to .77 p < .01, and correlations between imagery questionnaires ranged from .32 p < .01 to .57 p < .01. Blazhenkova and Kozhevnikov (2009) obtained a correlation of −.03 between the Mental Rotation Test ( Vandenber & Kuse, 1978) and the Vividness of Visual Imagery Questionnaire ( Marks, 1973). Similar results were reported by Campos, 2009 and Campos, 2012 and Dean and Morris (2003) for correlations between performance tests and imagery questionnaires, correlations between performance tests, and correlations between questionnaires. As for the scales of the Betts’ Questionnaire Upon Mental Imagery (Sheehan, 1967), correlations ranged from .44 p < .001 to 83 p < .001 for women, and .38 p < .001 to 82 p < .001 for men ( Ernest, 1983). Correlations between the Object-Spatial Imagery and Verbal Questionnaire scales (OSIVQ; Blazhenkova & Kozhevnikov, 2009) were weak, and only one was significant. The correlation between the object imagery scale and the spatial scale was −.03, between the object imagery scale and the verbal scale .12, and between spatial and verbal scales −.18, p < .05 ( Blazhenkova & Kozhevnikov, 2009). Few studies have assessed the correlations between tests according to gender, this study analyzed the correlations between tests in a group of women and a group of men to determine any significant differences in correlations between both genders. The results obtained for women and men on the image rotation task were rather consistent regardless of the test that was used (Blajenkova et al., 2006, Blazhenkova and Kozhevnikov, 2010, Campos, 2012, Dean and Morris, 2003, Parsons et al., 2004 and Vandenber and Kuse, 1978). Parsons et al. (2004) used 60 participants recruited from undergraduate and graduate schools who were administered the Mental Rotation Test (MRT; Vandenber & Kuse, 1978). MRT scores were significantly higher for men. Likewise, other studies using the same test have found significantly higher scores for men (Blajenkova et al., 2006, Blazhenkova and Kozhevnikov, 2010, Dean and Morris, 2003 and Vandenber and Kuse, 1978), and no study found higher scores for women. As for the Spatial Scale of the Primary Mental Abilities (PMA; Thurstone & Thurstone, 1962/2002), which is a mental imagery rotation test, significant gender differences were observed with men obtaining higher scores (Campos et al., 2004, Kail et al., 1984 and Stericker and LeVesconte, 1982). A recently designed imagery rotation test is the Measure of the Ability to Rotate Mental Images (MARMI: Campos, 2012), of which only one study has been published to date (Campos, 2012) that found significant differences between women and men, with scores being higher for the latter. Another recently published test is the Measure of the Ability to Form Spatial Mental Imagery (MASMI, Campos, 2009 and Campos, 2013), which is designed to measure the ability to form spatial imagery. The two studies that used the test to measure gender differences in university undergraduates found no significant differences. In contrast, in another spatial imagery test i.e., the Paper Folding Test (PFT, Ekstrom et al., 1976), Blazhenkova and Kozhevnikov (2010) found significantly higher scores in men than women. The Vividness of Visual Imagery Questionnaire-2 (VVIQ-2, Marks, 1995), is the latest version of the Vividness of Visual Imagery Questionnaire (VVIQ, Marks, 1973). Of the few studies that have applied the VVIQ-2, none have found gender differences, whereas the numerous studies that have applied the original version of the Vividness of Visual Imagery Questionnaire have reported contradictory results. McKelvie (1995), review of the literature revealed no significant gender difference on the VVIQ whereas Richardson (1995) review of the literature concluded women obtained slightly higher scores than men on this test. Likewise, studies examining the differences between women and men on the Betts’ Questionnaire Upon Mental Imagery (Betts’ QMI, Sheehan, 1967) were inconsistent i.e., White, Ashton, and Brown (1977) found higher scores for women than for men whereas other authors (Campos and Pérez-Fabello, 2005 and Sacco and Reda, 1998) found no significant differences, but no study found higher scores for men than for women (see Campos & Pérez-Fabello, 2005, for a review). Moreover, Sheehan (1967) examined gender differences in the seven sensory modalities of the Betts’ QMI, and reported no significant differences in any of the modalities. The Object-Spatial Imagery and Verbal Questionnaire (OSIVQ) was designed to distinguish between Object-Spatial-Verbal cognitive styles (Blazhenkova & Kozhevnikov, 2009). The test consists of a verbal scale (individuals scoring high on this scale have verbal cognitive styles); and two mental imagery scales where “object visualizers” are characterized by their ability to process the form, shape, color, and brightness of individual objects, and “spatial visualizers” who are characterized by their ability to process spatial relations among objects, moving objects, and objects in space (Blajenkova et al., 2006). Blazhenkova and Kozhevnikov (2009) found that men obtained higher spatial scores than women whereas women obtained higher object imagery scores than men, but no significant differences in terms of gender were found on verbal scores. Similar gender differences have been reported in children (Blazhenkova, Becker, & Kozhevnikov, 2011). Other studies using only the object imagery scale, and spatial scale found men obtained higher spatial scale scores than women whereas women obtained higher object imagery scores than men (Blajenkova et al., 2006 and Blazhenkova and Kozhevnikov, 2010). Previous studies have shown that each test casts different results, which is why when the mental imagery of men and women is compared, different results are obtained depending to some extent on the type of test that was used. Thus, the aim of this study was to obtain correlations among the tests, when they were scored by men and women i.e., the imagery vividness questionnaires, and the spatial and image rotation tests, and also to assess gender differences in each of the measures of imagery, and among the different correlations.

First, Pearson’s correlation was computed to analyze the correlations between the different measures under study. The results for the correlations shown in Table 1 reveal no strong correlations: the highest correlation (.52) was between the MASMI and the Spatial Scale of the PMA, and the lowest correlations (r = .00) were between the PMA and the Verbal Scale of the OSIVQ, and between the PMA and Betts’ QMI. Table 1. Correlations between measures of imagery. PMA MRT MARMI MASMI VVIQ-2 BETTS’ OBJECT SPATIAL MRT .50⁎⁎ MARMI .49⁎⁎ .51⁎⁎ MASMI .52⁎⁎ .46⁎⁎ .51⁎ VVIQ-2 .06 .05 .12 .08 BETTS’ .00 .01 .06 .03 −.48⁎⁎ OBJECT −.06 −.15⁎ −.11 −.08 .43⁎⁎ −.43⁎⁎ SPATIAL .39⁎⁎ .43⁎⁎ .36⁎⁎ .45⁎⁎ .09 .03 −.17 VERBAL .00 −.05 −.05 −.09 .02 .01 −.06 −.08 ⁎ p < .05. ⁎⁎ p < .01. Table options The correlations between tests for the group of women and the correlations between tests for the group of men are shown in Table 2. To determine significant differences in the correlations between women and men, a contrast of hypothesis between correlations for independent samples was undertaken (see Table 2). The analysis showed differences in the correlations between women and men in the performance test, and between the Betts’ QMI and the Objectscale of the OSIVQ. In all of the cases where significant differences were observed, men obtained higher correlations. Table 2. Inter-test correlation matrix for males (above diagonal), and females (below diagonal), and differences among correlations (NS = no significant, S = p < .05). PMA MRT MARMI MASMI VVIQ-2 BETTS’ OBJECT SPATIAL VERBAL PMA .58⁎⁎ (S) 51⁎⁎ (S) .52⁎⁎ (NS) .04 (NS) −.14 (NS) .02 (NS) .32⁎⁎ (NS) −.01 (NS) MRT .31⁎⁎ (S) .60⁎⁎ (S) .51⁎⁎ (S) .09 (NS) −.13 (NS) −.04 (NS) .34⁎⁎ (NS) −.08 (NS) MARMI .39⁎⁎ (S) .23⁎ (S) .48⁎⁎ (NS) .12 (NS) .06 (NS) −.07 (NS) .31⁎⁎ (NS) −.08 (NS) MASMI .45⁎⁎ (NS) .30⁎⁎ (S) .46⁎⁎ (NS) .01 (NS) .03 (NS) .01 (NS) .41⁎⁎ (NS) −.11 (NS) VVIQ-2 .07 (NS) −.01 (NS) .13 (NS) .15 (NS) −.52⁎⁎ (NS) .48⁎⁎ (NS) .06 (NS) .03 (NS) BETTS’ .08 (NS) .10 (NS) .05 (NS) .01 (NS) −.46⁎⁎ (NS) −.60⁎⁎ (S) .06 (NS) .01 (NS) OBJECT −.01 (NS) −.16 (NS) −.03 (NS) −.05 (NS) .39⁎⁎ (NS) −.29⁎⁎ (S) −.03 (NS) −.09 (NS) SPATIAL .31⁎⁎ (NS) .35⁎⁎ (NS) .22⁎ (NS) .36⁎⁎ (NS) .12 (NS) .01 (NS) −.21⁎ (NS) −.18 VERBAL −.04 (NS) −.07 (NS) −.08 (NS) −.13 (NS) .01 (NS) .02 (NS) .02 (NS) −.07 (NS) ⁎ p < .05. ⁎⁎ p < .01. Table options In order to determine difference in test scores between genders a Multivariate Analysis of Variance (MANOVA) was performed with gender as independent variable, and the scores of the tests as dependent variables. The means and standard deviations for each group are shown in Table 3. The MANOVA revealed significant differences between groups, Wilks’ lambda = .78, F(9, 192) = 5.53, p < .001, power = 1, View the MathML sourceηp2 = .22. Later, univariate analysis revealed significant differences between males and females in the Spatial Scale of PMA, F(1, 200) = 14.50, p < .001, power = .97, View the MathML sourceηp2 = .07. On the Spatial Scale of PMA men obtained higher scores than women. Moreover, men obtained higher scores on the MRT image rotation performance test than women, F(1, 200) = 22.92, p < .001, power = 1, View the MathML sourceηp2 = .11. Table 3. Means and standard deviations of measure of imagery. Measures Females Males Differences M SD M SD F p PMA 24.99 12.57 32.23 13.00 14.72 .001 MRT 7.04 4.43 10.31 4.86 22.92 .001 MARMI 7.39 7.18 13.39 11.93 18.57 .001 MASMI 20.30 11.52 26.82 13.90 12.32 .001 VVIQ-2 3.72 .55 3.71 .66 0.75 .989 OBJECT 3.74 .50 3.53 .60 6.90 .009 SPATIAL 2.32 .54 2.80 .61 32.21 .001 VERBAL 2.81 .44 2.89 .61 1.06 .305 BETTS’ 2.54 .80 2.60 .79 0.23 .630 Table options As for the MARMI image rotation test, significant gender differences were found with men obtaining higher score than women, F(1, 200) = 18.57, p < .001, power = .99, View the MathML sourceηp2 = .09. Similarly, men scored higher than women in spatial imagery, MASMI, F(1, 200) = 12.32, p < .001, power = .94, View the MathML sourceηp2 = .06. Whereas men obtained higher scores than women on the spatial imagery and image rotation performance tests, the results for the questionnaires did not show the same pattern. On the VVIQ-2, the measures of image vividness showed no significant differences in terms of gender, F(1, 200) = .01, p = .99, power = .05, View the MathML sourceηp2 = .01. Moreover, no significant gender differences were observed on the Betts’ QMI, F(1, 200) = .23, p = .63, power = .08, View the MathML sourceηp2 = .01. Gender differences were found on two scales of the OSIVQ i.e., women obtained higher scores than men on the Object scale of the OSIVQ, F(1, 200) = 6.90, p = .009, power = .74, View the MathML sourceηp2 = .04, whereas men obtained higher score on the Spatial scale, F(1, 200) = 32.21, p < .001, power = 1, View the MathML sourceηp2 = .15. As for the Verbal scale, no significant gender differences were found, F(1, 200) = 1.06, p = .31, power = .18, View the MathML sourceηp2 = .06.