Several researchers have investigated the phenomenon that those who are high in anxiety trait have a bias effect in their interpretation of ambiguous stimuli and situations. Eysenck, Macleod, and Mathews (1987) suggested that high anxiety is associated with interpreting ambiguous stimuli in a threatening fashion. They also suggested that anxious participants concentrate their attention more on sites vacated by emotionally negative stimuli. The results of these studies, as a whole, showed that anxiety – though it is assumed to evoke – also has a negative memory bias, and should be assumed to have greater impact on basic levels of processing (Stuchlikova, 2000). Eysenck (1992) proposed that the hypervigilance of high anxiety individuals involves a high rate of environmental scanning, a broadening of attention prior to the detection of threat-related or task-relevant stimuli, and a narrowing of attention when such stimuli are being processed.

The literature does provide some support for the assumption that aroused anxiety facilitates an analytical mode of processing (Cunningham, 1988; Kuhl, 1983; Salovey & Rodin, 1985; Sedikides, 1992). There also are some inconsistent findings, such as those gained by Cunningham and Ashley (2002), who examined the relationship of performance between intensity and direction of trait anxiety among students participating in a beginners’ collegiate class of golf, and found the participants’ scores to be dichotomized into those with negative perceptions of anxiety (debilitating interpreters) and those with positive perceptions (facilitating interpreters). As with attention orientation, habitual preferences for vigilance and cognitive avoidance are explained by the constructs of intolerance of uncertainty (vigilance) and intolerance of emotional arousal (avoidance).

Most researchers have indicated that anxious patients selectively attend to threatening stimuli. Reviewing available research in the field, Stewart, Conrod, Gignac, and Pihl (1998) designed two studies to assess whether high anxiety sensitive (AS) participants selectively process threat cues pertaining to the feared catastrophic consequences of anxiety, and to examine potential gender differences in selective processing of such threat cues among high versus low AS participants. They found significant gender differences in the principal feared consequences for AS participants, in that females with high AS scores selectively processed only physical threat cues relative to low AS females, and males with high scores of AS selectively processed only cues pertaining to social and psychological threat relative to low AS males. These differences among nonclinical high AS men and women were found to, respectively, resemble Mathews (1988) who reported that antinational bias towards threat may serve to maintain an anxiety state, because anxious individuals would be more likely to detect any potential source of danger in the environment.

Sinha and Sinha (1967) studied the effect of frustration on the extent of the Mueller-Lyer illusion of participants scoring high and low on the anxiety scale. Their results revealed that the mean error of the magnitude of illusion for the low anxious group of participants was greater than the mean error of the high anxious group of participants. Thus, participants scoring low on the anxiety scale were more susceptible to the Mueller-Lyer illusion than were those scoring high.

In a study in which anxiety-related bias in semantic activation was examined when processing threatening or neutral homographs, Richards and French (1992) stated that high trait anxiety participants were more likely to opt for the threat-related spelling compared to low-anxiety participants. This general finding was also supported by the finding of Mathews, Richards, and Eysenck (1989), in a study comparing anxious patients with a group who had recovered from clinical anxiety.

Since this study was also concerned with gender differences in perception of ambiguous figures, it is necessary to review also some of the studies devoted to investigating the relationships between gender differences and perception of the Mueller-Lyer illusion. Shehab (2001) found significant gender differences between Kuwaiti boys and girls aged 6 to 12 in perception of both Mueller-Lyer (Mueller-Lyer, 1889) and Horizontal-Vertical illusions. Boys exhibited higher illusion error scores than did girls. Walters (1942), Faisse and Vautrey (1956), and Stewart (1973), found significant sex-related differences in the perception of the Mueller-Lyer illusion, indicating that boys were more susceptible to the Mueller-Lyer and the horizontal-vertical (HV) illusions from ages 6 to 13 except the 8- and 9-year-olds, while girls were more susceptible from age 14 to 18. It was also found that boys were more susceptible to the H-V illusion than were girls from age 6 to 19 years old. Shehab (1992) reported in another study that girls were more susceptible to the Mueller-Lyer illusion in a Scottish sample, and that boys were more susceptible to the Horizontal-Vertical illusion in a Kuwaiti sample.

Brosvic, Dihoff, and Fama (2002) studied age-related differences as transition points in the sensitivity and responsivity indices for the horizontal-vertical illusion on 1- to 3-year-old children. They found a significant age-related trend for each illusion through the inverse relation between sensitivity and responsivity. The youngest participants had the lowest values on the sensitivity index and the highest values on the responsivity index, which the oldest participants had the highest values on the sensitivity index and the lowest values on the responsivity index.

The aim in the present study was to investigate the relationship between anxiety, gender, and perception in cognitive attention, such as in the Mueller- Lyer illusion. Studies of high anxiety have revealed evidence that highly anxious participants have a lack of integration, attention, and self-confidence generated by tension due to high anxiety. Thus, high anxiety can influence the perception of ambiguous stimuli, such as in the Mueller-Lyer illusion. We chose to examine the association between anxiety and gender in perceptual judgments related to the processing of visual stimuli (the Mueller-Lyer illusion).

Mueller-Lyer and Horizontal-Vertical illusions were used to assess perceptual judgment in processing a visual stimulus. Both forms of the Mueller-Lyer (ML) illusion were represented on the same board; the one with the incoming left shaft (MLL) and the one with the outgoing right shaft (MLR). The T form of the horizontal-vertical (HV) illusion was also used in this study. The bisected line was constant, while the other line was movable to the left side (HVL) and to the right side (HVR). The participant was asked to sit opposite the experimenter in the laboratory room, and to move the movable line until he/she felt that both lines (ML and HV illusion) were equal. Each participant was given only 10 trials for each side to minimize the practice effect. In half of the observations, the variable stimulus was on the participant’s left side (MLL and HVL), and in the other half on the right side of the participant (HVR and HVR). In half of these sets, MLL, HVL, and MLR, HVR, the trials were given with the variable stimulus placement obviously too great so that the movement of the arrow would be inwards. In the other half, the variable stimulus was set obviously too small so that the movement was outward. Figure 1 shows the direction of MLL and MLR. Figure 2 shows the direction of HVL and HVR.

Anxiety was assessed on the basis of the Kuwait University Anxiety Scale, (Abdel-Khalek, 2000). The scale comprises 20 brief statements, on a 4-point scale, ranging from 1 = rarely to 4 = always. Factor analysis yielded three highly loaded factors of cognitive/affective, behavioral/subjective, and somatic/anxiety, with moderate interfactor correlations, ranging from .27 to .74. Reliabilities ranged from .88 to .92 (alpha) and between .70 and .93 (test-retest), denoting good internal consistency and stability. Criterion-related validity of the scale ranged between .70 and .88 (5 criteria), while the loadings of the scale, on a general factor of anxiety, were .93 and .95 in two-factor analyses.

Although there were no significant relationships between anxiety and gender on ML and HV illusion perception, a close look at the data for the first half of the left (L) trials, and the other half of the right (R) trials of both illusions, using analyses of variance (ANOVAs) to evaluate group differences of the effect of anxiety and gender on perception of the six variables (ML, MLL and MLR, HV, left HVL and HVR), showed that there were some significant relationships between anxiety on perception of the illusion among both male and female participants for left and right sides. There were significant differences on the variables MLL for the male sample; the mean error of magnitude of the illusion for the high-anxious group was greater than the mean of the low-anxious group (F = 3.99, p > .05) and HVL (F = 3.72, p > .05). This implied that males with a higher score of anxiety had a higher mean illusion error score (MLL M = -88.00 mm, SD = 79.9 mm, HVL M = -147.00 mm, SD = 16.82 mm) than males with a low anxiety (MLL M = -188.00 mm, SD = 79.96, HVL M = -147.00 mm, SD = 16.82 mm) than males with low anxiety (MLL M = 121.98 mm, SD = 59.92 mm, HVL M = -59.77 mm, SD = 129.63 mm). Table 4 shows the means and standard deviations of males for MLL and MVL.

We found no significant relationship between gender and the Mueller-Lyer and Horizontal-Vertical illusion perception. This means that males and females perceived the Mueller-Lyer and horizontal-vertical illusion (for left and right sides together) with no differences in illusion error scores. The nonsignificant differences found for the Mueller-Lyer and horizontal-vertical illusion (as a total) in this study could be explained in terms of the perspective theory. The ecological hypothesis of an open area assumes that people in an open area, like open fields or deserts (such as in the State of Kuwait), where the apparent distance extends in front of them, would be more susceptible to the Mueller-Lyer and horizontal-vertical illusion than others would be. However, this result is inconsistent with Shehab’s (2001) findings, which showed significant gender differences between boys and girls (aged between 6 and 12 years old) on Mueller-Lyer and horizontal-vertical illusion perception. This inconsistency may be due to the fact that Shehab used a sample of children aged 6 to 12 years, whereas in this study we used graduate students.

The relationships between anxieties on the perception of illusion in this study were not clear. There were no significant differences between the groups with low and high anxiety in the perception of the Mueller-Lyer and horizontal- vertical illusions (as a total for both sides) respectively. However, there were some significant differences in anxiety on the other variables among males, such as MLL and HVL, and among females, HVR and HVL. The males with higher anxiety scores had higher mean error illusion on MLL and MVL (underestimation of the error of illusion). This implied that a male participant with high anxiety scores had a higher mean magnitude error of illusion on these variables than did a male with low anxiety. The results also showed that a female participant with high anxiety had a higher magnitude of illusion error score (underestimation of the error of illusion) of the HVR and HVL illusion than did the female with low anxiety scores. This finding is consistent with the results of the male data. The males with high anxiety had higher magnitude mean errors in the illusion; similarly, the females with high anxiety had higher mean errors in the illusion.

This finding is consistent with that of Sinha and Sinha (1967), who found that participants with low scores of anxiety were more susceptible to illusion than those scoring high, and with that of Pandey, Roy, and Pandey (1986) who found an effect of frustration on the Mueller-Lyer illusion, using 30 low and 30 high anxiety undergraduates, in that frustration enhanced the illusion in highly anxious participants and reduced it in low anxiety participants. The present finding is inconsistent with that of Stuchlikova (2000), who found that the magnitude of perceptual distortion was not closely related to trait anxiety. Therefore, increasing the level of anxiety for both males and females will increase the error in perceiving Mueller-Lyer, on the other hand, decreasing the level of anxiety will decrease the error in perceiving Mueller-Lyer. The findings of the present study could be explained by relationships of anxiety with the perception of the Mueller-Lyer and Horizontal-Vertical illusion in both males and females. Although the magnitude of perceptual illusion error scores was not closely related to trait anxiety on all variables, the visual tasks were used here to assess high and low anxiety and anxiety proneness was shown to influence perceptual learning. Highly anxious individuals seemed to need more rigidly tight criteria to judge or estimate the illusion and high anxiety is often held to foster an analytic mode of processing. Anxiety proneness may increase the tendency to use this type of processing whenever general ambiguity is increased. This explanation is consistent with that of Stuchlikova (2000). This difference can be also explained in terms of the lack of integration, attention and self-confidence on the part of the highly anxious participants generated by tension due to high anxiety (Pandey, Roy, & Pandey, 1986).

Another explanation for these results is that participants with high anxiety may have some worry, self-doubts, and experiences which result in procrastination and immobility in the face of possible failure. This explanation is supported by the findings of Thompson, Foreman, and Martin (2000) that participants with less satisfaction in their performance rated the success of their performance as low, expressed less confidence in their performance, reported greater negative affect, lower perceptual control, and reported greater anxiety both prior to and following the Stroop task (Stroop, 1935) on the State-Trait Anxiety Inventories 1 and 2 (STAI-1 and STAI-2) measures (Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983), respectively. They also reported that imposters’ anxiety is likely to be exacerbated in a situation characterized by many mistakes, in which imposters are particularly prone to overgeneralize a single mistake to mean total failure.

Results gained in the present study indicated greater tendency to succeed in solving the problems posed by the ML and HV illusions at the level of differentiating processes for low anxiety individuals.

Abdel-Khalek, A. (2000). The Kuwaiti Anxiety Scale psychometric properties. Psychological Reports, 87, 478-492.

Cunningham, M. R. (1988). What do you do when you’re happy or blue? Mood, expectancies and behavioral interest. Motivation and Emotion, 12, 309-331.

Cunningham, G. B., & Ashley, F. B. (2002). Debilitative and facilitative perceptions of trait anxiety among students in a college golf class. Perceptual and Motor Skills, 94, 739-742.

Brosvic, G., Dihoff, R., & Fama, J. (2002). Age-related susceptibility to the Mueller-Lyer and the Horizontal-Vertical illusion. Perceptual and Motor Skills, 94, 229-234.

Eysenck, M. (1992).Cognitive factors in clinical psychology: Potential relevance to therapy. In M. Briley & S. E. File (Eds.), New concepts in anxiety. London: MacMillan.

Eysenck, M. W., Macleod, C., & Mathews, A. (1987). Cognitive functioning and anxiety. Psychological Research, 49, 189-195.

Faisse, P., & Vautrey, P. (1956). The influence of age, sex, and specialized training on the Vertical-Horizontal illusion. Quarterly Journal of Experimental Psychology, 8, 114-120.

Kuhl, J. (1983). Emotion, kognition und motivation: L1. Die function bedeutung der emotionen fur das problemlosende denken und fur das konkrete handeln. Sprache and Kognition, 4, 228-253.

Mathews, A. (1988). Anxiety of the processing of threatening information. In V. Hamilton, G. Bower, & N. Frijda (Eds.), Cognitive perspectives on emotion and motivation. Dortrecht: Nijhoff.

Mathews, A., Richards, & Eysenck, M. (1989). Interpretation of homophones related to threat in anxiety states. Journal of Abnormal Psychology, 98, 31-34.

Mueller-Lyer, F. C. (1889). Dubloid-Reynolds Archive fur Anatomie und Physiologie. 263-270.

Pandey, R. R., Roy, L. K., & Pandey, N. K. (1986). The effect of frustration on the extent of Mueller-Lyer illusion in high and low anxious subjects. Indian Psychological Review, 30, 25-28.

Richards, A., & French, C. (1992). An anxiety related bias in semantic activation when processing threat/neutral homographs. Quarterly Journal of Experimental Psychology, 45, 501-525.

Salovey, P., & Rodin, J. (1985). Cognitions about the self-connecting feeling states and social behavior. Review of Personality and Social Psychology, 6, 143-166.

Sedikides, C. (1992). Mood as a determinant of antinational focus. Cognition and Emotion, 6, 129- 148.

Shehab, N. H. (1992). Developmental and environmental factors in geometrical illusion. Unpublished doctoral thesis, University of Stirling, Scotland.

Shehab, N. H. (2001). The effect of age and sex upon the Mueller-Lyer and Horizontal-Vertical illusion. Arab Journal for the Humanities, 19, 289-300.

Sinha, A. K., & Sinha, S. N. (1967). Mueller-Lyer illusion in subjects high and low in anxiety. Perceptual and Motor Skills, 24, 194.

Spielberger, C., Gorsuch, R., Lushene, R., Vagg, P., & Jacobs, G. (1983). State-Trait Anxiety Inventory (STAI) for adults. California: Consulting Psychologists’ Press.

Stewart, S., Conrod, P., Gignac, M. & Pihl, R. (1998). Selective processing biases in anxiety-sensitive men and women. Cognition and Emotion, 12 (1), 105-133.

Stewart, V. M. (1973). Test of the carpentered world by race and environment in America and Zambia. International Journal of Psychology, 8, 83-94.

Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643-662.

Stuchlikova, Iva (2000). Trait anxiety and perceptual distortion. Studia Psychologica, 42, 55-60.

Thompson, T., Foreman, P., & Martin, F. (2000). Impostor fears and perfectionist concern over mistakes. Personality and Individual Differences, 29, 629-647.

Walters, A. (1942). A genetic study of geometrical-optical illusion. Genetic Psychology Monographs, 25, 103-155.