Introduction: The underlying neural and/or perceptual mechanisms of different visual illusions are still unknown; thus, they continue to be the focus of many ongoing studies. Inconsistencies persist in the empirical findings for understanding how the perception of these illusions evolves over the course of development. Methods: We assessed 513 participants between 6.5 and 18.9 years of age, with 103 pairs of illusory and control images spanning five illusion types (Ebbinghaus, Müller-Lyer, Contrast, Moving Snake, and Subjective Contour). Misleading and helpful contexts were added when possible. Results: In general, we found that, except for the Ebbinghaus illusion, susceptibility changes with age: while for the Müller-Lyer it decreases, for the Contrast, Moving Snake, and Kanizsa, susceptibility increases. Across all illusory conditions, participants’ decision time decreased with age. Context also influenced the performance and choice latency. We also found a gender difference: boys were less susceptible than girls to Contrast and Moving Snake illusions and were faster to answer in Müller-Lyer illusion trials. Conclusion: The current study found that susceptibility to illusions changes in a manner that is age-specific and, in some cases, sex-specific. The different developmental trajectories of the perception of visual illusions support the idea of the lack of a common neural and/or perceptual process behind them. We can suggest that at least some of the cognitive processes and neural pathways involved develop heterochronically.

Inconsistencies between an observer’s visual percept and a stimulus’s physical characteristics are called visual illusions and are considered a result of long-term adaptation of the visual system to our environment [1, 2]. Visual illusions may represent a powerful window into the neurobiology of visual perception as inexpensive, non-invasive tools in basic and medical research, e.g., [3‒5]. Since the development of visual perception is heterochronic, i.e., different areas mature at different times [6], susceptibility to different visual illusions may change at different periods of development [7, 8]. Despite the extensive literature on how the perception of visual illusions changes with age and/or sex, there are mixed and contradictory results. The current research aimed to address this gap by using a wider range of visual illusions in a large sample size.

One of the main challenges of visual illusions is understanding how susceptibility changes with age. The organization of neuronal circuits in the visual cortex changes as an individual develops. Such changes lead to improvements in cognitive function. However, the development of various neural networks differs. Circuits which are in greater need during early childhood mature quicker and earlier [9‒11], while others develop much slower or at a later age [9, 10, 12, 13]. Coincidentally, these changes can result in growing or diminishing susceptibility to visual illusions [14‒16]. Exploring how the susceptibility to different types of visual illusions change with age can help better understand the different underlying processes [17‒19].

Mixed results regarding how visual illusions are perceived may stem partly from inconsistency between methodologies: small sample sizes, narrow age ranges, and a lack of variability in the type of illusion tested. For example, in the case of the Ebbinghaus illusion, it has been shown to be perceived by children as young as 5 years old to the same extent as adults [20, 21], while others reported that the illusion might be weaker or absent in children younger than seven [22‒24]; these studies limited the participant’s age range, which makes it hard to demonstrate a clear developmental trajectory. Other studies address this limitation by increasing the sample size and testing a wide age range but limiting the evaluation to one illusion (Müller-Lyer) [25] or a single type (motion illusions) [8]. Finally, Makris et al. [4] tested 37 different visual illusions in a wide age range, which did not include adults (6 to 14 years old) and had few neurotypical participants overall (n = 41).

Besides age, evidence suggests illusion perception also varies by sex. Several studies have found women are more susceptible; for example, to the Ebbinghaus [26], Vertical-Horizontal [27], Ponzo [28‒30], Müller-Lyer and Poggendorff illusions [31, 32]. Phillips et al. [26] attributed sex differences in perception to the differences between the ventral and dorsal stream of the visual system. They raised the possibility that the dorsal stream, which is less sensitive to context than the ventral one, might have a more dominant role in males. According to Miller [28], this discrepancy in the reliability of field independence is what may underlay sex differences. They conclude that rather than differences in skill, the results suggest differential strategies which diverge in the specific spatial skills males and females apply to particular tasks. In general, studies suggest sex differences in susceptibility of geometrical visual illusions are not so generalizable and mostly report them only to the younger population, and it seems to disappear in adulthood. Lo and Dinov [33] also reported differences between the sexes in the susceptibility of two motion illusions and other studies which also rely on luminosity or contrast effect; women were found to be more susceptible than men [34, 35].

The underlying perceptual mechanisms of the perception of illusions are still unknown. While most studies have tested the perception of a single type of visual illusion, there is potential in testing groups of illusions to better understand the individual differences in visual perception and the factors behind them [36]. Recent studies show that perception to different visual illusions by the same person can be very different [37‒39]. The mechanisms underlying the perception of different visual illusions are still unknown, but the weak inter-illusion correlations suggest that there is no unique common neural path behind all of them. These new findings suggest illusions even which considered belong to one category are processed by different mechanisms [8, 40]. In this sense, a test with multiple visual illusions can create a profile of the functioning of different neural pathways in the same subject.

In the current work, we address the methodological tradeoffs presented by some of the previous studies by testing a large, sex-balanced, sample of participants from a large age range (6–19 years of age) with a representative range of illusions. In order to test many illusions but make it so that it can be performed by young children, we produced a test battery which combines some of the most common illusions for each type. Our set includes Ebbinghaus [41] and Müller-Lyer [42] as geometrical illusions, the Simultaneous Contrast illusion (Contrast from here onward) [43], a black and white version of the Moving Snakes illusion [44] and the Kanizsa Illusory Contour illusion (Kanizsa from here onward) [45]. Our aim was to provide a description of how the perception of visual illusions changes over time, as well as by sex, clearing doubts regarding developmental trajectories for different illusion types.

Participants

We evaluated 513 (58% female) children, adolescents and young adults between 6.5 and 18.9 years of age (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000543308 of supplementary material) attending public and private schools in three medium-density suburban cities in the metropolitan area of Mexico City and the neighboring cities of Huixquilucan, State of Mexico and Cuernavaca, Morelos. Participants were low-middle-income people of Mexican origin. To take part in the study, participants had to assent to the test as well as provide a consent form signed by their parents. In addition to the age and sex of the participants, we also registered their handedness, and whether they had normal or corrected vision. The data collection was done between November 2021 and December 2023.

Equipment and the Visual Illusion Task

We constructed a test consisting of a series of two-alternative forced-choice trials (for example in [22, 26, 46, 47]) using the software PsychoPy. All tests were performed on one of the two identical Lenovo ThinkPad P15 15.6 screen laptops, with 1,920 × 1,080 screen resolution. During the test, two images were presented in each trial, one on each side of the screen on a white background. The task was to choose one according to the pre-recorded verbal and written instructions. For each participant, the test consisted in a total of 103 trials, each with an illusory picture pair belonging to one of the 5 illusions tested (24 Ebbinghaus, 24 Müller-Lyer, 20 Contrast, 10 Moving Snakes, and 25 Kanizsa). The Ebbinghaus, Müller-Lyer, and Contrast illusions were presented with Control, Misleading and Helpful contexts. In the Control conditions, the factors of the environment responsible for the illusory effect (i.e., direction of fins, size of surrounding circles or shade of the surrounding rectangle) were identical on both sides. In the Helpful context, the factors responsible for the illusory effect are arranged such that it facilitates correct choices, whereas in the Misleading context, the illusory effect acts against the correct choice. The Moving Snakes and Kanizsa illusions only had one illusion condition (Fig. 1; Table 1).

Fig. 1.

a Ebbinghaus illusion: circles in the center are the same size, but the one surrounded by large circles appears smaller. b Müller-Lyer illusion: lines are of equal length, but the one with the outward-pointing fins appears longer. c Simultaneous Contrast illusion: circles in the center of rectangle have the same luminosity, but the one with the less luminous background seems brighter. d Moving Snake illusion: the different luminosity and arrangement of the pieces give the snake the appearance of movement. e Kanizsa Subjective Contour illusion: evokes the perception of a white square while no physical edge is present.

Fig. 1.

a Ebbinghaus illusion: circles in the center are the same size, but the one surrounded by large circles appears smaller. b Müller-Lyer illusion: lines are of equal length, but the one with the outward-pointing fins appears longer. c Simultaneous Contrast illusion: circles in the center of rectangle have the same luminosity, but the one with the less luminous background seems brighter. d Moving Snake illusion: the different luminosity and arrangement of the pieces give the snake the appearance of movement. e Kanizsa Subjective Contour illusion: evokes the perception of a white square while no physical edge is present.

Close modal
Table 1.

Illusions used in the tests and their conditions

IllusionCondition
ControlHelpfulMisleading
Ebbinghaus Target circles of different sizes are surrounded by outer circles of the same size (smaller than both or larger than both) on each side. The larger target circle surrounded by smaller circles appears even larger. The smaller inner circle appears even smaller when surrounded by larger outer circles. The larger target circle surrounded by larger circles appears smaller. The smaller inner circle appears larger when surrounded by smaller outer circles. 
Müller-Lyer Target lines of different lengths, delimited by fins pointing inward or outward on both sides. The longer target line appears even longer with the fins pointing outward, the shorter target line appears even shorter with the fins pointing inward. The longer target line appears shorter with the fins pointing inward, the shorter target line appears longer with the fins pointing outward. 
Contrast Internal circles have different shades of gray, while their surroundings have the same luminance (darker than both or lighter than both). The darker internal circle presented against a light surrounding appears even darker, while the lighter internal circle presented against a dark surrounding appears even lighter. The darker internal circle presented against a dark surrounding appears lighter, while the lighter internal circle presented against a light surrounding appears darker. 
Moving Snakes A pair of images with various contrast differences between the segments comprising the ‘snake’. 
Kanizsa Four Pac-man figures on a 7 × 7 grid form the subjective contour of a square on one other side of the screen. 
IllusionCondition
ControlHelpfulMisleading
Ebbinghaus Target circles of different sizes are surrounded by outer circles of the same size (smaller than both or larger than both) on each side. The larger target circle surrounded by smaller circles appears even larger. The smaller inner circle appears even smaller when surrounded by larger outer circles. The larger target circle surrounded by larger circles appears smaller. The smaller inner circle appears larger when surrounded by smaller outer circles. 
Müller-Lyer Target lines of different lengths, delimited by fins pointing inward or outward on both sides. The longer target line appears even longer with the fins pointing outward, the shorter target line appears even shorter with the fins pointing inward. The longer target line appears shorter with the fins pointing inward, the shorter target line appears longer with the fins pointing outward. 
Contrast Internal circles have different shades of gray, while their surroundings have the same luminance (darker than both or lighter than both). The darker internal circle presented against a light surrounding appears even darker, while the lighter internal circle presented against a dark surrounding appears even lighter. The darker internal circle presented against a dark surrounding appears lighter, while the lighter internal circle presented against a light surrounding appears darker. 
Moving Snakes A pair of images with various contrast differences between the segments comprising the ‘snake’. 
Kanizsa Four Pac-man figures on a 7 × 7 grid form the subjective contour of a square on one other side of the screen. 

In the Ebbinghaus and Müller-Lyer trials, the target lines or circles differed by 1, 5, 10, 15, 20, or 30% in their length or area. In the Contrast trials, the luminosity of target circles differed by 1, 2, 4, 6, 10 units, based on a 100-step gray scale where 0 represents white and 100 represents black. In these, the participant’s task was to decide if the larger circle, longer line, darker circle was on the left or right side of the screen.

In the Moving Snakes trials, the concentric circles were constructed from parts of two shades of achromatic gray, separated by alternating black and white segments; the difference in luminance between the two gray parts creates the illusion of movement. We gradually reduced the contrast between the two shades of gray in five equal steps. On the last step, the two shades of gray were identical, resulting in no perceptual movement. Participants were asked to select the one that they perceived as moving faster.

The Kanizsa trials consisted of a 7 by 7 matrix of same-sized Pac-man-shape inducers on both sides but only on one side they formed a complete (illusory) square. We included five Pac-man sizes, with the largest radius being five times bigger than the smallest (1, 2×, 3×, 4×, 5×). The centers of the Pac-man shapes remained in the same position, resulting in less empty space between them as the Pac-man shapes got bigger, making it easier to find the contour of the square. The participant’s task was to select the side where they perceived the illusory square.

Trials for the same illusion were always presented together. The order of trials and the sequence in which the illusions were shown were randomized for each participant. To control laterality bias, for each participant, we chose one of two stimuli arrangements, alternating the side the correct answer was on.

Initially, probe trials were presented to familiarize participants with the equipment and to check that they understood the task, could distinguish basic differences (size, color, contrast) and could properly operate a computer mouse. If these probe trials were successfully answered, the participant could proceed to the visual illusion trials. Then, for each set, vocal and written instructions were given at the start of the set and repeated in short form for each trial. For the Ebbinghaus, Müller-Lyer, and Contrast illusion, for example, the program gave the instructions to “choose the image with the largest orange circle,” “longer line,” and “darker square,” respectively. For the Kanizsa trials, the instruction was to “choose the picture which contains the square,” and for the Moving Snake was to choose which “seems move more.”

Each trial lasted for a maximum of 15 s; if the participant did not make a choice, it moved on to the next one. No feedback was provided during the trials. The detailed description of the probe trials, the code, and the stimuli used for the study can be downloaded from the following URL: https://github.com/marcosrosetti/visual_illusion_task.

Test Procedures

All tests were performed during school hours between 8:00 and 17:00. Sessions lasted approximately 15–25 min. The participant was taken to a quiet classroom where the equipment was already set up. A brief verbal questionnaire was used to collect additional information, such as their handedness and whether they needed glasses. Participants were seated in front of a laptop computer in such a way that the visual angle was relatively constant, at approximately 60 cm, and the mouse location was adjusted to match the dominant hand. After that, the task was started by the experimenter. The experimenter gave a single instruction at the beginning of the task: “You are going to be presented with a series of pairs of images in which you have to choose the image according to the task instructions. Try to make your decision as fast as you can.” All other vocal instructions came from the computer to homogenize duration and exclude the experimenter’s influence. The brightness and volume of the computer were adjusted to 100% for all participants.

All methods were carried out in accordance with the approved guidelines. The Ethics Committees of the Instituto de Investigaciones Biomédicas of the Universidad Nacional Autónoma de México and the Hospital Psiquiátrico Infantil (II3/06/0721) approved the full procedure of the current research. Written informed consent was obtained from the participant and from the parents or guardians.

Data Analysis

Statistical analyses were performed with R version 4.3.0 (R Core Team, 2023). To assess the effect of age we used Generalized Linear (Mixed) Models with binomial distribution with the “logit” link on the outcome of the trials (correct or incorrect) using the package lme4 [48]. Decimal age (birthdate – test date) and sex of the participants were always included as factors, context when applicable; participant’s ID was added as a random factor. A sex*age interaction was originally included, but since it was not significant for any illusion or context, it was ultimately excluded. p values were extracted by Wald chi-square tests (type II). Kendall’s tau correlation was used to evaluate the relationship between illusions as it is less sensitive to ties than Spearman’s. For easier interpretation of the results, age categories are presented in tables and figures, where decimal ages of participants are rounded to the closest whole number. Since only two participants were younger than 6.5, their ages were rounded upward for plotting. Values in text are presented as mean ± standard deviation.

After reviewing participants’ performance, we excluded those whose responses in the control trials combined fell below (i.e., consistently chose the smaller figure) or did not surpass chance levels (two-sided exact binomial test, total contrast trials = 34, α = 0.5). Fifty-seven participants were excluded, of whom 55% were female, and the percentage of participants excluded from each age category ranged from 6.3% to 22.2%. The final sample size is 456. A finer description of the excluded cases and final sample can be found in online supplementary Table 1 of supplementary material.

Ebbinghaus Illusion

Trials in the Helpful context were mostly answered correctly in all age groups, displaying a ceiling effect (96% ± 14), followed by trials in the Control condition (77% ± 13), while those in the Misleading context had the most incorrect answers (12% ± 19). The latency to answer mirrored this pattern. Context had a very large effect (Table 2; Fig. 2) suggesting choices and reaction time behaved almost in an opposite manner when judging the stimuli in a Helpful (est ± se = −0.038 ± 0.109, p = 0.728, est ± se = −0.126 ± 0.011, p < 0.001), Misleading (est ± se = −0.056 ± 0.031, p = 0.069, est ± se = −0.16 ± 0.016, p < 0.001) or Control (est ± se = −0.004 ± 0.010, p = 0.700; est ± se = −0.176 ± 0.017, p < 0.001) contexts (correct choices; reaction time, respectively). When separated by context, we did not find any age or sex effect in the percent of correct answers for any of the three conditions; however, older participants answered significantly faster.

Table 2.

Results of the generalized linear mixed models on the performance and latency of the illusions

Illusion and conditionProportion of correct choicesLatency to choose
agesex (boys)agesex (boys)
bSDp valuebSDp valuebSDp valuebSDp value
Ebbinghaus 
 All conditions −0.016 0.010 0.099 0.039 0.062 0.523 −0.159 0.013 <0.001 −0.052 0.086 0.547 
 Control −0.004 0.010 0.700 −0.035 0.067 0.601 −0.176 0.017 <0.000 −0.120 0.108 0.265 
 Helpful −0.038 0.109 0.728 0.149 0.708 0.833 −0.126 0.011 <0.001 0.030 0.074 0.687 
 Misleading −0.056 0.031 0.069 0.269 0.199 0.177 −0.16 0.016 <0.001 0.003 0.102 0.973 
Müller-Lyer 
 All conditions 0.030 0.009 0.005 0.005 0.059 0.934 −0.129 0.011 <0.001 −0.197 0.073 0.007 
 Control 0.017 0.012 0.163 −0.026 0.077 0.738 −0.149 0.012 <0.001 −0.194 0.076 0.011 
 Helpful −0.031 0.115 0.784 −0.442 0.734 0.547 −0.110 0.011 <0.001 −0.208 0.074 0.005 
 Misleading 0.048 0.015 0.002 0.074 0.096 0.444 −0.110 0.016 <0.001 −0.161 0.104 0.027 
Contrast 
 All conditions 0.018 0.008 0.030 0.079 0.054 0.144 −0.133 0.015 <0.001 −0.052 0.095 0.588 
 Control 0.041 0.012 0.001 −0.028 0.077 0.713 −0.162 0.018 <0.001 −0.064 0.116 0.579 
 Helpful 0.178 0.067 0.008 −1.171 0.43 0.006 −0.089 0.014 <0.001 −0.077 0.089 0.386 
 Misleading −0.069 0.026 0.007 0.773 0.165 < 0.001 −0.117 0.018 <0.001 0.011 0.119 0.928 
 Moving snake 0.041 0.011 0.001 −0.182 0.068 0.008 −0.19 0.030 <0.001 0.106 0.197 0.588 
 Kanizsa 0.081 0.011 < 0.001 0.086 0.070 0.220 −0.169 0.017 <0.001 −0.015 0.113 0.893 
Illusion and conditionProportion of correct choicesLatency to choose
agesex (boys)agesex (boys)
bSDp valuebSDp valuebSDp valuebSDp value
Ebbinghaus 
 All conditions −0.016 0.010 0.099 0.039 0.062 0.523 −0.159 0.013 <0.001 −0.052 0.086 0.547 
 Control −0.004 0.010 0.700 −0.035 0.067 0.601 −0.176 0.017 <0.000 −0.120 0.108 0.265 
 Helpful −0.038 0.109 0.728 0.149 0.708 0.833 −0.126 0.011 <0.001 0.030 0.074 0.687 
 Misleading −0.056 0.031 0.069 0.269 0.199 0.177 −0.16 0.016 <0.001 0.003 0.102 0.973 
Müller-Lyer 
 All conditions 0.030 0.009 0.005 0.005 0.059 0.934 −0.129 0.011 <0.001 −0.197 0.073 0.007 
 Control 0.017 0.012 0.163 −0.026 0.077 0.738 −0.149 0.012 <0.001 −0.194 0.076 0.011 
 Helpful −0.031 0.115 0.784 −0.442 0.734 0.547 −0.110 0.011 <0.001 −0.208 0.074 0.005 
 Misleading 0.048 0.015 0.002 0.074 0.096 0.444 −0.110 0.016 <0.001 −0.161 0.104 0.027 
Contrast 
 All conditions 0.018 0.008 0.030 0.079 0.054 0.144 −0.133 0.015 <0.001 −0.052 0.095 0.588 
 Control 0.041 0.012 0.001 −0.028 0.077 0.713 −0.162 0.018 <0.001 −0.064 0.116 0.579 
 Helpful 0.178 0.067 0.008 −1.171 0.43 0.006 −0.089 0.014 <0.001 −0.077 0.089 0.386 
 Misleading −0.069 0.026 0.007 0.773 0.165 < 0.001 −0.117 0.018 <0.001 0.011 0.119 0.928 
 Moving snake 0.041 0.011 0.001 −0.182 0.068 0.008 −0.19 0.030 <0.001 0.106 0.197 0.588 
 Kanizsa 0.081 0.011 < 0.001 0.086 0.070 0.220 −0.169 0.017 <0.001 −0.015 0.113 0.893 

Factor estimates, χ2 and p values (obtained by type II Wald tests) are given.

Significant effects are bolded.

Fig. 2.

Proportion of correct choices and latency to choose in each illusion, separated by condition. For the Ebbinghaus, Müller-Lyer, and Contrast illusions, blue squares (with dotted lines) represent Helpful, red circles (dashed lines) Misleading, and green triangles (solid lines) Control contexts. Whiskers show standard error.

Fig. 2.

Proportion of correct choices and latency to choose in each illusion, separated by condition. For the Ebbinghaus, Müller-Lyer, and Contrast illusions, blue squares (with dotted lines) represent Helpful, red circles (dashed lines) Misleading, and green triangles (solid lines) Control contexts. Whiskers show standard error.

Close modal

Müller-Lyer Illusion

We observed a ceiling effect with children mostly answering correctly in the Helpful trials (98% ± 6), less so in Control (84% ± 11) and Misleading trials (21% ± 17) as they had mostly incorrect answers. We observed a significant effect of age on the performance (est ± se = 0.030 ± 0.009, p = 0.005) which could be attributed mainly to the increment in correct answers in the Misleading trials (est ± se = 0.048 ± 0.015, p = 0.002) (Table 2; Fig. 2) suggesting less susceptibility to the illusion with age. Latency to answer decreased with age (est ± se = −0.129 ± 0.011, p < 0.001); boys were faster to respond across all contexts (est ± se = −0.197 ± 0.073, p < 0.007), although it did not translate into a larger number of correct answers.

Contrast Illusion

Here, we also observed the largest percentage of correct answers in the Helpful trials (84% ± 29), followed by the Control (78% ± 15) and Misleading contexts (31% ± 31). The number of correct answers significantly increased with age in the Helpful (est ± se = 0.178 ± 0.067, p = 0.008) and the Control (est ± se = 0.041 ± 0.012, p = 0.001) trials but decreased for the Misleading context (est ± se = −0.069 ± 0.026, p = 0.007) (Table 2; Fig. 2). Also, boys answered correctly significantly more often than girls in Misleading trials (est ± se = −0.171 ± 0.43, p = 0.006), but girls outperformed boys in Helpful trials (est ± se = −0.089 ± 0.014, p < 0.0001). Latency to answer decreased with age in all conditions (est ± se = −0.133 ± 0.015, p < 0.0001).

Moving Snake and Kanizsa Illusions

The number of correct answers significantly increased with age in the trials of both illusions (Moving Snake 63% ± 17; est ± se = 0.041 ± 0.011, p = 0.001, Kanizsa: 82% ± 11; est ± se = 0.081 ± 0.011, p < 0.0001). Girls outperformed boys in the Moving Snake illusion (est ± se = −0.182 ± 0.068, p = 0.008). Latency to answer significantly decreased with age for both illusions (Moving snake: est ± se = −0.19 ± 0.030, p < 0.0001; Kanizsa: est ± se = −0.169 ± 0.017, p < 0.0001) (Table 2; Fig. 2).

Correlation between Illusions

We have calculated Kendall’s tau correlation between the individual performances in the control (Ebbinghaus, Müller-Lyer, Contrast), Helpful (Ebbinghaus, Müller-, Contrast, Moving snake, Kanizsa), and Misleading (Ebbinghaus, Müller-, Contrast) conditions. We found, in general, very low connections between the illusions (online suppl. material). The mean value of the correlation coefficients (τ) was 0.08 ± 0.07 (min = −0.02 max = 0.18, see details in online supplementary Table 2 in supplementary material).

We examined the change in the perception of several types of illusions in a large sample and a wide age range. First, we found that while for the Ebbinghaus illusion, susceptibility did not change with age, for the Müller-Lyer, it decreased and for the Contrast, the Moving Snake, and the Kanizsa illusions it increased as participants got older. In addition, participants’ decision time decreased with age across all illusory conditions, suggesting that choosing an option became easier for older participants. Second, we found a sex effect with boys showing less susceptibility to the Contrast and Moving Snake illusions. In terms of latency, boys answered faster when evaluating the Müller-Lyer illusion. Finally, context in all illusions (where it is applied) affected the proportion of choices and latency. Furthermore, a closer look reveals a large degree of heterogeneity between the results of each individual illusion, which we discuss below.

The review by Wincza et al. [7] suggests sensitivity to the Ebbinghaus illusion of 4-year-old children is the same as adults. However, agreement about the developmental trajectory of susceptibility to the Ebbinghaus illusion is not unanimous. The study by Doherty et al. [22] employed an alternative version of the Ebbinghaus illusion and showed that context sensitivity is still not fully developed by the age of 10, and it still develops beyond this age. They argued that both Misleading and Helpful contexts have large effects on adults, but not on children; thus, children are genuinely less context-sensitive. Here, we found no age effect in Misleading and Helpful context trials which is in line with the idea that children are susceptible to this illusion at the same magnitude as adults. In contrast, Bondarko and Semenov [49] reported a decreasing susceptibility to Ebbinghaus in 6- to 17-year-old participants. This discrepancy might be due to the different methodologies used (two-alternative forced-choice vs. adjustment) or variations in the stimuli; however, we can only make assertions regarding our implementations (see suggestions by [7]).

Droit-Volet et al. [50] showed that in a line-size-length discrimination task, 8-year-old children performed similarly to adults. Several studies have shown that susceptibility to the Müller-Lyer illusion decreases, rather than increases, with age, regardless of the method used; review in [7]. In the current sample, the number of correct choices in the misleading context increased with age, which suggests that perception of the Müller-Lyer illusion decreased, reaching an adult level of sensitivity around 14 years in line with other reports [25, 51].

Previous studies have also reported changes with age in the perception of the Müller-Lyer illusion but not for the Ebbinghaus; reviewed by [7]. There are several explanations for what appears to be a developmental heterochronicity between these illusions. One possibility is that the neural or cognitive operations underlying the processing of the illusions may be different [39, 52‒54] despite the illusions being classified in the same category [2]. Another explanation is that Müller-Lyer illusion may be cognitively and perceptually more demanding than the Ebbinghaus illusion and that is why it takes longer to reach an adult-like performance [7, 55].

The findings of the present study also have neuroanatomical implications. By 4-years of age, the development of V1 is characterized as “adult-like”; synaptic and dendritic architecture, cortical thickness, and feedback connections from extrastriate areas to V1 reach adult appearance [6]. Many illusion processes start in V1 and then continue in other areas of the brain. This might suggest that the neural substrate mediates the perception of the Ebbinghaus illusion is mainly based on V1, and it is well developed by 4 years; however, other brain areas which are involved in the perception of other illusions are still undergoing some change or refinement. This idea is supported by fMRI studies, which have shown that the strength of the Ebbinghaus illusion is negatively correlated with the surface of V1 [40, 56]. Also, the Moving Snake illusion not just activates the V1 which is known to be a motion-sensitive area too, but also activates the V5 (middle temporal area) [57, 58]. Additionally, mainly the lateral occipital cortex is responsible for shape and illusory contour processing [59, 60].

Research indicates that perception to contrast sensitivity increases with age [61, 62], although the precise point at which sensitivity reaches adult levels remains uncertain. Our findings align with this observation, as evidenced by a significant increase in correct responses with age across both the Control and Helpful context, and a decline in the Misleading context in the Contrast illusion. This finding coincides with that of Makris et al. [4], indicating that younger children perceive the Contrast illusory effects less strongly. Reduced susceptibility to contrast illusion could reflect an immature contrast sensitivity among younger participants [63].

In a similar way to Billino et al. [8], who reported that younger children perceive less illusory motion than young adults, we observed an enhanced susceptibility to the Moving Snake illusion as participants grow older. One possible explanation behind this might be the change in the frequency of small involuntary movements in the eyes, or microsaccades, during development. Troncoso et al. [64] found that a faster illusion perception of the Enigma illusion was driven by an increase in the frequency of microsaccades. During development the frequency increases, which may reflect in the stronger susceptibility of the Moving Snakes illusion [65].

Regarding the Kanizsa illusion, our results indicate an age-dependent improvement in perception and decline in latency of choice in this illusion, which is in line with previous findings [66]. However, since the task in the present study was more complex than what is regularly used (one-to-one comparison), it is important to note that this increment may not directly reflect a greater susceptibility to the illusion as it could also be influenced by an enhancement in organized visual search with age [67]. Therefore, it cannot be ruled out that the observed effect is simply the result of an improved ability to detect and recognize patterns in the Pac-man figures, rather than a genuine increase in susceptibility to the illusion itself. It would be beneficial to conduct further evaluations of this illusion using alternative methods, for example, reporting the thickness of the illusory figure [53] or a matching task [66], to confirm whether the increased susceptibility truly reflects changes in visual perception or is related to the development of more refined visual search skills.

In the present study, during the session of the contrast illusion, boys answered correctly more often than girls in the Misleading trials and less often in the Helpful trials. This means that boys are less influenced by the context of both types of trials and therefore less susceptible to this illusion. Additionally, boys were less able to choose the faster moving stimuli during the Moving Snake illusion trials. Since the strength of the Contrast and Moving Snake illusion depends on the luminosity, one might assume that the underlying mechanism behind these illusions may share neural substrates or cognitive processes [68], but the lack of correlation between the two contradicts this. Lo and Dinov [33] in an earlier study also showed that women are more susceptible to a motion illusion than men. Additionally, there are a few studies that align with this result as they found women were more susceptible to several visual illusions that rely on contrast as a mechanism [34, 35].

When faced with the Müller-Lyer illusion, boys answered faster than girls in all trials. Since measuring latency is not a common methodology in similar tasks, it makes this finding difficult to interpret. However, there are some previous studies where they reported that women are more susceptible to the Müller-Lyer illusion (or other similar type of illusions, e.g., Ponzo or Baldwin) than men [27‒30, 69, 70]. One possibility is that latency reflects a certain degree of susceptibility, with girls needing slightly more time to decide. However, the lack of sex differences in performance leaves this an open question. Another, not mutually exclusive explanation, is that girls have a different mental strategy to solve these types of tasks compared to boys, which takes more time but yields similar results, i.e., boys are more visually field independent, meaning that they are better in attending to visual stimuli which are embedded in a perceptually compelling, complex visual field as suggested by Miller [28]. A recent study suggests that the reaction time and the magnitude of the illusion perception correlate with each other, and it is negative in the case of the Ebbinghaus illusion [71]. This would suggest a diminished perception in girls; however, we did not find differences between the sexes in their performance.

Final notes on the control trials: we eliminated a large percentage of participants (11.1%) since their performance did not pass the chance level in the control trials. The excluded individuals’ age distribution was uniform (online suppl. material 1, Table 1), which suggests that the test was age appropriate. For the two geometric illusions, the Ebbinghaus and the Müller-Lyer, no age effect on performance was found in the control trials, suggesting that at 6 years, participants were already able to correctly judge size and length differences. This is in line with the results from Halberda and Feigenson [72], where they showed that 6 year olds can already discriminate between different quantities when the ratio reached 0.83, and it keeps improving to adulthood. Also, Droit-Volet et al. [50] showed that in a line size-length discrimination task, 8-year-old children performed similarly to adults. However, for the Contrast illusion, performance in the control condition improved with age, suggesting that perception of luminosity is still not mature at 6 and continues to improve with age.

Despite the significant results of the study, some limitations should be considered. Although the sample used is large and diverse in age, it is limited to participants of primary, middle, and high school age. As previous studies have shown that the perception of some illusions develops before this age (our results suggest the same for the Ebbinghaus illusion), it would be useful to include preschool-aged children in future studies. In addition, two illusions, Kanizsa and Moving Snake, did not have Helpful and Misleading trials. Thus, it would be useful to add additional tasks that could provide more detailed results on the perceptual development of these illusions. Furthermore, the lack of direct neurophysiological measurements in the current study prevents a full understanding of the mechanisms underlying the observed changes in illusion perception. Future studies may benefit from integrating neuroimaging or electrophysiological techniques to correlate behavioral changes with changes in brain activity, but these techniques are difficult to implement in children.

Susceptibility to visual illusions is a consequence of the adaptation of the visual system to a complex and dynamic world. Thus, the differences found in the susceptibility to certain illusions may reflect changes in more general/global cognitive processes during development. The lack of common patterns among illusions may indicate how underlying mechanisms change with the development of the visual system, and we can hypothesize that at least some of the cognitive processes and neural pathways involved develop heterochronically. The current study also found that changes in susceptibility to illusions are not only age-specific but, in some cases, sex-specific. Finally, this study confirms the idea that the use of visual illusions is a non-invasive and inexpensive tool for studying the neurobiology and development of vision, especially in young children where other neurophysiological methods are difficult to apply.

We thank Carolina Rojas for technical and bibliographical assistance and Dr. Alejandro Jiménez Orozco for recording the vocal instruction for the tests. We also thank the schools, Escuela Nacional Preparatoria número 4 “Vidal Castañeda y Nájera,” Escuela Secundaria Técnica N° 73 “Carlos Vallejo Márquez,” Escuela primaria “Guadalupe Victoria,” and the Colegio Montessori de Tepoztlán, to provide us technical and infrastructural help execute our experiments.

All methods were carried out in accordance with the approved guidelines. The Ethics Committees of the Instituto de Investigaciones Biomédicas of the Universidad Nacional Autónoma de México and the Hospital Psiquiátrico Infantil (II3/06/0721) approved the full procedure of the current research. Written informed consent was obtained from the participant and from the parents or guardians.

The authors have no conflicts of interest to declare.

This work was supported by the Dirección General de Asuntos del Personal Académico, Universidad Nacional Autónoma de México [DGAPA, Grant No. IA204222] and by a Cátedra grant to P.S. from the Consejo Nacional de Ciencia y Tecnología, México [CONACyT, Grant No. 691].

O.B. and M.R. served as lead for investigation, methodology, and supervision; served in a supporting role for formal analysis; and contributed equally to conceptualization, data curation, visualization, and writing. O.B. received financial support for the project. P.S. served as lead for data curation and formal analysis and contributed equally to investigation, methodology, supervision, validation, and writing. R.E.U. contributed equally to investigation, methodology, and writing. V.N. and V.M. contributed equally to investigation, methodology, and writing and served in a supporting role for visualization. M.A. contributed equally to investigation, writing, and data analysis.

The materials necessary to attempt to replicate the findings presented here are publicly accessible at the following URL: https://github.com/marcosrosetti/visual_illusion_task. The collected data necessary to reproduce the analyses presented here are available at https://doi.org/10.6084/m9.figshare.27058474. Further inquiries can be directed to the corresponding author.

1.
Howe
CQ
,
Purves
D
.
Natural-scene geometry predicts the perception of angles and line orientation
.
P Natl Acad Sci USA
.
2005
;
102
(
4
):
1228
33
.
2.
Gregory
RL
.
Visual illusions classified
.
Trends Cogn Sci
.
1997
;
1
(
5
):
190
4
.
3.
King
DJ
,
Hodgekins
J
,
Chouinard
PA
,
Chouinard
VA
,
Sperandio
I
.
A review of abnormalities in the perception of visual illusions in schizophrenia
.
Psychon Bull Rev
.
2017
;
24
(
3
):
734
51
.
4.
Makris
G
,
Pervanidou
P
,
Chouliaras
G
,
Stachtea
X
,
Valavani
E
,
Bastaki
D
, et al
.
Diverse patterns of vulnerability to visual illusions in children with neurodevelopmental disorders
.
Cogn Process
.
2021
;
22
(
4
):
659
73
.
5.
Eagleman
DM
.
Visual illusions and neurobiology
.
Nat Rev Neurosci
.
2001
;
2
(
12
):
920
6
.
6.
Siu
CR
,
Murphy
KM
.
The development of human visual cortex and clinical implications
.
Eye Brain
.
2018
;
10
:
25
36
.
7.
Wincza
R
,
Hartley
C
,
Fenton-Romdhani
J
,
Linkenauger
S
,
Crawford
T
.
The development of susceptibility to geometric visual illusions in children: a systematic review
.
Cogn Dev
.
2024
;
69
:
101410
.
8.
Billino
J
,
Hamburger
K
,
Gegenfurtner
KR
.
Age effects on the perception of motion illusions
.
Perception
.
2009
;
38
(
4
):
508
21
.
9.
Shankle
WR
,
Romney
AK
,
Landing
BH
,
Hara
J
.
Developmental patterns in the cytoarchitecture of the human cerebral cortex from birth to 6 years examined by correspondence analysis
.
Proc Natl Acad Sci U S A
.
1998
;
95
(
7
):
4023
8
.
10.
Shankle
WR
,
Landing
BH
,
Rafii
MS
,
Schiano
A
,
Chen
JM
,
Hara
J
.
Evidence for a postnatal doubling of neuron number in the developing human cerebral cortex between 15 months and 6 years
.
J Theor Biol
.
1998
;
191
(
2
):
115
40
.
11.
Deen
B
,
Richardson
H
,
Dilks
DD
,
Takahashi
A
,
Keil
B
,
Wald
LL
, et al
.
Organization of high-level visual cortex in human infants
.
Nat Commun
.
2017
;
8
(
1
):
13995
.
12.
Burkhalter
A
,
Bernardo
KL
,
Charles
V
.
Development of local circuits in human visual cortex
.
J Neurosci
.
1993
;
13
(
5
):
1916
31
.
13.
Burkhalter
A
.
Development of forward and feedback connections between areas V1 and V2 of human visual cortex
.
Cereb Cortex
.
1993
;
3
(
5
):
476
87
.
14.
Natu
VS
,
Gomez
J
,
Barnett
M
,
Jeska
B
,
Kirilina
E
,
Jaeger
C
, et al
.
Apparent thinning of human visual cortex during childhood is associated with myelination
.
Proc Natl Acad Sci U S A
.
2019
;
116
(
41
):
20750
9
.
15.
Norcia
AM
,
Gerhard
HE
,
Meredith
WJ
.
Development of relative disparity sensitivity in human visual cortex
.
J Neurosci
.
2017
;
37
(
23
):
5608
19
.
16.
Kovács
I
,
Kozma
P
,
Fehér
Á
,
Benedek
G
.
Late maturation of visual spatial integration in humans
.
Proc Natl Acad Sci U S A
.
1999
;
96
(
21
):
12204
9
.
17.
Page
JW
,
Crognale
MA
.
Differential aging of chromatic and achromatic visual pathways: behavior and electrophysiology
.
Vis Res
.
2005
;
45
(
11
):
1481
9
.
18.
Habak
C
,
Faubert
J
.
Larger effect of aging on the perception of higher-order stimuli
.
Vis Res
.
2000
;
40
(
8
):
943
50
.
19.
Weale
RA
.
Aging and vision
.
Vis Res
.
1986
;
26
(
9
):
1507
12
.
20.
Duemmler
T
,
Franz
VH
,
Jovanovic
B
,
Schwarzer
G
.
Effects of the Ebbinghaus illusion on children’s perception and grasping
.
Exp Brain Res
.
2008
;
186
(
2
):
249
60
.
21.
Hanisch
C
,
Konczak
J
,
Dohle
C
.
The effect of the Ebbinghaus illusion on grasping behaviour of children
.
Exp Brain Res
.
2001
;
137
(
2
):
237
45
.
22.
Doherty
MJ
,
Campbell
NM
,
Tsuji
H
,
Phillips
WA
.
The Ebbinghaus illusion deceives adults but not young children
.
Dev Sci
.
2010
;
13
(
5
):
714
21
.
23.
Káldy
Z
,
Kovács
I
.
Visual context integration is not fully developed in 4-year-old children
.
Perception
.
2003
;
32
(
6
):
657
66
.
24.
Weintraub
DJ
.
Ebbinghaus illusion: context, contour, and age influence the judged size of a circle amidst circles
.
J Exp Psychol Hum Percept Perform
.
1979
;
5
(
2
):
353
64
.
25.
Porac
C
,
Coren
S
.
Life-span age trends in the perception of the Mueller-Lyer: additional evidence for the existence of two illusions
.
Can J Psychol
.
1981
;
35
(
1
):
58
62
.
26.
Phillips
WA
,
Chapman
KLS
,
Berry
PD
.
Size perception is less context-sensitive in males
.
Perception
.
2004
;
33
(
1
):
79
86
.
27.
Fraisse
P
,
Vautrey
P
.
The influence of age, sex, and specialized training on the vertical-horizontal illusion
.
Q J Exp Psychol
.
1956
;
8
(
3
):
114
20
.
28.
Miller
RJ
.
Gender differences in illusion response: the influence of spatial strategy and sex ratio
.
Sex Roles
.
2001
;
44
(
3/4
):
209
25
.
29.
Shaqiri
A
,
Roinishvili
M
,
Grzeczkowski
L
,
Chkonia
E
,
Pilz
K
,
Mohr
C
, et al
.
Sex-related differences in vision are heterogeneous
.
Sci Rep
.
2018
;
8
(
1
):
7521
.
30.
Pressey
AW
,
Wilson
AE
.
Another look at age changes in geometric illusions
.
Bull Psychon Soc
.
1978
;
12
(
4
):
333
6
.
31.
Pressey
AW
,
Sweeney
O
.
Age changes in the Poggendorff illusion as measured by a method of production
.
Psychon Sci
.
1970
;
19
(
2
):
99
100
.
32.
Ling
J
,
Hamilton
C
,
Heffernan
TM
.
Sex differences in the Poggendorff illusion: identifying the locus of the effect
.
Percept Mot Ski
.
2006
;
102
(
1
):
142
6
.
33.
Lo
C
,
Dinov
I
.
Investigation of optical illusions on the aspects of gender and age
.
UCLA USJ
.
2011
;
24
:
5
10
.
34.
Chaplin
JP
.
Sex differences in the perception of autokinetic movement
.
J Gen Psychol
.
1955
;
52
(
1
):
149
55
.
35.
Conlon
EG
,
Power
GF
,
Hine
TJ
,
Rahaley
N
.
The impact of older age and sex on motion discrimination
.
Exp Aging Res
.
2017
;
43
(
1
):
55
79
.
36.
Tulver
K
.
The factorial structure of individual differences in visual perception
.
Conscious Cogn
.
2019
;
73
:
102762
.
37.
Axelrod
V
,
Schwarzkopf
DS
,
Gilaie-Dotan
S
,
Rees
G
.
Perceptual similarity and the neural correlates of geometrical illusions in human brain structure
.
Sci Rep
.
2017
;
7
(
1
):
39968
.
38.
Grzeczkowski
L
,
Clarke
AM
,
Francis
G
,
Mast
FW
,
Herzog
MH
.
About individual differences in vision
.
Vis Res
.
2017
;
141
:
282
92
.
39.
Cretenoud
AF
,
Karimpur
H
,
Grzeczkowski
L
,
Francis
G
,
Hamburger
K
,
Herzog
MH
.
Factors underlying visual illusions are illusion-specific but not feature-specific
.
J Vis
.
2019
;
19
(
14
):
12
.
40.
Schwarzkopf
DS
,
Song
C
,
Rees
G
.
The surface area of human V1 predicts the subjective experience of object size
.
Nat Neurosci
.
2011
;
14
(
1
):
28
30
.
41.
Ebbinghaus
H
.
Grundzüge der psychologie
.
Leipzig
:
Verlag von Viet & Co
;
1902
.
Vol. 1
.
42.
Müller-Lyer
FC
.
Optische urteilstäuschungen [optical misjudgments]
.
Arch Anat Physiol
.
1889
;
2
(
Suppl
):
263
70
.
43.
Chevreul
ME
.
De la loi du contraste simultané des couleurs et de l’assortiment des object colorés
.
Paris, France
:
Pitois-Levreault
;
1839
.
44.
Kitaoka
A
.
Rotating snakes
.
2003
. [Internet] [cited 2020 2020 May 20]; Available from: https://www.psy.ritsumei.ac.jp/akitaoka/rotsnakee.html
45.
Kanizsa
G
.
Margini quasi-percettivi in campi con stimolazione omogenea
.
Riv Psicol
.
1955
;
49
(
1
):
7
30
.
46.
Bremner
AJ
,
Doherty
MJ
,
Caparos
S
,
de Fockert
J
,
Linnell
KJ
,
Davidoff
J
.
Effects of culture and the urban environment on the development of the Ebbinghaus illusion
.
Child Dev
.
2016
;
87
(
3
):
962
81
.
47.
Manning
C
,
Morgan
MJ
,
Allen
CT
,
Pellicano
E
.
Susceptibility to Ebbinghaus and Müller-Lyer illusions in autistic children: a comparison of three different methods
.
Mol Autism
.
2017
;
8
(
1
):
16
.
48.
Bates
D
,
Mächler
M
,
Bolker
B
,
Walker
S
.
Fitting linear mixed-effects models using lme4
.
J Stat Softw
.
2015
;
67
(
1
):
1
48
.
49.
Bondarko
VM
,
Semenov
LA
.
Size estimates in Ebbinghaus illusion in adults and children of different age
.
Hum Physiol
.
2004
;
30
(
1
):
24
30
.
50.
Droit-Volet
S
,
Clément
A
,
Fayol
M
.
Time, number and length: similarities and differences in discrimination in adults and children
.
Q J Exp Psychol
.
2008
;
61
(
12
):
1827
46
.
51.
Brosvic
GM
,
Dihoff
RE
,
Fama
J
.
Age-related susceptibility to the Müller-Lyer and the horizontal-vertical illusions
.
Percept Mot Ski
.
2002
;
94
(
1
):
229
34
.
52.
Chouinard
PA
,
Noulty
WA
,
Sperandio
I
,
Landry
O
.
Global processing during the Müller-Lyer illusion is distinctively affected by the degree of autistic traits in the typical population
.
Exp Brain Res
.
2013
;
230
(
2
):
219
31
.
53.
Hadad
BS
.
Developmental trends in susceptibility to perceptual illusions: not all illusions are created equal
.
Atten Percept Psychophys
.
2018
;
80
(
6
):
1619
28
.
54.
Hamburger
K
.
Visual illusions based on processes: new classification system needed
.
Perception
.
2016
;
45
(
5
):
588
95
.
55.
Sperandio
I
,
Chouinard
PA
,
Paice
E
,
Griffiths-King
DJ
,
Hodgekins
J
.
Visual illusions in young people reporting psychotic-like experiences
.
J Behav Ther Exp Psychiatry
.
2023
;
79
:
101839
.
56.
Schwarzkopf
DS
,
Rees
G
.
Subjective size perception depends on central visual cortical magnification in human V1
.
PLoS One
.
2013
;
8
(
3
):
e60550
.
57.
Kuriki
I
,
Ashida
H
,
Murakami
I
,
Kitaoka
A
.
Functional brain imaging of the Rotating Snakes illusion by fMRI
.
J Vis
.
2008
;
8
(
10
):
1
10
.
58.
Ashida
H
,
Kuriki
I
,
Murakami
I
,
Hisakata
R
,
Kitaoka
A
.
Direction-specific fMRI adaptation reveals the visual cortical network underlying the “Rotating Snakes” illusion
.
Neuroimage
.
2012
;
61
(
4
):
1143
52
.
59.
Kourtzi
Z
,
Kanwisher
N
.
Representation of perceived object shape by the human lateral occipital complex
.
Science
.
2001
;
293
(
5534
):
1506
9
.
60.
Mendola
JD
,
Dale
AM
,
Fischl
B
,
Liu
AK
,
Tootell
RBH
.
The representation of illusory and real contours in human cortical visual areas revealed by functional magnetic resonance imaging
.
J Neurosci
.
1999
;
19
(
19
):
8560
72
.
61.
Leat
SJ
,
Yadav
NK
,
Irving
EL
.
Development of visual acuity and contrast sensitivity in children
.
J Optom
.
2009
;
2
(
1
):
19
26
.
62.
Nielsen
LS
,
Nielsen
SK
,
Skov
L
,
Jensen
H
.
Contrast sensitivity: an unnoticed factor of visual perception in children with developmental delay: normal data of the cambridge low contrast gratings test in children
.
J Child Neurol
.
2007
;
22
(
2
):
151
5
.
63.
Liu
R
,
Zhou
J
,
Zhao
H
,
Dai
Y
,
Zhang
Y
,
Tang
Y
, et al
.
Immature visual neural system in children reflected by contrast sensitivity with adaptive optics correction
.
Sci Rep
.
2014
;
4
(
1
):
4687
.
64.
Troncoso
XG
,
Macknik
SL
,
Otero-Millan
J
,
Martinez-Conde
S
.
Microsaccades drive illusory motion in the Enigma illusion
.
Proc Natl Acad Sci U S A
.
2008
;
105
(
41
):
16033
8
.
65.
Port
NL
,
Trimberger
J
,
Hitzeman
S
,
Redick
B
,
Beckerman
S
.
Micro and regular saccades across the lifespan during a visual search of “Where’s Waldo” puzzles
.
Vis Res
.
2016
;
118
:
144
57
.
66.
Nayar
K
,
Franchak
J
,
Adolph
K
,
Kiorpes
L
.
From local to global processing: the development of illusory contour perception
.
J Exp Child Psychol
.
2015
;
131
:
38
55
.
67.
Woods
AJ
,
Göksun
T
,
Chatterjee
A
,
Zelonis
S
,
Mehta
A
,
Smith
SE
.
The development of organized visual search
.
Acta Psychol
.
2013
;
143
(
2
):
191
9
.
68.
He
C
,
Nguyen
BN
,
Chan
YM
,
McKendrick
AM
.
Illusory motion perception is associated with contrast discrimination but not motion sensitivity, self-reported visual discomfort, or migraine status
.
Invest Ophthalmol Vis Sci
.
2020
;
61
(
8
):
43
.
69.
Barclay
JR
,
Comalli
PE
.
Age differences in perceptual learning on the Müller-Lyer illusion
.
Psychon Sci
.
1970
;
19
(
6
):
323
5
.
70.
Pratarelli
ME
,
Steitz
BJ
.
Effects of gender on perception of spatial illusions
.
Percept Mot Ski
.
1995
;
80
(
2
):
625
6
.
71.
Bressan
P
,
Kramer
P
.
Most findings obtained with untimed visual illusions are confounded
.
Psychol Sci
.
2021
;
32
(
8
):
1238
46
.
72.
Halberda
J
,
Feigenson
L
.
Developmental change in the acuity of the “number sense”: the approximate number system in 3-4-5-and 6-year-olds and adults
.
Dev Psychol
.
2008
;
44
(
5
):
1457
65
.