The influence of magnocellular and parvocellular visual information on global processing in White and Asian populations.

Humans have the remarkable ability to efficiently group elements of a scene together to form a global whole. However, cross-cultural comparisons show that East Asian individuals process scenes more globally than White individuals. This experiment presents new insights into global processing, revealing the relative contributions of two types of visual cells in mediating global and local visual processing in these two groups. Participants completed the Navon hierarchical letters task under divided-attention conditions, indicating whether a target letter "H" was present in the stimuli. Stimuli were either 'unbiased', displayed as black letters on a grey screen, or biased to predominantly process low spatial frequency information using psychophysical thresholds that converted unbiased stimuli into achromatic magnocellular-biased stimuli and red-green isoluminant parvocellular-biased stimuli. White participants processed stimuli more globally than Asian participants when low spatial frequency information was conveyed via the parvocellular pathway, while Asian participants showed a global processing advantage when low spatial frequency information was conveyed via the magnocellular pathway, and to a lesser extent through the parvocellular pathway. These findings suggest that the means by which a global processing bias is achieved depends on the subcortical pathway through which visual information is transmitted, and provides a deeper understanding of the relationship between global/local processing, subcortical pathways and spatial frequencies.

Introduction

Every day the human visual system is bombarded by vast amounts of visual information. Typically developing individuals are able to quickly and efficiently group individual features of a scene together to form a global whole, a phenomenon known as the ‘global precedence effect’ [1-3]. However, recent research has shown that the relative preference for global versus local distribution of attention may differ based on culture.

Comparisons between White (i.e., Canadians, Americans, British and Australians) and East Asian populations (i.e., Chinese and Japanese) show that East Asian individuals process objects and scenes more globally than White individuals ([4-7] although see [8-12]). For example, when individuals are asked to make absolute judgments (requiring local processing) and relative judgments (requiring global processing), East Asian individuals perform better than White individuals at the relative task while White individuals perform better than East Asian individuals at the absolute task [13]. Similarly, an fMRI study showed that achieving equivalent levels of behavioral performance required more sustained attentional effort for the absolute task in East Asian individuals and for the relative task in White individuals [14]. When the global processing bias is directly compared between White and East Asian individuals, a global advantage is found in East Asian relative to White individuals, extending through to a second generation of Asian-Australians [15]. An electrophysiological study examining the neural mechanisms and the temporal dynamics related to a global processing bias has shown a greater sensitivity to global congruency in East Asian relative to White individuals, as indexed by an early P1 component [16].

Studies involving eye movements have suggested that perceptual differences may be driven by culture-specific tuning towards visual spatial frequency information [17]. Face recognition studies show that White individuals preferentially process high spatial frequency information from foveal vision and are more biased to local processing in hierarchical stimuli, while East Asian individuals preferentially process contextual information by relying on extra-foveal vision during face recognition, favoring globally-based holistic stimulus processing [17]. East Asian individuals also show more of a reliance on extra-foveal vision than White individuals when detecting low level visual stimuli [18] and for change detection of complex real-world stimuli [5], indicating that East Asian individuals allocate their attention more broadly than White individuals.

Numerous psychophysical studies have shown that the global precedence effect in typically developing individuals is mediated by low spatial frequencies [19-21]. Studies have also shown that stimuli presented without low spatial frequency information do not result in a global precedence effect [22, 23], but rather result in a local precedence effect [24]. Since White and East Asian individuals both show a global precedence effect, it is unlikely that differences in global processing are due to an inability to process low spatial frequency information. Based on previous research showing that culture-specific tuning towards visual spatial frequency information can influence perception, we explore the notion that differences in global processing are driven by differences in the means by which low spatial frequency information is conveyed through the visual system.

Before visual information makes its way from visual cortex to extrastriate visual areas by the dorsal stream, which governs the visual control of action, and the ventral stream, which governs visual perception, it initially passes from the retina to the primary visual cortex via the relay station called the lateral geniculate nucleus [25, 26]. The lateral geniculate nucleus consists of two pathways, magnocellular and parvocellular, that operate relatively independently in early visual processing and encode contrasting information. Specifically, magnocellular pathway neurons are visual cells that are not responsive to color and are known for processing achromatic, low contrast stimuli [27, 28], which is critical for global processing [20, 24]. In contrast, parvocellular pathway neurons are visual cells that are highly responsive to opposing colours (red-green or blue-yellow) [29] and high spatial frequencies [30, 31], requiring much higher contrast (~8% at least) when detecting achromatic stimuli [32]. While both magnocellular and parvocellular cells are present in both dorsal and ventral processing streams [27, 28], magnocellular cells are primarily conveyed through the dorsal stream, responsible for global processing, and parvocellular cells are primarily conveyed through the ventral stream, responsible for local processing.

A powerful way to gain insight into the way global information is processed in the visual system is to examine how visual information is processed through the magnocellular and parvocellular streams. For example, when stimuli are biased so that they are processed primarily through either the magnocellular or parvocellular pathway, global processing abilities differ between the two pathways in both individuals with simultanagnosia [33] and autism spectrum disorder [34]. This biasing can be achieved by manipulating the spatial frequencies conveyed through the parvocellular pathway. Although an increased sensitivity for high spatial frequencies is crucial for local processing via the parvocellular pathway specifically, the parvocellular pathway’s sensitivity to color information can also be selectively stimulated to act as a low-pass filter, conveying low spatial frequency information (required for global processing) when stimulated with isoluminant color stimuli [19, 35]. Although the magnocellular pathway is considered the primary pathway through which global information is transmitted, biasing both pathways to convey global information provides new insights into the relative contribution of these cell types to visual perception. Here we investigate the relative contribution of these two types of visual cells to the global precedence effect across Asian and White groups.

To examine the contribution of magnocellular and parvocellular cells to the global precedence effect, we used the Navon letters task [3] to compare global/local processing differences between White and Asian groups. Participants completed the Navon task under divided-attention conditions, indicating whether a target letter “H” was present in the hierarchical stimuli. Based on previous research [15] we expected that overall, Asian participants would show a stronger global advantage than White participants as indexed by faster reaction times, lower number of errors made, and lower inverse efficiency scores when the target was presented at a global compared to a local level. We also used psychophysical techniques to examine the mechanism underlying differences in global/local processing between groups by biasing stimuli to test the indirect contribution of the dorsal and ventral pathways in mediating global and local visual processing. For each participant we first established the achromatic contrast threshold and chromatic isoluminance threshold [36]. Next, to test global and local visual processing, we used these thresholds to dynamically generate magnocellular- and parvocellular-biased stimuli from a set of ‘unbiased’ hierarchical letter stimuli. Considering the parvocellular system’s ability to convey low spatial frequencies when stimulated with isoluminant color stimuli, both magnocellular and parvocellular systems were biased to activate in response to the same range of spatial frequencies to test the relative contributions of the dorsal and ventral pathways in mediating global visual processing in White compared to Asian participants. Since global processing relies on low spatial frequency information and previous research [17] has shown that White individuals preferentially process high spatial frequency information, which generally relies on the parvocellular system, we hypothesized that global processing in White participants would be more influenced by the parvocellular system than the magnocellular system when the ventral stream is biased to convey low spatial frequency information via the parvocellular pathway. By filtering out the high-spatial frequency information (required for local processing) from the parvocellular pathway, such that mainly low-frequency information (required for global processing) would pass through, we hypothesized that in the parvocellular-biased condition White participants would process the target faster and more accurately at the global level than the local level. Similarly, since Asian individuals have been shown to rely more on low contrast stimuli, which involves mainly the magnocellular pathway, we expected that in the magnocellular-biased condition Asian participants would process the target faster and more accurately at the global level than the local level. By filtering out high-spatial frequency information from the parvocellular stimuli, this allowed us to compare global processing abilities overall when stimuli are biased to one of the two subcortical pathways (magnocellular/parvocellular) as a way to examine if any differences in processing ability between the two subcortical pathways are responsible for the global processing advantage commonly found in the research.

Method

Participants

This study tested two groups: (a) 27 White participants (17 females) with a mean age of 23.5 years (18–36 years old) and (b) 25 Asian participants (15 females) with a mean age of 21.4 years (18–28 years old). The number of participants was determined based on comparable research examining cultural differences in global processing [15, 16] and magnocellular functioning [37]. Based on this resulting sample size, a sensitivity analysis for the key prediction (i.e., the group x condition x level interaction) was conducted. G*Power 3.1. [38] was used to conduct an F-test for repeated measures ANOVA, specifying a within-between interaction with 2 groups (Asian, White) and 6 measurements (2 levels x 3 conditions). Assuming an error probability of .05 and a nonsphericity correction of 1, the study was 80% powered to detect an effect size ηp2 = .021.

All participants were right-handed young adults with normal or corrected-to-normal vision. Ethnicity was determined by self-report and all participants were Introduction to Psychology students studying at an English-language university. All White participants and 11 Asian participants indicated English as their first language, while 14 Asian participants indicated Chinese as their first language. However, based on the rigorous English language testing required before being accepted into a University of Manitoba degree program, all participants were expected to be highly fluent at identifying single English letters, although no direct information is available regarding each participant’s specific English language ability. Participants were excluded from the study if they failed a colorblindness test administered before the start of the experiment (N = 7). Participants signed an informed consent form in writing before taking part in the study, which was approved by the Research Ethics Board (REB1) at the University of Manitoba.

General procedure

Participants signed a consent form and completed a short demographics questionnaire. They then completed the experiment, which began with a color-blindness test, followed by a luminance contrast thresholding task of which the results were used to create the magnocellular-biased stimuli and an isoluminance task of which the results were used to create the parvocellular-biased stimuli. The final portion of the experiment involved a computerized task using the hierarchical letters task [3].

Psychophysical thresholds

To determine the luminance contrast for each participant, a multiple staircase procedure was used to find the luminance threshold. Participants were shown light gray hierarchical stimuli overlaid on a dark gray background and asked if the stimuli was detected following each stimulus presentation. On 25% of the trials no stimulus was presented, which served as catch trials. Each stimulus was presented for 1500 ms, during which a response was made. If the stimuli was detected, the contrast between the stimuli and background was decreased in the next trial by modifying the luminance of the stimuli. Otherwise the contrast was increased. A commonly used luminance threshold-finding algorithm [36] was used to compute the mean of the turnaround points above and below medium-gray background. A luminance (~3.5% Weber contrast) value was then computed from this threshold for the grayscale stimuli and was used in the low-luminance-contrast (magnocellular-biased) condition.

For the chromatically defined stimuli in the isoluminant (parvocellular-biased) condition, the isoluminance point was found using heterochromatic flicker photometry [39, 40] with stimuli consisting of the same hierarchical letters from the luminance contrast task displayed in alternating colors (pure red and green). By alternating the two colors in the range of 12–20 Hz, the flicker disappears for a small range of luminance values. The color values at the point where the two colors appear to fuse together and the stimulus appears steady is the participant’s isoluminance interval. In this task participants used the up and down arrow keys to adjust the green color to the point at which the stimulus appeared steady. Depending on the response, the output of the green color was adjusted up or down such that the participant passed over the isoluminant point many times, and the average of the values in the narrow range when the participant reported a steady stimulus was computed as the isoluminance value for that participant. This luminance contrast threshold and isoluminance value was then used to dynamically create the magnocellular- and parvocellular-biased stimuli respectively for the experiment that followed.

Stimuli and procedure

All experiments were programmed using the Python programming language (Python Software Foundation, https://www.python.org/). Estimation of the psychophysical thresholds and the subsequent experiments were conducted in a low-lit room with an enclosure around the monitor to ensure lighting remained consistent for all participants. Stimuli were presented on a 20-inch color monitor (resolution: 1920 x 1200 pixels; refresh rate 120 Hz) placed 50 cm in front of the participant and responses were recorded using the left and right arrow keys on a keyboard. A chin rest was used to stabilize the viewing distance and all participants made their responses with the right hand. All stimuli were presented in a pseudorandom order to ensure that identical stimuli were not presented consecutively.

Participants were instructed to indicate by key press as quickly and accurately as possible whether the target letter “H” was observed. The target could appear either as the small local letter or the large global letter (Fig 1), allowing the individual’s implicit (i.e., uninstructed) preference for one level or the other to be assessed. The experiment began with one block of practice trials followed by the experimental trials. One third of the trials contained the target letter at the global level, one third of the trials contained the target letter at the local level, and one third of the trials served as catch trials with the target absent at both levels of processing. The stimuli were coded as Global (H made up of smaller distractor letters, either R or S), Local (H displayed as small letters forming a global S or R) and Neither (R or S made up of smaller distractor letters, either S or R respectively) for a total of 6 stimuli. The background of the hierarchical stimuli subtended 7.84° horizontally and 9.26° vertically, the global letters subtended a visual angle of 4.7° x 6.42° and the local letters subtended .73° x 1°, respectively. The distance between the local letters was .2° visual angle. Initially these stimuli were presented as ‘unbiased’ (i.e., not biased towards the magnocellular or parvocellular pathways) hierarchical letter stimuli displayed as black letters on a grey screen. These ‘unbiased’ stimuli were converted into achromatic magnocellular-biased stimuli and red-green isoluminant parvocellular-biased stimuli using the psychophysical thresholds outlined previously, resulting in three conditions: unbiased, magnocellular-biased and parvocellular-biased. Each of the 6 stimuli were presented in each of the 3 conditions for a total of 18 stimuli.

Fig 1

The three types of stimuli used to bias processing.

This example shows incongruent stimuli in which the local letters (R) combine to form the global letter (H). The contrast and luminance properties of the magnocellular- and parvocellular-biased stimuli have been altered to make the stimuli more discernible to viewers.

In the practice block, each of the six stimuli were presented twice as unbiased stimuli for a total of 12 trials. For the experimental trials, the 18 stimuli were presented 10 times for a total of 180 trials. Each trial began with a fixation cross for 1000 ms, after which the stimuli were presented until a response was indicated. Reaction time (RT) and accuracy (ACC) were collected as dependent measures, and to account for both measurements an adjusted RT measure called inverse efficiency score (IES; [41, 42]) was calculated as: IES = RT/ACC and also used as a dependent measure. While conventional reaction time and accuracy measures provide statistics regarding the speed and error rates independent of one another, the addition of inverse efficiency score to the analysis provides a comprehensive summary of the findings by combining both measures together. IES takes into consideration differences in speed-accuracy trade-offs by adjusting reaction time performance for sacrifices in accuracy that might have been made in favor of speed. A mean reaction time achieved with a high accuracy will have a lower IES than the same reaction time achieved at the cost of more errors. To determine whether there were any perceptual differences between White and Asian participants attributed to relative differences in global versus local distribution of attention, 2 (Group: White, Asian) x 2 (Level: Global, Local) x 3 (Condition: Magnocellular, Parvocellular, Unbiased) ANOVAs were carried out on log-transformed reaction time and inverse efficiency scores. Reaction times and inverse efficiency scores were log-transformed to account for non-normality in the data. Since accuracy data was positively skewed and transformation of the data did not address concerns related to normality, a Poisson regression was used to analyze the number of errors made. All post-hoc pairwise comparisons were performed using Bonferroni correction and alpha = .05.

Results

Reaction time analysis

Results of the ANOVA for log-transformed reaction time showed significant main effects of level, F(1,50) = 25.487, p < .001, ηp2 = .338, and condition, F(2,100) = 401.206, p < .001, ηp2 = .889, and a significant level x condition interaction, F(2,100) = 10.998, p < .001, ηp2 = .180. There was no significant main effect of group, F(1,50) = 1.714, p = .196, ηp2 = .033, and no group x level x condition interaction, F(2,100) = .197, p = .659, ηp2 = .004. Post-hoc t-tests on the level x condition interaction revealed faster identification of the target when it was presented globally compared to locally in the parvocellular condition (p < .001). Post-hoc tests also revealed that when the target was presented at the global level, it was identified faster in the unbiased condition compared to the magnocellular condition (p < .001), and in the parvocellular condition compared to the unbiased condition (p < .001). When the target was presented at the local level, it was identified faster in the unbiased condition compared to the parvocellular (p < .001) and magnocellular (p < .001) conditions. The target was also identified faster in the parvocellular compared to the magnocellular condition (p < .001).

Accuracy analysis

A Poisson regression was run to predict the number of errors made based on whether the target was located in the global or local configuration (Level: Global/Local) and the condition type (Condition: Unbiased/Magnocellular/Parvocellular). Results of the Poisson regression using the Wald Chi-Square statistic showed significant main effects of level, X2(1) = 6.81, p = .009, and condition, X2(2) = 6.7.46, p <, 001, and significant interactions between level x condition, X2(2) = 13.56, p = .001, and level x condition x group, X2(2) = 6.297, p = .043. There was no significant main effect of group, X2(1) = 1.128, p < .001 and no level x group interaction, X2(1) = .736, p = .391.

Results of the level x condition interaction revealed the less errors were made when the target was presented at the global level compared to the local level in the magnocellular (p = .015) condition. No differences in accuracy were found in the parvocellular or unbiased conditions. For both global and local levels, the number or errors made was the highest in the magnocellular condition compared to the parvocellular (Global: p = .001; Local: p < .001) and unbiased (Global: p = .001; Local: p < .001) conditions. No accuracy differences were found between parvocellular and unbiased conditions.

Results of the three-way group x level x condition interaction revealed that for both Asian and White participants, when the stimuli was presented at the local level the number of errors made was found to be higher in the magnocellular condition compared to the parvocellular (Asian: p = .03; White: p = .037) and unbiased (Asian: p = .002; White: p = .016) conditions (Fig 2). No significant differences between conditions were found for either group when the target was presented at the global level.

Fig 2

Accuracy results.

More errors were made identifying the target in the magnocellular condition compared to the parvocellular and unbiased conditions when it was presented at the local level for both White and Asian individuals. Error bars show +/- 1 SEM.

Adjusted reaction time analysis using inverse efficiency score (IES)

Results of the ANOVA on IES showed significant main effects of level, F(1,50) = 27.665, p < .001, ηp2 = .356, and condition, F(2,100) = 202.826, p < .001, ηp2 = .802, and significant interactions between level x group, F(1,50) = 5.305, p = .025, ηp2 = .096, condition x level, F(2,100) = 11.191, p < .001, ηp2 = .183, and group x condition x level, F(2,100) = 2.977, p = .05, ηp2 = .056. There was no significant main effect of group, F(1,50) = 1.700, p = .198, ηp2 = .033.

Results of the level x condition interaction revealed faster identification of the target when it was presented at the global level compared to the local level in both the magnocellular (p = .001) and parvocellular (p < .001) conditions. No differences were found in the unbiased condition. For both global and local levels the target was identified slower in the magnocellular condition compared to the parvocellular (Global: p < .001; Local: p < .001) and unbiased (Global: p < .001; Local: p < .001) conditions. A difference was also found between parvocellular and unbiased conditions at the local level (p < .001) but not at the global level. The level x group interaction revealed that within the Asian group, the target at the global level was processed faster than at the local level (p < .001). No differences were found between global and local processing for White participants.

Results of the three-way group x level x condition interaction revealed that in the parvocellular condition, when the target was presented at the global level, White participants responded more quickly than Asian participants (p = .05; Fig 3). In the unbiased condition, when the target was presented at the local level, White participants responded more quickly than Asian participants (p = .045). Results also showed faster processing for White participants when the target was presented at the global level compared to the local level (p < .001) in the parvocellular condition, while Asian participants showed faster processing at the global compared to the local levels in both magnocellular (p < .001) and parvocellular (p < .001) conditions. Finally, at the global level White participants showed slower processing of the target in magnocellular compared to parvocellular (p < .001) and unbiased (p < .001) conditions and for unbiased compared to parvocellular conditions (p = .032). At the local level, White participants also showed slower processing for magnocellular compared to parvocellular (p < .001) and unbiased (p < .001) conditions and for parvocellular compared to unbiased conditions (p < .001). At the global level, Asian participants showed slower processing in the magnocellular compared to parvocellular (p < .001) and unbiased (p < .001) conditions. No significant differences were found between parvocellular and unbiased conditions. At the local level, Asian participants showed slower processing in the magnocellular compared to parvocellular (p < .001) and unbiased (p < .001) conditions, as well as for parvocellular compared to unbiased conditions (p < .001).

Fig 3

Inverse efficiency score results.

Between groups, the target was identified faster at the global level compared to the local level in the parvocellular condition for White participants and in both the parvocellular and magnocellular conditions for Asian participants. Error bars show +/- 1 SEM.

Discussion

Previous studies suggest that East Asian individuals process scenes more globally than White individuals [15, 16]. Perceptual differences may be driven by culture-specific tuning towards visual spatial frequency information, as demonstrated in studies showing that White individuals preferentially process high spatial frequency information, while East Asian individuals preferentially process low spatial contextual information [17, 18]. Within the visual system, high spatial frequency information, required for local processing, is mainly conveyed via the parvocellular pathway and low spatial frequency information, required for global processing, is mainly conveyed via the magnocellular pathway. From this, we hypothesized that perceptual differences between White and Asian participants might be driven by biases towards one pathway or the other. The aim of this study was to test the potential mechanism underlying these differences by examining the relative contributions of the magnocellular and parvocellular pathways in mediating global and local visual processing in Asian and White participants.

Overall, our comparison of White and Asian participants revealed little evidence for cultural differences. In the unbiased condition, which would most closely represent the stimuli used in comparable previous studies, White participants performed similarly to those of Asian participants. Even more surprising, no global bias was found within each group. The lack of robust and consistent evidence is unexpected, given the original reports of statistically significant cultural differences. However, there are a number of reasons that could explain these differential findings.

First, stimulus presentation may have weakened our ability to see cultural differences. In previous research stimuli has been presented laterally rather than centrally [15]. While location of the stimuli was not found to influence global/local response times in their study, it is possible that faster reaction times for Asian compared to White participants were due to a broader allocation of attention in the Asian group rather than a global processing bias [5, 18]. In the current study, the stimuli were always presented centrally, so it may be the case that stimulus presentation was more advantageous for White participants, who have been shown to preferentially process information from foveal vision [17]. However, the number of errors made were low in both groups, suggesting that Asian participants were not that affected by the location of the stimulus.

A more plausible explanation is that differences in behavioral performance were masked by other factors such as attention. For example, if attention is viewed as a spotlight where stimuli falling within the beam of the spotlight are processed preferentially, then examining how the spotlight differs between groups may provide insight into why the two groups differ if there are characteristics that are unique to one group compared to the other, like size of the spotlight. While previous behavioral studies have shown that Asian individuals perform better on a task requiring global processing and White individuals perform better on a task requiring local processing [13], an fMRI study showed that equal levels of behavioral performance were achieved by allocating more sustained attentional effort for the local task in Asian individuals and the global task in White individuals [14]. In the latter case, this finding corresponds with activation in the frontal and parietal regions of the brain, which typically show greater activation for more demanding tasks and are thought to mediate cognitive control over working memory and attention [14]. As such, it is possible that White participants were recruiting more attentional resources to produce the same behavioral outcome as evident in the Asian group. Future research using a combined fMRI/behavioral approach will help determine the extent to which attention influences global/local processing.

It is possible that individual biases may have influenced the extent to which global processing was observed. Previous research has shown that the degree of individual bias toward global information can vary based on stimulus parameters, such as the aspect ratio of local to global items [43, 44], the overall visual angle [45], or the amount of time participants have to view stimuli [46]. Additionally, older individuals [47], individuals induced into a state of negative affect [48], individuals from remote cultures [49] and musicians [50] all tend to show a local compared to global processing bias. Conversely, individuals from collectivist cultures [15] and individuals induced into a state of positive affect [51] tend to show a preference for global processing. Thus, a global bias can be influenced by participant characteristics and is not absolute. As such, culture is only one of many possible influencers of global processing, and future research involving larger sample sizes in each group and controlling for a wider breadth of participant characteristics will help to untangle the relationship between culture and global/local processing.

Finally, factors related to the two samples themselves may have limited the extent to which global processing was observed. While participants were asked to self-report their ethnicity and familiarity with the English language, other potentially important contributing information such as where participants were born, how long they had been residing in an English-speaking country, as well as factors related to socioeconomic status, possible influences from reading disorders, and familiarity of experience with digital technology were not collected, limiting the extent to which we can generalize these findings to the general population. While these results are promising and relevant, providing a deeper understanding of the potential mechanisms underlying a global processing advantage, considering the impact these factors may have on the current findings suggests that these findings should be considered preliminary and a foundation upon which future studies should be carried out.

Although no differences were observed in the unbiased condition, differences were found within the magnocellular and parvocellular conditions. Differences in reaction times and accuracy scores (measured by the number of errors made) emphasize the importance of considering both variables in the same measure, which was achieved using an inverse efficiency score. In this case, a mean reaction time achieved with a high accuracy will have a lower inverse efficiency score than the same reaction time achieved at the cost of more errors. A direct comparison between groups indicated lower inverse efficiency scores in White compared to Asian participants when processing the target at the global level, compared to the local level, in the parvocellular condition specifically. Individual group differences also showed that White participants processed the target with lower inverse efficiency scores at the global level compared to the local level in the parvocellular condition only, while no significant differences between global and local processing were observed in the magnocellular condition. Asian participants, however, processed the target with lower inverse efficiency scores at the global level compared to the local level in the magnocellular condition. In the parvocellular condition, Asian participants also showed lower inverse efficiency scores at the global level compared to the local level. Together, this suggests that global/local processing in White participants is influenced more by the parvocellular stream, and in Asian participants by the magnocellular stream (and to a lesser extent the parvocellular stream).

Based on our knowledge that the parvocellular pathway typically transmits high spatial frequency information and the magnocellular pathway is biased to convey low spatial frequency information, we suggest that the potential mechanism underlying global/local processing in White individuals relies more heavily on information from the parvocellular pathway, while the mechanism in Asian individuals relies more heavily on information from the magnocellular pathway (and to a lesser extent the parvocellular pathway). This would explain why some research has found that White individuals show a local processing bias in global/local processing tasks (see [52]), and why White individuals show a less robust global processing bias compared to Asian individuals. While the current study biased parvocellular stimuli to convey low spatial frequency information so that global processing between the two pathways could be examined more directly, this strongly limited the extent to which we could interpret the results as the significant advantage shown by White compared to Asian individuals for processing global information when low spatial frequency information was conveyed through the parvocellular pathway may not have been as evident if high spatial frequencies were not filtered out of the stimuli. Consequently, these results can only be interpreted in the context of low spatial frequency stimuli, and future studies will be required to determine the extent to which White individuals process parvocellular stimuli when they are not filtered to isolate low spatial frequency information. Neuroimaging studies will also be an important contributor in verifying the extent to which these differences are observed in the brain.

A stronger influence of the magnocellular stimuli on global processing in the Asian group may also explain why these individuals show an early sensitivity to global information coding [16]. Magnocellular pathway neurons, critical for global processing, transmit information much faster than parvocellular pathway neurons from the lateral geniculate nucleus to the primary visual cortex. If visual processing in Asian individuals is more influenced by the magnocellular pathway than it is in White individuals, this suggests that a more robust global processing bias in Asian individuals may be driven by a stronger influence of the magnocellular pathway over the parvocellular pathway. This is also in line with previous research showing that Asian individuals rely more on low spatial, extra foveal vision.

Visual saliency between global and local features has also been suggested as a potential explanation for why White individuals process scenes less globally than Asian individuals. Previous work has found that White individuals were less efficient at detecting global compared to local feature changes, while Asian individuals performed equally well on both conditions, suggesting that the behavioral disadvantage of White individuals in the global task stemmed from differences in visual saliency between global and local features [16]. The rationale is that since visual processing of global features precedes the analysis of local information, an initial preference for global processing would conflict with local information, inhibiting the ability to detect local features [3, 53, 54]. Further research showed that participants identified local targets slower in the presence of a global shape, even when the global information was irrelevant [54]. From this, previous research suggests that when White individuals are required to detect changes in local information, the presence of global features is more disruptive for them than for Asian individuals, who seem to benefit from a top-down attention control to global features [16]. As such, the visual saliency induced by the global feature change did not seem to disturb processing in Asian individuals to the same extent as in White individuals. However, our results suggest that saliency cannot fully account for this difference in processing between the two groups since White individuals showed a global precedence in the parvocellular-biased condition using the same stimuli as was used in the unbiased and magnocellular-biased conditions. If visual saliency is the driving force behind a global processing difference, then Asian individuals should show the same global processing advantage over White individuals regardless of condition. It is more likely that the notion that White individuals process scenes less globally than Asian individuals is due to the way that visual information is conveyed by the two pathways. This is not to say that White individuals do not benefit from information transmitted via the magnocellular pathway, but rather suggests that the way in which visual information is transmitted via the two streams differs between groups.

Another potential factor that could influence magnocellular functioning is experience with digital technology. For example, several studies involving individuals with developmental dyslexia, in which magnocellular visual functioning is selectively deficient, have shown that reading difficulties associated with this visual deficit can be improved by playing action video games [55, 56]. This type of video game in particular involves a specific set of qualitative features in order to be successful in playing the game: extraordinary speed, an ability to take on a high degree of perceptual, cognitive and motor load to accurately maneuver through the game, keeping multiple action plans in memory and assessing each event presented in the game in the context of these action plans, and peripheral processing abilities [57]. As such, if individuals in the current study were avid video gamers, it is possible that differences in magnocellular functioning could be a result of familiarity with this type of gaming technology, a potential confound that should be considered in future studies.

It should also be noted that while the magnocellular and parvocellular stimuli were biased in a certain way, both types of cells may have still been responding. While magnocellular and parvocellular information conveyed from the lateral geniculate nucleus to the primary visual cortex does remain, to some extent, functionally segregated, once this information projects beyond V1 there is a considerable amount of mixing of magnocellular and parvocellular signals. From the primary visual cortex information is still conveyed via two largely functionally distinct streams, the dorsal and ventral streams, however each stream is composed of a mixture of magnocellular and parvocellular signals [58]. Although information in one stream is influenced by the other, lesion studies involving the magnocellular and parvocellular pathways provide evidence for a distinct relationship between the response properties of the magnocellular and parvocellular cells and the functions of the cortical regions along the dorsal and ventral streams. For example, lesions to the parvocellular pathway results in deficits in chromatic vision, texture perception, pattern perception, acuity and a loss in contrast sensitivity at low temporal and high spatial frequencies [59, 60]. Lesions to the magnocellular pathway have been found to cause deficits in flicker and motion perception [59]. Together, this demonstrates that while the dorsal and ventral streams receive input from both magnocellular and parvocellular cells, the effect these cells have on each of the two streams differs. As such, in the current study we do not assume that biasing the stimuli towards the magnocellular and parvocellular pathways only activates that individual pathway, but rather that the manipulation biases processing in one pathway over the other.

Conclusion

We have demonstrated that White individuals can be biased to process scenes more globally than Asian individuals as long as low spatial frequency information is conveyed through the parvocellular pathway. Asian individuals also show a global precedence effect when low spatial frequency information is conveyed through the magnocellular pathway, and to a lesser extent through the parvocellular pathway. These findings suggest that a global processing advantage can be altered when stimuli is biased towards one of two subcortical pathways: the magnocellular or parvocellular pathway. That is, White individuals may depend more on ventrally-based information transmitted through the parvocellular pathway in global/local processing, while Asian individuals may depend more on dorsally-based information transmitted through the magnocellular pathway. However, since the parvocellular stream conveys high spatial frequency information useful for local processing when stimuli are not isoluminant, this may explain why research often finds that Asian individuals process scenes more globally than White individuals. Here we examined the more intricate mechanisms underlying global and local processing by looking individually at the two subcortical pathways that together form our overall visual perception–the magnocellular and parvocellular pathways–demonstrating that when the parvocellular pathways is biased to only convey low-frequency information, this global advantage in Asian individuals can be reduced such that White individuals show a global advantage compared to Asian individuals. Of course, as humans we typically view visual stimuli in our environment as a combination of the inputs sent through both the magnocellular and parvocellular pathways, but in understanding the mechanism by which this global advantage is formed, it is important to consider the contribution of each pathway and how different types of spatial frequency information can alter visual output.

22 Dec 2021

I have also read the manuscript very carefully and I suggest adding in the rationale and in the discussion relevant information to the literature that does not support the theory with respect to the differences between magno and parvocellular segregation when dealing with behavioral methods. Alternative explanations of the results should also be given for completeness to the one provided. Moreover, I think that the significant results obtained are promising and relevant but, because of the small sample size recruited, I suggest to be more careful in the interpretation in terms of a characteristic of a general population (Asian vs Caucasian). Some additional limitations should be considered in the final discussion: are the sample matched for socio-economical status? and/or for the familiarity and experience of digital technology? Several studies highlight that magnocellular functions could improve significantly by the use of action video games, therefore it could be relevant to include this information in the discussion. In the light of the above mentioned limitations I suggest including in the discussion session that the results could be considered as preliminary.

Please submit your revised manuscript by Feb 03 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Mariagrazia Benassi

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please note that according to our submission guidelines (http://journals.plos.org/plosone/s/submission-guidelines), outmoded terms and potentially stigmatizing labels should be changed to more current, acceptable terminology. For example: “Caucasian” should be changed to “white” or “of [Western] European descent” (as appropriate), including in the title and abstract. In addition, please change "female” or "male" to "woman” or "man" as appropriate, when used as a noun (see for instance https://apastyle.apa.org/style-grammar-guidelines/bias-free-language/gender).

3. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

4. Thank you for stating the following in the Acknowledgments Section of your manuscript:

This research was supported by a grant from the Natural Science and Engineering Research Council of Canada (NSERC) (Grant no. 04964-2018) held by J.J.M.

Please note that funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

This research was supported by a grant from the Natural Science and Engineering Research Council of Canada (NSERC) (Grant no. 04964-2018) held by J.J.M.

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

5. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is an interesting study that addresses the culture differences in visual perception. Results show that Caucasians prefer to globally process low spatial frequency information that is presented via the parvocellular pathway, while Asians show a global processing when low spatial frequency information is presented via the magnocellular pathway. These findings suggest the association between global/local processing and subcortical pathways, advancing our understanding of the underlying mechanisms of culture differences in visual perception.

My concerns are listed below.

1) Please further explain how to distinguish magnocellular stimuli from parvocellular stimuli.

2) There are only 6 stimuli for each condition. The number of stimuli is not enough to achieve a reliable measure.

3) Why use IES? Please provide the detailed explanation for the use of such statistical method.

Reviewer #2: REVIEW

The study is interesting, well conducted and the manuscript is well written. Yet, there are some issues (and missing information) with the rationale behind the study and with the methodology which make it difficult to evaluate the actual soundness of the design and its publishable quality.

Methods:

The main question arising is: if comparing Caucasian and Asian participants who (probably) are readers of two (or more) different writing systems, why did the authors chose Navon’s letter task and not Navon’s geometric shapes task, which would have offered the same opportunities but not have suffered from any bias due to reading habits?

The absence of certain crucial conditions makes interpretation of the results very uncertain. For instance, a condition where P-function including low- and high-frequency information is missing (as also stated by the authors in the Discussion, suggesting that future research should provide this information). Thus, any hypothesis about what the results would have been with a “normal” P-condition are purely speculative.

Participants:

The information provided on participants’ characteristics is very poor and insufficient to evaluate their actual comparability. Where were the EA participants from? And the Caucasian ones? What language were they speaking and reading/writing? How long had they been in an English-speaking country? Were any dyslexic students among the participants? It is not sufficient to “assume” they were fluent in reading English: reading fluency should have been measured, compared with the Caucasians’ fluency and, if different, used as a covariate.

Rationale:

The rationale behind the study is not clear enough. The cultural and physiological differences seem to be addressed mixing up causes and effects and the reasoning is sometimes circular: how does culture affect physiology, and how does physiology affect culture? It is not clear whether the authors are proposing that physiological differences constrain functional patterns or that culture has influenced physiology through habits.

Minor points:

line 138-139: “we expected that in the parvocellular-biased condition Caucasians would process the target faster and more accurately at the global level than the local level”: the logic of this expectation should be explained more clearly

line 153: the sentence is redundant and circular (power of .8 and 80% power are the same information)

line 158: were expected to be highly fluent: you should state that no direct information is available

line 228: delete “are”

lines 370-372: what is the hypothesized role of sustained attention? I would guess that selective attention should be involved instead.

Lines 416-418: the absence of this condition with no filtering of low special frequency information is not an “optional addition” but a strong limitation of the study (making interpretation of results more difficult) and it should be described as such.

Lines 479-480: what do the authors mean by “depends on”?

Lines 486-488: the causal relationship (if any is hypothesized) should be made explicit – if no hypothesis is proposed, or if a parallelism without any causal relationship is hypothesized, it should be clearly stated.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Apr 2022

Editor Recommendations:

1. About the statistics, I suggest controlling for the distribution of the data (in particular the accuracy seems to be not normally distributed) and using non-parametric model in the case of non-normal distribution of the data.

To control for the distribution of the data, reaction time and IES data were log-transformed and the analysis was re-run using the log transformed data. All results remained the same, and the statistics and figures have been updated to reflect the log-transformed data (p. 12-16).

Since accuracy in the form of error rate is considered count data and fails to fix the normality issue using a simple log transformation, a Poisson regression was used to analyze the count data (eg. The number of errors made). Since our data is positively skewed and contains a large proportion of zeros (eg. 0 errors made), the Poisson distribution is considered an appropriate statistical technique to address our normality issues. Most of the results remained the same, however, the level x group interaction was no longer significant after this analysis. This along with the corresponding pairwise comparisons have been removed, although this does not affect our results since they are based on the more comprehensive inverse efficiency score analysis.

2. I suggest adding in the rationale and in the discussion relevant information to the literature that does not support the theory with respect to the differences between magno and parvocellular segregation when dealing with behavioral methods. Alternative explanations of the results should also be given for completeness to the one provided. Moreover, I think that the significant results obtained are promising and relevant but, because of the small sample size recruited, I suggest to be more careful in the interpretation in terms of a characteristic of a general population (Asian vs Caucasian). Some additional limitations should be considered in the final discussion: are the sample matched for socio-economical status? and/or for the familiarity and experience of digital technology? Several studies highlight that magnocellular functions could improve significantly by the use of action video games, therefore it could be relevant to include this information in the discussion. In the light of the above mentioned limitations I suggest including in the discussion session that the results could be considered as preliminary.

We agree that more demographic information would allow for better control over potential confounds in the analysis, and have noted that a lack of these variables, including information about socio-economic status and familiarity of experience with digital technology, along with factors outlined by reviewer 2 (where participants were from, how long they had been residing in an English speaking country, and more specific information regarding any disorders that could have impacted reading ability), is a limitation to the study in the discussion section, emphasizing the preliminary nature of these results (p. 18-19).

Additionally, we have included a discussion about digital technology, with an emphasis on action video gaming, as an alternative explanation that could explain magnocellular functioning in the discussion (p. 22).

3. Please note that according to our submission guidelines, outmoded terms and potentially stigmatizing labels should be changed to more current, acceptable terminology. For example: “Caucasian” should be changed to “white” or “of [Western] European descent” (as appropriate), including in the title and abstract. In addition, please change "female” or "male" to "woman” or "man" as appropriate, when used as a noun (see for instance https://apastyle.apa.org/style-grammar-guidelines/bias-free-language/gender).

Caucasian has been changed to White individual (Asian has been changed to Asian individual for consistency).

4. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

Updated to include “in writing” (p.8)

5. Please note that funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form. Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows: This research was supported by a grant from the Natural Science and Engineering Research Council of Canada (NSERC) (Grant no. 04964-2018) held by J.J.M. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Funding information has been removed from the acknowledgments section. The funding statement as stands is correct.

Reviewer 1:

1. Please further explain how to distinguish magnocellular stimuli from parvocellular stimuli.

Further clarification of how magnocellular and parvocellular information were distinguished is provided at the end of the introduction (p.7).

2. There are only 6 stimuli for each condition. The number of stimuli is not enough to achieve a reliable measure.

The current study designed was modelled off of a prior study in which only 6 stimuli were used for each condition (magnocellular, parvocellular, unbiased) in Experiment 1, and 4 stimuli were used for each condition in Experiment 2 (Thomas et al., 2012). Even with their small sample size (5 healthy controls and 2 patients for Experiment 1 and 10 healthy participants for Experiment 2), a significant difference between conditions was observed. More recent studies have also employed a similar experimental set up (Guy et al. 2018: 4 stimuli: 2 congruent/2 incongruent; Primativo et al. 2020: 6 stimuli for each condition (parvocellular/unbiased); Baisa et al. 2021: 4 stimuli: 2 congruent/2 incongruent for each condition (small/medium/large).

References:

Guy, J., Mottron, L., Berthiaume, C., & Bertone, A. (2019). A developmental perspective of global and local visual perception in autism spectrum disorder. Journal of Autism and Developmental Disorders, 49(7), 2706-2720.

Primativo, S., Crutch, S., Pavisic, I., Yong, K., Rossetti, A., & Daini, R. (2020). Impaired mechanism of visual focal attention in posterior cortical atrophy. Neuropsychology, 34(7), 799.

Baisa, A., Mevorach, C., & Shalev, L. (2021). Hierarchical processing in ASD is driven by exaggerated salience effects, not local bias. Journal of Autism and Developmental Disorders, 51(2), 666-676.

3. Why use IES? Please provide the detailed explanation for the use of such a statistical method.

While conventional reaction time and accuracy measures provide statistics regarding the speed and error rates independent of one another, the addition of inverse efficiency score to the analysis provides a comprehensive summary of the findings by combining both measures together. IES takes into consideration differences in speed-accuracy trade-offs by adjusting reaction time performance for sacrifices in accuracy that might have been made in favor of speed. A mean reaction time achieved with a high accuracy will have a lower IES than the same reaction time achieved at the cost of more errors.

This explanation has been updated in the methodology section on p. 11-12.

Reviewer 2:

1. Methods: The main question arising is: if comparing Caucasian and Asian participants who (probably) are readers of two (or more) different writing systems, why did the authors chose Navon’s letter task and not Navon’s geometric shapes task, which would have offered the same opportunities but not have suffered from any bias due to reading habits?

While it is entirely possible that Asian or Caucasian participants could have been proficient in two or more different writing systems, all participants were Introduction to Psychology students studying at an English language university (University of Manitoba) where applicants are required to demonstrate sufficient mastery of English before being accepted into a program in order to meet the demands of classroom instruction, written assignments and participations in oral discussions. Based on this rigorous requirement, all participants were expected to be highly fluent at identifying single English letters. By using the Navon’s letter task rather than the Navon’s geometric shapes task, this provided more consistency with prior studies, allowing for comparisons to be made between our study and studies that have examined a global advantage in Asian populations in the past (namely, McKone et al., 2010).

2. The absence of certain crucial conditions makes interpretation of the results very uncertain. For instance, a condition where P-function including low- and high-frequency information is missing (as also stated by the authors in the Discussion, suggesting that future research should provide this information). Thus, any hypothesis about what the results would have been with a “normal” P-condition are purely speculative.

P.20 of the discussion has be re-written to acknowledge the absence of this condition with no filtering as a strong limitation to the study.

3. Participants: The information provided on participants’ characteristics is very poor and insufficient to evaluate their actual comparability. Where were the EA participants from? And the Caucasian ones? What language were they speaking and reading/writing? How long had they been in an English-speaking country? Were any dyslexic students among the participants? It is not sufficient to “assume” they were fluent in reading English: reading fluency should have been measured, compared with the Caucasians’ fluency and, if different, used as a covariate.

All participants were asked to indicate on their demographics form their ethnicity, as well as whether English was their first language and the languages they were fluent in. All Caucasians indicated English as their first language, while 11 Asians indicated English as their first language and 14 Asians indicated Chinese as their first language (Demographics updated on p. 8). All analyses presented in the manuscript were also carried out using a grouping variable in which Asians were split into two groups according to whether they spoke English as a first language or not, along with the Caucasian group, showing similar results to those outlined in the manuscript. Importantly, no significant differences were observed between the two Asian groups. However, the limitation to including these groups with much smaller sample sizes separately in the analysis resulted in decreased power and effect size. As such, the two groups were combined together into one Asian group to improve the power of our study.

While dyslexia was not queried specifically, as part of the demographics questionnaire, participants were asked to indicate whether they were diagnosed with any medical/psychological disorders. None were noted and thus no participants were excluded from the study.

We agree that more demographic information would allow for better control over potential confounds in the analysis, and have noted that a lack of these variables, including information about where participants were from, how long they had been residing in an English-speaking country, and more specific information regarding any disorders that could have impacted reading ability, is a limitation to the study in the discussion section (p. 18-19).

4. Rationale: The rationale behind the study is not clear enough. The cultural and physiological differences seem to be addressed mixing up causes and effects and the reasoning is sometimes circular: how does culture affect physiology, and how does physiology affect culture? It is not clear whether the authors are proposing that physiological differences constrain functional patterns or that culture has influenced physiology through habits.

Here we are proposing that physiological differences influence behavior. While many cross-cultural studies point to habits and environment as driving forces behind physiological differences in global/local processing, here we aimed to look at more fundamental mechanisms underlying global and local processing as a potential explanation for differences in global/local processing between Asians and Caucasians. Since low-frequency spatial information is required for global processing, we hypothesized that a Caucasian preference for processing information through the parvocellular pathway (most notably involved in processing high-spatial frequency information important for local processing) is responsible for a lesser global processing response compared to Asians, and that this global processing ability could be strengthened by filtering out high-spatial frequency information. Since previous studies have found that Asians rely more on low contrast stimuli, involving the magnocellular pathway (important for global processing), by filtering out high-spatial frequency information from the parvocellular stimuli, this allowed us to compare global processing abilities overall and when stimuli are biased to one of the two subcortical pathways (magnocellular/parvocellular) as a way to examine if any differences in processing ability between the two subcortical pathways are responsible for the global processing advantage commonly found in the research. This clarification has been added to the end of the introduction on p. 7.

Minor points:

5. line 138-139: “we expected that in the parvocellular-biased condition Caucasians would process the target faster and more accurately at the global level than the local level”: the logic of this expectation should be explained more clearly

We have re-written this statement to clarify the rationale behind this expectation. (p.7)

6. line 153: the sentence is redundant and circular (power of .8 and 80% power are the same information)

“power of .8” has been removed to avoid redundancy

7. line 158: were expected to be highly fluent: you should state that no direct information is available

This information has been updated on p. 8.

8. line 228: delete “are” - deleted

9. lines 370-372: what is the hypothesized role of sustained attention? I would guess that selective attention should be involved instead.

In the context of the fMRI study to which the sustained attention refers to, the hypothesized role of sustained attention is thought to correspond with the activation found in the frontal and parietal regions of the brain, which typically show greater activation for more demanding tasks and are thought to mediate cognitive control over working memory and attention. In this case it is hypothesized that sustained attention is involved rather than selective attention because the stimuli in this study do not involve selecting and focusing on a particular aspect of the stimuli while simultaneously suppressing irrelevant or distracting information, but rather involve directly focusing on the specific stimuli. The difference in the amount of sustained attention allocated to stimuli between Asian and White individuals is the variable of interest in this case. This hypothesized role of sustained attention has been clarified on p. 17-18.

10. Lines 416-418: the absence of this condition with no filtering of low special frequency information is not an “optional addition” but a strong limitation of the study (making interpretation of results more difficult) and it should be described as such.

P.20 of the discussion has be re-written to acknowledge the absence of this condition with no filtering as a strong limitation to the study.

11. Lines 479-480: what do the authors mean by “depends on”?

In this case, “depends on” refers to an underlying factor influencing a global processing bias: the magnocellular or parvocellular pathway. To clarify, this sentence has been re-written to address this ambiguity (p.24).

12. Lines 486-488: the causal relationship (if any is hypothesized) should be made explicit – if no hypothesis is proposed, or if a parallelism without any causal relationship is hypothesized, it should be clearly stated.

Reframed the end of the conclusion to be more specific as to our intent behind the study (p.24).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

10 Jun 2022

The influence of magnocellular and parvocellular visual information on global processing in White and Asian populations

PONE-D-21-29948R1

Dear Dr. Carther-Krone,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Mariagrazia Benassi

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

**********

16 Jun 2022

PONE-D-21-29948R1

The influence of magnocellular and parvocellular visual information on global processing in White and Asian populations

Dear Dr. Carther-Krone:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Mariagrazia Benassi

Academic Editor

PLOS ONE