Edited by: Pietro Spataro, Mercatorum University, Italy
Reviewed by: Takahiko Masuda, University of Alberta, Canada; Xiaoqian Yan, Université de Lorraine, France
This article was submitted to Cognition, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Across cultures, there are marked differences in visual attention that gradually develop between 4 and 6 years of age. According to the social orientation hypothesis, people in interdependent cultures should show more pronounced context sensitivity than people in independent cultures. However, according to the differential familiarity hypothesis, the focus on the salient object should also depend on the familiarity of the stimulus; people will focus more on the focal object (i.e., less context sensitivity), if it is a less familiar stimulus. To examine the differences in visual attention between interdependent and independent cultures while taking into account stimulus familiarity, this study used an eye-tracking paradigm to assess visual attention of participants between 4 and 20 years who came from urban middle-class families from Germany (
The way in which people attend to their visual field differs strongly between cultures. This has been demonstrated by various cross-cultural studies that have mainly focused on educated urban middle-class participants in Western and Eastern cultural contexts (e.g.,
Cultural differences in attentional styles have been reported consistently across a variety of tasks, such as picture description and recognition tasks (
Empirical evidence indicates that cross-cultural differences in attentional styles start to develop in the late preschool years, around 4 to 6 years, and become further pronounced in the years thereafter (
A recent study extended the research on the development of culture-specific attentional styles to cultures other than educated urban middle-class families from the USA or East Asia.
Familiarity has been considered as an important aspect in memory and cognitive processing (
The present study set out to close this empirical gap by examining how different types of stimuli, varying in familiarity, affect cross-cultural differences in context sensitivity in an eye-tracking task. Specifically, we were interested in, first, whether perceptual styles are consistent across stimulus categories and, second, whether the familiarity of the stimuli influences the cross-cultural differences in visual attention.
To analyze how the familiarity of stimuli affects cross-cultural differences in analytic and holistic attention, we selected two cultural communities, in particular a middle-class sample from Münster, urban Germany, and a sample from the Nso culture, living in a mainly subsistence-based farming ecology in Kumbo, rural Cameroon. The main reasons for selecting these samples were three-fold: (1) the samples differ in social orientations as described above: they represent an independent and interdependent cultural orientation, respectively; (2) the visual environments in these contexts are highly different; and (3) data exists on the attentional styles for both cultural communities (
The urban German middle class represents a prototype of an independent (
Children from the Nso culture in rural Nsoland in the North West region of Cameroon grow up in large, extended family settings. Livelihood depends on subsistence-based farming which plays a central role for family alimentation. Most parents are farmers and engage their children in household tasks and fieldwork from early on (
Given that attentional styles gradually develop during childhood (
Children and adolescents from both cultural contexts participated in an eye-tracking task assessing their spontaneous attention to four different sets of pictures varying in familiarity. The first two sets are similar to the stimuli used in the study by
Example stimuli used in the different sets:
Given the social orientation hypothesis – the assumption that in a cultural context, the social orientation is associated with the dominant cognitive style (
The final sample consisted of 53 participants of middle-class families from Münster in urban Germany and 50 participants of subsistence-based farming families of the Nso-culture from Kumbo in rural Cameroon. The age range was between 4 and 20 years. All participants had normal or corrected to normal visual acuity. In Münster, families were contacted via a database from the local university. In Kumbo, children were recruited by word of mouth. Informed written consent was obtained from parents in both contexts, and children gave informed assent. For their participation, families in Kumbo received financial compensation of 1000 CFA, which was equivalent to 1.50 € at that time. Participants in Münster received cinema vouchers. The type and amount of compensation have been discussed with local assistants to determine a locally appropriate compensation for the time spent.
An additional 12 participants in Kumbo and 15 participants in Münster were not included in the analyses because they did not meet the defined criteria. These were, first, that participants had no more than one degree of deviation when calibrating the eye-tracker and, second, that they were looking at the monitor for at least 70% of the presentation time (indicated by the tracking ratio). Three additional children were excluded because in one case the child did not feel well (Kumbo:
The gender distribution did not systematically differ between both samples (64.2% females in Münster, 56.0% females in Kumbo,
Participants attended one experimental session, and each participant saw all four sets of stimuli. In sum, each participant saw 120 pictures: 40 standard, 40 non-semantic, 20 culture-specific matched (either from Münster or Kumbo) and 20 simple pictures. In Kumbo, the laboratory was set up in a quiet room of a house, whereas in Münster, participants visited the university laboratory with a parent. The four stimuli sets differed in familiarity (as described below), and for each set the participant received the instruction to “…watch the pictures as you like…”
This set consisted of 40 real pictures displaying a focal object (animals and vehicles) in front of an urban or nature background (see
Further, 40 non-semantic pictures with abstract objects in front of abstract backgrounds were shown (see
These equivalent sets were designed to be similarly familiar in Münster and Kumbo. We compiled two sets of 20 real pictures with animals, vehicles, and buildings as focal objects. One set of pictures was taken in Münster and, therefore, should be highly familiar to children in Münster, and the other set was taken in and around Kumbo. The participants only saw the set with the stimuli from their respective cultural context. Pictures were matched over the two sets in the sense that they depicted the same kind of object in front of an equally complex background (e.g., a typical domestic animal on a typical field, see
Finally, we presented a simple set of 20 pictures with common, everyday objects in front of a simple background (e.g., a garlic clove on a wooden table, see
In order to back up the assumptions concerning the familiarity of the stimuli sets, the pictures of the standard, culture-specific matched and the simple set were rated by experts that were highly familiar with the respective environment. These experts, namely adolescents or adults living in urban Germany (
Assessments started with either the standard or the culture-specific matched set, with the respective other set being second. Then, participants saw either the non-semantic set or the simple set, with the respective other set being last. The order of the sets was counterbalanced, and pictures were randomized within each set and separated by a blank screen. Following studies that conducted a similar task with children (
Fixations were exported for further analyses in MATLAB (Version 2013a). The BeGaze software was used to define areas of interest (AOIs) around the focal object of each picture to distinguish the focal object from the background. To quantify participants’ visual attention to the object relative to their visual attention to the context, we computed an object score separately for each of the four sets. For each picture, the total duration of all fixations made into the AOI of the object was divided by the duration of all fixations on the picture (object and background area) within the 5 s of stimulus presentation. As a consequence, pictures that were not fixated at all, were not included in the mean score. Based on the object score for each picture, we then calculated the mean object score for each set separately. A score of 1 would mean that the participant only looked at the object, while a score of 0 would indicate that a participant only looked at the background. The lower the mean object score of a participant is, the higher is his or her context sensitivity.
For the culture-specific matched set, where different stimuli were used for the two samples, the average size of the AOIs was slightly larger in the German picture set (
To test for the cross-cultural and age differences in attentional style and whether these depend on the familiarity of the stimuli used, the object scores, defined as the relative gaze duration on the focal object, for each set were entered as the dependent variable in separate multiple regression analyses. As independent variables, we included culture (1 = Münster, 0 = Kumbo), age (z-standardized) and, to test whether the effect of age was moderated by culture, the interaction term culture × age.
For the standard set, the regression model was significant,
Object scores across culture and age in the (A) standard set, (B) non-semantic set, (C/D) culture-specific matched set, and (E) simple set.
For the non-semantic set, there was a significant regression model,
In the culture-specific matched set, we compared the gaze behavior between participants from Kumbo and Münster on their respective culture-specific set. We found a significant regression model,
For the simple set, there was no significant regression model,
In order to analyze whether the different sets of stimuli index participants’ attentional style more generally, we calculated the internal consistency between the object scores of the four sets. Cronbach’s α for age-corrected residuals was
The present study aimed at investigating how the familiarity of stimuli affects cross-cultural differences in context sensitivity expected along the social orientation hypothesis. For this purpose, participants from Münster in urban Germany and Kumbo in rural Cameroon saw four different sets of stimuli varying in familiarity while their spontaneous gaze behavior was recorded in an eye-tracking paradigm.
Overall, we found support for the differential familiarity hypothesis and partial support for the social orientation hypothesis: When the stimuli were highly familiar for both cultural contexts, we found a significantly higher object score for participants from Münster than for participants from Kumbo. These results are in line with the social orientation hypothesis that predicts higher context sensitivity in interdependent than in independent cultures (
This finding – which is unexpected based on the social orientation hypothesis alone – is similar to the findings from
However, this does not explain the findings for the simple set: although the stimuli are familiar in both cultural contexts – according to the expert judgments, somewhat more so for the Kumbo participants – context sensitivity did not differ between cultures. Two explanations are plausible: first, it might be that – due to the fact that the pictures were taken in Kumbo – the depicted backgrounds, but not the objects, were more typical for Kumbo participants, which led the Münster participants to explore the backgrounds more. As a consequence, the two effects (familiarity and social orientation) neutralized each other, leading to similar results on both cultures. Second, as the background is quite simple in this set (e.g., a tabletop) and does not provide much potential for exploration, the saliency of the object compared to the background might have been even higher. Possibly, culture-specific patterns of context sensitivity may only come into effect if stimuli, including the context, are sufficiently complex. Based on these data, the potential explanations cannot be further scrutinized, but both suggest that the cross-cultural differences in context sensitivity are highly susceptible to the stimulus material used.
Furthermore, we expected age-related changes in attentional styles, namely that context sensitivity would increase with age (
Finally, we explored whether perceptual styles are consistent across the different stimulus sets. The internal consistencies of the gaze behavior across sets show that all four sets capture similar aspects of participants’ attention. While there was no correlation between the context sensitivity measured with different tasks (i.e., eye-tracking, picture description, and optical illusion) in the study by
It should be noted that it would have been of additional interest to run a fully-balanced design with both groups of participants observing the stimuli of their own and the other culture (i.e., both culture-specific sets). By doing so, it would have been possible to compare the participants’ gaze behavior when looking at the respective unfamiliar set to further back up our results. However, here our main focus was to compare the culture-specific matched set, that is highly familiar in a given culture, to a standard set of stimuli used in other studies analyzing the development of cross-cultural differences in attentional styles, which allows important conclusions about the validity of earlier findings and the influence that stimulus familiarity might have on spontaneous visual attention. A further limitation concerns the unequal AOI sizes of the different sets: When inspecting the absolute object scores across sets, it is important to keep in mind that these cannot be compared directly, but the different AOI sizes between sets may have influenced the respective absolute object scores (e.g., object scores were smallest in the non-semantic set, which also had the smallest AOIs). Having said this, the rationale of this study remains unaffected since the question whether the familiarity of the stimuli affects gaze behavior is addressed by analyzing whether the cross-cultural differences vary by stimulus familiarity.
Taken together, these findings provide considerable potential for further research as they highlight the importance of the stimulus materials and tasks used to assess the concept of context sensitivity across different cultural contexts. For future research, it would be advisable to carefully develop appropriate stimuli reflecting the actual lifeworld of participants from different cultural contexts. In this respect, it should be taken into account that the overall scene of a stimulus, including object and background, is equally familiar across the cultures being compared. In most of the previous studies on analytic and holistic visual attention (e.g.,
Furthermore, in this study we only considered the effect of stimulus familiarity on spontaneous gaze behavior. It remains an open question how stimulus familiarity affects other tasks capturing attentional styles. For example,
This study has shown that stimulus familiarity is a central aspect to take into account when investigating cross-cultural differences in context sensitivity. Specifically, stimulus familiarity should be taken into account when analyzing other aspects that influence perceptual patterns in order to understand the mechanisms behind cross-cultural differences, such as cultural socialization practices. For example,
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethical approval was not provided for this study on human participants because all details of the ethical guidelines have been followed. This research was conducted in accordance with the Declaration of Helsinki and the Ethical Principles of the German Psychological Society (DGPs), the Association of German Professional Psychologists (BDP), and the American Psychological Association (APA). It involved no invasive or otherwise ethically problematic techniques and no deception and therefore, according to National jurisdiction, does not require a separate vote by a local Institutional Review Board; see the regulations on freedom of research in the German Constitution [§ 5 (3)], and the German University Law (§ 22). Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.
JK and MK designed the study and the stimuli. MK and RY recruited the sample and conducted the study. SJ analyzed the data. SJ, JK, and MK wrote the manuscript. JK supervised the research. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We would like to thank Ruth Baumann, Lena Knüwer, Marius Gruber and Tatah Gaston Fonyuy for their contributions to task development and data assessment. We thank all the children and adolescents for their participation. We acknowledge support from the Open Access Publication Fund of the University of Muenster.
1