Identify: sampling method, what measurement did the article used?, Are there any validity and reliability measures present in the article?

PSYCHOLOGICAL SCIENCE Research Article VOL. 14, NO. 3, MAY 2003 Copyright © 2003 American Psychological Society 201 PERCEIVING AN OBJECT AND ITS CONTEXT IN DIFFERENT CULTURES: A Cultural Look at New Look Shinobu Kitayama, 1 Sean Duffy, 2 Tadashi Kawamura, 1 and Jeff T. Larsen 3 1 Kyoto University, Kyoto, Japan; 2 University of Chicago; and 3 Texas Tech University Abstract— In two studies, a newly devised test (framed-line test) was used to examine the hypothesis that individuals engaging in Asian cul- tures are more capable of incorporating contextual information and those engaging in North American cultures are more capable of ignor- ing contextual information. On each trial, participants were presented with a square frame, within which was printed a vertical line. Partici- pants were then shown another square frame of the same or different size and asked to draw a line that was identical to the ﬁrst line in ei- ther absolute length ( absolute task) or proportion to the height of the surrounding frame ( relative task). The results supported the hypothe- sis: Whereas Japanese were more accurate in the relative task, Ameri- cans were more accurate in the absolute task. Moreover, when engaging in another culture, individuals tended to show the cognitive characteristic common in the host culture. Although perception depends on sensory input, it also involves a variety of top-down processes that are automatically recruited to ac- tively construct a conscious percept from the input. According to this thesis, called New Look in the 1950s, the percept is signiﬁcantly mod- iﬁed by expectations, values, emotions, needs, and other factors that are endogenous to the perceiver (Bruner, 1957; Bruner & Goodman, 1947). Exogenous factors, such as physical properties of the imping- ing stimulus, cannot account, in full, for the emerging percept. Al- though initial demonstrations evoked considerable controversy and skepticism (e.g., Postman, Bronson, & Gropper, 1953), the basic idea has proved quite viable (Erdelyi, 1974; Niedenthal & Kitayama, 1994; Zajonc, 1980), and has since taken a strong hold in the mainstream of cognitive and social psychology (Higgins & Bargh, 1987). In large part, however, this literature has so far ignored culture. This omission is both surprising and unfortunate. As a pool of ideational re- sources (e.g., lay theories, images, scripts, and worldviews) that are em- bodied in public narratives, practices, and institutions of given geographic regions, historical periods, and groups, whether ethnic, religious, or oth\ er- wise (Kitayama, 2002), culture may be expected to be one source, and perhaps the most fundamental source, of each person’s values, expecta- tions, and needs. The purpose of the current work, then, was to take a re- newed look at the New Look from a cultural point of view. CULTURE AND COGNITION A number of recent studies focusing on cultural variation in cogni- tive processes have already shed some light on this issue. As a whole, these studies suggest that different cultures foster quite different modes of cognitive processing (Kitayama, 2000; Markus & Kitayama, 1991; Nisbett, Peng, Choi, & Norenzayan, 2001). In particular, people engag- ing in North American cultures (North Americans, in short) are assumed to be relatively more attuned to a focal object and less sensitive to con- text. North Americans are thus described as analytic (Nisbett et al., 2001) or ﬁeld independent (Witkin & Berry, 1975) in cognitive style.

Conversely, persons engaging in Asian cultures (Asians, in short) are hypothesized to be attuned more to contextual information—namely, in- formation that surrounds the focal object. Asians are thus described as holistic or ﬁeld dependent in cognitive style. These culturally divergent cognitive characteristics have been examined with several different measures, such as attitude attribution (e.g., Masuda & Kitayama, 2002; Miyamoto & Kitayama, 2002), performance in a rod-and-frame task (RFT; Ji, Peng, & Nisbett, 2000; Witkin & Berry, 1975), a Stroop inter- ference effect (Ishii, Reyes, & Kitayama, 2003; Kitayama & Ishii, 2002), and context-dependent memory (Masuda & Nisbett, 2001). A reasonable conjecture from this emerging literature is that the cross-cul- turally divergent modes of cognitive processing must be differentially advantageous, depending on the demands of a particular task.

Speciﬁcally, some tasks require ignoring contextual information when making a judgment about a focal object. For example, a judgment about another person may often be tainted by wrong stereotypes associ- ated with a group of which that person is a member. In these circum- stances, it is necessary to discount any such stereotypes. Such tasks may be called absolute tasks in that the focal judgment must be uninﬂuenced or unchanged by any contextual information. In these tasks, perfor- mance should be better for North Americans than for Asians. Using the RFT (Witkin & Berry, 1975), Ji et al. (2000) recently provided evidence supporting this prediction. Participants viewed a tilted frame with a ro- tating line placed at the center. The participants’ task was to rotate the line so that it was orthogonal to the earth’s surface (or it was aligned to the direction of gravity) while ignoring the frame. Ji et al. found that Americans were more accurate than Chinese in aligning the line (hence indicating their superior ability to ignore contextual informa- tion). This evidence is noteworthy because the RFT has no obvious so- cial elements.

In contrast, other tasks require incorporating contextual informa- tion. For example, a judgment about another person often beneﬁts from attention duly given to the speciﬁc social situation in which that person behaves. These tasks may be called relative tasks in that the fo- cal judgment must change in accordance with the nature of the rele- vant context. One might expect that Asians with contextual sensitivity would have an advantage over North Americans in performing such tasks. The evidence supporting this prediction comes exclusively from social domains. Thus, it is well known that North Americans often fail to give proper weight to signiﬁcant contextual information in drawing a judgment about a focal person. This bias, called the fundamental at- tribution error , is typically substantially weaker in Asian cultures (e.g., Miyamoto & Kitayama, 2002; Morris & Peng, 1994). Address correspondence to Shinobu Kitayama, Faculty of Integrated Human Studies, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan; e-mail:

[email protected]. at UNIV TEXAS AT TYLER on August 22, 2016 pss.sagepub.com Downloaded from PSYCHOLOGICAL SCIENCE Cultural Look 202 VOL. 14, NO. 3, MAY 2003 THE PRESENT RESEARCH The available evidence indicates that there is substantial cognitive difference across cultures (Nisbett et al., 2001). Furthermore, this dif-\ ference can be demonstrated with both tasks that are obviously social (e.g., Miyamoto & Kitayama, 2002) and those that are minimally so- cial (e.g., Ji et al., 2000). Nevertheless, there are some signiﬁcant lim- itations that have hampered the further development of theory on cultural variation in cognitive competences. First, with the important exception of the work by Ji et al. (2000), virtually no studies have evaluated performance against any objective criterion. This makes it difﬁcult to draw any conclusions on the nor- mative status of the cognitive biases that are suggested in the litera- ture. Second, in all the existent studies examining relative tasks (e.g., attitude attribution), participants were not instructed to use contextual information, and therefore it is uncertain whether the cross-cultural differences were due to Asians’ greater propensity to attend to the con- text, their greater competence to incorporate information in the con- text, or both. Third, though the RFT ﬁndings of Ji et al. suggest that North Americans’ greater ability to ignore context extends from social to nonsocial tasks, it is not clear whether Asians’ greater ability to in- corporate context extends to nonsocial tasks. Fourth, and perhaps most important, in all the existent studies, very different domains, such as line alignment and social perception, were used in deﬁning the two theoretical types of tasks. This makes it impossible to draw any mean- ingful comparison between performance in an absolute task and per- formance in a relative task. In an effort to address these limitations inherent in the current evi- dence, we developed a new test called the framed-line test (FLT). The FLT is speciﬁcally designed to assess both the ability to incorporate a\ nd the ability to ignore contextual information within a single domain that is arguably nonsocial. Further, this assessment can be made in reference to an objective standard of performance. Speciﬁcally, on each trial, par- ticipants are presented with a square frame, within which is printed a vertical line. The participants are then shown another square frame of the same or different size and asked to draw a line that is identical to the ﬁrst line in either absolute length (absolute task) or proportion t\ o the height of the surrounding frame (relative task). In the absolute task, the participants have to ignore both the ﬁrst frame (when assessing the length of the line) and the second frame (when reproducing the line). Hence, North Americans should perform this task better than Asians. In the relative task, the participants have to incorporate the height information of the surrounding frame in both encoding and reproducing the line. Hence, Asians should perform this task better than North Americans. Moreover, one major advantage of the FLT is that it allows an assessment of the relative ease or difﬁculty of the two tasks. We predicted that whereas Asians would be more ac- curate on the relative task than the absolute task, the reverse would be the case for North Americans.

STUDY 1: THE FLT IN JAPAN AND THE UNITED STATES Method Participants Twenty undergraduates at Kyoto University, in Kyoto, Japan (8 males and 12 females), and 20 undergraduates at the University of Chicago (9 males and 11 females) volunteered to participate in the study. All the Japanese undergraduates were native Japanese, and all the Ameri- can undergraduates were of European descent.

Materials and procedure Upon arrival in a lab, participants were told that they would per- form simple cognitive tasks. They were given both the absolute task and the relative task in a counterbalanced order, receiving speciﬁc in- structions for each task right before they performed it. In both tasks, on each trial they were shown a square frame, within which a vertical line was printed. The line was extended downward from the center of the upper edge of the square (see Fig. 1). The participants were then moved to a different table placed in the opposite corner of the lab (so as to ensure that iconic memory played no role) and shown a second square frame that was either larger than, smaller than, or the same size as the ﬁrst frame. The task was to draw a line in the second frame. In the absolute task, the participants were instructed to draw a line that was the same absolute length as the line in the ﬁrst frame. In the rel\ a- tive task, the participants were instructed to draw a line whose propor- tion to the size of the second frame was the same as the proportion of the ﬁrst line to the size of the ﬁrst frame. We took care to ensure that the participants understood the tasks by using concrete examples, such as the ones given in Figure 1. The stimuli were prepared such that there were ﬁve different com- binations of the relative sizes of the two frames and the line in the ﬁrst Fig. 1. Example of the framed-line test (FLT) used in these studies.

Participants were shown a square frame with a vertical line, and asked to draw a line in a new square of the same or different size. The line was to be drawn so that it was identical to the ﬁrst line in absolute length (absolute task) or so that the proportion between the length of\ the line and the height of its frame was identical to the proportion be- tween the line and frame in the original stimulus (relative task). at UNIV TEXAS AT TYLER on August 22, 2016 pss.sagepub.com Downloaded from PSYCHOLOGICAL SCIENCE S. Kitayama et al.

VOL. 14, NO. 3, MAY 2003 203 frame (see Table 1). In two combinations the ﬁrst frame was smaller than the second, and in two other combinations the ﬁrst frame was larger than the second. Furthermore, in half of these cases, the ﬁrst line was longer than one half the height of the ﬁrst square, and in the remaining half, the ﬁrst line was shorter than half the height of the ﬁrst square. Finally, in the ﬁfth combination, the ﬁrst and the second frames were identical in size. This last case is of interest because the correct response would be identical in the relative and the absolute tasks. The ﬁve combinations were presented in a random order. The same stimuli were used in the relative and absolute tasks.

Results and Discussion An inspection of the data showed that overestimation and underes- timation occurred to a nearly equal extent for each of the ﬁve stimulus combinations. Accordingly, in order to assess performance in the two tasks, we measured the lines drawn by the participants and calculated the absolute difference between these lengths and the correct lengths.

Because the absolute size of error increased somewhat as the lines be- came longer, we also analyzed the percentage of the error relative to the correct line length. The results were no different whether we ana- lyzed absolute error or percentage of error. We therefore discuss only the results for absolute error. The mean error scores for the two tasks are summarized in Table 1.

These means were submitted to an analysis of variance (ANOVA) with one between-subjects variable (culture of participant: Japanese vs. Amer- ican) and two within-subjects variables (task: absolute vs. relative; stimu- lus version: the ﬁve combinations of frames and line). A preliminary analysis had shown that the effects did not depend on either gender of the participants or the order in which the two tasks were given. As predicted, the interaction between culture and task proved signiﬁ- cant, F(1, 38) 24.41, p .0001. The pertinent means are plotted in Fig- ure 2. As predicted, Japanese performed the relative task signiﬁcantly more accurately than the absolute task ( Ms 6.05 vs. 4.52), t(38) 2.56, p .02. In contrast, Americans performed the absolute task more accu- rately than the relative task ( Ms 6.35 vs. 3.71), t(38) 4.57, p .01.

Moreover, Americans performed the absolute task signiﬁcantly better than Japanese, t(38) 3.92, p .01, and the reverse was true for performance in the relative task, t(38) 3.06, p .01. The relative ease of the two tasks varied across the ﬁve stimulus combinations, as indicated by a sig- niﬁcant main effect of version and a signiﬁcant interaction between ver- sion and task, F(4, 152) 8.95, p .001, and F(4, 152) 9.44, p .001. However, the three-way interaction involving culture, task, and ver- sion was negligible ( F 1). Thus, the pattern in Figure 2 emerged over all the ﬁve combinations. The same pattern emerged even when the two frames were identical in size, and the correct length was identical in the two tasks. Speciﬁcally, for this stimulus combination, Japanese performed the relative task signiﬁcantly better than the absolute task, t(38) 2.53, p .02, although the difference in performance was considerably attenuated for Americans, t 1. We return to this issue later. STUDY 2: VARIABILITY AND MALLEABILITY OF FLT PERFORMANCE In Study 2, we sought to replicate Study 1 and to extend it by testing both Americans and Japanese in both Japan and the United States. This effort was motivated by a concern with the variability and malleability of cross-cultural variations. If, for example, the cross-cultural difference we observed is both relatively uniform within each culture and relatively stable and traitlike over time, then the cross-cultural variation should be entirely a function of the cultural origins of each participant: Americans (or Japanese) should show a prototypically American (or Japanese) pat- tern more or less uniformly regardless of where they are tested. If, how- ever, the cognitive abilities at issue are both variable and malleable, there ought to be considerable variation as a function of both the cultural origins of participants and the speciﬁc location in which they are tested.

Speciﬁcally, participants in a foreign culture should show a pattern of cognitive biases that resembles the pattern typical in the host culture.

Table 1. Mean error scores (in millimeters) in the absolute and relative line-drawing tasks in Study 1 Height of ﬁrst frame Length of line Height of second frame Absolute task Relative task Japanese Americans Japanese Americans 89 62 179 6.8 (5.0) 2.2 (2.1) 6.3 (4.8) 10.6 (4.1) 102 29 153 3.6 (4.9) 2.5 (2.5) 5.7 (5.9) 6.2 (4.0) 127 53 127 7.1 (4.7) 3.8 (4.0) 4.0 (3.3) 4.5 (3.5) 153 87 102 9.0 (4.9) 6.1 (4.6) 4.2 (3.4) 6.6 (5.0) 179 31 89 3.6 (2.9) 3.8 (3.6) 2.3 (2.7) 3.7 (2.5) Note. Standard deviations are in parentheses. Fig. 2. Results from Study 1: Japanese and American participants’ mean error (in millimeters) in the two line-drawing tasks of th e framed-line test. The error bars represent standard errors. at UNIV TEXAS AT TYLER on August 22, 2016 pss.sagepub.com Downloaded from PSYCHOLOGICAL SCIENCE Cultural Look 204 VOL. 14, NO. 3, MAY 2003 Method Participants Four groups of individuals (total N 111) volunteered for the study.

The Japanese in Japan were 32 undergraduates (20 males and 12 females) at Kyoto University. They were tested by a Japanese experimenter, and in- structions were given in Japanese. The Americans in Japan were 18 ex- change students (8 males and 10 females) who were temporarily staying at the Kansai Institute for Foreign Languages. They had stayed in Japan for 4 months at most, and their Japanese proﬁciency was quite limited.

They were tested by a Japanese experimenter, and their instructions were given in English. The Americans in the United States were 40 undergrad- uates (21 males and 19 females) at the University of Chicago, and the Jap- anese in the United States were 21 Japanese undergraduates (13 males and 8 females) who were temporarily studying at the University of Chi- cago. These Japanese had stayed at the university for a varying length of time, from 2 months up to 4 years. The Americans and the Japanese in the United States were both tested by an American experimenter, and instruc- tions were given in English. Materials and procedure Six different combinations of the relative sizes of the two frames and the line in the ﬁrst frame were prepared (see Table 2). They were similar to the combinations used in Study 1, except that some of the ratios of the size of the two frames were somewhat changed and the pattern with the two frames of the identical size was run in two varia- tions. The participants were tested individually using the same proce- dure as in Study 1.

Results and Discussion As in Study 1, there was no systematic tendency for over- or under- estimation in any of the stimulus combinations. Also as in Study 1, the results were no different whether we analyzed absolute error or per- centage of error, so we discuss only the results for absolute error. Pre- liminary analysis showed no signiﬁcant effects involving the gender of participants. Although past research tended to show females to be more context sensitive than males (Cross & Madson, 1997), this effect appears to be less robust than the cultural difference. Also as in Study 1, we did not ﬁnd any reliable effect of the order in which the two tasks were given. The means were submitted to a 2 2 2 6 ANOVA, with two between-subjects variables (cultural origin of participant: Japanese vs.

American; testing location: Japan vs. United States) and two within- subjects variables (task: absolute vs. relative; stimulus version: the six combinations of frames and line). The mean error scores are summa- rized in Table 2. The size of error varied systematically across the four groups of participants, as indicated by a signiﬁcant main effect for testing location and a signiﬁcant interaction between testing locatio\ n and cultural origin, F(1, 105) 15.97, p .0001, and F(1, 105) 8.25, p .01. Further, the size of error was larger for the absolute task than for the relative task, F(1, 105) 39.56, p .0001. However, as in Study 1, the critical interaction between task and cultural origin proved to be highly signiﬁcant, F(1, 105) 10.18, p .002. More- over, we also found a highly reliable interaction between task and test- ing location, F(1, 105) 56.19, p .0001.

The pertinent means are summarized in Figure 3. As in Study 1, Japa- nese participants in Japan proved to be much more accurate in the relative task than in the absolute task, t(105) 9.90, p .01. In contrast, Ameri- cans in the United States were signiﬁcantly more accurate in the abso\ lute task than in the relative task, t(105) 2.15, p .05. The remaining two groups of participants showed an effect that strongly resembled the effect of the host culture. Thus, the pattern for Americans in Japan was closer to the pattern for Japanese in Japan than to the pattern for Americans in the United States. Likewise, the pattern for Japanese in the United States was closer to the pattern for Americans in the United States than to the pattern for Japanese in Japan. Examined from a different angle, performance in the relative task was signiﬁcantly better for Japanese in Japan than for Americans in the United States, t(105) 7.84, p .001, but performance in the absolute task was better for Americans in the United States than for Japanese in Japan, t(105) 4.73, p .001. In both cases, the data in the remaining two groups fell in between. It is noteworthy that essentially the same pattern was observed across the six combinations of stimuli (see Ta- ble 2). In particular, as in Study 1, the same cross-cultural difference in er- ror pattern was observed even when the two frames were equal in size. GENERAL DISCUSSION Sociocultural Shaping of Attention and Perception The current work is consistent with recent theorizing on cultural varia- tion in cognition (Nisbett et al., 2001) in showing that Japanese are more capable of incorporating contextual information in making a judgment on Table 2. Mean error scores (in millimeters) in the absolute and relative line-drawing tasks in Study 2 Height of ﬁrst frame Length of line Height of second frame Absolute task Relative task Japanese in Japan Americans in Japan Japanese in the U.S. Americans in the U.S. Japanese in Japan Americans in Japan Japanese in the U.S. Americans in the U.S.

81 68 162 9.2 (5.7) 6.6 (5.7) 4.1 (3.0) 3.4 (2.9) 4.3 (3.5) 3.6 (2.6) 5.7 (3.2) 7.7 (7.4) 108 22 162 5.1 (5.1) 3.2 (2.7) 3.2 (2.3) 4.1 (2.6) 3.1 (2.0) 5.8 (4.4) 3.0 (2.4) 3.9 (3.4) 101 28 101 8.1 (6.7) 5.4 (7.0) 4.0 (2.8) 4.2 (3.0) 3.1 (2.0) 2.7 (1.8) 2.9 (2.6) 4.4 (2.9) 141 102 141 14.5 (10.1) 14.0 (9.1) 3.3 (2.9) 6.5 (4.7) 3.9 (2.8) 5.1 (4.1) 4.7 (3.3) 8.4 (5.6) 108 73 81 8.0 (4.4) 7.8 (3.5) 4.8 (3.3) 5.3 (3.8) 3.0 (2.2) 4.2 (4.8) 2.8 (2.4) 5.6 (4.0) 162 30 81 6.7 (7.0) 4.6 (3.6) 3.9 (2.4) 4.2 (3.2) 2.0 (1.9) 2.8 (1.7) 2.7 (2.0) 3.9 (2.2) Note. Standard deviations are in parentheses. at UNIV TEXAS AT TYLER on August 22, 2016 pss.sagepub.com Downloaded from PSYCHOLOGICAL SCIENCE S. Kitayama et al.

VOL. 14, NO. 3, MAY 2003 205 a focal object, but North Americans are more capable of ignoring contex- tual information. We demonstrated this cross-cultural variation in a nonso- cial task that is speciﬁcally designed to simultaneously assess the t\ wo cognitive competences of ignoring or incorporating context. The FLT may be an important source of information about a nonsocial, basic cognitive capacity that is recruited in making more complex, contextualized judg- ments and perceptions. Future research should clarify both the social or\ i- gins of the nonsocial cognitive and attentional skills and processes involved in the FLT and the contribution of these processes to judgments and inferences about social objects and events. We repeatedly found the same cross-cultural variation in error pat- tern even when the two frames were identical. This ﬁnding contradicts the notion that Asians (or Americans) are predisposed to perform the relative (or the absolute) task even when instructed to do otherwise. If our participants had such predispositions, errors should have been minimal when the two frames were identical because, under these conditions, the correct answers in the two tasks converge. In view of the current evidence, we suggest that errors were due, in part, to difﬁ- culty in accurately encoding the central line. That is, whereas Japa- nese may have had difﬁculty releasing attention from the frame and then refocusing it on the line in the absolute task, Americans may have had difﬁculty releasing attention from the central line and shifting it to the frame in the relative task. 1 Future work should examine the role of attentional and other mechanisms in greater detail.

Limitations The effects of cultural origins of participants and test locations in Study 2 are quite suggestive, but should be interpreted with caution. One provocative interpretation is that the effect of test location was caused by immersion in a new culture. That is, cognitive and attentional capacities and tendencies may be modiﬁed in accordance with new demands and af- fordances of living in a new host culture (Kitayama, Markus, Matsumoto, & Norasakkunkit 1997). Such a modiﬁcation might be permanent once it\ takes place or relatively temporary and reversible; these different possibil- ities have different implications for what might be expected to happen when people return to their home country. However, other interpretations of our data are also possible. Be- cause of a variety of practical constraints associated with cross-cul- tural experimental research, we could not exercise the degree of experimental control that we would have preferred. In particular, we depended entirely on convenience samples. We wish to acknowledge three resulting difﬁculties in interpretation. First, the average length of stay in the host culture turned out to be very different between the Americans in Japan and Japanese in the United States. Although this variable did not correlate with the size of error in the two tasks, a more balanced sampling would have been desirable.

Second, the language used for instructions was also chosen for convenience. In particular, in the United States not only Americans but also Japanese were tested in English, but in Japan participants were tested in their native languages. This might have had an unknown de- gree of inﬂuence on the results because evidence indicates that lan- guage can prime the associated culture (Hoffman, Lau, & Johnson, 1986; but see Ishii et al., 2003, for an important caveat). Fig. 3. Results from Study 2: Mean error scores (in millimeters) in the two line-drawing tasks of the framed-line test. Results are shown separately for Japanese and Americans in the two cultural locations, Japan and the United States. The error bars represent standard errors. 1. It is of note that the errors in the two tasks of the FLT were largely inde- pendent within each experimental group in both studies ( .08 r .44, median .06), indicating that the two attentional abilities are mostly separate. at UNIV TEXAS AT TYLER on August 22, 2016 pss.sagepub.com Downloaded from PSYCHOLOGICAL SCIENCE Cultural Look206 VOL. 14, NO. 3, MAY 2003 Third, and most intriguing, both the Japanese participants in the United States and the American participants in Japan were people who voluntarily moved to the other culture. Though the obtained effects on cognitive biases may have been due to immersion in a new culture, the possibility of selection bias is equally viable. That is, only those peo- ple who have psychological afﬁnities to another culture may ﬁnd themselves living in this other culture. Although these explanations for the effect of test location are not mutually exclusive, future work should empirically address their relative signiﬁcance.

Conclusion Culture is a source of generic expectations; default goals, desires, and needs; and overarching values. Cultural variations in attention, per- ception, and cognition, then, would enable researchers to take a renewed look at the New Look (Bruner, 1994). The current work suggests that different cultures’ practices and beliefs encourage very divergent cogni- tive and attentional capacities involved in either incorporating or ignor- ing context while making a judgment about a focal object. Future work along the lines proposed here may reveal a degree of so- ciocultural shaping of attention and perception that is substantially greater than has so far been assumed in the psychological literature. If so, this evidence would provide a solid basis for reconceptualizing human psychological processes and structures as fully embedded in and thus signiﬁcantly constituted by the collectively shared practices, values, and beliefs of culture (Kitayama, 2002). Indeed, if properly analyzed, the\ the- sis of the New Look will be instrumental in breaking a self-imposed shell of the traditional psychological discipline and broadening its horizon t\ o include society, culture, and history in its territory of investigation. REFERENCES Bruner, J. (1957). On perceptual readiness. Psychological Review, 64 , 123–152.

Bruner, J. (1994). The view from the heart’s eye: A commentary. In P.M. Niedenthal & S.

Acknowledgments— This research was supported by a grant-in-aid from the Japanese Ministry of Science and Education (B-14310056). We thank members of the cultural psychology labs at both Kyoto University and the University of Chicago for comments on an earlier draft. Kitayama (Eds.), The heart’s eye: Emotional inﬂuences in perception and attention (pp. 269–286). New York: Academic Press.

Bruner, J., & Goodman, C.C. (1947). Value and need as organizing factors in perception. Journal of Abnormal and Social Psychology , 42, 33–44.

Cross, S.E., & Madson, L. (1997). Models of the self: Self-construals \ and gender. Psycho- logical Bulletin , 122 , 5–37.

Erdelyi, M.H. (1974). A new look at the new look: Perceptual defense and vigilance. Psy- chological Review , 81 , 1–25.

Higgins, E.T., & Bargh, J.A. (1987). Social cognition and social perception. Annual Re- view of Psychology , 38 , 369–425.

Hoffman, C., Lau, I., & Johnson, D.R. (1986). The linguistic relativity of person cogni- tion: An English-Chinese comparison. Journal of Personality and Social Psychol- ogy , 51 , 1097–1105.

Ishii, K., Reyes, J.A., & Kitayama, S. (2003). Spontaneous attention to word content ver- sus emotional tone: Differences among three cultures. Psychological Science, 14 , 39–46.

Ji, L.J., Peng, K., & Nisbett, R.E. (2000). Culture, control, and perc\ eption of relationship in the environment. Journal of Personality and Social Psychology , 78 , 943–955.

Kitayama, S. (2000). Cultural variations in cognition: Implications for aging research. In P.C. Stern & L.L. Carstensen (Eds.), The aging mind: Opportunities in cognitive re- search (pp. 218–237). Washington, DC: National Academy Press.

Kitayama, S. (2002). Culture and basic psychological processes: Toward a system view of culture. Psychological Bulletin , 128 , 189–196.

Kitayama, S., & Ishii, K. (2002). Word and voice: Spontaneous attention to emotional speech in two cultures. Cognition and Emotion, 16 , 29–59.

Kitayama, S., Markus, H.R., Matsumoto, H., & Norasakkunkit, V. (1997). Individual and collective processes in the construction of the self: Self-enhancement in the Uni\ ted States and self-criticism in Japan. Journal of Personality and Social Psychology , 72 , 1245–1267.

Markus, H., & Kitayama, S. (1991). Culture and the self: Implications \ for cognition, emo- tion, and motivation. Psychological Review , 98 , 224–253.

Masuda, T., & Kitayama, S. (2002). Perceiver-imposed constraint and attitude attribution in Japan and the US: A case for culture-dependence of correspondence bias. Un- published manuscript, University of Michigan, Ann Arbor.

Masuda, T., & Nisbett, R.E. (2001). Attending holistically vs. analytically: Comparing the context sensitivity of Japanese and Americans. Journal of Personality and Social Psychology , 81 , 922–934.

Miyamoto, Y., & Kitayama, S. (2002). Cultural variation in correspondence bias: The crit- ical role of attitude diagnosticity of socially constrained behavior. Journal of Per- sonality and Social Psychology , 83 , 1239–1248.

Morris, M.W., & Peng, K. (1994). Culture and cause: American and Chinese attributions for social and physical events. Journal of Personality and Social Psychology , 67 , 949–971.

Niedenthal, P .M., & Kitayama, S. (Eds.). (1994). The heart’s eye: Emotional inﬂuences in perception and attention. New York: Academic Press.

Nisbett, R.E., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of thought: Holistic vs. analytic cognition. Psychological Review, 108 , 291–310.

Postman, L., Bronson, W.C., & Gropper, G.L. (1953). Is there a mechanism of perceptual defense? Journal of Abnormal and Social Psychology , 48 , 215–224.

Witkin, H.A., & Berry, J.W. (1975). Psychological differentiation in cross-cultural per- spective. Journal of Cross-Cultural Psychology , 6, 4–87.

Zajonc, R.B. (1980). Feeling and thinking: Preferences need no inferen\ ce. American Psy- chologist , 35 , 151–175. (RECEIVED 2/13/02; R EVISION ACCEPTED 6/26/02) at UNIV TEXAS AT TYLER on August 22, 2016 pss.sagepub.com Downloaded from