JFK, the Vietnam War, and U.S. Political Culture
This dimension reflects the consensus both within a culture and within each member of this culture of the association between an odor and a label. The notion of codability thus seems to be applicable to odors even though, unlike colors, there is no specific vocabulary that has been extracted to describe odors. At a quantitative level, we found that codability was globally lower in Vietnam than in the two other cultures. Two non exclusive explanations may be put forward for this effect. These accessory terms were frequently added to the odor labels in our study, but not in a consistent way within and between participants.
The inconsistent use of accessory terms might have contributed to the general low codability scores for the Vietnamese group. The second explanation is linked to the familiarity of the odors.
Although we made sure to include odors encountered in Vietnam in the set of odorants, our odorants were manufactured in France and were probably closer to familiar odors present in western countries. During the identification task, Vietnamese participants often used general terms e. This result is consistent with previous work with these three cultures Ly Mai, and also with cross-cultural studies which found a large effect of culture on odor label quality Aubaile Sallenave, ; Ayabe-Kanamura et al. These cross-cultural differences may again reflect cross-cultural differences in the frequency of exposure to specific odors.
In agreement with this interpretation, some data collected from a questionnaire in the French, American, and Vietnamese populations Chrea, showed a link between the frequency of encounter of odorant objects and performance on an identification task. Odorant objects that are often encountered in a given culture are more likely to be consensually used to identify an odor. Moreover, codability of the 40 stimuli was clearly prone to cross-cultural variations. These variations in turn make it possible to investigate in Experiment 2 the link between codability and odor recognition memory, and its stability across the three cultures.
All participants were born and raised in the culture where the experiment took place. They had not taken part in Experiment 1 and did not have previous experience with odor experiments. None of them reported to have any problem with their sense of smell. As our aim was to determine whether codability could predict recognition of odors, the main criterion for the constitution of the target group was the codability score of the odors.
Revisiting Vietnam by Julia Bleakney (ebook)
We thus selected 20 odors which included 1 odors with a high codability score in all three cultures, 2 odors with a low codability score in all three cultures, and 3 odors with a high codability score in only one or two cultures. A secondary criterion of interest was to balance the groups of targets and distractors in terms of pleasantness and perceptual similarity. Both variables were assessed in a previous study on similar French, American, and Vietnamese populations where participants rated pleasantness on a 7-point rating scale, and similarity was determined by means of a free sorting task Chrea et al.
In the first session, participants were asked to smell each of the 20 target odorants and try to memorize them in order to recognize them during the second session of the task. This set was composed of the 20 target odorants and the 20 distracting odorants.
We first examined the potential predictive role of codability on recognition memory within each cultural group and then cross-culturally. The independent variable was the codability of the target odors, determined as the score of the odors on the first factor of the PCA carried out in Experiment 1. The dependent variable was the hit rate. Regressions between codability and hit rate see Fig. A Duncan test significance level: 0.
Risk preferences and development revisited: A field experiment in Vietnam
Results are presented for the 20 target odors in France A and the United States B , for 19 target odors without cinnamon , dot in brackets in Vietnam C. The results showed that American and Vietnamese codability respectively did not predict hit rates in the other cultures. This result supports the existence of a relation between linguistic information and non linguistic behavior at least in the French group. In order to compare our results with results for colors, Spearman rank-order correlation were computed between codability scores and hit rates.
Thus, in contrast to the previous study by Lawless and Cain , codability seems to be an influential factor in the memorization process of chemosensory stimuli to the same extent as it is for other sensory modalities. The remaining question now is: Why did codability predict recognition memory in France but not in the United States or in Vietnam? Our results indicate that, although discrepancies in recognition performance emerged between the three cultures, the level of recognition did not seem to interact in the relationship between codability and recognition memory.
Indeed the link between codability and recognition memory was observed only in the French group which elicited recognition performance intermediary between the American and the Vietnamese groups. If the link between codability and recognition memory implied superiority in recognition abilities, we should have found higher recognition performance in the French group. A more plausible explanation is linked to the distribution of the codability scores in the three cultural groups. Indeed, codability scores were more distributed in France than in the two other cultural groups, and because a larger spread allowed for a better prediction, this may explain that the link between codability and recognition memory was observed only in the French group.
Our first objective was to evaluate whether codability was a meaningful dimension to understand the cognitive processes involved in odor memory representations. For that purpose we evaluated the relevance and the consistency of several indices that were used to measure codability of visual stimuli. In Experiment 1, we showed that codability measured by both inter- and intrapersonal agreement indices has some relevance for odors.
Moreover, we found agreement in the different indices we measured and this agreement reflects the validity of the codability construct. Finally, as expected, we found that the codability of an odor was partially dependent on the culture.
Our second objective in Experiment 2 was to evaluate the extent to which codability could predict a non-linguistic behavior such as odor recognition memory. We found that the French codability scores predicted recognition memory performance in all three cultures. By contrast, the American and Vietnamese codability scores predicted none of the recognition performances.
The remaining question is to know what could be the mechanisms underlying the odor codability effect? These two opposed views are still debated, as demonstrated by the impressive number of papers investigating color naming and categorical perception in different cultures see e.
Indeed, we found that French codability not only predicted the French odor recognition memory but that it also predicted recognition memory for the other two cultural groups. A plausible interpretation is that the codability effect we observed is not due to the properties of language, but to perceptual properties of the odors themselves. In fact, as was suggested by Heider to account for universals in color naming and color categorization, it is possible that odor codability is determined by the perceptual salience of odors.
In This Section
To explain the cultural differences we found, we can imagine that odors which were consensually labeled in France were also perceptually salient in the three cultural groups. And because the odors were manufactured in France and thus were probably more familiar to French participants, these latter participants were better able to pin down their perceptual salience in words, regardless the correctness of the names, than participants from the two other groups.
This interpretation suggests that some universal invariants might exist in odor representation. In agreement with this idea, Chrea et al. The next question would be then: Why are some odors perceptually more salient than others? A possibility is that odor salience has a physiological origin as it was suggested for color salience Heider, Recent work on odor-structure relationships suggests that some receptors may be more selective than others and may respond to a specific class of odorant substances e.
It is tempting to speculate, as suggested by Holley , that the odors to which these high selectivity receptors are preferentially tuned, might be more salient and thus would be both codable and well-recognized. According to the universalist view, color salience influences the formation and the coherence of linguistic categories across cultures. This is unlikely to appear for odors because odor codability refers to the consensus in a concrete name of a unique odorant object and not to the consensus in an abstract name of a linguistic category as for colors.
A factor that may be of higher relevance for odors is the functional context in which odors are encountered. For example, the same odor of wintergreen can be associated to a candy by Americans or to a traditional medicine by Vietnamese Chrea et al. Therefore, if we give sense to these odor-specific factors, what would drive the organization of odors in memory is not the name itself, as for colors, but the function linked to the odorant source.
Codability would thus measure in this context the consensus in the function associated to the odor rather than in the name itself. Aubaile Sallenave, F. Ayabe-Kanamura, S. Differences in perception of everyday odors: a Japanese-German cross-cultural study. Chemical senses , 23 , Berlin, B. Basic Color Terms: their Universality and Evolution. Berkeley: University of California Press. Brown, R. A study in language and cognition.
Journal of Abnormal and Social Psychology, 49 , Chrea, C. Culture and odor categorization: agreement between cultures depends upon the odors. Food Quality and Preference, 15 , Semantic, typicality and odor representation: A cross-cultural study.
Chemical Senses, 30 , Cohen, J. PsyScope: a new graphic interactive environment for designing psychology experiments. Good judgment of meaning and significance will involve interpretative and historical methods, as well as an understanding of both material and ideological cause and effect. Despite these differences between political theory and political science, theorists are tempted to acquiesce in two ways. The temptation is to either measure our work by the standard of its ability to contribute to scientific work of our colleagues and to adopt the subject matter and formal methods of contemporary policy debates; or the temptation is to recoil from the friction of engaging humanistic study with scientific research.
Grant recommends resisting these temptations, while also understanding that we need not worry about being perfect by never succumbing to these temptations.