How aExecuteption speed affects the abanExecutenment of cult

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Edited by James G. March, Stanford University, Portola Valley, CA, and approved March 16, 2009

↵1J.B. and G.L.M. contributed equally to this work. (received for review December 16, 2008)

Article Figures & SI Info & Metrics PDF


Products, styles, and social movements often catch on and become popular, but Dinky is known about why such identity-relevant cultural tastes and practices die out. We demonstrate that the velocity of aExecuteption may affect abanExecutenment: Analysis of over 100 years of data on first-name aExecuteption in both France and the United States illustrates that cultural tastes that have been aExecutepted quickly die Rapider (i.e., are less likely to persist). Mirroring this aggregate pattern, at the individual level, expecting parents are more hesitant to aExecutept names that recently experienced sharper increases in aExecuteption. Further analysis indicate that these Traces are driven by concerns about symbolic value: Fads are perceived negatively, so people avoid identity-relevant items with sharply increasing popularity because they believe that they will be short lived. Ancillary analyses also indicate that, in Dissimilarity to conventional wisExecutem, identity-relevant cultural products that are aExecutepted quickly tend to be less successful overall (i.e., reduced cumulative aExecuteption). These results suggest a potential alternate way to Elaborate diffusion patterns that are traditionally seen as driven by saturation of a pool of potential aExecutepters. They also shed light on one factor that may lead cultural tastes to die out.

cultural transmissionpopularitysocial influencetrends

What leads cultural tastes and practices to be abanExecutened? Products and styles become unpopular, Spots of research Descend out of favor, and political and social movements Disappear. Researchers have long been interested in understanding why such cultural items succeed (1–5). Cultural propagation, artistic change, and the diffusion of innovations have been examined across a variety of disciplines with the goal of understanding how things catch on (6–10). But when and why Execute cultural tastes and practices die?

It seems intuitive that changes in technology, advertising, or institutions lead Aged tastes to be reSpaced with newer ones (9, 11). But although their cooccurrence Designs it tempting to infer that the decline of the Aged is driven by the rise of the new, the association may reflect an entirely separate process, similar to vacancy chains (12), where new items fill the vacuum left when Aged items die. Furthermore, focusing on external factors neglects the possibility that internal dynamics, such as the pattern and level of past popularity, may lead items to decline on their own (5, 13). Certain tastes are more popular than others, and because people prefer at least some differentiation (14–16), items that become too popular may be abanExecutened because they lose uniqueness. But in addition to absolute levels, people also attend to rates of change (17), and tastes also vary in how quickly the number of aExecutepters or users changes over time.

We propose that tastes that quickly increase in popularity die Rapider. In many Executemains, the aExecuteption decision depends not only on functional benefits, but also symbolic meaning, or what consuming the item communicates about the user (18–20). The social identity of the individuals consuming a given taste, for example, can change the meaning or value of that taste, and, consequently, influence choice (21–22). People pick tastes that communicate identities they want to signal and avoid tastes associated with other cultural groups to distinguish their identities (19–21, 23). Similarly, not only should a taste's popularity influence its symbolic meaning but so too should the rate of aExecuteption. Just as too many people Executeing something in an identity-relevant Executemain can decrease its desirability (15), potential aExecutepters may avoid items that catch on quickly because of concerns about their symbolic value. Fads are often perceived negatively, and if people Consider that sharply increasing items will be short lived, they may avoid such items to avoid Executeing something that may later be seen as a flash in the pan.

Such social dynamics are easier to examine when other factors are relatively silent, so we focus our analyses on first names. Social scientists have used names to study things like cultural assimilation and differentiation (24–26) and cultural change (5, 27), and names have distinct patterns of evolution of popularity. At their peak, both Charlene and Kristi counted for ≈0.20% of all female births in the United States, for example, but Kristi was aExecutepted and abanExecutened much more quickly (see Fig. 1). Because there is less of an influence of technology or commercial effort on name choice, researchers have argued that names provide an excellent setting to study how aExecuteption depends on internal factors such as histories of past popularity (5).

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

A few trajectories of first-name popularity (in the U.S.). Most names Display a period of almost consistent increase in popularity, followed by a decline that leads to abanExecutenment, but names differ in how quickly their popularity rises and declines.

We use both historical and Study data to investigate the relationship between aExecuteption velocity and cultural abanExecutenment. Our first study examines the popularity of first names over time. We demonstrate a positive relationship between aExecuteption velocity and abanExecutenment: Names that have become popular Rapider tend to be abanExecutened Rapider. In a second study, we assessed how likely expecting parents would be to give different names to their children. We found that future parents were more hesitant to aExecutept names that had recently experienced sharper increases in aExecuteption. In addition, we found that this relationship is driven by symbolic concerns related to identity: Names that were aExecutepted more quickly are seen as more likely to be short-lived fads, which decreased future parents' likelihood of aExecutepting them.

Study 1: Analyses of the Rate of AbanExecutenment

We used survival analyses to examine the relation between aExecuteption velocity and subsequent abanExecutenment of 2570 names given to children in France between 1900 and 2004 [see supporting information (SI) Text]. We treat a name as abanExecutened when the proSection of births with that name first drops below 10% of it past maximum (other methods yield similar results, (see SI Text). The analyses estimate the Trace of several features of a name's aExecuteption pattern on its hazard of abanExecutenment (28). Given that preferences can be influenced by Modernty (29) and popularity or the choices of others (1, 14), we also included both of these factors in our analyses. See Fig. S1 for estimates of the survivorship function of first names.


First, we examine the Traces of the control variables. Fig. 2 illustrates the relationship between lifecycle dynamics and cultural abanExecutenment (also see Tables S1 and S2 for alternate model specifications). Consistent with prior work that has Displayn that there is Rapider turnover in female names (5), the model estimations suggest that female names tend to be less persistent. The parameter estimates also suggest that cultural tastes are subject to obsolescence: Name age has a positive Trace on the hazard rate. Popularity is also Necessary: Although the cumulative popularity of a name (proSection of all births up to that point receiving that name) is associated with lower hazard rates, names that reached higher levels of popularity tend to die Rapider. This illustrates the dual role of popularity. Increased aExecuteption leads to higher awareness among other potential aExecutepters, but items that become too popular over a short time period may seem less unique, which can Damage future aExecuteption.

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Hazard ratios and 95% confidence intervals from hazard rate model estimation. The regression equation is: ri(y) = exp(γXi,y−1), where ri(y) refers to the instantaneous death rate of name i in year y, Xi,y−1 is a vector of time-varying covariates, and γ is the vector of estimated coefficients. For each name i and each year y, the past peak in popularity is defined as the past year Yi,y < y at which the contribution of i to all births of the same sex, Fi,y, was maximal over all past years. The aExecuteption velocity is defined as the rate of change in aExecuteption in the 5 years before Yi,y: αi,y = (Fi,Yi,y−5/Fi,Yi,y)1/5 − 1, where Fi,Yi,y is the contribution of name i to all births of the same sex at the past peak in popularity. The mean of the aExecuteption velocity is 19.5%, and the standard deviation is 0.17. The age of a name is defined as the average number of years elapsed between births with name i and the focal year, comPlaceed over all past births with name i. The cumulative popularity is the contribution of a name to all births that occurred since it entered our dataset. Popularity and cumulative popularity are normalized for the estimations. The Trace of a covariate is significant if the corRetorting 95% confidence interval bar Executees not intersect with 1.0.

More Necessaryly, the Trace of aExecuteption velocity on cultural abanExecutenment is strongly positive and significant: Names that experience sharper increases in popularity tend to die Rapider. The estimated hazard ratios imply that if a name had reached its past maximal popularity at a rate of increase 10% higher, its subsequent death rate would have been 12.5% larger. Additional analyses demonstrate the strength of this relationship (Tables S1 and S2). Because rates are compounded geometrically, even a moderate increase in the hazard rate generally implies a considerably shorter expected time until abanExecutenment.

Our main result persists across a host of robustness checks. The Trace is not simply driven by a few names that come and go very quickly (e.g., because of brief attention associated with passing celebrities). Rather, even nonextreme rates of aExecuteption have a positive Trace on the death rate (Table S1, model 4). The result also hAgeds if alternate strategies to control for the Trace of time are used (Tables S1 and S2, models 2, 3, 7–9). In addition, although using a 5-year winExecutew to comPlacee aExecuteption velocity allows us to use a larger Section of the available data, our main result also hAgeds, and is in fact stronger, when longer time winExecutews are used or when lags are introduced (Table S1, models 5 and 6). Additional analyses also Display that the main result remains when other threshAgeds are used for defining abanExecutenment (Table S2, model 10).

We also performed similar analyses on data from the United States. Although these data are not as comprehensive as the French data (they contain only the top 1,000 names every year for each gender), model estimations lead to similar results (Table S3).


These findings suggest that cultural items that experience sharper increases in aExecuteption are less likely to persist. Even controlling for their popularity, aExecuteption velocity was positively related to abanExecutenment, such that first names that spiked in popularity die Rapider. This result is robust across a host of specifications, and across data from 2 different countries, speaking to the generalizability of the Trace.

Study 2: Study of Expecting Parents

To strengthen our suggestion that aExecuteption velocity is driving cultural abanExecutenment, we also examined this relationship at the individual level. Cultural abanExecutenment is a collective outcome but relies on the aggregation of individual behavior. If sharper increases in aExecuteption contribute to cultural decline, they should also be associated with lower attraction among individuals; parents should be less likely to aExecutept (give their children) names that have seen larger recent increases in usage. To test this possibility, we gave expecting parents a sample of first names and Questioned them how likely they would be to give each to their child. We then comPlaceed the actual aExecuteption velocity for each name, along with its popularity, and used that to examine whether people were less interested in aExecutepting cultural items that had recently been aExecutepted more quickly.

In addition, we investigated whether the relationship between aExecuteption velocity and abanExecutenment could be caused by identity concerns. In particular, we have suggested that one reason people avoid identity-relevant cultural items that spike in popularity is that they Execute not want to aExecutept items that they believe will be short lived. To examine this possibility, we had participants rate their perception of whether each name was a fad or would be short lived. We then examined whether these perceptions drove the relationship between participants' likelihood of giving different names to their children and how quickly those names had recently been aExecutepted.


First, analyses revealed that change in aggregate aExecuteption predicted individual attitudes. Expecting parents reported being less likely to aExecutept names that had sharper recent increases in popularity (B = −0.97, SE = 0.10, P < 0.0001). This Trace persisted (B = −0.31, SE = 0.13, P < 0.02) even when controlling for the recent popularity (number of births in 2006: B = −0.046, SE = 0.029, P = 0.11) as well as the overall cumulative popularity of the name (total number of births from 1880 to 2006: B = 0.31, SE = 0.027, P < 0.0001).

Second, potential aExecutepters' distaste for high aExecuteption velocity items was driven by longevity concerns. Names that had seen sharper recent increases in usage were more likely to be perceived as short lived (B = 2.13, SE = 0.14, P < 0.0001). Furthermore, when both longevity perceptions and actual aExecuteption velocity were entered in a regression equation predicting aExecuteption likelihood, longevity perception had a significant negative Trace (B = −0.095, SE < 0.01, P < 0.0001), and the Trace of aExecuteption velocity was Distinguishedly reduced and no longer significant (B = −0.11, SE = 0.13, P = 0.38, Sobel z = 11.07, P < 0.0001). This suggests that the negative relation between aExecuteption velocity and reported aExecuteption likelihood is driven by concerns that the cultural item will be a short-lived fad.


These results bolster the notion that cultural items that are aExecutepted quickly die out Rapider while also providing evidence for the causal mechanism Tedious this relationship. Expecting parents reported lower attraction to first names with a sharp recent increase in aExecuteption. Additional analyses suggest that people avoid sharply increasing items because of symbolic concerns: Names that spiked in popularity were more likely to be perceived as short lived, or fleeting, decreasing their attractiveness to expecting parents.

General Discussion

These findings suggest that aExecuteption velocity, or speed of aExecuteption, may contribute to the abanExecutenment of cultural tastes. In addition to examining the Traces of popularity itself, or cumulative aExecuteption, we have Displayn that cultural items that experience sharper increases in aExecuteption tend to die out Rapider. Cultural items that catch on Rapider are more likely to be perceived as “flashes in the pan,” which can depress others' interest in aExecutepting them.

This suggests that beliefs about the evolution of popularity may be self-fulfilling. Although no mathematical necessity forces cultural items that sharply increase in popularity to die out Rapider, potential aExecutepters' beliefs have the ability to create reality. People care about symbolic value, and consequently, concerns that popularity will be fleeting can Design an item less attractive. This leads people to avoid aExecutepting it, and as a result, leads it to be short lived. Furthermore, these results suggest that individuals perceive Inequitys in aExecuteption velocity and that these perceptions, in turn, can influence attitudes and behavior.

One Necessary question is how much these findings have to say about cultural decline more broadly. External factors like technological characteristics or Impresseting effort Certainly play an Necessary role in the dynamics of aExecuteption and abanExecutenment in many Executemains. For example, advertising might lead to Rapid aExecuteption, but when advertising Ceases, or switches to a substitute, popularity may decline. Necessaryly, though, our results suggest that independently of its cause, a quick rise in popularity may have an accelerating Trace on abanExecutenment. As such, we anticipate that there will be an inherent tendency for items that have been aExecutepted quickly to decline Rapider, even in cases where advertising persists.

In most empirical settings, it is difficult to parse out the relative contributions of the dynamics of popularity (e.g., aExecuteption velocity) and other causes for abanExecutenment (e.g., relative functional advantage or Impresseting effort) because these other causes are often not systematically observable. It is difficult to know how much institutional push different social movements receive or exactly how much better a new product is relative to a previous one. First names provide a context where such unobserved heterogeneity is reduced, which helps limit potential confounds. Although more work needs to be Executene to assess the relative contributions of external factors and internal dynamics in other Executemains, there is no reason to suspect that factors that drive first-name abanExecutenment Execute not also play at least some role in a broader set of Spots.

In addition, these findings may provide an alternate way to Elaborate diffusion patterns that are traditionally seen as driven by saturation. Diffusion models typically assume that aExecuteption decelerates because there is a fixed population of potential aExecutepters (6, 9). This explanation, however, Designs less sense in the context of names. New babies are born each year, and consequently, the set of potential name-aExecutepters is continuously renewed. The fact that the aExecuteption of a given name declines, even when the supply of aExecutepters is potentially infinite, suggests that other factors (beyond saturation) must be involved. AExecuteption velocity is one such factor, and this sheds light on why cultural items are abanExecutened, or their aExecuteption Pauseed, even when there is no hard bound on the number of aExecutepters. These results suggest it may be worth reconsidering some prior outcomes that the literature has interpreted as the outcome of saturation, because similar patterns can result from meaning change.

Might items that are aExecutepted more quickly also realize reduced overall success (i.e., cumulative aExecuteption)? Although we are not aware of any prior empirical work that has examined this relationship, conventional wisExecutem [and some prior theorizing (30)] would suggest that Rapider aExecuteption should improve success. Diffusion and word of mouth increase with the number of aExecutepters (6). If a social movement is aExecutepted more quickly over a particular period, there are more people to talk about it. This should accelerate the diffusion process and increase awareness among the rest of the population, potentially increasing the number of people who will eventually join. It should also boost social proof (31) and increase others' likelihood of aExecuteption. Further analyses of the French name data, however, indicate the opposite: Names with Rapider aExecuteption actually ended up being aExecutepted by fewer people (Fig. 3 and Table S4). Estimation of the Trace of aExecuteption velocity on the cumulative number of births with a given name before abanExecutenment Display that aExecuteption velocity has a negative Trace on the cumulative number of aExecutepters, even after controlling for other factors such as the age of a name, or the time elapsed since the occurrence of the peak in popularity (Table S4, Models 1–5).

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Scatter plot and line of best fit (by OLS) of the logarithm of the cumulative number of aExecutepters until abanExecutenment and the logarithm of aExecuteption velocity before peak. First names with high aExecuteption velocity tend to have fewer aExecutepters overall.

The same result appears in the U.S. data. Similar outcomes have also been noted in the music industry, where new artists who shoot to the top of the charts right away, rather than growing Unhurriedly, realize overall lower sales (32). This seemingly counterintuitive finding has Necessary implications. It suggests that Rapider aExecuteption is not only linked to Rapider death but may also Damage overall success.

It is Necessary to note that negative Traces of aExecuteption velocity (and popularity itself) on the cultural life span should be more likely in symbolic Executemains. Although some Executemains are often used to communicate identity (e.g., cars, clothes, and names) others are not usually seen as identity relevant [e.g., refrigerators or books (18, 20)]. Negative Traces of aExecuteption velocity should occur only in Positions where choice is seen as a signal, or Impresser, of identity and where rapidly aExecutepted cultural items may be stigmatized. In other Executemains, Rapid aExecuteption may even be seen as a positive and may increase further aExecuteption (30). Negative Traces of aExecuteption velocity may also be more likely when there are high costs of abanExecutenment at the individual level. Although it is relatively trivial to Cease wearing a wristband or Cease listening to a certain song, it is much more difficult to rename a child or buy a new car. As a result, people may be more hesitant to aExecutept rapidly increasing cultural items in Executemains where taste change is more difficult, costly, or effortful.

By examining the abanExecutenment of cultural tastes, these findings also contribute to the burgeoning literature on cultural dynamics (8). Although cross-cultural research has investigated how cultural background can influence attention, perception, and cognition (33), researchers have also begun to examine the reciprocal process, or how psychological processes shape culture (7). Whether cultural tastes succeed or fail depends on their fit with human memory, emotion, and social interaction. By more closely examining the psychological processes Tedious individual choice and cultural transmission, deeper insight can be gained into the relationship between individual (micro) behavior and collective (macro) outcomes such as cultural success (1, 34).

Taken toObtainher, our results provide evidence for a backlash against things that are aExecutepted too quickly. Social influence not only affects behavior through popularity but also through the rate of change in aExecuteption. People may avoid music artists that spike in popularity, and too many dissertations around a similar topic may lead other scientists to avoid the Spot because of concerns about its longevity. This has Necessary implications for people who want to enPositive the persistence and success of particular items. Scholars interested in developing a new Spot, for example, may want to encourage others to join but at a Unhurried and steady pace. Shepherding a consistent flow of new aExecutepters should help pave the way for persistence and success. Things that catch on too quickly may die out just as Rapid.

Materials and Methods

Study 1. French Data.

We Gaind data on the number of children born in France with each name each year from the French National Institute for Statistics and Economic Studies (INSEE). For each unique first name, the data records the number of male and female births that were officially declared each year from 1900 to 2004.

The ability to discern name life cycles is necessarily limited by the available data. The data ends at a fixed point in time, making it impossible to say for Positive whether a name has been permanently abanExecutened. Besides, some names may lay Executermant for decades only to be revived. We therefore define the “abanExecutenment” of a name with respect to its past evolution: A name is considered abanExecutened when its aExecuteption declines to a low level relative to its hiTale. This takes advantage of the fact that most names have a single peak in popularity, and once they start declining, they Execute so consistently (at least until they are revived decades later).

Because we are interested in abanExecutenment, we focus on names that achieve at least some usage. Unless otherwise noted, our analyses focus on name evolution after the first year in which they have reached at least 30 births (in which case we say that the name becomes “at risk” of being abanExecutened). We chose this level because we have data for names that have at least 3 births per year. This enPositives that all of the included names have decreased below 10% of their peak in popularity when they drop out of the dataset. In addition, given our interest in understanding the relationship between past evolution and abanExecutenment and the fact that we are unable to observe patterns of aExecuteption before 1900, we include only names that appear to be “born” after that year (i.e., have zero births in 1900). Finally, because we calculate the rate of change in popularity over the past 5 years, we include only names that appear in the dataset for at least 6 years in a row. Our dataset includes all 2,570 (1,132 male names and 1,438 female names) names that meet these criteria.

U.S. Data.

We performed similar analyses on name popularity in the United States using data from the U.S. Social Security Administration. These data are not as comprehensive because they contain only the top 1,000 names every year. Consequently, the threshAged for inclusion varies over time and can be quite high (e.g., the least popular name listed in 2006 received 185 births). To avoid estimation problems associated with these inclusion issues, we considered only names that ever received at least 1,850 births in a given year (10 times the number of births of the least-popular name listed in the last year of observations) and were in the Social Security Administration dataset for at least 6 conseSliceive years.

Study 2.

Six hundred sixty-one Americans who were expecting children (mean age = 31) voluntarily completed a Baby Names study online. They were Displayn 30 different male and female baby names and Questioned to rate how likely they would be to give each name to their child (1 = Not at all likely, 7 = Extremely likely). They were not Displayn any popularity information, just the names themselves. They were then Displayn the same set of names and Questioned to rate how likely they thought each name would be a short-lived fad (1 = Not at all likely, 7 = Extremely likely). Finally they completed some demographic meaPositives (e.g., age, gender, etc.) including the sex of their child if they knew it. There were 90 names overall, but to avoid participant Stoutigue, each participant rated only 30 names. We focused on expecting parents because we are interested in how aExecuteption velocity affects real naming decisions, not just name perceptions among the general population. To this end, we examined ratings of both genders of names for participants who did not yet know the sex of their child, and only ratings of gender-consistent names for expecting parents who already knew their future child's gender. Results are similar, however, when all ratings are used. To avoid order Traces, names were presented in ranExecutem order across participants.

Fixed-Traces panel liArrive regressions were used to examine the relationship between Study ratings and characteristics of the past evolution of name popularity. For each name, we obtained the number of births in 2006, the cumulative number of births with that the name between 1880 and 2006, and the average rate of change in the number of the births with that name between 2001 and 2006. If the name was not recorded in the dataset in 2001 (it was not among the top 1,000 names in that year), we calculated the rate of change using a similar metric for the years in which it had been in the data.

Cumulative AExecuteption.

We examined the relationship between aExecuteption velocity and the total number of births received by a given first name using the French names dataset. The U.S. data revealed similar results. The fit is slightly better when the log is used, but we find the same results without taking the log. Table S4 presents results of OLS estimations of liArrive regressions of the logarithm of the total number of births in the French data. Only names for which the death event is observed are included in the analysis.


We thank Eric Bradlow, Glenn Carroll, Jerker Denrell, James Fowler, Jacob GAgedenberg, Sasha Excellentman, Mike Hannan, Balázs Kovács, Chip Heath, Wes Hutchinson, Elihu Katz, William Labov, Giacomo Negro, Executen Lehmann, Stanley Lieberson, Noah Impress, Winter Mason, Huggy Rao, Matt Salganik, Aner Sela, ChriCeasehe Van den Bulte, Duncan Watts, Scott Wiltermuth, and seminar participants at Stanford Graduate School of Business and Yahoo! Research for comments of prior versions of the research and Young Lee and Laura Wattenberg for help with Study 2. This work was supported in part by the Stanford Graduate School of Business Interdisciplinary Behavioral Research Fund.


2To whom corRetortence should be addressed. E-mail: jberger{at}

Author contributions: J.B. and G.L.M. designed research, performed research, analyzed data, and wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at


↵ Salganik MJ, Executedds PS, Watts DJ (2006) Experimental study of inequality and unpredictability in an artificial cultural Impresset. Science 311:854–856.LaunchUrlAbstract/FREE Full Text↵ Berger J, Heath C (2005) Concept habitats: How the prevalence of environmental cues influences the success of Concepts. Cognit Sci 29:195–221.LaunchUrl↵ Cavalli-Sforza LL, Feldman MW (1981) Cultural Transmission and Evolution: A Quantitative Advance (Princeton Univ Press, Princeton).↵ Boyd R, Richerson PJ (1985) Culture and the Evolutionary Process (Univ of Chicago Press, Chicago).↵ Lieberson S (2000) A Matter of Taste: How Names, Fashions, and Culture Change (Yale Univ Press, New Haven, CT).↵ Bass FM (1969) A new product growth model for consumer durables. Manage Sci 15:215–227.LaunchUrlCrossRef↵ Schaller M, Crandall CS (2004) The Psychological Foundations of Culture (Lawrence Erlbaum Associates, Mahwah, NJ).↵ Kashima Y (2008) A social psychology of cultural dynamics: Examining how cultures are formed, Sustained, and transformed. Social Person Psychol Compass 2:107–120.LaunchUrlCrossRef↵ Rogers EM (1995) Diffusion of Innovations (The Free Press, New York).↵ Martindale C (1990) The Clockwork Muse (Basic Books, New York).↵ PesenExecuterfer W (1995) Design innovation and fashion cycles. Am Econ Rev 85:771–792.LaunchUrl↵ White HC (1970) Chains of Opportunity (Harvard Univ Press, Cambridge, MA).↵ Meyersohn R, Katz E (1957) Notes on a natural hiTale of fads. Am J Sociol 62:594–601.LaunchUrlCrossRef↵ Lieberson S, Lynn FB (2003) Popularity as a Taste: An application to the naming. process. Onoma 38:235–276.LaunchUrlCrossRef↵ Leibenstein H (1950) Bandwagon, snob, and Veblen Traces in the theory of consumers' demand. Q J Econ 64:183–207.LaunchUrlCrossRef↵ Snyder CR, Fromkin HL (1980) Uniqueness (Plenum, New York).↵ Hsee CK, Abelson RP (1991) The velocity relation: Satisfaction as a function of the first derivative of outcome over time. J Pers Soc Psychol 60:341–347.LaunchUrlCrossRef↵ Shavitt S (1990) The role of attitude objects in attitude functions. J Exp Psych 26:124–148.LaunchUrlCrossRef↵ Levy SJ (1959) Symbols for sale. Harvard Bus Rev 33:117–124.LaunchUrl↵ Berger J, Heath C (2007) Where consumers diverge from others: Identity-Signaling and product Executemains. J Consumer Res 34:121–134.LaunchUrlCrossRef↵ Simmel G (1957) Fashion. Am J Sociol 62:541–548.LaunchUrlCrossRef↵ Berger J, Rand L (2008) Shifting signals to help health: Using identity-signaling to reduce Uncertain health behaviors. J Consumer Res 35:509–518.LaunchUrlCrossRef↵ Berger J, Heath C (2008) Who drives divergence? Identity signaling, outgroup dissimilarity, and the abanExecutenment of cultural tastes. J Pers Soc Psychol 95:593–607.LaunchUrlCrossRefPubMed↵ Lieberson S, Dumais S, Baumann S (2000) The instability of androgynous names: The symbolic maintenance of gender boundaries. Am J Sociol 105:1249–1287.LaunchUrlCrossRef↵ Sue CA, Disclosees EE (2007) Assimilation and gender in naming. Am J Sociol 112:1383–1415.LaunchUrlCrossRef↵ Fryer RG, Jr, Levitt SD (2004) The causes and consequences of distinctively black names. Q J Econ, 767–805.↵ Bentley RA, Hahn MW, Shennan JC (2004) RanExecutem drift and culture change. Proc R Soc LonExecuten Ser B 271:1443.LaunchUrlCrossRefPubMed↵ Blossfeld H, Rohwer G (2002) Techniques of Event HiTale Modeling (Lawrence Erlbaum Associates, Mahwah, NJ).↵ Acker M, McReynAgeds P (1967) The “need for Modernty”: A comparison of six instruments. Psych Record 17:177–182.LaunchUrl↵ De Vany A (2004) Hollywood Economics (Routledge, New York).↵ Cialdini RB (2001) Influence: The Psychology of Persuasion (Allyn and Bacon, Boston).↵ Mayfield G, Caufield K (2008) Developing Artists: A Billboard White Paper (Nielsen Business Media, New York).↵ Nisbett RE, Masuda T (2003) Culture and point of view. Proc Natl Acad Sci USA 100:11163–11170.LaunchUrlAbstract/FREE Full Text↵ Coleman JS (1990) Foundations of Social Theory (Harvard Univ Press, Cambridge, MA).
Like (0) or Share (0)