Published multiple files

src/site/notes/Bookmarks/Health, Mind and so on/Is Noise Always Bad_ Exploring the Effectsof Ambient Noise on Creative Cognition.md

@@ -0,0 +1,681 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/health-mind-and-so-on/is-noise-always-bad-exploring-the-effectsof-ambient-noise-on-creative-cognition/","tags":["ideas","interesting","pdf","scientific"]}
+---
+
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.001.png)
+
+Is Noise Always Bad? Exploring the Effects of Ambient Noise on Creative Cognition Author(s): Ravi Mehta, Rui (Juliet) Zhu and Amar Cheema![ref1]
+
+Source: _Journal of Consumer Research_ , Vol. 39, No. 4 (December 2012), pp. 784-799 Published by: Oxford University Press
+
+Stable URL: https://www.jstor.org/stable/10.1086/665048![ref1]
+
+JSTOR is a [not-for-profit service that helps scholars, researchers, and students discover, use, and](https://www.jstor.org/stable/10.1086/665048) build upon a wide range of content in [a trusted digital archive. We use information technology and tools to increase](https://www.jstor.org/stable/10.1086/665048) productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
+
+Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms
+
+_Oxford University Press_ is collaborating with JSTOR to digitize, preserve and extend access to ![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.003.png)_Journal of Consumer Research_
+
+**Is Noise Always Bad? Exploring the Effects of![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.004.png) Ambient Noise on Creative Cognition**
+
+RAVI MEHTA
+
+RUI (JULIET) ZHU AMAR CHEEMA
+
+This paper examines how ambient noise,animportantenvironmentalvariable,can affect creativity. Results from five experiments demonstrate that a moderate (70 dB) versus low (50 dB) level of ambient noise enhances performance on creative tasks and increases the buying likelihood of innovative products. A high level of noise (85 dB), on the other hand, hurts creativity. Process measures reveal that amoderate(vs.low)levelofnoiseincreasesprocessingdifficulty,inducingahigher construal level and thus promoting abstract processing, which subsequently leads to higher creativity. A highlevel ofnoise,however,reducestheextentofinformation processing and thus impairs creativity.
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+
+Cthe one hand, we as consumers engage in everyday
+
+reativity is ubiquitous in the realm of consumption. On creative behavior such as home decor, fashion, or planning
+
+meals with limited resources (Burroughs and Mick 2004; Burroughs, Moreau, and Mick 2008). On the other hand, many businesses thrive on consumers’ ability and desire to be creative. For example, consumers’ ability to understand and appreciate creative and metaphorical persuasive mes- sages is an essential element of any successful creative ad- vertising campaign. Similarly, consumers’ desire to be cre- ative has a significant impact on the success of many products, including play kits (e.g., model trains, paint-by- numbers kits), how-to guides (e.g., cookbooks, landscaping; Dahl and Moreau 2007), and many other innovative new products.
+
+Because creativity permeates the consumption environ-
+
+Ravi Mehta (mehtar@illinois.edu) is assistant professor of business administration ![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.005.png)atthe University ofIllinoisatUrbana-Champaign,350Woh- lers Hall, Champaign, IL 61820. Rui (Juliet) Zhu (juliet.zhu@sauder .ubc.ca) is Canada Research Chair in Consumer Behavior and associate professor of [marketing at the Sauder ](mailto:mehtar@illinois.edu)School of Business, University of British Columbia, Vancouver, BC V6T 1Z2. Amar Cheema (cheema@ virginia.edu) is associate professor of marketing at the [McIntire School of Commerce,](mailto:juliet.zhu@sauder.ubc.ca) University of Virginia, Charlottesville, VA 22904. The authors thank Jim Burroughs for helpful comments. Financial support from the Social Sciences and Humanities Research Council of Canada is[ gratefully acknowledged.](mailto:cheema@virginia.edu)
+
+_John Deighton and Laura Peracchio served as editors and Jaideep Sen- gupta served as associate editor for this article._
+
+_Electronically published March 21, 2012_
+
+ment, it is not surprising that a great deal of research has explored factors that can affect consumers’ creative ability and performance, including external constraints (Moreau and Dahl 2005), involvement (Burroughs and Mick 2004), an- alogical thinking (Dahl and Moreau 2002), systematic train- ing (Goldenberg, Mazursky, and Solomon 1999), and life experiences (Maddux and Galinsky 2009). However, extant research in this domain has largely ignored the impact of physical environment on an individual’s creativity (for ex- ceptions, see Mehta and Zhu 2009; Meyers-Levy and Zhu 2007). The current study attempts to fill this gap in the literature by investigating the effects of an important en- vironmental variable—ambient noise—on creativity.
+
+Although extensive research has examined the impact of noise on human cognition and behavior in general, little research has focused on the effects of noise on creativity per se. Furthermore, this limited research shares two key weaknesses. First, studies examining the noise-creativity re- lationship have yielded inconclusive findings. While most studies find that noise hurts creativity (Hillier, Alexander, and Beversdorf 2006; Kasof 1997; Martindale and Gree- nough 1973), there is some evidence that for highly original individuals, moderate noise may lead to improved creative performance (Toplyn and Maguire 1991). Second, although researchers have proposed different reasons as to why noise may affect creativity, such as arousal (Martindale and Gree- nough 1973; Toplyn and Maguire 1991), stress (Hillier et al. 2006), attention span (Kasof 1997), very little research has actually examined these mechanisms empirically. Thus, there is no clear understanding of why noise affects creativity.
+
+In this research we examine the underlying mechanism
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+
+784
+
+`  `2012 by JOURNAL OF CONSUMER RESEARCH, Inc. ● Vol. 39 ● December 2012 All rights reserved. 0093-5301/2012/3904-0008$10.00. DOI: 10.1086/665048
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY 789
+
+through which ambient noise affects creative cognition. We theorize that a moderate (vs. low) level of ambient noise is likely to induce processing disfluency or processing diffi- culty, which activates abstract cognition and consequently enhances creative performance. A high level of noise, how- ever, reduces the extent of information processing, thus im- pairing creativity. A series of five experiments offers sys- tematic support to our theory. In addition, findings from our last experiment extend our theorizing by showing that a moderate level of noise also increases buying likelihood of innovative products.
+
+This research promises to make several contributions. First, it contributes theoretically to the noise literature by demonstrating an inverted-U relationship between noise and creativity, thus reconciling the mixed findings observed in the current literature. Second, it adds to the creativity lit- erature by identifying ambient noise as an important factor affecting creative cognition and by providing process evi- dence for this relationship. Finally, it adds support to the growing recognition that subtle cues in our physical envi- ronment can indeed affect human cognition and behavior. The results from this research also have important practical implications in terms of inducing consumer creativity and encouraging adoption of new products.
+
+**THEORETICAL BACKGROUND** Noise and Creativity
+
+By definition, any unwanted sound is called “noise.” A _sound_ is defined as a vibration, or a traveling wave that is an oscillation of pressure transmitted through a medium (solid, liquid, or gas). The pressure of these vibrations within a given frequency range stimulates sensation in the ears and enables hearing. Hearing is thus sensitive to the sound pres- sure level, or “sound level,” measured in decibels (dB). (See appendix table A1 for a list of sound sources and their sound levels.) It is worth noting that sound level is not equivalent to loudness; the latter is a psychological correlate and a subjective measure of sound level. There is a complex re- lationship between the two, such that a 10 dB increase in sound level approximately corresponds to a twofold increase in loudness (“Noise Pollution,” _The Columbia Encyclope- dia_, 6th ed. [2008]).
+
+Although considerable research has examined the effects of noise on human cognition and behavior (Hamilton and Copeman 1970; Hockey 1969, 1970a, 1970b; Hygge, Evans, and Bullinger 2002; Nagar and Pandey 1987; O’Malley and Poplawsky 1971; Weinstein 1974), there has been littlefocus on the effects of noise on creativity. Furthermore, this lim- ited area of research has not only produced inconclusive results but also proposed different process mechanisms through which noise might affect creativity. The most com- mon finding is that high levels of noise hurt creativity. Re- searchers have focused on primarily white noise and pink noise in this line of research. White noise is a sound that is artificially created by combining all audible frequencies (i.e., every frequency within the range of human hearing,
+
+generally from 250 Hz to 8,000 kHz) in equal amounts. White noise sounds like a gentle hiss. Pink noise is also artificially created and is a variant of white noise. Pink noise sounds something like the buzz on an empty television sta- tion. For example, Martindale and Greenough (1973) dem- onstrate that a high level of white noise reduces performance on the Remote Associates Test (RAT), a task commonly used to measure creativity; they conjecture that high arousal induced by the high level of white noise is responsible for the reduced creativity they observed. Kasof (1997) dem- onstrates that creative performance in writing poetry is im- paired by exposure to a high level of pink noise, and spec- ulates that the high noise level may have hurt creativity by narrowing attention. Hillier et al. (2006) argue that stress induced by a high level of white noise is responsible for reduced performance on a creative task (RAT). None of these studies, however, actually testedtheirproposedprocess mechanism.
+
+One exception in this line of research is the finding that for highly creative individuals, a moderate noise level may lead to higher creative performance relative to both low and high noise levels (Toplyn and Maguire 1991). Toplyn and Maguire had participants complete a number of creativity tasks and used their performance on one such task (the RAT) to assess their baseline creativity level. Theyfoundthathighly creative individuals (defined as those who scored high on the RAT) exhibited greater creativity on other tasks when pre- sented with a moderate level of white noise than when the noise level was either high or low. Toplyn and Maguire spec- ulate that arousal may underlie this effect. For less creative individuals, on the other hand, no significant difference was observed among low, moderate, and high levels of noise.
+
+The above review of the extant literature on the impact of noise on creativity thus reveals a number of problems. First, this literature not only has produced inconclusive re- sults but also lacks rigorous testing of the proposed mech- anisms through which noise affects creativity. In the current paper, therefore, we empirically test the cognitive mecha- nism through which we propose ambient noise affects cre- ativity. Second, most extant research has employed non- realistic noise stimuli that are neither common nor sustainable in typical consumption contexts, such as white noise (Hillier et al. 2006; Martindale and Greenough 1973; Toplyn and Maguire 1991) and pink noise (Kasof 1997). In our research we therefore focus on ambient noises that are much more common in daily life (e.g., background noise in a restaurant). Finally, existing research is silent on how noise may influence individuals’ acceptance of creative ideas. We examine this question in one of our studies by looking at how noise affects consumers’ responses to in- novative products.
+
+The Proposed Process through Which Noise Can Affect Creativity
+
+We argue that noise distracts people but that the degree of distraction induced by various noise levels will affect
+
+creativity differently. A high level of noise may cause a great deal of distraction, causing individuals to process in- formation to a lesser extent and therefore to exhibit lower creativity. A moderate (vs. low) level of noise, however, is expected to distract people without significantly affecting the extent of processing. Further, we reason that such a moderate distraction, which induces processing difficulty, enhances creativity by prompting abstract thinking. We pre- dict, in sum, that a moderate level of noise will enhance creativity relative to both high and low levels of noise.
+
+To elaborate, we theorize that the distraction caused by a moderate level of noise will lead to processing difficulty or disfluency (we use these terms interchangeably in this paper). _Processing disfluency_ has been defined as the lack of “the subjective experience of ease or speed in processing information” (Oppenheimer 2008, 237). Of particular rel- evance to our research is the finding that processing dis- fluency induces a higher construal level, such that individ- uals engage in abstract thinking (Alter and Oppenheimer 2008). Alter and Oppenheimer (2008) presented participants with city names (e.g., New York) in either hard-to-read (i.e., disfluent) or easy-to-read (i.e., fluent) fonts, asking them to judge how far away the target city was relative to their current location and to describe the target city. People who saw disfluent fonts adopted a higher construal level, judging the city to be farther away and describing it using more abstract terms than those who saw fluent fonts.
+
+Further, there is evidence relating a higher construal level (or abstract thinking) to greater creativity. Smith (1995) sug- gests that when people are thinking abstractly, they are less likely to fixate, and thus more creative, than those who are thinking concretely. Echoing this idea, Smith, Ward, and Schumacher (1993) had participants generate new product ideas (e.g., a spill-proof coffee cup or a new toy) and found that fixating people with a few examples before the idea- generation task decreased the abstraction and the creativity level of the generated ideas. Similarly, Forster, Friedman, and Liberman (2004) find that priming people with a distant- future time perspective, which prompts a higher construal level and increased abstraction, enhances creativity.
+
+Based on the above research, we predict that the distrac- tion caused by a moderate (vs. low) level of noisewillinduce processing difficulty, leading to abstract processing and, consequently, to greater creativity. A different mechanism, however, is proposed for the high noise level. Although we expect that a high (vs. moderate) noise level will also lead to reduced creativity, we argue that this reduction is driven largely by the reduced extent of information processing. Specifically, while a high noise level should also prompt a higher construal level, this positive effect on creativity is likely to be counteracted by the reduced information pro- cessing that issimultaneouslyinducedbyhigh(butnotmod- erate) levels of noise. This reduced processing can prevent individuals from thinking divergently, for example, creating new links, which is necessary for creative thinking (Wood- man, Sawyer, and Griffin 1993). In fact, brain-imaging stud- ies have shown that the brain areas responsible for atten-
+
+tional processes, which indicate the extent of information processing, are also responsible for cognitive processes that lead to higher creativity (Fink et al. 2010). On the basis of the above arguments, we hypothesize that a high (vs. mod- erate) level of noise will lead to a lower creativity level due to reduced information processing.
+
+Finally, we posit that the effects hypothesized above will apply not only to the generation of creative ideas but also to the adoption of innovations. Prior research suggests that innovative consumers are more likely to adopt novel prod- ucts (Hirschman 1980; Houston and Mednick 1963; Im, Bayus, and Mason 2003). If a moderate level of noise can enhance creativity, it should also enhance consumers’ ap- preciation for novel products (Thompson, Hamilton, and Rust 2005). Thus, we expect that a moderate (vs. high or low) level of noise should increase the adoption of inno- vative products.
+
+We test the above hypotheses in five experiments. The first experiment demonstrates the basic effect that a mod- erate level of noise enhances creativity relative to both high and low levels of noise. Experiments 2 and 3 provide evi- dence that construal level and process disfluency indepen- dently mediate the noise-creativity relationship, and at the same time rule out a number of alternative explanations. Experiment 4 tests for the complete process mechanism through which a moderate (vs. low) level of noise enhances creativity. The final experiment, a field study, examines the effect of ambient noise on the adoption of innovative prod- ucts and also examines an important moderator of this effect, namely, individuals’ baseline level of innovativeness.
+
+**EXPERIMENT 1**
+
+Method
+
+_Stimuli._ To create ambient noise reflecting typical con- sumption contexts, we blended a combination ofmulti-talker noise in a cafeteria, roadside traffic,and distant construction noise to create a soundtrack of constantly varying back- ground noise. All noises were first independently recorded at real-life venues (e.g., in a cafeteria, near a construction site) and then superimposed electronically to create the final digital soundtrack. Noise manipulation was accomplished by playing this digital soundtrack on an MP3 player plugged into two stereophonic speakers while participants were com- pleting the task. The volume of the speakers was adjusted as needed to generate low (50 dB), moderate (70 dB), and high (85 dB) levels of noise (Nagar and Pandey 1987). To add a baseline for comparison purposes, we also included a control condition in this experiment, in which about one- fourth of participants completed the focal task while the soundtrack was not played. In this condition, the average ambient noise level for each session in our lab setting varied between 39 dB and 44 dB, with an overall average of 42 dB.
+
+To assess creative performance, we used the Remote As- sociates Test (RAT; Mednick 1962), which has been widely used to assess creative thinking in both psychology and
+
+marketing research (Griskevicius, Cialdini, and Kenrick 2006, 2007; Van den Bergh, Dewitte, and Warlop 2008). Each RAT item consists of three or four stimulus words that are in some way related to a fourth or fifth unreported target word. Participants are given the stimulus words, and their task is to determine the target word. For example, for the stimulus words “Shelf,” “Read,” and “End,” the correct re- sponse is “Book.” We included eight RAT items in this experiment. We expected that participants in the moderate- noise condition would perform better on this test than those in all other conditions (i.e., high-noise, low-noise, and con- trol conditions).
+
+_Procedure._ Sixty-fiveundergraduatestudents(46women) from the University of British Columbia participated in the “Restaurant Experience Study” in exchange for a course credit. The experiment was run in small groups of no more than four people per session. Each session was randomly assigned to one of the four noise conditions. Upon arrival, participants were asked to take one of the four available desks, which were strategically placed on the arc of a semi- circle. Two stereophonic speakers on stands were positioned in the center of the circle, so that all desks were equidistant to the speakers. For the high-, moderate-, and low-noise conditions, the noise level was measured using a sound- level meter before each session and was kept constant (≈85 dB, 70 dB, or 50 dB; variation due to changes in noise content was approximately 3 dB) at each desk. The setup was identical for the control condition, except that no noise soundtrack was played.
+
+All experiments were computer based. In experiment 1, the instruction screen explained that the researchers were studying people’s experiences in different kinds of restau- rants and therefore, to create an appropriate ambience, back- ground noise such as one would usually hear while dining at a roadside restaurant might be present during the exper- imental session. The speakers were then either turned on at the 85 dB, 70 dB, or 50 dB level or left turned off, depending on the condition. All participants then completed eight RAT items, presented one at a time on the computer screen. The program recorded each participant’s responses and his or her response time for each RAT item.
+
+Next, participants rated their current feelings in response to six adjectives, using a 7-point scale (1 p not at all, 7 p very much). Three were positive mood items (happy, cheer- ful, joyful) and three were negative mood items (sad, de- pressed, glum); the presentation order of the six items was randomized. The experiment concluded with some demo- graphic questions.
+
+Results
+
+As anticipated, an analysis of variance (ANOVA) re- vealed a significant main effect of noise level on RAT per- formance (_F_(3, 61) p 3.21, _p_ ! .05), such that respondents in the moderate-noise condition (_M_ p 5.80) generated more correct answers than those in the low-noise (_M_ p 4.29, _t_(61) p 2.34, _p_ ! .05), high-noise (_M_ p 3.88; _t_(61) p
+
+2.94, _p_ ! .01), or control conditions (_M_ p 4.48, _t_(61) p 2.06, _p_ ! .05). The differences among the latter three conditions were not significant (all _t_ ! 1).
+
+We then analyzed the average response time for each RAT item, and again found a significant main effect of noise level (_F_(3, 61) p 3.48, _p_ ! .05). Although there was no significant difference among the participants in moderate-noise (_M_ p 14.09 seconds), low-noise (_M_ p 14.94 seconds), and control (_M_ p 13.29 seconds) conditions (all _t_ ! 1), those in the high-noise condition (_M_ p 9.51 seconds) spent significantly less time on each item than participants in each of the other three conditions (moderate noise: _t_(61) p 2.47, _p_ ! .05; low noise: _t_(61) p 3.02, _p_ ! .01; control: _t_(61) p 2.10, _p_
+
+- .05). This finding is consistent with our theorizing that a high level of noise reduces the extent of information pro- cessing.
+
+  Next, we analyzed participants’ responses on the mood items. The positive and negative items were averaged to create a positive mood index (a p .90) and a negative mood index (a p .75). There were no significant mood effects across conditions (positive: *M*High p 2.96, *M*Moderate p 2.62, *M*Low p 2.78, *M*Control p 2.88; negative: *M*High p 1.54, *M*Moderate p 1.91, *M*Low p 1.78, *M*Control p 1.59; all _F_ ! 1).
+
+  Discussion
+
+  Results from experiment 1 provide support for our basic proposition that a moderate level of background noise en- hances creativity relative to high-, low-, and no-noise con- ditions. As noted above, although the control condition did not include any active manipulation of noise, there was al- ways some ambient noise present; the average ambient noise across all control-condition sessions was measured as 42 dB, which is close to our manipulated low-noise condition (50 dB). Therefore, it is not surprising that there was no significant difference in respondents’ creativity levels be- tween these two conditions. In addition, the nonsignificant results from the mood measures rule out a potential expla- nation, that is, that mood might have contributed to our findings. Finally, we observed that the time spent on the focal task was lower in the high-noise condition than the other three conditions, indicating reduced information pro- cessing under the high noise condition. While we believe this findingsupports our theorizing that a high level of noise leads to reduced cognitive capacity to process, it may also imply a motivation account, such that a high level of noise reduces processing motivation. In the next experiment we try to tease apart these two competing accounts and provide further evidence for the reduced cognitive capacity account.
+
+**EXPERIMENT 2**
+
+This experiment aims to provide theoretical replication of the results of experiment 1. In addition, it is intended to (1) test whether construal level underlies the beneficial effect of moderate (vs. low) levels of noise on creativity and (2) test whether reduced capacity of processing, rather than re-
+
+duced processing motivation, is responsible for the impaired creativity in the high- (vs. moderate-) noise condition.
+
+Method
+
+_Stimuli._ We used the same noise manipulation as em- ployed in experiment 1, except that the control condition (i.e., no noise) was dropped from this and all subsequent experiments because it was similar to the low-noise con- dition (42 dB vs. 50 dB) and the two did not produce any statistically significant difference in results, as shown in experiment 1.
+
+An idea-generation task was used as the focal task in this study. Participants were asked to imagine themselves as a mattress manufacturer looking for creative ideas for a new kind of a mattress; that is, their task was to come up with creative ideas for a new mattress. They were also told that the ideas could be geared toward either new features or a completely new product. We recorded both the ideas gen- erated by each participant and the amount of time spent by him/her on this task. The quality of ideas was used to mea- sure creativity, whereas the number of ideas and the time spent on the task were used to measure the extent of pro- cessing.
+
+We also measured participants’ construal level in order to test whether this variable can explain our results. The 25- item Behavioral Identification Form (BIF; Vallacher and Wegner 1987) was used to measure individuals’ situational construal level. The BIF presents individuals with a series of behaviors and offers two different ways of identifying each behavior; for example, “making a list” could be iden- tified as “getting organized” (an abstract, high-level iden- tification) or as “writing things down” (a concrete, low-level identification). Individuals must select which of the two identifications best describes the behavior for them at the current moment. Participants’ responses to all 25 behaviors are summed to create a construal-level index; higher values indicate a higher construal level.
+
+Finally, to measure participants’ processing motivation, we asked three questions. Specifically, participants indicated on a 7-point scale (1 p not at all, 7 p very much) the extent to which they were motivated to complete the study, enjoyed doing the task, and thought the study was inter- esting.
+
+_Procedure._ Sixty undergraduate students (36 women) at the University of British Columbia participated in the ex- periment in exchange for $10. Experimental sessions were run in groups of no more than four people per session. Each group of participants was randomly assigned to the high-, moderate-, or low-noise condition. The cover story, seating arrangement, and equipment setup were exactly the same as in experiment 1. Once participants had settled down and after the noise was started, they first answered some de- mographic questions, which took no more than 2 minutes to complete. Then all participants were presented with the BIF items. Next they completed the idea-generation task, which asked them to generate as many creative ideas as they
+
+could think of for a new kind of a mattress and type them into the computer. No time limit was imposed for this task. The computer program recorded the ideas generated by each participant and the time taken to generate these ideas. Fi- nally, participants answered the three questions assessing their processing motivation. The experiment concluded with some demographic questions.
+
+Results
+
+_Number of Ideas Generated._ Participants generated a to- tal of 211 ideas, for an average of 3.52 ideas per participant (SD p 2.44). Noise level had a marginally significant effect on this measure (_F_(2, 57) p 2.41, _p_ p .10): those in the high-noise condition (_M_ p 2.5) generated fewer ideas than those in the low-noise (_M_ p 4.10; _t_(57) p 2.07, _p_ ! .05) and moderate-noise (_M_ p 3.82, _t_(57) p 1.74, _p_ p .09) conditions. The difference between the latter two conditions was not significant (_t_ ! 1).
+
+_Time Spent on Generating the Ideas._ To further assess the extent of processing, we next analyzed the time taken by participants to generate their ideas. One-way ANOVA revealed a significantmain effect of noise level (_F_(2, 59) p 3.70, _p_ ! .05). Participants in the high-noise condition (_M_ p 98.06) spent significantly less time on this task than both those in the low- (_M_ p 140.04; _t_(57) p 2.32, _p_ ! .05) and moderate-noise (_M_ p 141.24; _t_(57) p 2.44, _p_ ! .05) conditions. Again, the difference between thelattertwocon- ditions was not significant (_t_ ! 1).
+
+_Processing Motivation._ To test whether our noise ma- nipulation changed participants’ processing motivation, we averaged each participant’s responses to the threemotivation questions detailed above (a p .77). A one-way ANOVA revealed no significant treatment effect of noise level on participants’ processing motivation (*M*High p 4.22, *M*Moderate p 4.18, *M*Low p 4.02; _F_ ! 1).
+
+_Creativity of the Ideas Generated._ To assess the creativ- ity of participants’ ideas, we first identified all unique ideas generated. A total of 122 unique ideas were identified in the set of all 211 ideas. Next, 12 independent judges, hired from the same population as the study participants, rated how creative they thought each of the 122 unique ideas was on a 7-point scale (1 p not at all, 7 p very much; Dahl, Chattopadhyay, and Gorn 1999; Goldenberg et al. 1999). The judges were shown only the unique ideas, rather than all ideas, to control for frequency effects (i.e., more fre- quently presented ideas might be judged as more or less creative). Ratings from 12 judges were then averaged for each unique idea (a p .81) to obtain the average judge rating for that idea. These average ratings for all ideas gen- erated by each participant were then averaged (i.e., summed and then divided by the total number of ideas generated by that participant) to obtain a mean creativity score for each participant.
+
+Replicating the results from experiment 1, one-way ANOVA revealed a significant main effect of noise level
+
+on mean creativity score (_F_(2, 57) p 3.59, _p_ ! .05), such that participantsin themoderate-noisecondition(*M*p 4.01) generated ideas that were more creative than those generated in eitherthelow-noisecondition(*M*p 3.57;_t_(57)p 2.33, _p_ ! .05) or the high-noise condition (_M_ p 3.58; _t_(57) p 2.25, _p_ ! .05). No difference was observed between high- and low-noise conditions (all _t_ ! 1).
+
+_Construal Level._ We created a construal-level index (a p .75) by summing each participant’s responses to the 25 BIF items. One-way ANOVA with the construal-level index as the dependent variable revealed a significant main effect of noise level (_F_(2, 57) p 3.65, _p_ ! .05). Specifically, participants in the moderate-noise condition (_M_ p 42.32) were operating at a higher construal level than those in the low-noise condition (_M_ p 39.60; _t_(57) p 2.18, _p_ ! .05), and participants in the high-noise condition were also op- erating at a higher construal level (_M_ p 42.83) than those in the low-noise condition (_M_ p 39.60; _t_(57) p 2.47, _p_
+
+- .05). However, no difference in construal levels was ob- served between those in the moderate- and those in the high- noise condition (_t_ ! 1).
+
+  _Mediation Analyses._ Two sets of analyses were con- ducted in order to test whether (1) construal level mediates the beneficial effect of moderate (vs. low) levels of noise on creativity and (2) the reduced capacity of processing is responsible for the lower creativity observed in high (vs. moderate) levels of noise.
+
+  For the first analysis, following the procedure recom- mended by MacKinnon, Lockwood, and Williams (2004), we used the bootstrapping approach to assess the mediation effect. The 95% bias-corrected bootstrap confidence inter- vals (CIs) were obtained for each of the two contrasts (mod- erate- vs. low-noise conditions and moderate- vs. high-noise conditions) using 5,000 bootstrap samples. The results sup- port our proposition, demonstrating that the 95% confidence interval for the moderate-low noise contrast ( .42 to .02) did not include zero, which indicates that construal level indeed mediates the effect of moderate (vs. low) noise levels on creativity. However, the 95% CI obtained for moderate- high noise contrast ( .15 to .27) did include zero, which suggests that the indirect effect of construal level was absent for this contrast.
+
+  The second analysis examined whether the capacity of processing underlies the impaired creativity observedathigh (vs. moderate) levels of noise. The bootstrap approach was again used to test the mediation model. Time spent on the creative task was used as the measure of processing capacity, such that less time spent meant reduced capacity to process information. The 95% bias-corrected bootstrap CI was ob- tained using 5,000 bootstrap samples for the moderate-high noise contrast. The mediation was tested only for this con- trast, as the capacity-of-processing measure did not differ between the moderate- and low-noise conditions. As hy- pothesized, we observed a significant indirect effect of the processing capacity on the reduction in creativity from the
+
+  moderate to the high noise level; that is, a 95% bias-cor- rected CI did not include zero ( .51 to .06).
+
+  Discussion
+
+  Results from experiment 2 theoretically replicated those of experiment 1, that is, that a moderate level of noise leads to higher creativity than either a low or a high level of noise. In addition, experiment 2 provides evidence that construal level underlies the beneficial effect of a moderate (vs. low) level of noise on creativity. Although a high level of noise also leads to a higher construal level, relative to a low level of noise, comparable to the high construal level induced by a moderate level of noise, this positive influence on crea- tivity is counteracted by the reduced capacity of processing also induced by the high, but not by the moderate, level of noise. Thus, experiment 2 shows that the reduced processing capacity was responsible for the impaired creativity ob- served at high (vs. moderate) noise level. Finally, results on the measure of processing motivation confirmed our expec- tation that while a high (vs. moderate) level of noise led to reduced capacity of processing, it did not affect the moti- vation to process.
+
+  Up to this point, we have argued that a moderate level of noise induces processing disfluency, leading to abstract processing and thus to higher creativity. Yet an alternative argument, as speculated by Toplyn and Maguire (1991), is that a moderate level of noise induces a moderate level of arousal, thus enhancing creativity. In fact, it is plausible that a moderate (vs. low) level of noise may induce both pro- cessing disfluency and arousal. We argue, however, that if noise is present for a longer period, people become accus- tomed to it physiologically (i.e., their arousal level will nor- malize) but not cognitively (i.e., their level of distraction, and hence of processing disfluency, will remain high).Thus, if our theorizing is correct, we should observe that a mod- erate level of noise leads to greater creativity regardless of whether the task is administered at the beginning of the experiment or later on (i.e., whether noise has just begun or has been present for a while). If, on the otherhand, arousal underlies the effect, we should observe the beneficial effect of a moderate level of noise only at the beginning of the study. Our next experiment tests thesecompetinghypotheses and examines the role of processing disfluency in the noise- creativity relationship.
+
+**EXPERIMENT 3**
+
+Method
+
+_Stimuli and Design._ Experiment 3 used the same noise manipulation as before. However, as we were particularly interested in the process mechanism underlying the bene- ficial effect of a moderate (vs. low) level of noise on cre- ativity, we dropped the high-noise condition. The focal task asked participants to list as many creative uses of a brick as they could think of (Friedman and Forster 2001). Partic- ipants completed this task either shortly afterthebackground
+
+noise started to play or after a delay. Thus, this experiment used a 2 (noise level: low vs. moderate) # 2 (timing of task: immediate vs. delayed) between-subjects design. To assess arousal level, we took two physiological measures: heart rate and blood pressure.
+
+To assess processing disfluency, we measured the extent to which participants felt distracted during the study. Past research has measured processing fluency by simply asking participants about the difficulty they experienced in com- pleting a given task. However, most of this research has manipulated processing fluency by varying the difficulty level of the focal task, for example, by employing easy or difficult to read fonts in a reading task (Oppenheimer 2008; Song and Schwarz 2008). In contrast, because the focal task in experiment 3 involved generating unusual ideas, we de- liberately refrained from using the difficulty of completing the task as a measure of processing fluency. This is because, in our context, the perceived difficulty of the focal task could represent either processing disfluency induced by the noise manipulation or ease of retrieval in completing the idea- generation task (Tsai and McGill 2011). For example, a person who was able to generate more ideas might perceive the task as easy to complete; this is not the kind of noise- induced processing disfluency that we are trying to assess. To avoid such a potentially confounding measure, we mea- sured processing disfluency indirectly, by assessing the level of distraction, which has been shown to affect processing difficulty (Jacoby et al. 1988; Schwarz 2004). Specifically, we measured participants’ level of distraction via three items, each rated on a 7-point scale (1 p not at all, 7 p very much): (1) How distracting did you find the room ambience while completing the study? (2) How well were you able to concentrate while completing the study? (reverse coded); and (3) How comfortable was the experimental room to complete the study? (reverse coded). Measures of both arousal and processing disfluency were taken immediately after participants completed the focal task.
+
+_Procedure._ Ninety-five undergraduate students (60 wom- en) from the University of British Columbia participated in this experiment, one person at a time, in exchange for course credit. The study setup and noise manipulations were iden- tical to those described in experiment 1, except that only low (50 dB) or moderate (70 dB) levels of noise wereplayed. After the participant had settled down and the background noise had started, s/he first answered some demographic questions that took about 2 minutes. Upon completing the demographic questions, half the participants were presented with the brick task: they were told to generate as many creative uses for a brick as they could think of, but to refrain from listing both typical uses and uses that are virtually impossible. Following Friedman and Forster (2001), partic- ipants were given 2 minutes to generate their list. Once the participants completed the brick task, they answered the three disfluency questions. Next, the study administrator measured the participant’s heart rate and blood pressure. The sequenceof takingphysiologicalmeasuresandanswering
+
+disfluency questions for all the participants was counterbal- anced.
+
+The other half of the participants, upon finishing the de- mographic questions, worked on some unrelated tasks for about 12–15 minutes before doing the brick task. The un- related task included answering various but unrelated in- dividual difference scales. After completing the brick task, participants answered the disfluency questions and their physiological measures were taken, with the order of these two measures counterbalanced.
+
+Results
+
+_Number of Ideas Generated._ A total of 480 ideas were generated by all participants, for an average of 5.05 (SD p 2.50) per person. Neither the interaction between noise level and timing of the idea-generation task (_F_(1, 91) p 1.09, _p_ p .30) nor the main effects were found to be statistically significant (all _F_ ! 1).
+
+_Arousal Level._ Arousal level was assessed through non- invasive measures of participants’ heart rate and blood pres- sure. For the heart-rate measure, the main effects of noise level (_F_(1, 91) p 11.34, _p_ ! .01) and timing of task (_F_(1, 91) p 5.44, _p_ ! .05) were both significant, along with a significant two-way interaction (_F_(1, 91) p 5.31, _p_ ! .05). Contrast analysis revealed that in the moderate-noise con- dition, heart rate was significantly higher when taken shortly after the experiment began than when taken after a delay (_M_ p 78.12 vs. 69.5; _t_(91) p 3.38, _p_ ! .01). No such difference was observed in the low-noisecondition(*M*immediate p 67.60, *M*delay p 67.55; _t_ ! 1; see fig. 1*A*). Analysis of the other two contrasts revealed that when heart rate was measured shortly after the experiment began, it was signif- icantly higher in the moderate-noise than in the low-noise condition (_t_(91) p 4.14, _p_ ! .001). No such difference was observed when heart rate was measured later on in the experiment (_t_ ! 1).
+
+Similar results were observed for the blood-pressure mea- sure. ANOVA revealed significant main effects of noise level (_F_(1, 91) p 9.96, _p_ ! .01) and timing of task (_F_(1, 91) p 4.29, _p_ ! .05), which were qualified by a marginally significant two-way interaction (_F_(1, 91) p 3.16, _p_ p .08). Further contrast analysis indicated that in themoderate-noise condition, blood pressure was higher when taken shortly after the experiment began than when taken after a delay (_M_ p 115.92 vs. 107.40; _t_(91) p 2.81, _p_ ! .01). No such difference was observed in the low-noisecondition(_M_
+
+p 105.00, *M*delay p 104.35; _t_ ! 1; see fig. 1*B*). Examiinmmeatidiatone of the other set of contrasts revealed that when blood pres-
+
+sure was measured shortly after the experiment began, it was significantly higher in the moderate- than in the low- noise condition (_t_(91) p 3.60, _p_ ! .01); no such difference was observed when blood pressure was measured later on in the experiment (_t_ ! 1).
+
+_Processing Disfluency._ Each participant’s responses to the three questions assessing processing disfluency were av-
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY 791
+
+**FIGURE 1**
+
+EXPERIMENT 3: _A_, HEART RATE AS A FUNCTION OF NOISE LEVEL AND TIMING OF TASK. _B_, BLOOD PRESSURE AS A FUNCTION OF NOISE LEVEL AND TIMING OF TASK
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.006.jpeg)
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY
+
+eraged to create an index (a p .78). For this measure, as expected, the two-way ANOVA revealed only a significant main effect of noise level (_F_(1, 91) p 19.12, _p_ ! .001; see fig. 2). The two-way interaction (_F_(1, 91) p 1.46, _p_ p .23) and the main effect of task timing were nonsignificant (all _F_ ! 1). We therefore collapsed the data across the timing- of-task variable and ran a one-way ANOVA. As hypothe- sized, we found that the moderate (vs. low) level of noise led to greater processing disfluency (_M_ p 5.37 vs. 4.29; _F_(1, 93) p 20.26, _p_ ! .001).
+
+_Creativity of the Ideas Generated._ To assess the creativ- ity of the ideas generated by participants, we used the same procedure described in experiment 2. We first screened all 480 ideas and identified 198 unique ideas. Next, we hired 12 judges from the same population as our study participants and asked them to rate the creativity of each unique idea on a 7-point scale (1 p not at all, 7 p very much). We then used these ratings to create a mean creativity score for each participant, as detailed in experiment 2. A two-way ANOVA revealed only a significant main effect of noise level (_F_(1, 91) p 13.8, _p_ ! .001; see fig. 3). We therefore collapsed the data across the timing-of-task variable and ran a one-way ANOVA. We found that the moderate (vs. low) level of noise prompted participants to generate ideas that were rated as more creative (_M_ p 4.70 vs. 4.16; _F_(1, 93) p 14.48, _p_ ! .001).
+
+Discussion
+
+Results from this experiment show that a moderate (vs. low) level of noise induces both higher arousal (as indicated by higher heart rate and blood pressure) and processing disfluency. With the passage of time, however, people seem to become physiologically accustomed to the moderate noise level (i.e., their arousal level normalizes). On the other hand,
+
+the high processing disfluency level induced by moderate noise appeared to persist over the course of the experiment. Importantly, we found that a moderate level of noise en- hances creativity regardless of the timing of the task, which suggests that processing disfluency, as opposed to arousal, drives this effect. To further test this proposition, we ran a mediation test using the bootstrap approach, with noise level as independent variable, mean creativity score as dependent variable, and disfluency as the mediator in the model. We obtained a 99% CI of ( 4.06 to .014), indicating that disfluency did indeed mediate the noise-creativity relation- ship.
+
+Up to this point, we have shown that both processing dis-
+
+**FIGURE 2**
+
+EXPERIMENT 3: PROCESSING DISFLUENCY AS A FUNCTION OF NOISE LEVEL AND TIMING OF TASK
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.007.jpeg)
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY
+
+**FIGURE 3**
+
+EXPERIMENT 3: CREATIVITY MEASURE AS A FUNCTION OF NOISE LEVEL AND TIMING OF TASK
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.008.jpeg)
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY 793
+
+fluency and construal level independently mediate the noise- creativity relationship. However, as theorized earlier,wepro- pose that processing disfluency and construal level should simultaneously mediate the relationship between ambient noise and creativity, such that a moderate (vs. low) level of noise induces higher processing disfluency, which further prompts a high construal level, thus leading to higher cre- ativity. We test this chain of process mechanism in the next experiment.
+
+Furthermore, all our previous experiments have examined creativity as a one-dimensional construct. However, extant research on creativity suggests that creativity may be treated as a multidimensional concept with two main components: originality and appropriateness (Burroughs et al. 2008; Mo- reau and Dahl 2005). To be creative, an idea must be dif- ferent from what is already known (the originality dimen- sion) and must also be appropriate in solving the problem at hand (the appropriateness dimension). In other words, an original but bizarre idea is not a creative idea (Lubart 1994). In experiment 4, therefore, we examined both the originality and the appropriateness dimensions of creativity.
+
+**EXPERIMENT 4**
+
+Method
+
+_Stimuli._ Experiment 4 used the same noise manipulation described in experiment 3. We used theshoe-polishproblem- solving task from Burroughs and Mick (2004), as the focal task in this experiment. Participants were told to imagine the following scenario:
+
+You are going out to a banquet held by your new employer
+
+and will probably be called up front and introduced to the rest of the company. You put on a black outfit and are all ready to leave for the dinner when you realize that your shoes are all scuffed up and the scuffs are definitely noticeable. You have completely run out of polish and these shoes are the only ones that can go with your outfit, and there is really no other outfit you can wear. You have to leave in the next 2 minutes if you want to be on time. All the stores in your part of the town are closed for the evening. Although there is one shopping mall that is still open, it would mean an extra 5 miles of freeway driving.
+
+After reading the scenario, participants were asked to gen- erate as many solutions as they could think of for the given problem. To assess the underlying process, we measured construal level using the BIF scale, as in experiment 2, and processing disfluency using the same three items as in Ex- periment 3.
+
+_Procedure._ Forty-two undergraduate students (27 wom- en) at the University of British Columbia participated in this experiment in exchange for $10. The experiment was run in small groups of no more than four people per session. The study setup and noise manipulation remained identical to experiments 1–3. We presented participantswiththeshoe- polish problem (Burroughs and Mick 2004) after they had settled down and the background noise had started. Partic- ipants were asked to generate as many solutions to the prob- lem as they could think of; once they had finished generating solutions, they completed the 25 BIF items (Vallacher and Wegner 1987) and the processing-disfluency measures.
+
+Results
+
+_Number of Ideas Generated._ A total of 188 ideas were generated, for an average of 4.48 (SD p 2.09) ideas per person. The noise level did not affect thenumberofsolutions generated (_M_ p 4.30, _M_ p 4.64; _F_ ! 1), nor was
+
+any differenceModerateobserved in the Lowamount of time (in seconds) taken to complete the focal task (*M*Moderate p 251.61, *M*Low
+
+p 237.73; _F_ ! 1).
+
+_Originality of the Ideas Generated._ We first screened all 188 ideas and identified 61 unique ideas, then hired 12 judges from the same population as our study participants. These 12 judges rated the originality, novelty, and inno- vativeness of each of the 61 unique ideas on a 7-point scale (1 p not at all, 7 p very much). We then averaged the 12 judges’ ratings to obtain the mean judges’ originality score (a p .62), mean judges’ novelty score (a p .67), and mean judges’ innovativeness score (a p .66) for each unique idea. These scores were used to calculate the mean originality, novelty, and innovativeness score for each participant. For example, to obtain the mean novelty score for a participant, we summed the mean judges’ novelty scores for each idea generated by that participant, then divided this sum by the total number of ideas generated by that person. The mean originality, novelty,andinnovativenessscoresloadedonone factor with high reliability (a p .95) and werethenaveraged to create an overall originality index.
+
+One-way ANOVA revealed a significant effect of noise on this originality index, such that ideas generated by par- ticipants in the moderate- (vs. low-) noise condition were rated as more original (_M_ p 3.87 vs. 3.66; _F_(1, 40) p 4.76, _p_ ! .05).
+
+_Appropriateness of Ideas._ Another set of 14 judges rated each of the 61 unique ideas on its appropriateness (a p .82), usefulness (a p .81), and practicality (a p .80); these ratings were then used to create an overall appropriateness index (a p .98), using the same procedure described above. One-way ANOVA also revealed a significant effect of noise on this appropriateness index, such that ideas generated by respondents in the moderate-noise (vs. low-noise) condition were rated as more appropriate (_M_ p 4.48 vs. 4.20; _F_(1, 40) p 5.34, _p_ ! .05).
+
+_Processing Disfluency and Construal Level._ Participants’ responses to the three processing-disfluency questions were averaged to create a processing-disfluencyindex (a p .74). As expected, one-way ANOVA revealed a main effect of noise on processing disfluency, such that moderate noise (_M_ p 4.02) led to higher processing disfluency than low noise (_M_ p 3.12; _F_(1, 40) p 9.16, _p_ ! .01). Similarly, partici- pants’ responses to the 25 construal-level items were summed to create a construal-level index (a p .85). A significant effect of noise level was also observed for this index, such that a moderate noise level induced a higher construal level (_M_ p 4.02) relative to a low noise level (_M_ p 39.41; _F_(1, 40) p 8.46, _p_ ! .01).
+
+**FIGURE 4**
+
+EXPERIMENT 4: MULTISTEP MULTIPLE-MEDIATION MODEL FOR THE ORIGINALITY INDEX
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.009.jpeg)
+
+NOTE.—Coefficients are unstandardized and _t_-statistics are in parentheses. \*\* _p_ ! .01.
+
+_Multiple Mediation._ Next, we conducted mediation anal- ysis to examine whether processing disfluency and construal level simultaneously mediate the noise-creativity relation- ship. Because we hypothesized a causalrelationshipbetween processing disfluency and construal level, we conducted a test of multiple mediation using the multiple-step multiple- mediator model (Hayes, Preacher, and Myers 2011; Preacher and Hayes 2008). Using this model, we tested for the pres- ence of a multiple-mediation effect, such that a moderate (vs. low) noise level induces higher processing disfluency, leading to a higher construal level and consequently en- hancing both the originality and the appropriateness dimen- sions of creativity.
+
+We conducted our first set of multiple-mediation analyses to test for a mediating effect of processing disfluency and construal level on the relationship between noise level and originality of ideas generated. We therefore included noise, the processing-disfluency index, the construal-level index, and the mean originality score in the model. A 5,000-re- samples bootstrap approach generated a 95% CI of (.010 to .161) for the multiple mediators’ indirect effect, indicating a significant multiple-mediation effect at the _p_ ! .05 level. Analysis of individual paths in the model provided further interesting information about the multiple-mediation effect. A separately run individual set of regressions indicated sig- nificant direct effects of noise on the originality of ideas (b p .21, _t_ p 2.18, _p_ ! .05), processing disfluency (b p .90, _t_ p 3.03, _p_ ! .01), and construal level (b p 4.34, _t_ p 2.91, _p_ ! .01). When both processing disfluency and con- strual level were included in the multistep multiple-mediator model, however, only the individual paths from noise to processing disfluency (b p .90, _t_ p 3.03, _p_ ! .01), from processing disfluency to construal level (b p 2.48, _t_ p 3.53, _p_ ! .01), and from construal level to originality (b p .03, _t_ p 3.01, _p_ ! .01) remained significant, while all other paths became nonsignificant. This result confirms that the moderate level of noise induced higher processing dis- fluency, which then induced a higher construal level, leading to increasingly original ideas (see fig. 4).
+
+Next, we ran the same multistep multiple-mediator model as described above but replaced the originality index with
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY
+
+the appropriateness index. A 5,000-resamples bootstrap ap- **FIGURE 5**
+
+proach generated a 95% CI of (.004 to .173) for the multiple
+
+mediators’ indirect effect, indicating a significant multiple- EXPERIMENT 4: MULTISTEP MULTIPLE-MEDIATION MODEL mediation effect at the _p_ ! .05 level. Again, further analysis FOR THE APPROPRIATENESS INDEX
+
+of individual paths in the model demonstrated that although
+
+the direct effect of noise on appropriateness was significant ![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.010.jpeg)initially (b p .29, _t_ p 2.31, _p_ ! .05), when both processing disfluency and construal level were included in the multiple- mediation model only the individual paths from noise to processing disfluency (b p .90, _t_ p 3.03, _p_ ! .01), from processing disfluency to construal level (b p 2.48, _t_ p 3.53, _p_ ! .01), and from construal level to appropriateness
+
+(b p .04, _t_ p 2.78, _p_ ! .05) remained significant, while NOTE.—Coefficients are unstandardized and _t_-statistics are in all other paths became nonsignificant. Thus, the mediation parentheses. \*\* _p_ ! .01.
+
+confirms that a moderate level noise led to higher processing
+
+disfluency, which then induced a higher construal level, thus
+
+producing more appropriate ideas (see fig. 5).
+
+hidden from the participants, was switched on before each participant began the experimental task and paused as soon
+
+Discussion as the participant left. The settings of the sound-level meter Results from this experiment provide crucial support for gave us an average noise level for the period during which
+
+our theory by demonstrating the chain of underlying pro- it was switched on.
+
+cesses through which noise affects creativity. Specifically, _Stimuli._ We constructed eight pairs of different products we demonstrate that moderate (vs. low) levels of noise in- for this experiment. Each product pair offered two options duce higher processing disfluency, which induces a higher from the same product category, one of which was new and construal level and abstract processing, and consequently innovative while other was more traditional (see appendix enhances both the originality and the appropriateness di- fig. A1 for an example of such a pair). Full-color pictures, mensions of creativity. along with some product information for the two options in
+
+Our next and final experiment aims to extend earlier stud- each pair, were presented together on the same screen. Par- ies in three ways. First, although in previous experiments ticipants indicated their likelihood of buying the innovative we manipulated noise to a high (85 db), moderate (70 dB), option over the traditional one on a 7-point scale (1 p not or low (50 dB) level, we recognize that in real lifeconsumers at all, 7 p very much).
+
+encounter a wide range of noise intensities, from low to Because the study involved innovative product adoption, high. In our final experiment, therefore, we operationalized individual differences in creativity were measured using a the noise factor as a continuous variable by measuring it in user innovativeness scale (Price and Ridgway 1983). This a natural setting. Second, we employed innovation adoption scale measures individuals’ tendency to use products crea- as the focal task in this study, as our theorizing suggests tively to solve problems and includes items such as “I enjoy that noise should affect not only creative production but also thinking of new ways to use old things around the house” adoption of innovative products. Finally, we explored how and “I take great pleasure in adapting products to new uses individuals’ baseline level of innovativenessmightmoderate that the manufacturer never intended.” Participants com- the effect of noise on innovation adoption. pleted this measure, along with some other individual dif-
+
+ference measures, at the beginning of the term. Their re-
+
+**EXPERIMENT 5** sponses to this user innovativeness scale were later used for
+
+experiment 5 data analysis.
+
+Method
+
+_Procedure._ Sixty-eight undergraduate students (44 wom- In this experiment we aimed to study the effect of noise en) at the University of British Columbia participated in this on innovation adoption in a real-life setting. We conducted experiment, one at a time. The sessions were run throughout the experiment in a student lounge area equipped with var- the day, every day for 5 days. Upon arrival, the participant ious appliances (microwave, fridge, water cooler, coffee ma- was asked to sit in front of a computer inside the cubicle. chine, and oven). This lounge is used by graduate students, Once the participant was ready to begin, the research as- and the noise level varies through the day. Typically it is sistant turned on the hidden sound-level meter and asked quite noisy during the lunch hour, moderately noisy during the participant to complete the survey at his/her own pace. coffee breaks, and rather quiet for the rest of the day. We The focal task involved presenting the participant with eight set up a cubicle with a desk and computer in a corner of pairs of traditional-innovative products one at a time. For this lounge, such that the study participant in the cubicle each pair, the participant rated his/her likelihood of buying could hear noise but could not see movement in the lounge. the innovative product as compared to the traditional option, A sound-level meter placed near the desk in this cubicle, as described above. Once the participant completed the task,
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY 795
+
+the research assistant paused the sound meter and noted the noise level for that session.
+
+Results
+
+Average noise levels across all sessions ranged from 38.40 dB to 71.50 dB, with an overall average of 51.35 dB (SD p 7.59 dB). The average noise level for each session was treated as a continuous variable in the subsequent anal- yses. Note that the highest noise participants were exposed to in this real-life setting corresponds to the moderate-noise condition in experiments 1–4. Thus, based on our theorizing and findings so far, we expected a positive linear relationship between noise and willingness to buy innovative products.
+
+_Buying Likelihood._ Each participant’s buying-likelihood scores for the eight product categories were averaged to create a buying-likelihood index; higher scores indicated a greater likelihood of adopting innovation. A linear regres- sion with the buying-likelihood index as the dependent var- iable and the continuous noise measure as the independent variable revealed a significant positive coefficient for the noise variable (b p .25, _t_(66) p 2.08, _p_ ! .05). As the noise level increased, respondents indicated a higher like- lihood of buying the innovative products (buying likelihood at 1 SD of mean noise level p 4.61; buying likelihood at 1 SD of mean noise level p 4.15).
+
+_Individual Difference in Creativity._ Of the 68 partici- pants who completed this experiment, we were able tomatch user-innovativeness scale responses for 62 people. There- fore, we used the data for only these 62 participants to examine whether individual differences in creativity mod- erate the effect of noise on innovation adoption. All data were analyzed in accordance with theAiken andWest(1991) approach. We regressed buying likelihood on noise, mean- centered user-innovativeness scale, and the interaction term. A marginally significant two-way interaction emerged for the measure of participants’ buying likelihood (b p .05, _t_ p 1.77, _p_ p .08). Next, we plotted the graphs at one standard deviation above and below the mean of the user- innovativeness scale (see fig. 6). Consistent with prior re- search (Toplyn and Maguire 1991), we found that as noise level increased, buying likelihood for the innovative prod- ucts increased only for highly creative people (i.e., at 1 SD on the user-innovativeness scale; _M_ 1 SD noise level p 4.08, _M_ 1 SD noise level p 5.09; b p .06, _t_ p 3.15, _p_ ! .01). For less creative people (i.e., at 1 SD on the user-innovative- ness scale), no difference was observed as noise level in- creased (_M_ p 4.13, _M_ p 4.33; b p
+
+.01, _t_ ! 1). 1 SD noise level 1 SD noise level
+
+Discussion
+
+Results from this experiment provideconvergentevidence for our theory. They support our hypothesis that moderate levels of noise not only lead to higher creative output but also enhance people’s adoption of innovative products. These results also support previous findings in the literature
+
+**FIGURE 6**
+
+EXPERIMENT 5: BUYING LIKELIHOOD AS A FUNCTION OF NOISE LEVEL AND INDIVIDUAL DIFFERENCES IN USER INNOVATIVENESS
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.011.jpeg)
+
+to the effect that increasing noise to a moderate level helps highly creative people to be more creative but may not be of value for people whose baseline creativity level is low (Amabile 1983).
+
+**GENERAL DISCUSSION**
+
+While ambient noise is omnipresent, our understanding of its impact on human cognition, particularly creative cog- nition, remains limited. In this study, through a series of five experiments, we demonstrate how and why ambient background noise can affect creativity.Specifically,weshow that a moderate (vs. low) level of ambient noise induces processing disfluency, which leads to abstract cognition and consequently enhances creativity. A high level of noise, however, impairs creativity by reducing the extent of in- formation processing.
+
+Findings from this research make several theoretical con- tributions. First, they contribute to the noise literature by providing valuable insights into the noise-creativity rela- tionship. Previous research has reported inconclusive find- ings with respect to the effect of noise on creativity: while the majority of prior studies suggest that high noise levels hurt creativity, some have found that moderate noise can enhance creativity. In addition, prior research has primarily employed noise stimuli rarely found in consumer environ- ments (e.g., white noise, pink noise). Our study helps to reconcile the mixed findings in the extant literature by dem- onstrating an inverted-U relationship between noise level and creativity. Using background noise that is commonly
+
+present in consumers’ lives (in this case, ambient noise in a roadside restaurant), we show that while a moderate level of background noise enhances creativity relative to a low noise level, a rather high level of noise impairs creativity.
+
+Second, we uncover the process through which ambient noise affects creative cognition. We find that increasing lev- els of noise induce distraction, leading to a higher construal level. That is, both moderate and high noise levels lead to more abstract processing as compared to a low noise level. This higher construal level then induces greater creativity in the moderate-noise condition; however, the very high level of distraction induced by the high-noise condition, although it prompts a higher construal level, also causes reduced information processing, thus impairing creativity. In other words, while a moderate level of noise produces just enough distraction to induce disfluency, leading to higher creativity, a very high level of noise induces too much dis- traction so as to actually reduce the amount of processing, leading to lower creativity.
+
+As discussed above, a clear understanding of how noise affects creativity is lacking in the extant literature. Different scholars conjecture different mechanisms (e.g., arousal, stress, attention) but do not provide rigorous empirical evi- dence. For example, Toplyn and Maguire (1991) speculate that a moderate level of background noise induces higher arousal and that this enhances creativity. In our research, however, arousal does not appear to be the driving force underlying the effect of noise on creativity. Consistent with our findings, other researchers have also documented null effects of arousal on creativity. For example, Van den Bergh et al. (2008) demonstrate that the activation of the reward circuitry, not the arousal induced, by exposure to sex cues enhances performance on the RAT task (i.e., a creative task). Although arousal appears to be an intuitive explanation for the effect of noise on creativity, it was not supported by our findings. We believe further research is needed to examine whether in other contexts, such as with different types of noise and among various segments of consumers, arousal might play a role in affecting creativity.
+
+Another interesting finding from our experiment 5 was that the main effect (i.e., a moderate level of noise enhances creativity) was present only among highly creative people. Although these results are in line with prior findings (Toplyn and Maguire 1991), they merit further attention. A logical question that arises, given this proposition, is why we ob- served similar results for the general population of our par- ticipants in experiments 1–4. While a dedicated inquiry is needed to fully address this question, we lay out a possible explanation for this observation. Amabile (1983) suggests that an individual will be incapable of producing work that is considered creative if creativity-relevantskillsarelacking. Thus, a person must have certain basic skills before his/her creativity can be enhanced through subtle manipulations such as background noise. This proposition is supported by our results from experiment 5. With respect to the presence of our main effect for the general participant population through experiments 1–4, we draw support from Alba (2000),
+
+who points out that college students, who are regularly used as research participants, are preselected on the basis of their cognitive skills. These individuals, on an average, arebound to have above-average innate competence or creativity-rel- evant skills. In fact, the user-innovativeness measure as ob- tained in experiment 5 supports this argument, in that the average score of all participants was significantly above the midpoint of the scale (_M_ p 4.20, _t_(61) p 2.43, _p_ ! .05). Hence, it may not be surprising that the manipulation of background noise affected creativity among our overall par- ticipant population.
+
+Finally, this research also contributes theoretically to the literature on creativity and innovation adoption. We docu- ment that ambient noise, an incidental environmental cue, is an important antecedent of creative cognition. A moderate level of noise not only enhances creative production but also leads to greater adoption of innovative products.
+
+In addition to the preceding theoretical contributions of our study, valuable practical implications also follow for both marketers, who typically strive to increase adoption rates of new and innovative products, and consumers, who look for creative solutions to their everyday problems. For example, in order to encourage adoption of new and in- novative products, marketers might consider equipping their showrooms with a moderate level of ambient noise. For individuals looking for creative solutions to daily problems, such as planning a dinner menu based on limited supplies or generating interesting research topics to study, our find- ings imply that instead of burying oneself in a quiet room trying to figure out a solution, walking out of one’s comfort zone and getting into a relatively noisy environment (such as a cafe´) may trigger the brain to think abstractly, and thus generate creative ideas.
+
+While our findings are intriguing, they also offer avenues for future research. First, future research might investigate whether different types of noise will produce similar effects on creativity. For example, does the valence of noise, in ad- dition to its decibel level, influence creativity? The findings from our work confirm that the disfluency or distraction in- duced by multi-talker noise in the background can enhance creativity. However, what about more pleasant types of noise in the background (e.g., serene music)? Will they affect cre- ativity? And, if so, in what direction? It may be possible that pleasant noise will actually increase processing fluency and thus prompt more concrete processing, consequently hurting creativity.
+
+A second avenue for future research would be to examine the effect of ambient noise on different types of creative tasks. Although we focused on creative tasks of the problem- solving type, our theory can be extended to open-ended or divergent creative tasks, such as art and music.
+
+Third, future research can investigate how background noise might affect consumers’ assessment of neutral ideas or products. While we show that a moderate level of noise enhances adoption of innovative products, it seems plausible that even a seemingly ordinary idea might be assessed as
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY 797
+
+more creative/innovative in the context of a moderate (vs. creative cognition in a similar manner as noise. What about low) noise level. whether the conversation takes place in our native language or
+
+Finally, future research can examine whether other types of in a foreign language? We hope that our research willstimulate distracting variables, such as background conversations, affect further investigation in this fascinating domain.
+
+**APPENDIX**
+
+**TABLE A1**
+
+EXAMPLES OF SOUND SOURCES AND INTENSITY
+
+| Sound source                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            | Intensity (dB)                                                                                                                                  |
+| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------- |
+| <p>Rocket launch equipment acoustic tests</p><p>Threshold of pain</p><p>Hearing damage during short-term effect</p><p>Jet engine, 100 m distant</p><p>Jackhammer, 1 m distant/Discotheque</p><p>Hearing damage from long-term exposure</p><p>_High noise condition in present studies_</p><p>Traffic noise on major road, 10 m distant</p><p>_Moderate noise condition in present studies_</p><p>Moving automobile, 10 m distant</p><p>TV set at typical home level, 1 m distant</p><p>_Low noise condition in present studies_</p><p>Normal talking, 1 m distant</p><p>Very calm room</p><p>Quiet rustling leaves, calm human breathing</p><p>Auditory threshold at 2 kHz for undamaged human ears</p> | <p>Approx. 165 134 Approx. 120 110–40 Approx. 100 Approx. 85 _85_</p><p>80–90</p><p>_70_ 60–80 Approx. 60 _50_</p><p>40–60 20–30 10</p><p>0</p> |
+
+**FIGURE A1**
+
+SAMPLE PRODUCT PAIR USED IN EXPERIMENT 4
+
+![](/img/user/_resources/Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.012.jpeg)
+
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+THE EFFECTS OF NOISE ON CREATIVITY 799
+
+**REFERENCES**
+
+Aiken, LeonaS., and StephenG.West(1991),_MultipleRegression_,
+
+Newbury Park, CA: Sage.
+
+Alba, Joseph W. (2000), “Presidential Address: Dimensions of Consumer Expertise or Lack Thereof,” in _Advances in Con- sumer Research_, Vol. 27, ed. Stephen J. Hoch and Robert J. Meyer, Provo, UT: Association for Consumer Research, 1–9. Alter, Adam L., and Daniel M. Oppenheimer (2008), “Effects of
+
+Fluency on Psychological Distance and Mental Construal (or
+
+Why New York Is a Large City, but New York Is a Civilized
+
+Jungle),” _Psychological Science_, 19 (February), 161–67.
+Amabile, Teresa M. (1983), _The Social Psychology of Creativity_,
+
+New York: Springer.
+
+Burroughs, James E., and David Glen Mick (2004), “Exploring
+
+Antecedents and Consequences of Consumer Creativity in a Problem-Solving Context,” _Journal of Consumer Research_, 31 (September), 402–11.
+
+Burroughs, James E., Page C. Moreau, and David Glen Mick
+
+(2008), “Toward a Psychology of Consumer Creativity,” in _Handbook of Consumer Psychology_, ed. Curtis P. Haugtvedt, Paul M. Herr, and Frank R. Kardes, New York: Erlbaum, 1011–38.
+
+Dahl, Darren W., Amitava Chattopadhyay, and Gerald J. Gorn
+
+(1999), “The Use of Visual Mental Imagery in New Product Design,” _Journal of Marketing Research_, 36 (February), 18–28.
+
+Dahl, Darren W., and Page Moreau (2002), “The Influence and
+
+Value of Analogical Thinking during New Product Ideation,”
+
+_Journal of Marketing Research_, 39 (February), 47–60.
+——— (2007), “Thinking inside the Box: Why Consumers Enjoy
+
+Constrained Creative Experiences,” _Journal of Marketing Re-_
+
+_search_, 44 (August), 357–69.
+
+Fink, Andreas, Roland H. Grabner, Daniela Gebauer, Gernot
+
+Reishofer, Karl Koschutnig, and Franz Ebner (2010), “En-
+
+hancing Creativity by Means of Cognitive Stimulation: Evi-
+
+dence from an fMRI Study,” _NeuroImage_, 52 (4), 1687–95. Forster, Jens, Ronald S. Friedman, and Nira Liberman (2004),
+
+“Temporal Construal Effects on Abstract and ConcreteThink-
+
+ing: Consequences for Insight and Creative Cognition,” _Jour-_
+
+_nal of Personality and Social Psychology_, 87 (August),
+
+177–89.
+
+Friedman, Ronald S., and Jens Forster (2001), “The Effects of
+
+Promotion and Prevention Cues on Creativity,” _Journal of_
+
+_Personality and Social Psychology_, 81 (December), 1001–13. Goldenberg, Jacob, David Mazursky, and Sorin Solomon (1999),
+
+“Toward Identifying the Inventive Templates of New Prod-
+
+ucts: A Channeled Ideation Approach,” _Journal of Marketing_
+
+_Research_, 36 (May), 200–210.
+
+Griskevicius, Vladas, Robert B. Cialdini, and Douglas T. Kenrick
+
+(2006), “Peacocks, Picasso, and Parental Investment: The Ef- fects of Romantic Motives on Creativity,” _Journal of Per- sonality and Social Psychology_, 91 (July), 63–76.
+
+——— (2007), “The Muse Effect: When RomanticMotivesCreate
+
+Creativity,” in _Advances in Consumer Research_, Vol. 34, ed. Gavan Fitzsimons and Vicki Morwitz, Duluth, MN: Asso- ciation for Consumer Research, 15.
+
+Hamilton, P., and A. Copeman (1970), “The Effect of Alcohol and
+
+Noise on Components of a Tracking and Monitoring Task,” _British Journal of Psychology_, 61 (May), 149–56.
+
+Hayes, Andrew F., Kristopher J. Preacher, and Teresa A. Myers
+
+(2011), “Mediation and the Estimation of Indirect Effects in Political Communication Research,” in _Sourcebook for Po- litical Communication Research: Methods, Measures, and Analytical Techniques_, ed. ErikP.BucyandR.LanceHolbert, New York: Routledge, 434–65.
+
+Hillier, Ashleigh, Jessica K. Alexander, and David Q. Beversdorf
+
+(2006), “The Effect of Auditory Stressors on Cognitive Flex- ibility,” _Neurocase_, 12 (4), 228–31.
+
+Hirschman, Elizabeth C. (1980), “Innovativeness, Novelty Seek-
+
+ing, and Consumer Creativity,” _Journal of Consumer Re- search_, 7 (December), 283–95.
+
+Hockey, G. R. J. (1969), “Noise and Efficiency:The Visual Task,”
+
+_New Scientist_, 1 (May), 244–46.
+
+——— (1970a), “Changes in Attention Allocation in a Multi-
+
+component Task under Loss of Sleep,” _British Journal of Psychology,_ 61 (November), 473–80.
+
+——— (1970b), “The Effect of Loud Noise on Attentional Se-
+
+lectivity,” _Quarterly Journal of Experimental Psychology_, 22 (1), 28–36.
+
+Houston, John P., and Sarnoff A. Mednick (1963), “Creativity and
+
+the Need for Novelty,” _Journal of Abnormal and Social Psy- chology_, 66 (2), 137–41.
+
+Hygge, Staffan, Gary W. Evans, and Monika Bullinger (2002), “A
+
+Prospective Study of Some Effects of Aircraft Noise on Cog- nitive Performance in Schoolchildren,” _Psychological Sci- ence_, 13 (September), 469–74.
+
+Im, Subin, Barry L. Bayus, and Charlotte H. Mason (2003), “An
+
+Empirical Study of Innate ConsumerInnovativeness, Personal Characteristics, and New-Product Adoption Behavior,” _Jour- nal of the Academy of Marketing Science_, 31 (December), 61–73.
+
+Jacoby, Larry L., Lorraine G. Allan, Jane C. Collins, and Linda
+
+K. Larwill (1988), “Memory Influences Subjective Experi- ence: Noise Judgments,” _Journal of Experimental Psychol- ogy: Learning, Memory, and Cognition_, 14 (2), 240–47.
+
+Kasof, Joseph (1997), “Creativity and Breadth of Attention,” _Cre-_
+
+_ativity Research Journal_, 10 (4), 303–15.
+
+Lubart, Todd I. (1994), “Creativity,” in _Handbook of Perception_
+
+_and Cognition: Thinking and Problem Solving_, ed. Robert J.
+
+Sternberg, New York: Academic Press, 289–332. MacKinnon, David P., Chondra M. Lockwood, and Jason Williams
+
+(2004). “Confidence Limits for the Indirect Effect: Distri-
+
+bution of the Product and Resampling Methods,”_Multivariate_
+
+_Behavioral Research_, 39 (January), 99–128.
+
+Maddux, William W., and Adam D. Galinsky (2009), “Cultural
+
+Borders and Mental Barriers: The Relationship between Liv- ing Abroad and Creativity,” _Journal of Personality and Social Psychology_, 96 (May), 1047–61.
+
+Martindale, Colin, and James Greenough (1973), “The Differential
+
+Effect of Increased Arousal on Creative and Intellectual Per-
+
+formance,” _Journal of Genetic Psychology_, 123 (December),
+
+329–35. Mednick,SarnoffA.(1962),“TheAssociativeBasisoftheCreative
+
+Process,” _Psychological Review_, 69 (3), 220–32.
+
+Mehta, Ravi, and Rui (Juliet) Zhu (2009), “Blue or Red? Exploring
+
+the Effect of Color on Cognitive Task Performances,” _Science_, 323 (5918), 1226–29.
+
+Meyers-Levy, Joan, and Rui (Juliet) Zhu (2007), “The Influence
+
+of Ceiling Height: The Effect of Priming on the Type of Processing That People Use,” _Journal of Consumer Research_, 34 (August), 174–86.
+
+Moreau, Page C., and Darren W. Dahl (2005), “Designing the
+
+Solution: The Impact of Constraints on Consumers’ Creativ-
+
+ity,” _Journal of Consumer Research_, 32 (June), 13–22.
+Nagar, Dinesh, and Janak Pandey (1987), “Affect and Performance
+
+on Cognitive Task as a Function of Crowding and Noise,”
+
+_Journal of Applied Social Psychology_, 17 (February), 147–
+
+57\.
+
+O’Malley, J. J., and A. Poplawsky (1971), “Noise-Induced Arousal
+
+and Breadth of Attention,” _Perceptual and Motor Skills_, 33 (3), 887–90.
+
+Oppenheimer, Daniel M. (2008), “The Secret Life of Fluency,”
+
+_Trends in Cognitive Science_, 12 (June), 237–41.
+
+Preacher, Kristopher J., and Andrew F. Hayes (2008), “Asymptotic
+
+and Resampling Strategies for Assessing and Comparing In- direct Effects in Multiple Mediator Models,” _Behavior Re- search Methods_, 40 (August), 879–91.
+
+Price, Linda L., and Nancy M. Ridgway (1983), “Development of a Scale to Measure Use Innovativeness,” in _Advances in Con- sumer Research_, Vol. 10, ed. Richard P. Bagozzi and Alice M. Tybout, Ann Arbor, MI: Association for Consumer Re- search, 679–84.
+
+Schwarz, Norbert (2004), “Metacognitive Experiences in Con-
+
+sumer Judgment and Decision Making,” _Journal of Consumer Psychology_, 14 (4), 332–48.
+
+Smith, Steven M. (1995), “Fixation, Incubation and Insight in Memory and Creative Thinking,” in _The Creative Cognition Approach_, ed. Steven M. Smith, Thomas B. Ward,andRonald A. Finke, Cambridge, MA: MIT Press, 135–56.
+
+Smith,StevenM.,ThomasB.Ward,andJayS.Schumacher(1993),
+
+“Constraining Effects of Examples in a Creative Generation
+
+Task,” _Memory and Cognition_, 21 (November), 837–45.
+Song, Hyunjin, and Norbert Schwarz (2008), “If It’s Hard to Read,
+
+It’s Hard to Do: Processing Fluency Affects Effort Prediction
+
+and Motivation,” _Psychological Science_, 19 (October),
+
+986–88.
+
+Thompson, Debora V., Rebecca W. Hamilton, and Roland Rust
+
+(2005), “Feature Fatigue: When Product Capabilities Become Too Much of a Good Thing,” _Journal of Marketing Research_, 42 (November), 431–42.
+
+Toplyn, Glenn, and William Maguire (1991), “The Differential
+
+Effect of Noise on Creative Task Performance,” _Creativity Research Journal_, 4 (4), 337–47.
+
+Tsai, Claire I., and Ann L. McGill (2011), “No Pain, No Gain?
+
+How Fluency and Construal Level Affect Consumer Confi- dence,” _Journal of Consumer Research_, 37 (February), 807–21.
+
+Vallacher, Robin R., and Daniel M. Wegner (1987), “What Do
+
+People Think They’re Doing? Action Identification and Hu-
+
+man Behavior,” _Psychological Review_, 94 (January), 3–15. Van den Bergh, Bram, Siegfried Dewitte, and Luk Warlop (2008),
+
+“Bikinis Instigate Generalized Impatience in Intertemporal
+
+Choice,” _Journal of Consumer Research_, 35 (June), 85–97. Weinstein, Neil D. (1974), “Effect of Noise on Intellectual Per-
+
+formance,” _Journal of Applied Psychology_, 59 (October),
+
+548–54.
+
+Woodman, Richard W., John E. Sawyer, and Ricky W. Griffin
+
+(1993), “Toward a Theory of Organizational Creativity,” _Academy of Management Review_, 18 (2), 293–321.
+This content downloaded from
+
+`            `146.70.225.216 on Fri, 20 Sep 2024 19:44:39 UTC All use subject to https://about.jstor.org/terms
+
+[ref1]: Aspose.Words.db40b7bd-7a9a-483c-b00d-85372625ccc6.002.png
+
+[[IMG-2[[IMG-20241106232519962.pdf\|Open: Mehta-NoiseAlwaysBad-2012.pdf]]#psychology
+![[IMG-202411062325199![[IMG-20241106232519962.pdf\|_resources/Is Noise Always Bad? Exploring the Effectsof Ambient Noise on Creative Cognition/869af34bba3313a062f60f04d14b9b6f_MD5.pdf]]

src/site/notes/Bookmarks/Ideas and Theories/The Gentle Singularity.md

@@ -0,0 +1,61 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/ideas-and-theories/the-gentle-singularity/","tags":["ai","societies","algorithm","capitalism","ethics"]}
+---
+
+
+We are past the event horizon; the takeoff has started. Humanity is close to building digital superintelligence, and at least so far it’s much less weird than it seems like it should be.
+
+Robots are not yet walking the streets, nor are most of us talking to AI all day. People still die of disease, we still can’t easily go to space, and there is a lot about the universe we don’t understand.
+
+And yet, we have recently built systems that are smarter than people in many ways, and are able to significantly amplify the output of people using them. The least-likely part of the work is behind us; the scientific insights that got us to systems like GPT-4 and o3 were hard-won, but will take us very far.
+
+AI will contribute to the world in many ways, but the gains to quality of life from AI driving faster scientific progress and increased productivity will be enormous; the future can be vastly better than the present. Scientific progress is the biggest driver of overall progress; it’s hugely exciting to think about how much more we could have.
+
+In some big sense, ChatGPT is already more powerful than any human who has ever lived. Hundreds of millions of people rely on it every day and for increasingly important tasks; a small new capability can create a hugely positive impact; a small misalignment multiplied by hundreds of millions of people can cause a great deal of negative impact.
+
+2025 has seen the arrival of agents that can do real cognitive work; writing computer code will never be the same. 2026 will likely see the arrival of systems that can figure out novel insights. 2027 may see the arrival of robots that can do tasks in the real world.
+
+A lot more people will be able to create software, and art. But the world wants a lot more of both, and experts will probably still be much better than novices, as long as they embrace the new tools. Generally speaking, the ability for one person to get much more done in 2030 than they could in 2020 will be a striking change, and one many people will figure out how to benefit from.
+
+In the most important ways, the 2030s may not be wildly different. People will still love their families, express their creativity, play games, and swim in lakes.
+
+But in still-very-important-ways, the 2030s are likely going to be wildly different from any time that has come before. We do not know how far beyond human-level intelligence we can go, but we are about to find out.
+
+In the 2030s, intelligence and energy—ideas, and the ability to make ideas happen—are going to become wildly abundant. These two have been the fundamental limiters on human progress for a long time; with abundant intelligence and energy (and good governance), we can theoretically have anything else.
+
+Already we live with incredible digital intelligence, and after some initial shock, most of us are pretty used to it. Very quickly we go from being amazed that AI can generate a beautifully-written paragraph to wondering when it can generate a beautifully-written novel; or from being amazed that it can make live-saving medical diagnoses to wondering when it can develop the cures; or from being amazed it can create a small computer program to wondering when it can create an entire new company. This is how the singularity goes: wonders become routine, and then table stakes.
+
+We already hear from scientists that they are two or three times more productive than they were before AI. Advanced AI is interesting for many reasons, but perhaps nothing is quite as significant as the fact that we can use it to do faster AI research. We may be able to discover new computing substrates, better algorithms, and who knows what else. If we can do a decade’s worth of research in a year, or a month, then the rate of progress will obviously be quite different.
+
+From here on, the tools we have already built will help us find further scientific insights and aid us in creating better AI systems. Of course this isn’t the same thing as an AI system completely autonomously updating its own code, but nevertheless this is a larval version of recursive self-improvement.
+
+There are other self-reinforcing loops at play. The economic value creation has started a flywheel of compounding infrastructure buildout to run these increasingly-powerful AI systems. And robots that can build other robots (and in some sense, datacenters that can build other datacenters) aren’t that far off. 
+
+If we have to make the first million humanoid robots the old-fashioned way, but then they can operate the entire supply chain—digging and refining minerals, driving trucks, running factories, etc.—to build more robots, which can build more chip fabrication facilities, data centers, etc, then the rate of progress will obviously be quite different.
+
+As datacenter production gets automated, the cost of intelligence should eventually converge to near the cost of electricity. (People are often curious about how much energy a ChatGPT query uses; the average query uses about 0.34 watt-hours, about what an oven would use in a little over one second, or a high-efficiency lightbulb would use in a couple of minutes. It also uses about 0.000085 gallons of water; roughly one fifteenth of a teaspoon.)
+
+The rate of technological progress will keep accelerating, and it will continue to be the case that people are capable of adapting to almost anything. There will be very hard parts like whole classes of jobs going away, but on the other hand the world will be getting so much richer so quickly that we’ll be able to seriously entertain new policy ideas we never could before. We probably won’t adopt a new social contract all at once, but when we look back in a few decades, the gradual changes will have amounted to something big.
+
+If history is any guide, we will figure out new things to do and new things to want, and assimilate new tools quickly (job change after the industrial revolution is a good recent example). Expectations will go up, but capabilities will go up equally quickly, and we’ll all get better stuff. We will build ever-more-wonderful things for each other. People have a long-term important and curious advantage over AI: we are hard-wired to care about other people and what they think and do, and we don’t care very much about machines.
+
+A subsistence farmer from a thousand years ago would look at what many of us do and say we have fake jobs, and think that we are just playing games to entertain ourselves since we have plenty of food and unimaginable luxuries. I hope we will look at the jobs a thousand years in the future and think they are very fake jobs, and I have no doubt they will feel incredibly important and satisfying to the people doing them.
+
+The rate of new wonders being achieved will be immense. It’s hard to even imagine today what we will have discovered by 2035; maybe we will go from solving high-energy physics one year to beginning space colonization the next year; or from a major materials science breakthrough one year to true high-bandwidth brain-computer interfaces the next year. Many people will choose to live their lives in much the same way, but at least some people will probably decide to “plug in”.
+
+Looking forward, this sounds hard to wrap our heads around. But probably living through it will feel impressive but manageable. From a relativistic perspective, the singularity happens bit by bit, and the merge happens slowly. We are climbing the long arc of exponential technological progress; it always looks vertical looking forward and flat going backwards, but it’s one smooth curve. (Think back to 2020, and what it would have sounded like to have something close to AGI by 2025, versus what the last 5 years have actually been like.)
+
+There are serious challenges to confront along with the huge upsides. We do need to solve the safety issues, technically and societally, but then it’s critically important to widely distribute access to superintelligence given the economic implications. The best path forward might be something like:
+
+Solve the alignment problem, meaning that we can robustly guarantee that we get AI systems to learn and act towards what we collectively really want over the long-term (social media feeds are an example of misaligned AI; the algorithms that power those are incredible at getting you to keep scrolling and clearly understand your short-term preferences, but they do so by exploiting something in your brain that overrides your long-term preference).
+
+Then focus on making superintelligence cheap, widely available, and not too concentrated with any person, company, or country. Society is resilient, creative, and adapts quickly. If we can harness the collective will and wisdom of people, then although we’ll make plenty of mistakes and some things will go really wrong, we will learn and adapt quickly and be able to use this technology to get maximum upside and minimal downside. Giving users a lot of freedom, within broad bounds society has to decide on, seems very important. The sooner the world can start a conversation about what these broad bounds are and how we define collective alignment, the better.
+
+We (the whole industry, not just OpenAI) are building a brain for the world. It will be extremely personalized and easy for everyone to use; we will be limited by good ideas. For a long time, technical people in the startup industry have made fun of “the idea guys”; people who had an idea and were looking for a team to build it. It now looks to me like they are about to have their day in the sun.
+
+OpenAI is a lot of things now, but before anything else, we are a superintelligence research company. We have a lot of work in front of us, but most of the path in front of us is now lit, and the dark areas are receding fast. We feel extraordinarily grateful to get to do what we do.
+
+Intelligence too cheap to meter is well within grasp. This may sound crazy to say, but if we told you back in 2020 we were going to be where we are today, it probably sounded more crazy than our current predictions about 2030.
+
+May we scale smoothly, exponentially and uneventfully through superintelligence.
+

src/site/notes/Bookmarks/Ideas and Theories/Would software engineers be considered working class_.md

@@ -0,0 +1,41 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/ideas-and-theories/would-software-engineers-be-considered-working-class/","tags":["capitalism","coding","communism","internet","money","societies"]}
+---
+
+
+> [!NOTE]
+> Almost all receive their compensation in the form of a base salary. However, more “high level” software engineers will also be paid in equity of some sort, effectively owning part of the company and having some stake in it
+>
+> This seems like a weird middle ground. Is there a more definitive definition of what a working class person is?
+
+See also [[Bookmarks/Ideas and Theories/The GNU Manifesto\|The GNU Manifesto]]
+
+## Answer 1
+
+We are workers, but members of the labor aristocracy. It's a well established phenomenon of imperialist countries, particularly for skilled workers, that far predates the tech industry. From the wiki:
+
+> In Marxist theory, those workers (proletarians) in the developed countries who benefit from the superprofits extracted from the impoverished workers of developing countries form an "aristocracy of labor". The phrase was popularized by Karl Kautsky in 1901 and theorized by Vladimir Lenin in his treatise Imperialism, the Highest Stage of Capitalism. According to Lenin, companies in the developed world exploit workers in the developing world where wages are much lower.
+>
+> The increased profits enable these companies to pay higher wages to their employees "at home" (that is, in the developed world), thus creating a working class satisfied with their standard of living and not inclined to proletarian revolution. It is a form of exporting poverty, creating an "exclave" of lower social class. Lenin contended that imperialism had prevented increasing class polarization in the developed world and argued that a workers' revolution could only begin in one of the developing countries, such as Russia.
+
+## Answer 2
+
+Yes, SEs are working class.
+
+I'm one of those people. I want to share with you an experience I've had.
+
+Many years ago, I was working at a startup as a Software Engineer. As you described, my compensation was in the form of combined salary + equity.
+
+It was a great job. We had a small team of very talented engineers who worked together _wonderfully_. The work itself was interesting and exciting, and our product had _tons_ of room for growth, both in the market and in regards to technical work.
+
+A few years after I joined the company, the CEO sold the company to one of our competitors.
+
+No one on the engineering team wanted this, but our opinions didn't need to be considered. We were being thrown from a team of 5 engineers to a company with 5,000 engineers, into a company with different technology preferences/stacks, and with mountains of legal red tape that were suddenly going to be part of our day-to-day.
+
+But hey, we had equity. So at least we were going to get a hefty payment to dry our tears with, as that equity was immediately transformed into cold-hard-cash.
+
+Each engineer received a lump sum in the 10s of thousands.
+
+The CEO, CFO, and COO each received lump sums in the 10s of millions.
+
+Those three C-level individuals retired in their early 30s. I bought myself a new car to replace my 15-year-old toyota and had to find a new job.

src/site/notes/Bookmarks/Literature, Writing, Poetry/Can Reading maake you Happier_.md

@@ -0,0 +1,46 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/literature-writing-poetry/can-reading-maake-you-happier/","tags":["books","health","literature","mind","techniques"]}
+---
+
+
+# [Can Reading Make You Happier?](https://www.newyorker.com/culture/cultural-comment/can-reading-make-you-happier?ref=dailydev)
+
+## Summary
+
+> [!NOTE]
+>
+> The article discusses the practice of bibliotherapy, which involves using literature as a form of therapy to help people cope with various emotional and psychological issues. The author highlights the work of two bibliotherapists Berthoud and Elderkin, who have been practicing this method for over twenty years.
+>
+> Berthoud and Elderkin met at Cambridge University and bonded over their shared love of books. They began recommending to each other as a way to cope with personal issues, such as heartbreak or career uncertainty. As their friendship developed, they started prescribing novels to friends and family members who were struggling with similar problemsThe two women eventually pitched the idea of running a bibliotherapy clinic to Alain de Botton, a philosopher and friend from Cambridge. De Botton was interested in starting the School of Life, and saw the potential for bibliotherapy as a valuable addition to his program.
+>
+> Berthoud and Elderkin's approach to bibliotherapy is centered on using fiction as a transformative experience. They believe that literature has power to heal and inspire people, and they have developed a network of trained bibliotherapists who work with clients around the world.
+>
+> The most common issues that people bring bibliotherapists are to life transitions, such as career uncertainty or relationship problems. Many retirees also seek help adjusting to their new role in life, and some parents look for guidance on how to navigate the challenges of parenthood.
+>
+> Berthoud and Elderkin have developed a range of strategies for recommending novels that can help people cope with specific issues. For example, they might recommend "Room Temperature" by Nicholson Baker for someone with the responsibilities of fatherhood, or "To Kill a Mockingbird" for someone looking for guidance on how to be a good parent.
+>
+> Overall, the article highlights the potential of bibliotherapy as form of therapy that can help people cope with a range of emotional and psychological issues. By using literature as a tool for healing and growth, bibliotherapists like Berthoud and Elderkin are people to find new ways of thinking about themselves and their place in the world.
+>
+> The article also mentions the history of bibliotherapy, which dates back to ancient Greece and was later used by Sigmund during psychoanalysis sessions. After World War I, libr were trained to give books to veterans as a form of therapy, and the practice has continued to evolve over time.
+>
+> Today, there is a of trained bibliotherapists working around the world, affiliated with the School of Life. They work with clients who are struggling with a range of issues, from career uncertainty to relationship problems, and they use literature as a tool for healing and growth.
+
+## Original article
+
+Several years ago, I was given as a gift a remote session with a bibliotherapist at the London headquarters of the School of Life, which offers innovative courses to help people deal with the daily emotional challenges of existence. I have to admit that at first I didn’t really like the idea of being given a reading “prescription.” I’ve generally preferred to mimic Virginia Woolf’s passionate commitment to serendipity in my personal reading discoveries, delighting not only in the books themselves but in the randomly meaningful nature of how I came upon them (on the bus after a breakup, in a backpackers’ hostel in Damascus, or in the dark library stacks at graduate school, while browsing instead of studying). I’ve long been wary of the peculiar evangelism of certain readers: You must read this, they say, thrusting a book into your hands with a beatific gleam in their eyes, with no allowance for the fact that books mean different things to people—or different things to the same person—at various points in our lives. I loved John Updike’s stories about the Maples in my twenties, for example, and hate them in my thirties, and I’m not even exactly sure why.
+
+But the session was a gift, and I found myself unexpectedly enjoying the initial questionnaire about my reading habits that the bibliotherapist, Ella Berthoud, sent me. Nobody had ever asked me these questions before, even though reading fiction is and always has been essential to my life. I love to gorge on books over long breaks—I’ll pack more books than clothes, I told Berthoud. I confided my dirty little secret, which is that I don’t like buying or owning books, and always prefer to get them from the library (which, as I am a writer, does not bring me very good book-sales karma). In response to the question “What is preoccupying you at the moment?,” I was surprised by what I wanted to confess: I am worried about having no spiritual resources to shore myself up against the inevitable future grief of losing somebody I love, I wrote. I’m not religious, and I don’t particularly want to be, but I’d like to read more about other people’s reflections on coming to some sort of early, weird form of faith in a “higher being” as an emotional survival tactic. Simply answering the questions made me feel better, lighter.
+
+We had some satisfying back-and-forths over e-mail, with Berthoud digging deeper, asking about my family’s history and my fear of grief, and when she sent the final reading prescription it was filled with gems, none of which I’d previously read. Among the recommendations was “The Guide,” by R. K. Narayan. Berthoud wrote that it was “a lovely story about a man who starts his working life as a tourist guide at a train station in Malgudi, India, but then goes through many other occupations before finding his unexpected destiny as a spiritual guide.” She had picked it because she hoped it might leave me feeling “strangely enlightened.” Another was “The Gospel According to Jesus Christ,” by José Saramago: “Saramago doesn’t reveal his own spiritual stance here but portrays a vivid and compelling version of the story we know so well.” “Henderson the Rain King,” by Saul Bellow, and “Siddhartha,” by Hermann Hesse, were among other prescribed works of fiction, and she included some nonfiction, too, such as “The Case for God,” by Karen Armstrong, and “Sum,” by the neuroscientist David Eagleman, a “short and wonderful book about possible afterlives.”
+
+I worked my way through the books on the list over the next couple of years, at my own pace—interspersed with my own “discoveries”—and while I am fortunate enough to have my ability to withstand terrible grief untested, thus far, some of the insights I gleaned from these books helped me through something entirely different, when, over several months, I endured acute physical pain. The insights themselves are still nebulous, as learning gained through reading fiction often is—but therein lies its power. In a secular age, I suspect that reading fiction is one of the few remaining paths to transcendence, that elusive state in which the distance between the self and the universe shrinks. Reading fiction makes me lose all sense of self, but at the same time makes me feel most uniquely myself. As Woolf, the most fervent of readers, wrote, a book “splits us into two parts as we read,” for “the state of reading consists in the complete elimination of the ego,” while promising “perpetual union” with another mind.
+
+Bibliotherapy is a very broad term for the ancient practice of encouraging reading for therapeutic effect. The first use of the term is usually dated to a jaunty 1916 article in _The Atlantic Monthly_, “A Literary Clinic.” In it, the author describes stumbling upon a “bibliopathic institute” run by an acquaintance, Bagster, in the basement of his church, from where he dispenses reading recommendations with healing value. “Bibliotherapy is…a new science,” Bagster explains. “A book may be a stimulant or a sedative or an irritant or a soporific. The point is that it must do something to you, and you ought to know what it is. A book may be of the nature of a soothing syrup or it may be of the nature of a mustard plaster.” To a middle-aged client with “opinions partially ossified,” Bagster gives the following prescription: “You must read more novels. Not pleasant stories that make you forget yourself. They must be searching, drastic, stinging, relentless novels.” (George Bernard Shaw is at the top of the list.) Bagster is finally called away to deal with a patient who has “taken an overdose of war literature,” leaving the author to think about the books that “put new life into us and then set the life pulse strong but slow.”
+
+Today, bibliotherapy takes many different forms, from literature courses run for prison inmates to reading circles for elderly people suffering from dementia. Sometimes it can simply mean one-on-one or group sessions for “lapsed” readers who want to find their way back to an enjoyment of books. Berthoud and her longtime friend and fellow bibliotherapist Susan Elderkin mostly practice “affective” bibliotherapy, advocating the restorative power of reading fiction. The two met at Cambridge University as undergraduates, more than twenty years ago, and bonded immediately over the shared contents of their bookshelves, in particular Italo Calvino’s novel “If on a Winter’s Night a Traveller,” which is itself about the nature of reading. As their friendship developed, they began prescribing novels to cure each other’s ailments, such as a broken heart or career uncertainty. “When Suse was having a crisis about her profession—she wanted to be a writer, but was wondering if she could cope with the inevitable rejection—I gave her Don Marquis’s ‘Archy and Mehitabel’ poems,” Berthoud told me. “If Archy the cockroach could be so dedicated to his art as to jump on the typewriter keys in order to write his free-verse poems every night in the New York offices of the _Evening Sun,_ then surely she should be prepared to suffer for her art, too.” Years later, Elderkin gave Berthoud,who wanted to figure out how to balance being a painter and a mother, Patrick Gale’s novel “Notes from an Exhibition,” about a successful but troubled female artist.
+
+They kept recommending novels to each other, and to friends and family, for many years, and, in 2007, when the philosopher Alain de Botton, a fellow Cambridge classmate, was thinking about starting the School of Life, they pitched to him the idea of running a bibliotherapy clinic. “As far as we knew, nobody was doing it in that form at the time,” Berthoud said. “Bibliotherapy, if it existed at all, tended to be based within a more medical context, with an emphasis on self-help books. But we were dedicated to fiction as the ultimate cure because it gives readers a transformational experience.”
+
+Berthoud and Elderkin trace the method of bibliotherapy all the way back to the Ancient Greeks, “who inscribed above the entrance to a library in Thebes that this was a ‘healing place for the soul.’ ” The practice came into its own at the end of the nineteenth century, when Sigmund Freud began using literature during psychoanalysis sessions. After the First World War, traumatized soldiers returning home from the front were often prescribed a course of reading. “Librarians in the States were given training on how to give books to WWI vets, and there’s a nice story about Jane Austen’s novels being used for bibliotherapeutic purposes at the same time in the U.K.,” Elderkin says. Later in the century, bibliotherapy was used in varying ways in hospitals and libraries, and has more recently been taken up by psychologists, social and aged-care workers, and doctors as a viable mode of therapy.
+
+There is now a network of bibliotherapists selected and trained by Berthoud and Elderkin, and affiliated with the School of Life, working around the world, from New York to Melbourne. The most common ailments people tend to bring to them are the life-juncture transitions, Berthoud says: being stuck in a rut in your career, feeling depressed in your relationship, or suffering bereavement. The bibliotherapists see a lot of retirees, too, who know that they have twenty years of reading ahead of them but perhaps have only previously read crime thrillers, and want to find something new to sustain them. Many seek help adjusting to becoming a parent. “I had a client in New York, a man who was having his first child, and was worried about being responsible for another tiny being,” Berthoud says. “I recommended ‘Room Temperature,’ by Nicholson Baker, which is about a man feeding his baby a bottle and having these meditative thoughts about being a father. And of course 'To Kill a Mockingbird,' because Atticus Finch is the ideal father in literature.”

src/site/notes/Bookmarks/Management and Work related/Is sending Factorio to your competitors' engineers a cost-effective means of sabotage_.md

@@ -0,0 +1,7 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/management-and-work-related/is-sending-factorio-to-your-competitors-engineers-a-cost-effective-means-of-sabotage/","tags":["behavior","interesting","productivity","work","wow"]}
+---
+
+
+[[_resources/Is sending Factorio to your competitors' engineers a cost-effective means of sabotage?/3db17e71c7ceb0bef41431379c2d0716_MD5.pdf\|Open: Is sending Factorio to your competitors' engineers a cost-effective means of sabotage_.pdf]]
+![[_resources/Is sending Factorio to your competitors' engineers a cost-effective means of sabotage?/3db17e71c7ceb0bef41431379c2d0716_MD5.pdf\|_resources/Is sending Factorio to your competitors' engineers a cost-effective means of sabotage?/3db17e71c7ceb0bef41431379c2d0716_MD5.pdf]]

src/site/notes/Bookmarks/Tech/How to store data on paper_.md

@@ -0,0 +1,138 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/tech/how-to-store-data-on-paper/","tags":["interesting","wow","internet","art","algorithm"]}
+---
+
+How to print my voice? How to put music on paper? How to print an executable program? All this boils down to storing digital data on paper. This is also called [paper data storage](https://en.wikipedia.org/wiki/Paper_data_storage). Arrived in this domain from poetry, I’ve been investigating the area for some time. Here is what I found.
+
+## What to print on paper?
+
+[![_resources/Untitled 1/9e2303582f8661b77607d9a599270772_MD5.png](/img/user/_resources/Untitled%201/9e2303582f8661b77607d9a599270772_MD5.png)](https://user-images.githubusercontent.com/803666/102592241-19e7cf00-410b-11eb-9c3b-97185c910aa8.png)  
+Printing an executable program (encoded in base64)
+
+[![_resources/Untitled 1/d42a4e39bd563f29d4bd18041b901dd0_MD5.png](/img/user/_resources/Untitled%201/d42a4e39bd563f29d4bd18041b901dd0_MD5.png)](https://user-images.githubusercontent.com/803666/82190179-2b03ff80-98e0-11ea-90ec-61c471535a49.png)  
+Printing a secret message (encrypted with [GPG-AES256](https://en.wikipedia.org/wiki/Advanced_Encryption_Standard), encoded as stacked qrcode).
+
+[![_resources/Untitled 1/fb11b5c83a23a6bd839179e237cd000c_MD5.png](/img/user/_resources/Untitled%201/fb11b5c83a23a6bd839179e237cd000c_MD5.png)](https://user-images.githubusercontent.com/803666/79379449-78dac000-7f4e-11ea-8747-b66a13ee1527.png)  
+Printing a 21 seconds [soundscape of a rugby match](https://encyclopediedelaparole.org/fr/documents/joue-tous-les-ballons) (in 64kpbs MP3 encoded with Optar)
+
+[![_resources/Untitled 1/a6944973231def10cc5eb277bcfe6557_MD5.png](/img/user/_resources/Untitled%201/a6944973231def10cc5eb277bcfe6557_MD5.png)](https://user-images.githubusercontent.com/803666/79601528-773d0380-80d8-11ea-88f9-0bc8536ad6b5.png)  
+Printing a 7-page scientific paper on a single page (a PDF file encoded with colorsafe)
+
+How to transform digital data to a printable format? You need a format that transforms a sequence of bytes into an image with a specific encoding. Then, you need two pieces of software, one encoder (data->image) and one decoder (image->decoder). We now review different kinds of encodings.
+
+## Character-based encodings
+
+TL;DR; OCR digital data is a very hard problem, one can get 2.5 kilobytes per A4 page that is decodable afterwards, see [“How to get perfect OCR for digital data?”](https://www.monperrus.net/martin/perfect-ocr-digital-data).
+
+With character-based encoding, the idea is to use a [binary-to-text encoding](https://en.wikipedia.org/wiki/Binary-to-text_encoding) and then to use optical character recognition (OCR). The biggest advantage of character-based encodings is that they can be decoded by humans (as opposed to dot-based encodings), which means that you don’t need a camera or a scanner to recover the data. This is the most robust assumption for extreme conditions.
+
+The information capacity primarily depends on the size of the alphabet and the font size (plus some variations due to the font being used). Yet, those numbers are theoretical in the sense that they assume perfect OCR.
+
+**[base16/hexadecimal](https://en.wikipedia.org/wiki/Hexadecimal)**: I can get 3.5 kilobytes per A4 page (font size 12pt, font Inconsolata) and OCR it with gocr at 400DPI. With smaller fonts, we have for example 4.6 kilobytes per A4 page (font size 9pt, font Inconsolata), but I cannot OCR it well (confusion between “7” and “1”), see [“How to get perfect OCR for digital data?”](https://www.monperrus.net/martin/perfect-ocr-digital-data).
+
+[![_resources/Untitled 1/109eb057c36ec2566cdd2e3f1a67514d_MD5.png](/img/user/_resources/Untitled%201/109eb057c36ec2566cdd2e3f1a67514d_MD5.png)](https://user-images.githubusercontent.com/803666/97078209-f6584a80-15d9-11eb-85ea-adea8d950e78.png)  
+Printing an encrypted GPG file encoded in base16
+
+**[base32](https://en.wikipedia.org/wiki/Base32)**: I can get 5.8 kilobytes per A4 page (font size 9pt, font Inconsolata), but it is impossible for me to OCR it later without errors, see [“How to get perfect OCR for digital data?”](https://www.monperrus.net/martin/perfect-ocr-digital-data).
+
+**[base64](https://en.wikipedia.org/wiki/Base64)**: 9.3 kilobytes per A4 page at font size 8t, even 17 kilobytes per A4 page at font size 6pt (confirming [this post](https://bitcoin.stackexchange.com/a/13494), but this is theoretical, no OCR can handle it).
+
+[![_resources/Untitled 1/9e2303582f8661b77607d9a599270772_MD5.png](/img/user/_resources/Untitled%201/9e2303582f8661b77607d9a599270772_MD5.png)](https://user-images.githubusercontent.com/803666/102592241-19e7cf00-410b-11eb-9c3b-97185c910aa8.png)  
+Printing an executable program encoded in base64
+
+**[bocr32](https://www.monperrus.net/martin/perfect-ocr-digital-data)**: I have myself invented a character set of 32 characters, called bocr32, optimized for OCR, see [“How to get perfect OCR for digital data?”](https://www.monperrus.net/martin/perfect-ocr-digital-data).
+
+[![_resources/Untitled 1/efdfe785906d9bde69566da5b402ffb3_MD5.png](/img/user/_resources/Untitled%201/efdfe785906d9bde69566da5b402ffb3_MD5.png)](https://user-images.githubusercontent.com/803666/102592774-de99d000-410b-11eb-8c07-6aa1fa566e48.png)  
+Example of bocr32 encoded data
+
+**[bip39](https://github.com/bitcoin/bips/blob/master/bip-0039.mediawiki)**, I can get 1kb per A4 page (font size 12pt, font Inconsolata), and decode it, see [“How to get perfect OCR for digital data?”](https://www.monperrus.net/martin/perfect-ocr-digital-data).
+
+[![_resources/Untitled 1/4d0d823de4f8fae752005a2b358530c4_MD5.png](/img/user/_resources/Untitled%201/4d0d823de4f8fae752005a2b358530c4_MD5.png)](https://user-images.githubusercontent.com/803666/97078330-28b67780-15db-11eb-9d0f-14f92645dc2f.png)  
+Printing binary data encoded in bip39
+
+## Black-and-white Dot Encodings
+
+TL;DR; Up to 70k-100K per A4 page. Stacked QRCodes work well with off-the-shelf software. Optar is OK if one decreases the default density (XCROSSES/YCROSSES).
+
+In a dot-based based encoding, the information capacity is a function of how small the dots are. In other words, it depends on your printer and your scanner (or camera if you take a picture of it). More technically, it means that the information capacity heavily depends on the conjunction of dots per inch (DPI) resolution at printing time and at scanning time.
+
+**With a 600 DPI scanning resolution** (all those experiments are done on a Konica Minolta c550)
+
+**QRCode** [(wikipedia)](https://en.wikipedia.org/wiki/QR_code): The maximum size of QR-Code is 2953 bytes, but you can stack several QRCodes on a single page (up to 24 Symbol-40 in my experiments). I decode them using [zbar-tools](https://packages.debian.org/zbar-tools). With 24 QRcode and UUencoding, I can put 47k of binary data. With direct UTF8 text, I can put 70k of text.
+
+[![_resources/Untitled 1/b4cc93eb3104fd1398af991ee09c7f40_MD5.png](/img/user/_resources/Untitled%201/b4cc93eb3104fd1398af991ee09c7f40_MD5.png)](https://user-images.githubusercontent.com/803666/79479934-007f0800-7ffd-11ea-9d0e-7a9e22c78e1a.png)
+
+24 QrCodes on a page, with text, totaling 9549 words over 71 kilobytes, successfully decoded
+
+**Optar** [origin](http://ronja.twibright.com/optar/) [(unofficial Github repository)](https://github.com/colindean/optar/) works well, with which I can get 100kilobytes per A4 page, with the following tweaks: 1) XCROSSES=45 and YCROSSES=65 2) using TIF scanner output instead of JPG 3) by decreasing the scanning DPI resolution (from 600 to 400 counter-intuitive) I managed to encode 90 kilobytes of MP3 in Optar(52x74) decoded with 0.68% irreparable bits, resulting in a few audible glitches.
+
+I could also get some reasonable results with stacked [DataMatrix](https://en.wikipedia.org/wiki/Data_Matrix) Same idea as QRCode, we have a maximum of 3116 bytes per matrix and one can pack several datamatrices on a page. However, the default decoder software on Linux ([dmtxread](https://packages.debian.org/dmtx-utils)) is not as effective as zbarimg (I succeeded to only pack 9 datamatrix per A4 page).
+
+Other formats (failed or untested):
+
+[paperback](http://ollydbg.de/Paperbak/) ([Windows version](http://ollydbg.de/Paperbak/)) ([linux version](https://git.teknik.io/scuti/paperback-cli)) I tested the linux version with a 600 DPI scanner and it fails ([issue](https://git.teknik.io/scuti/paperback-cli/issues/35)). Note that their 500kb-per-page claim is not reproducible.
+
+## Color Dot Encodings
+
+TL;DR; beautiful, but I haven’t thoroughly investigated them yet, only got some small success with jabcode.
+
+Adding color allows on to code more information per dot (3x more with three colors). The drawback is that colorized dot encodings requires color printers, which are more expensive.
+
+[![_resources/Untitled 1/b7393c889e2b83438cfd5a2b18303091_MD5.png](/img/user/_resources/Untitled%201/b7393c889e2b83438cfd5a2b18303091_MD5.png)](https://user-images.githubusercontent.com/803666/80257504-986d9900-8670-11ea-8ebb-af9ff18e3921.png)  
+Example of Jabcode
+
+Colorized paper formats (failed or untested):
+
+-   [jabcode](https://github.com/jabcode/jabcode) is done by a German government organization, still maintained.
+-   [colorsafe](https://github.com/colorsafe/colorsafe/) is full Python, can be installed through pip, claims up to 224kB of data per page. (fails, decoding fails after full-cycle)
+-   [MrYakobo/tk](https://github.com/MrYakobo/tk) is for fun
+-   [elm-hccb](https://github.com/canadaduane/elm-hccb) for [High Capacity Color Barcode (HCCB)](https://en.wikipedia.org/wiki/High_Capacity_Color_Barcode) (does not contain a decoder)
+-   [blots](https://github.com/lf94/blots/) (not tested, does not contain a decoder)
+
+## Drawing encoding
+
+Artcodes allows users to draw encoded information, see [https://www.artcodes.co.uk/artcodes/](https://www.artcodes.co.uk/artcodes/).
+
+## Discussion
+
+### Theoretical information density maxima
+
+The theoretical maxima is when you assume perfect scanning, alignment, etc. It depends on the scanning density and the color scale. At 300 DPI and black white, a page is 2480x3508 pixels, so it contains 8.7M bits which is 1100kB (1.1MB).
+
+### Long term storage
+
+Once you have some kind of encoded data, we store on paper. There exists different paper quality, incl. papers specifically designed for archival (and certified as such, for instance by a [Swedish government agency](https://www.ri.se/en/what-we-do/services/paper-testing-of-archival-and-permanent-paper)). Archival paper with archival ink is meant to [last for centuries](https://en.wikipedia.org/wiki/Print_permanence).
+
+Now, it is also possible to put the encoded data on other media than paper, such as [stone](https://www.sys-teco.com/photo-engraving/qr-codes-grave-stone/), [lego bricks](https://hvitis.dev/jab-code-and-everything-you-need-to-know-about-color-bar-code) (claimed to last at least 1300 years), and [solid nickel](https://blocksandfiles.com/2021/10/25/totenpass-stores-data-on-credit-card-sized-gold-plated-nickel-slab/). Bitcoiners have gone quite far in falsifying the [durability claims of digital crypo keys on metal plates](https://blog.lopp.net/metal-bitcoin-seed-storage-stress-test/).
+
+Also based on optical storage, one can use microfiches, as once done by [Foto-Mem](https://en.wikipedia.org/wiki/Foto-Mem).
+
+### Redundancy and error correction
+
+In a full cycle printing-transport-scanning there is a real risk that some dots in the image become corrupted. If the file format and use case allow for corruption (eg music in MP3), that’s fine. If the exact bit-per-bit content is required, this risk can be mitigated by [error correction](https://en.wikipedia.org/wiki/Error_correction_code).
+
+For instance, QRCodes use [Reed-Solomon encoding](https://en.wikipedia.org/wiki/Reed%E2%80%93Solomon_error_correction). In QRcode, there are four levels of error correction, low, medium, quartile and high, where the latter allows for up to 30% of the information to be lost.
+
+It is always possible to use off-the-shelf error correction before encoding such as [rsbep](https://manpages.ubuntu.com/manpages/precise/man1/rsbep.1.html) or [PAR2](https://en.wikipedia.org/wiki/PAR2).
+
+### Hand writing digital data
+
+Not all encodings are appropriate for handwriting digital data. The character based encodings are OK if one has good OCR for one’s writing. Otherwise, the linear encodings (1D barcode) based on height (and not based on width) can be drawn easily. Examples of such height based encodings are the [US Postal format](https://en.wikipedia.org/wiki/Intelligent_Mail_barcode) and [Spotify codes](https://techcrunch.com/2017/05/05/spotify-codes/).
+
+[![_resources/Untitled 1/ed8fae4bb8a18c15826bca41d21fe5c0_MD5.jpg](/img/user/_resources/Untitled%201/ed8fae4bb8a18c15826bca41d21fe5c0_MD5.jpg)](https://www.monperrus.net/martin/spotify-code-manual.jpeg)  
+Hand-written Spotify barcode (credits Léonie Monperrus)
+
+### Transportation of paper-stored data
+
+You can take the sheets with you, send them by post, or even [attach them to homing pigeons](https://en.wikipedia.org/wiki/IP_over_Avian_Carriers)
+
+## See also
+
+-   Papers
+    -   In [Robust and Fast Decoding of High-Capacity Color QR Codes for Mobile Applications (2017)](https://arxiv.org/pdf/1704.06447.pdf) Yang et al. claim to put 7.7kB on 38×38 mm2, which means ~300kb per A4 page. For experimenting, their [tool is on Github](https://github.com/cuhk-mobitec/HiQ-Robust-and-Fast-Decoding-of-High-Capacity-Color-QR-Codes).
+    -   In [Paper encryption technology (2009)](https://www.fujitsu.com/global/documents/about/resources/publications/fstj/archives/vol46-1/paper20.pdf), Anan and colleagues do the same thing as GPG plus paper encoding, as presented above.
+-   [The SKOR codex](http://societeanonyme.la/) contains 306 pages of printed images and printed sound based on a [homegrown encoding](https://archive.bleu255.com/bleu255.com-log/2012/07/03/electronic-publishing-one-bit-at-a-time/index.html).
+-   [Wikinaut’s notes](https://hydra.wikinaut.de/bhntwiki/index.php/BHNT67/Paperback) is a gold mine with tons of links, in particular related to paper-storage of GPG and bitcoin cryptographic keys.
+-   Wikipedia:
+    -   [Paper data storage](https://en.wikipedia.org/wiki/Paper_data_storage)
+    -   [2D bar code](https://en.wikipedia.org/wiki/Barcode)

src/site/notes/Bookmarks/Tech/What is an AI Agent_.md

@@ -0,0 +1,66 @@
+---
+{"dg-publish":true,"permalink":"/bookmarks/tech/what-is-an-ai-agent/","tags":["ai","blog","dev","explanation","llm","tools"]}
+---
+
+
+# [What is an agent?](https://blog.langchain.dev/what-is-an-agent/)
+
+_“What is an agent?”_
+
+See also [[Bookmarks/Tech/Understanding LLMs from scratch using middle school math\|Understanding LLMs from scratch using middle school math]] to understand how LLMs work.
+
+- [What does it mean to be agentic?]()
+- [Why is “agentic” a helpful concept?]()
+- [Agentic is new](ic-is-new)
+
+I get asked this question almost daily. At LangChain, we build tools to help developers build LLM applications, especially those that act as a reasoning engines and interact with external sources of data and computation. This includes systems that are commonly referred to as “agents”.
+
+Everyone seems to have a slightly different definition of what an agent is. My definition is perhaps more technical than most:
+
+An agent is a system that uses an LLM to decide the control flow of an application.
+
+Even here, I’ll admit that my definition is not perfect. People often think of agents as advanced, autonomous, and human-like — but what about a simple system where an LLM routes between two different paths? This fits my technical definition, but not the common perception of what an agent should be capable of. It’s hard to define _exactly_ what an agent is!
+
+That’s why I really liked Andrew Ng’s [tweet last week](https://x.com/AndrewYNg/status/1801295202788983136?ref=blog.langchain.dev). In it he suggests that “rather than arguing over which work to include or exclude as being a true agent, we can acknowledge that there are different degrees to which systems can be agentic.” Just like autonomous vehicles, for example, have levels of autonomy, we can also view agent capabilities as a spectrum. I really agree with this viewpoint and I think Andrew expressed it nicely. In the future, when I get asked about what an agent is, I will instead turn the conversation to discuss what it means to be “agentic”.
+
+## What does it mean to be agentic?
+
+I gave a TED talk last year about LLM systems and used the slide below to talk about the different levels of autonomy present in LLM applications.
+
+<img alt="" src="https://blog.langchain.dev/content/images/2024/06/Screenshot-2024-06-28-at-7.33.10-PM.png" height="1150" width="1794" />
+
+A system is more “agentic” the more an LLM decides how the system can behave.
+
+Using an LLM to route inputs into a particular downstream workflow has some small amount of “agentic” behavior. This would fall into the `Router` category in the above diagram.
+
+If you do use multiple LLMs to do multiple routing steps? This would be somewhere between `Router` and `State Machine`.
+
+If one of those steps is then determining whether to continue or finish - effectively allowing the system to run in a loop until finished? That would fall into `State Machine`.
+
+If the system is building tools, remembering those, and then taking those in future steps? That is similar to what the [Voyager paper](https://arxiv.org/abs/2305.16291?ref=blog.langchain.dev) implemented, and is incredibly agentic, falling into the higher `Autonomous Agent` category.
+
+These definitions of “agentic” are still pretty technical. I prefer the more technical definition of “agentic” because I think it’s useful when designing and describing LLM systems.
+
+## Why is “agentic” a helpful concept?
+
+As with all concepts, it’s worth asking why we even need the concept of “agentic”. What does it help with?
+
+Having an idea of how agentic your system can guide your decision-making during the development process - including building it, running it, interacting with it, evaluating it, and even monitoring it.
+
+The more agentic your system is, the more an orchestration framework will help. If you are designing a complex agentic system, having a framework with the right abstractions for thinking about these concepts can enable faster development. This framework should have first-class support for branching logic and cycles.
+
+The more agentic your system is, the harder it is to run. It will be more and more complex, having some tasks that will take a long time to complete. This means you will want to run jobs as background runs. This also means you want durable execution to handle any errors that occur halfway through.
+
+The more agentic your system is, the more you will want to interact with it while it’s running. You’ll want the ability to observe what is going on inside, since the exact steps taken may not be known ahead of time. You’ll want the ability to modify the state or instructions of the agent at a particular point in time, to nudge it back on track if it’s deviating from the intended path.
+
+The more agentic your system is, the more you will want an evaluation framework built for these types of applications. You’ll want to run evals multiple times, since there is compounding amount of randomness. You’ll want the ability to test not only the final output but also the intermediate steps to test how efficient the agent is behaving.
+
+The more agentic your system is, the more you will want a new type of monitoring framework. You’ll want the ability to drill down into all the steps an agent takes. You’ll also want the ability to query for runs based on steps an agent takes.
+
+Understanding and leveraging the spectrum of agentic capabilities in your system can improve the efficiency and robustness of your development process.
+
+## Agentic is new
+
+One thing that I often think about is what is _actually new_ in all this craze. Do we need new tooling and new infrastructure for the LLM applications people are building? Or will generic tools and infrastructure from pre-LLM days suffice?
+
+To me, the more agentic your application is, the more critical it is to have new tooling and infrastructure. That’s exactly what motivated us to build [LangGraph](https://www.langchain.com/langgraph?ref=blog.langchain.dev), the agent orchestrator to help with building, running, and interacting with agents, and [LangSmith](https://www.langchain.com/langsmith?ref=blog.langchain.dev), the testing and observability platform for LLM apps. As we move further on the agentic spectrum, the entire ecosystem of supportive tooling needs to be reimagined.