RoomSense Final

In ITP, Quantified Self

Click here for a quick look at the website I created for the RoomSense and click here for the slides we presented in class. The full research document prepared by my partner, Matt Goral, is below.

 

what we hope to find

We are hoping to discover a direct correlation between the temperature, humidity, light and sound of a social environment and the sentiment or mood of its patrons. We hypothesize that these elements will have a significant effect on the interpersonal behavior of individuals, as well as each individual’s perception of the bar’s quality of atmosphere.

 

Why we are interested

From an artistic perspective, and as professional designers, we have studied to great lengths the effect that aesthetic experience has on human psychology and social behavior. Our individual work and research has brought us to the conclusion that the aesthetic influences in social and business settings, such as bars and restaurants, are insufficiently understood or manipulated to improve or stabilize the common social sentiment of patrons. Better dynamic systems are necessary to improve the prosperity of these businesses via the social behavior of their customers.

 

Precedent and Relevant Works

Effects of Bright Artificial Light Mood of Shift Work Nurses

  • subjects are all nurses
  • all female subjects
  • hospital environment
  • “Each subject was asked to make a mark across a 100-mm line, which defined the extremes of a bipolar scale”
  • “This study indicated that bright artificial light tends to improve eagerness and reduce tension in shift work nurses. Moreover, several psychological symptoms, such as vigor, eagerness, appetite and impairment (the latter only on the second night), improved significantly or nearly significantly due to bright artificial light during the night, but not the evening, shifts. It should be noted, however, that this picture was not necessarily consistent for all the symptoms we studied

Effects of Indoor Color on Mood and Cognitive Performance

  • study of wall color (therefore reflected light)
  • accounts for gender/age of participants
  • violet and yellow = colors used
  • café and store
  • 245 patrons canvassed for qualitative response

Noise Pollution: Non-auditory Effects on Health

  • “Studies of occupational and environmental noise exposure suggest an association with hypertension, whereas community studies show only weak relationships between noise and cardiovascular disease. Aircraft and road traffic noise exposure are associated with psychological symptoms but not with clinically defined psychiatric disorder. In both industrial studies and community studies, noise exposure is related to raised catecholamine secretion. In children, chronic aircraft noise exposure impairs reading comprehension and long-term memory and may be associated with raised blood pressure.”
  • defines noise as “unwanted sound”
  • evidence of link between noise and hypertension

A Review of Environmental Noise and Mental Health

  • inconclusive

The Influence of Crowd Density on the Sound Environments of Commercial Pedestrian Streets

  • specific to outdoor commercial pedestrian streets
  • acknowledges that traffic noise can cause some variation in outcomes
  • supports hypothesis that noise/ambient sound levels can be used to determine crowd density

Hot and Crowded: Influences of Population Density and Temperature on Interpersonal Behavior

  • shows correlation between high temperatures and unpleasant feelings
  • shows correlation between high population density and unpleasant feelings
  • findings indicate that people are less attracted to one another in hot, dense environments

what we are using

We are using a simple, singular device to record variations in temperature, humidity, ambient light and audio. For temperature and humidity, we are using a DHT sensor; for ambient light, a simple photocell; and for audio, an unamplified microphone sensor. All of which are of Radioshack make.The data acquired from each sensor will be managed by an Arduino Micro microcontroller, which will then write the data, in comma-separated values, to a CSV file on a MicroSD card via an Adafruit SD breakout board. The device uses a 9 volt battery, regulated to 5 volts, as a power source. The device is less than 10 inches in length, 4 inches in width, and 2 inches in height. The fritzing below was from a previous iteration that included a piezo vibration sensor, a PIR motion sensor and a potentiometer for calibration purposes.

 

sentiment circuit_bb

Screen Shot 2015-05-08 at 5.16.59 PM

 

What we hope each sensor will indicate

We have chosen these variables, and the sensors to record them, through an intense process of literature review, discussion with peers and professionals, and trial and error. Audio, light, and temperature have been used widely to study human social behavior, and as tools for psychological therapy. These three elements in combination are capable of creating a wide variety of environments, from soothing to aggravating. We hope our research will illuminate the individual effects of each of these influences on social behavior to inform further research on their manipulation for the improvement and stabilization of crowd sentiment.

 

How we will measure sentiment

Our measurements are taken in one hour long sessions at a bar that we have chosen for its popularity and diversity of patrons. E’s Bar is located in the Upper West Side of New York City. It is recently established and has a spectrum of visitors ranging from leisurely afternoon drinkers to late-night revelers. Subjectively, by our standards, it is just a good bar. We strategically chose times of day to reflect different social atmospheres. For example, between the hours of 3pm and 6pm, we are expecting that most customers at E’s Bar will be eating dinner, or stopping for casual, after-work drinks; at 10pm, we expect that most customers will be interested in heavier drinking, dancing, and excited conversation. During our sessions, we will be making observations of the atmosphere and crowd’s general sentiment every quarter-hour. We have decided to focus on two attributes, sentiment and activity. Sentiment will be measured on a scale of 0 to 10, where 0 is extreme negative sentiment and 10 is extreme positive sentiment. Activity will similarly be measured on a scale of 0 to 10, where 0 is no activity and 10 is high activity. Through our differing and subjective attempts at unbiased observation, we hope to derive a qualitative analyses that reflects the social atmosphere of the bar to a reasonable degree of accuracy.

 

Visualizing the Data

Here is the website we built to illustrate the data we collected throughout our process. We built a website using HTML, CSS, JQuery and Chart.js. To see check out the live website, click here.

 

Screen Shot 2015-05-08 at 5.19.43 PM

 

what correlations are anticipated

The qualitative measurements of sentiment and activity will serve as axes for a point graph that will better illuminate trends in sentiment. For example, it is possible that higher activity is frequently associated with higher sentiment. This, however, is not anticipated. We expect to see a bell curve that shows sentiment rising with activity, but only to a certain threshold, at which sentiment would drop off significantly as activity continues to rise. We hope to find correlations between these observations and our measurements of light, audio, temperature and humidity. We expect that temperature and humidity will steadily rise with activity, which will correlate to an eventual drop in sentiment. We expect to see several telling things in our analysis of audio levels and ambient light levels, which we predict will drop and rise in an inversely proportionate manner. These are all assumptions, however, as our insufficient data can not at this time support any hypotheses.

 

What causation would mean

We are confident that our research will lead us to find evidence of several things — specifically, that the attributes of the physical atmosphere that we have chosen to observe will have a noticeable and significant causal effect on the social atmosphere and behavior of patrons. Such a conclusion would lead us to continue our research of the effects of these elements on other various social settings, such as public spaces like parks, museums and libraries. We would then use our findings to inform the development of a dynamic system through which these atmospheric elements could be better controlled. This would mean extending our research into light and color therapy, audio therapy, anxiety, social behavior, and the effects of heat and humidity on bodily and psychological stress.

 

Further research

Moving forward, our goal is to continue our one hour long sessions at E’s Bar until noticeable, significant trends can be found in our data. Once our hypothesis — that these correlations will arise — has been proven, we will extend our research to more bars and restaurants, adding another degree of variation to our experiment. This will require more volunteers to conduct the research (which we do not think should be too hard, as the experiments are conducted at bars), which will in turn require multiple sensor devices. We will need to develop a streamlined system by which we can consolidate our data and visualize it efficiently.

 

Sources

  1. Iwata, Noboru, Sadaaki Ichii, and Kazumichi Egashira. “Effects of Bright Artificial Light on Subjective Mood of Shift Work Nurses.” Industrial Health (1997): 41-47. Print.
  2. Yildirim, K., A. Akalinbaskaya, and M. Hidayetoglu. “Effects Of Indoor Color On Mood And Cognitive Performance.” Building and Environment 42.9 (2007): 3233-240. Print.
  3. Stansfeld, S. A. “Noise Pollution: Non-auditory Effects On Health.” British Medical Bulletin 68.1 (2003): 243-57. Print.
  4. Stansfeld, SA, MM Haines, M. Burr, B. Berry, and P. Lercher. “A Review of Environmental Noise and Mental Health.” Noise & Health 2.8 (2000): 1-8. Print.
  5. Meng, Qi, and Jian Kang. “The Influence of Crowd Density on the Sound Environment of Commercial Pedestrian Streets.” Science of The Total Environment 511 (2015): 249-58. Print.
  6. Griffitt, William, and Russel Veitch. “Hot and Crowded: Influences of Population Density and Temperature on Interpersonal Affective Behavior.” Journal of Personality and Social Psychology 17.1 (1971): 92-98. Print.

 

Submit a comment