Monday, March 28, 2011

Ethnography Week #7

Life at the helpdesk has been very calm this week. We are nearing the end of our study and feel very confident in the amount of data we have collected and the conclusions we have come to.

We have been considering the application we will design for our group and believe we have something which could really be of great assistance to them. We hope that it will actually be used, as we plan to implement it for our final project.

For the most part we have observed the same things as previous weeks, occasionally a friend of an employee or someone from another department will stop by and socialize. But mostly, we see the culture as a place to pass time, socialize and occasionally perform tech duties. It is noteable that almost everyone working in the center has a personal interest in technology and the job is not difficult or challenging to them.

Thursday, March 24, 2011

Microblogs: Why We Make Mistakes

Chapter 0 & 1

The first chapter provided a bit of background for the book as well as some motivation for studying the topic. It was asserted that humans are not necessarily to blame for their errors, rather design choices may be to blame. A number of interesting experiments and studies concerning our ability to recognize our motivation or to miss seeing something were cited.


This book seems like it will be a wealth of interesting information concerning mistakes or more generally human tendencies to miss portions of the world around them. I’ve found the studies cited to be very interesting, and would like more details on some of them, rather than just the shocking facts.


Chapter 2 & 3

Meaningful things are easier to remember. We are great at storing away connections between objects and ideas or events. Examples like remembering faces better than names and using stories to remember something are given. We are also prone to mistakes when things seem to represent something we know, but don't fully. A discussion of how judgements are made on visual attributes is given. These are denoted as "invisible" attributes that affect our choices.


These chapters were enjoyable and pointed out many of the things I have put into practice in my own life without realizing why. For example, I always create stories or relationships even if they are fake when I need to memorize a large amount of terminology for an exam. Overall these chapters remind me that I hate taking exams which aren't essay style.


Chapter 4 & 5

Wearing rose colored glasses refers to the idea that we tend to see ourselves in a very positive light when we recall an event. We always describe our actions as making us look the best. This effect is not intentional. Having knowledge of events, i.e. looking back on the past, has an effect on our memory. We know the outcomes and we will fill in the gaps with our imagination to make things look good. Next, multitasking is examined. The claims is that much like a single processor computer, we can only do time slicing, never truly proces two events at once. The idea of inattentional blindness is explored through examples.


I really enjoy the amount of examples that Hallinan provides. Often times I find myself wanting to look into the particular studies or incidents more. So far this book has really captured many aspects of human nature. The ideas of rose colored glasses and multitasking are not surprising. Though I would debate the idea that we cannot multitask. Many of our body systems run in parallel and though they may not be upper cognitive processes, they are present. For example, you don't need to do any context switching to think about breathing or to dodge a flying object -- it is a natural background process.


Chapter 6 & 7

The first chapter deals with the idea of framing and anchoring. The idea is that if we frame a situation wrong we will make the wrong choices and interpretations. Interestingly if a person expects a loss in a situation they are more likely to take risk. The second chapter discusses the fact that we don't often read and parse the entirety of anything. We hit key points and interpret the rest. The better you are at something, the more likely you will miss details.


The framing and anchoring issues are tremendously interesting. Especially from a computer science standpoint. In developing true AI we need to be able to frame a situation correctly otherwise we will get strange and incorrect results. If humans suffer from framing problems, will machines ever overcome them? The chapter on skimming is another general truth about the way we operate. It is interesting that experts miss more details, though it seems plausible.


Chapter 8 & 9

The first chapter explores the idea that we often smooth out or simplify memories or details in order to make them easier to understand. A number of examples of this are given. The human mind organizes in a hierarchical fashion.The second chapter explores how men and women differ in the ways they arrive at errors. The book asserts that men are more confident and women are less likely to take a risk.


I thought the idea of childhood having an impact on confidence in mapping was very intriguing and true. There is definitely a lot to be said for childhood experiences shaping the decision making processes of individuals. This would be an interesting area to explore by itself. The smoothing of memories and storytelling are not surprising.


Chapter 10 & 11

Only those who are depressed have a view of themselves that is less than average. Many people have taken advantage of humans' overconfidence in order to achieve financial gain. We also have tendency to believe more information will help us, when in fact it can do the opposite. In order to become an expert at something it is estimated that about ten years of refinement and practice is required. The idea of mappings in the mind by experts is explained. Additionally, people become set in their ways.


It was sad to consider that only the depressed have an accurate world view. I feel like a tiny bit of overconfidence is an important motivating factor (proof: how many depressed people are highly motivated?) in our lives. Again I appreciate the examples given in these chapters which seem to back up the claims. I am not sure about the information overflow claims. It seems like a large amount of information acquired over a decent period of time would be beneficial.


Chapter 12 & 13 & Conclusion

People die because of simple errors which could be easily preventable. The argument is that giving meaningful attributes to things can result in increased performance and less human error. Studies of the FAA improvements in comparison to the lack of improvements in the medical community are given. Finally, the idea that the grass is always greener on the other side is explored. This fact is proven to be true, illustrating that people often perceive a increase in happiness will occur in a differing situation. The book concludes with summarizing the major findings of the book and the main ideas. The point that aspiring for different situations constantly will lead to unhappiness is made.


These chapters reminded me of Norman's concepts and ideas. Overall it is interesting to see the simple steps we can take to avoid human error. The FAA studies were particularly encouraging to see. Overall the conclusion provided a good summarization and take away message from the book.


Paper Reading #16: Performance optimizations of virtual keyboards for stroke-based text entry on a touch-based tabletop

Comments

Stuart

Jeremy


Reference

Performance optimizations of virtual keyboards for stroke-based text entry on a touch-based tabletop

Jochen Rick, The Open University

UIST 2010 - New York


Summary

Entering text into devices without physical keyboards can be difficult. A common method for text entry is the use of shape writing, where the user strokes through the letters of the word without lifting their finger, in order to input text into a device. This paper investigated the keyboard layouts which would be optimal for text entry using the shape writing method. A large review of keyboard history and various types of keyboard was given. Additionally, the author performed two user studies to determine the optimal keyboard design. Based on Fitt’s law the author constructed a model of stroking through multiple points. This model was later used to evaluate the effectiveness of various keyboard designs. The results indicate that the typical QWERTY keyboard gives a 17.3% increase in speed, while the OPTI II gives 29.5%. The author concludes by saying it is unlikely this keyboard will gain widespread acceptance.



Analysis

This paper was a great survey of keyboard technology and an interesting study of new shape writing methods. The author performed his analysis thoroughly and fully. While, I don’t see this new keyboard design becoming used by mainstream users, I think having it as an option would be nice. I also appreciate the author’s concession that most people will continue to use the QWERTY keyboard despite it not being the most efficient. I do share the optimism that since we are using more and more software keyboards it may become easier for users to try out a new keyboard and eventually switch to it.


Wednesday, March 23, 2011

Obedience To Authority - Full Blog


Reference
Obedience to Authority
Stanley Milgram
Harper Perennial Modern Classics - 2009

Summary
This book provides a detailed account of the planning and execution of several experiments designed by Stanley Milgram to test the effects of authority. The main premise of the experiment is to see how long someone will obey authority despite their belief that they are causing harm to another human being. The setup includes an actor pretending to be injured by a shock administered by a participant during a learning game. The authority figure instructions the participant to continue. Results show that most people continue even when the actor expresses severe pain and a desire to quit.Special attention is given to providing a complete set of information detailing the experimental setups and all possible variations of the experiment. Raw data is presented as well. Additionally, Milgram works to define the theory of the mechanisms which are in action during these experiments. Milgram explains the hierarchical structure of obedience from a cybernetic viewpoint of people moving from functioning autonomously to being an active part of an organization. Finally, Milgram works to address critics of his experiments by citing reasons their claims cannot be true.

Discussion
I enjoyed reading the details of Milgrams experiments. Up until this point I did not realize the magnitude of the many variations and validations he performed. I particularly enjoyed the presentation of the dialogue between the participant and the researcher. I also appreciated the methodical way that Milgram planned his experiments and validated them. In terms of Computer Science I think this first shows us how important user studies can be, and reminds us that we should spend the extra time to validate our results. It is also interesting to me, as my research is in rescue robotics, where we may want to attribute an authority role to a robot for a number of reasons. It is both challenging and interesting to try and apply Milgram's findings to this area.

Ethnography Week #6

As we continue our ethnography we have become a regular part of the environment in our area of study. The employees expect us to be present at our scheduled times and expect us to follow them to their tech calls. For the past week or so calls have been limited to repairs which must be made after classes. Because of this we have spent a lot of time observing leisure life in the tech office. An interesting aspect of the job is that their are constraints on when work can actually be done, so often times there is very little to do other than work on class assignments or socialize with others in the room.

We have not been back on site since Spring Break (we make our visits on Friday's). Spring break was the time that a number of installation projects were to take place, so we are curious to hear the results and see if there are any remaining issues. We have also started to consider the application we would like to develop for our ethnography group for Project 3. Since our study group is already very technologically oriented, they do have a number of custom made software systems in place for their employees to use. We will need to focus on areas that are not already covered by the existing systems, or perhaps focus on improving a particularly frustrating system.

For the most part we have observed the same things as previous weeks, occasionally a friend of an employee or someone from another department will stop by and socialize. But mostly, we see the culture as a place to pass time, socialize and occasionally perform tech duties. It is noteable that almost everyone working in the center has a personal interest in technology and the job is not difficult or challenging to them.

Monday, March 21, 2011

Obedience To Authority - Microblogs

Chapter One
Summary
Obedience is described as a method which connects the actions of people and a purpose. Milgram also explains the basics of the experiments that will be chronicled in the book. The motivation for his experiments is found in Nazi Germany.

Discussion
This seems to be a basic introduction to the motivation behind the research area and the experimental setup that will be used. I appreciate the clarity with which this is written so far, it doesn't leave much to interpretation.

Chapter Two
Summary
A very detailed explanation of the experimental design is given. Consideration is given to who should participate and how to keep the true meaning a secret. Other artifacts like advertisements for the study are disclosed.

Discussion
It's nice to see such a thorough experimental design set out in advance. When I first read about these experiments they seemed ad-hoc and mad-scientist like, but now they seem very scientific and methodically planned.

Chapter Three
Summary
Milgram begins to form a hypothesis about the outcomes of the experiment. He polls people (without telling them about the actor) and they all believe the teacher will stop administering shocks.

Discussion
I really like the idea of polling for a hypothesis. So often we just use our own intuition and hope to validate it, but this is a good idea. I don't see this style used very often in the area I study.

Chapter Four
Summary
The variations of the experiment are detailed. Proximity plays a role in altering the setup. In some designs the actor is very close, in others he is completely obscured. The basic outcome is that the closer victim lowers obedience to authority.

Discussion
It's interesting to see proximity playing a role in obedience. The idea that simply being physically closer to the actor made participants less likely to shock them is intriguing. Why is this? Is the theatrical aspect just more powerful or do they feel more connected?

Chapter Five
Summary
The various factors which Milgram views as effecting a participant's response are discussed. Several detailed accounts of interactions are given. A strong focus to accounts of the interaction is given.

Discussion
I really enjoyed reading the accounts of the interactions. In this case having that kind of raw data is much more interesting to me than the numbers. This also answered questions I had about people having special knowledge about electric shock and their actions. I was glad to see they did not obey.

Chapter Six
Summary
The experiment was repeated several times with varying factors. The location was changed for example. Various expansions of the experiment were made (women as participants, choice of schock level, etc.). The results were not very different.

Discussion
The validation of the study was crucial and Milgram definitely proves his point with all of these checks. Once again it was interesting to observe the effects of proxemics in action.

Chapter Seven
Summary
More accounts of the interactions during the experiment are given.

Discussion
These chapters are very enjoyable to read. I appreciate the inclusion of this data very much, as it really helps me understand the justifications participants had for their actions. Knowing the outcome is one thing, but understanding the reasoning is another-- this gives that extra viewpoint.

Chapter Eight
Summary
More variations are applied. Many manipulations concerning roles of authority and interactions between the actor and participant are described.

Discussion
This experiment must have never ended-- I can't figure out if Milgram was just hoping to invalidate his results eventually or was obsessed with detail. In any case it is great to have this much data to analyze.... this is useful for those wishing to look back at the study in detail.

Chapter Nine
Summary
Milgram begins to look at the effects of forming peer pressure type groups and monitoring participant reactions. For the most part the responses of the pressure group seems to be mimicked by the participant.

Discussion
This seems expected: if someone takes the initiative most everyone in an unfamiliar situation will follow their lead. This feels more like social pressure to conform rather than obey authority.

Chapter Ten
Summary
A substantial discussion concerning the place of obedience in society and the necessity of it evolves throughout this chapter. The basic argument is that a hierarchical structure of authority is required to keep social order. Discussion of agnetic state and binding factors takes place.

Discussion
This was more of the actual model or theory that Milgram was using to give reason and purpose to the results of the experiment. It was interesting, but a bit involved and heavy on terminology. Overall it is necessary to understand the broader implications.

Chapter Eleven
Summary
More detail is given about the theory behind authority. Explanations of binding factors and antecedent conditions, like family and rewards are discussed. The preservation of the agnetic state is explored through binding factors.

Discussion
More the theory is exposed here, but with a bit more of application and example. This was a bit easier to read than the previous chapter and seemed to reinforce the same ideas

Chapter Twelve
Summary
Milgram explores strains and the mechanisms that connect it to obedience. Milgram explains that increased strain will cause disobedience.

Discussion
I am not sure I agree with Milgrams viewpoint concerning the strain being some sort of force which completely decides the outcome. I would argue that knowledge could also cause disobedience.

Chapter Thirteen
Summary
Is it possible that people just wanted to hurt someone with no penalty? Milgram doesn't think so. People consistently chose lower voltage levels when given the option, indicating they weren't in it to torture another human.

Discussion
Agression is considered here and rightfully so. In my mind it would be rare that so many people could possible fall into an agression category should it exist, but this is a good check to have.

Chapter Fourteen
Summary
The skeptics viewpoint is considered and argued against in this chapter. Milgram discredits many alternative theories as to why the participants obeyed. Ultimately he compares this to Nazi Germany.

Discussion
It was good to see these criticisms addressed formally by Milgram. I still believe it was a factor of knowledge and trust which caused participants to obey.

Opening Skinner's Box - Full Blog


Reference
Opening Skinner's Box: Great Psychological Experiments of the Twentieth Century
Lauren Slater
W.W. Norton & Company, 2008

Summary
Opening Skinner's Box is a survey of the greatest, most influential and controversial psychological experiments of the twentieth century. The book focuses not only on the content of the research, but also the surrounding setting and the lasting effects of the findings. Beginning with B.F. Skinner himself, the author examines the concept of conditioning animals to perform based on a reward system and positive reinforcement. Continuing forward, Stanley Milgram's famous obedience to authority experiments are examined, with attempts to contact participants and examine the outcomes of their lives as a result of the experiment. As the book continues, Slater examines Rosenhan's experiments of gaining entry into a mental health facility with a limited set of symptoms. Slater is also sure to entertain readers by participating in the topics she is studying as much as possible, so much so as to try to be admitted into a mental health facility herself. Travelling into frightening territory, the bystander effect is explored, illustrating that the larger the group of people observing, the less likely individuals are to give assistance. Slater then invades the world of cult living as she brings the idea of cognitive dissonance to the foreground. Following this, the true nature of love is envisioned through the eyes of Harlow's monkey attachment experiments. Progressing onward the concept of addiction to drugs is presented from two viewpoints. Finally, the finer points of the human memory and brain are brought into the light with the idea of imagination filling in false memories and the physical aspects of the brain being examined.

Analysis
Skinner's Box is one of the most interesting reads I have ever encountered. The collection of fascinating researchers and their visions and projects into one volume proves to be quite enjoyable. Additionally, Slater adds interest to the facts by exploring the surrounding settings and even interviewing people connected to the research which was being conducted. Some of the most interesting topics were focused around the ideas of addiction and satisfaction. Computer Scientist can take away many things from Slater's examination of great psychological experiments. Of most importance may be to consider a human's complex nature when designing interfaces. Following design concepts wich account for the psychology of a human will greatly aide us in creating usable systems.