Documentation & Reflection

What have I done?

With the prototype, I was focusing on user testing.

There are two users have tested the prototype and I have found a few problems that I need to improve about.

The rough test plan:

Designer explain the concept to testers. Then let the tester clearly know the task.

Observe the interaction and tester's behaviour. Write down the notes about the mistake tester make, the facial expression of the tester, what the tester say and do during the task and the task completion situation.

Interview the tester after. (experience, opinion, confusion, etc)

One of the notes

Findings:

1. Without the visual physical design (flower), user do not know where to speak to. So when the feedback is not accurate, the user tends to think that is the voice not loud enough.
2. The keywords are hard to be recognized when they are in a sentence. User needs to speak the keywords separately to trigger the LED light.
3. The keywords are not enough. There are always more answers then the keywords pool contains.
4. The LED light is weak and hard to be noticed that they are enlightened.

What to do next?

Based on the findings above, I need to improve the prototype with the coding parts which can solve the findings 2 and 3.

Also, the visual design of the prototype will be built which can solve the problem from finding 1 and 4.

After the above step completed, more user test will be conducted which provide me with more alternatives before the final exhibition.

Will start to build my portfolio from early next week.

I have been working on the Arduino development and the coding parts. I set up two keywords pool which consists of negative/positive words. The voice recognition module can recognize the keywords and match with the keywords pool. The LED lights can react to the matches and show different lighten-up combination.

So far I have done “most” of the function parts in my concept. User speak "I'm home" to start the interaction process. The MP3 module broadcast "How's your day?" every time triggered by the voice keyword command " I'm home". Based on what user answer the question, the keyword contained can be analysed and matched with either key words pool. The positive key words pool matched lead to three GREEN lights up; the negative key words pool matched lead to three RED lights up.

The left is I may add extra output through the mp3 module which response to the feedback from the prototype testing that my flower should help the user to change their negative emotion to the positive one. I am still brainstorming about the solution but it won’t be hard to develop once I have the solution. Also, I will be working on how to combine the LED lights from Arduino with a flower. I got some material that I can shape a flower so I won’t be hard either.

What have I done?

I received three comments on my individual work. Here is the key points that be mentioned up:

1. Temper the bad emotions and transfer them into good feelings.
2. Cannot only use volume as a trigger
3. Should include more emotions that can be detected
4. Should set a proper detection distance
5. How to use one flower to represent individual's emotion
6. How to maintain the family functioning through it

The most confusing point in my prototype is that I use volume to simulate the anger level which further represents the emotion. Which I will include the keyword recognition feature in my next prototype to finalize this function.

What's the next step

The point that how to transfer the negative emotion to positive one remind me of my original idea that the flower can be a voice-changing toy which can amuse users. In my next step, I will 1. use keyword recognition feature to build two databases which stand for positive and negative emotion. 2. Use the mp3 module in Arduino to set the feedback from user's voice trigger 3. Connect the keyword database to LED light which different keyword can trigger a different reaction of LED lights.

It has been a long week because my team has to re-allocate the job and restart to build the Arduino again.

What have I done?

Plan changing !!!

Because three of our teammates are working on the same project together but separate the job instead of developing our own version, it caused the problem that we need to fully communicate about our allocation and focus. For the beginning, I was assigned to build the prototype based on the "boundary" function, which the Arduino can respond to the user's speech. I focused on the voice recognition technique which I thought that was my focus. However, in the end of last week, through the discussion and exchanging the progress and findings with other teammates, I found that the voice recognition in my part is repetitive with the key words recognition technique for Anna. Then we have to discuss again and finally decide to split the work which she takes charge of the voice recognition and I focus on the Arduino building, and ideally, our parts can be combined later. So I basically gave her advice about the findings on voice recognition I have got and started to build the Arduino.

Arduino building

My work is to prototype the function that LOME responds to the user when they arrive home and the petal colour can change based on the conversation happened. In this stage, I use the Sound Sensor module to enable the LED light to respond to the user’s speech.

I set a certain decibel which only when the sound’s decibel exceeds this number, the LED light will on. This set certain decibel number is used to simulate when the user is shouting out the negative emotion, and the red LED light will turn on. Otherwise, if the user’s emotion is plain or happy, the low decibel voice will not trigger the LED light.

What have I done

1. Setting the plan

By discussing with teammates, my part of work is to build the prototype of the 'boundary“ function, which was my focus from the previous study. The next step was to scope the function and decide to which step I will build, and to which step I need to simulate.

The "boundary" function is used to welcome the user home. It is to set a clear boundary between work to study life outside and the family life, which helps to prevent the user from bringing the external negative emotion into the family. Mixing up the emotion happened outside and inside the family can cause unnecessary conflict due to the inappropriate communication method. With this feature, LOME can give an initial evaluation of each user’s emotion after their work/study experience outside and set an indicator for the user him/herself and other family members to realize.

1. Research about the solution

For this week, I did much amount of research about what's the best platform or method to build the prototype. The key point of my part of work is to let the Arduino responded to voice. To realize the voice recognition, I studied different platforms and languages such as Unity and Python, also the API which can be called. It is still in progress but I found that Google API highly matched what I need, and using unity to call the voice recognition feature which built-in with PC system is also effective.

The next step

Next step I need to do more brainstorm about how I can realize the simulated function through Arduino which can be physically interacted with on a certain standard. The simulation has to be highly matched with the actual function which can make sure the testing of the prototype can provide useful and high-quality feedback on the future developing. Also, the interaction method has to be same or similar to the real one which gives me the foundation that I can later build on for the final delivery.

I have been searching for the existing technology that can help us with the voice changing part of our design. It seems like there is a bunch of toys that can be referenced. For the first prototype, Anna and I will work on the flower part together which we plan to simulate the voice changing and color changing of petals functions.

I will also work on the user experience testing, feedback evaluation and alternatives building. I will draft the user testing plan with the application of a suitable methodology for each prototype. This becomes challenging as we cannot physically meet, which is the main constraint for most of the user testing method. However, on the other hand, some online testing can be applied, which can provide us with qualitative data. I need to explore more about how to set up a most adaptive user testing plan so our prototype can be widely tested and have enough data for us to investigate more alternatives in our team domain.

1. The representation of the scale

As there are few feedback mentions about the death of flower will lead to the worse mental condition for the user, we use the colour changement of petals to represent the family conversation situation.

Red petal represents the negative emotion, and the green petal represents positive. Lome can recognize pre-set keywords which indicate negative emotion. Once ten keywords have been captured, one of the petals will turn into red. Six petals in total mean six levels which represent weather the family conversation is running positive and healthy.

2.Passive guidance

According to the first two directions concluded from the feedback which is “more interaction” and “ passive encourage/teach the user to be positive”, LOME can proactively invite the user to talk to it and physically interact with it.

Once the petals tend to be entirely red ( remains two green petals), LOME will broadcast “ Seems like you are not in a good mood, please come talk to me.”, such suggests to encourage the user to vent or relieve the stress. To start the vent/pour out, the user can remove one of the red petals and plug it onto the sprout. The sprout will keep growing during the vent/pour out process, which represents the emotion is being re-born. When vent/pours out complete, the user needs to move the petal from the sprout and place it back to the flower. The petal will get back to be green on the flower again, which represent the emotion has been transferred.

3. Encourage user to release their stress properly, but help them to transfer the emotion.

Noticed from the feedback that we shouldn’t limit the method for the individual to relief their stress. Shouting/swear can be an effective method for a certain group of users, so LOME respects this method. Once the user starts to talk to the flower, the flower will become a funny toy which can twist while replaying users’ talk in a funny way (change the tone). This can amuse the user and transfer their negative emotion effectively.

4. Set up a boundary between work and study (outside) and family (inside)

From the study of work-to-family conflict, we use LOME as a boundary to remind users that they have entered the family life and should not bring the emotion from work/study to home.

This function includes two main parts:

Firstly, the user needs to say “ I’m home” when they come home, and LOME will recognize and talk to the user to ask for an emotion check-up. For example, LOME will say “ Welcome home, how’s your day today?”, and the user will respond with a short sentence for LOME to analyse whether the user is in a good mood or not. Reacted to it, the flower will only have two conditions, fully green which represent the user is in a good mood or half green half red which means the user is suffering negative emotion from work/study.

Secondly, every time that the boundary conversation triggered, it is served as a re-start for LOME. When the family have multiple members, this process will be repeated. The condition of the flower can indicate the emotion condition for the other member when the user enters home.

For example, the wife coming home first, and she is in a good mood with work. After the boundary talks, the flower will turn into fully green, and the wife will notice that. Then the husband is coming home, but he had a bad day. Accordingly, the flower will turn into half green and half red, but the husband will notice that the flower was entirely green, which means his wife had a good day. The flower colour change can also be seen by the wife who let her know that the husband had a bad day.

This function can encourage family conversation, which is also pointed out in the feedback.

By reading the feedback, we classify the opinions into several big themes. We combined several similar opinions into some new possible features, for example, flower touching. We also critically consider some of the opinions but decide to not adopt it into the design for some realistic reason, for example, customize the incense or use the real flower.

In conclusion, there are four main directions for us to further develop on based on these feedbacks:

1. More interaction

There are a lot of feedback mentions increasing the interactive function such as touching or conversation. By considering the difficulty to develop the conversation function, we decided to add in more physical interaction features.

2. Passively encourage/teach the user to be positive

Few opinions mention about when users being mad or sad, the dead flower will make them feel worse which cannot provide positive support. In this case, the design should be more passive and offer the opportunity or attention for users to come and interact with it.

We are thinking of combining the above two points together. The new feature can be triggered by the sound/colour/smell that coming out from the flower to drive users' attention. The passive signal from the flower can contain the invitation of an interactive process with users which is a continually tangible/physical interaction.

3. Shouting/swear can be an effective method for a certain group of users. We cannot use a flower to replace this method.

This point is the most delightful one for me as it was me attached “negative" onto this kind of stress relief method. By reflecting on this point, I realized that the shouting/swear can be the most effective and efficient method for some people to express the stress. It may not be a negative or unhealthy way for them compares to hiding the emotion inside or take it out on someone else.

It reminds me of the research I did on the last step, the negative emotion is easy to be taken into the family life. I realized that the flower condition is lead to the monitoring of daily family life, but we ignored the part for users to express their stress/emotion from their study/work. We can do something to help the user express these emotions to prevent them bring these negative emotion into family life and messed around the scale of family functioning.

We can do something to encourage users to express their negative emotion from outside (work/study) first before they enter family life. They can use whatever method they are used to, but we can do something to help them transfer the emotion faster.

4. The wiser use of colour

Feedback reminds me that the use of colours can be more meaningful and detailed. We only thought about the four conditions of flower can be represented by different colour states, but the colours are not especially mean anything. However, colours actually have meaning. Red stands for anger, blue stands for peace, etc. We can wisely use this colour in our design as our design is about emotion.

We read through the poster from World Cafe, captured the key insights from the opinions on it. We grouped the insights into themes that provide further information for the next round of ideation after the group pitch.

Following characteristics are raised from the well-established consensus [1] which influencing family functioning:

· Parents actively build self-esteem in their children through compliments and demonstrating affection

· Family members actively listen to each other

· Family members argue without wounding each other (i.e. relying on the merits of an argument rather than personal attacks)

· Engage in family activities that all members enjoy

· Parents have time to spend with their children

A healthy language environment and a quality company contribute to a positive family functioning. It is interesting that from the above characteristics, we can see both “compliments and demonstrate” and “ argue” which are opposite but also compatible with each other. The argument is not easy to be controlled compares to compliments, which asks for awareness of the boundary of a positive argument for family members. It is the first point that I want to work on, which can be adapted in my project aims to assist family members in managing and controlling the argument level.

From the study of work-to-family conflict [2], which “boundary theory” suggests a cognitive, physical, and/or behavioral boundaries (“fences”) may exist between individuals’ work and family domains. It proves that without a clear boundary, family members will easy to bring the affections from work to home. It leads to a magnification of the details of despair in family life. It is the second point that I want to include in my project. I want to help users prevent amplifying the negative emotion which brings from work at home.

[1]National Drug Strategy (Australia), Australia, and Department of Health and Ageing, Training frontline workers: young people, alcohol & other drugs. Australia: publisher not identified, 2004.

[2] Z. Chen, G. N. Powell, and J. H. Greenhaus, “Work-to-family conflict, positive spillover, and boundary management: a person-environment fit approach,” Journal of Vocational Behavior, vol. 74, no. 1, pp. 82–93, Feb. 2009, doi: 10.1016/j.jvb.2008.10.009.

Feedback

Following points are the common content captured from the peer critique:

Good visualization

• Help people get aware

• Display emotions

• Get remind or alert

• Help to indicate how we treat others

• Decoration with interesting functions related to family moods.

Hard to implement

• Sound categorization

• Measurement

• Distinguish positive/negative loud noise

Concept

• Good idea to ensure home life is healthy

• Like it which try to capture different kinds of emotions

After reading through all this feedback, I realized that form and visualization of my concept is a pass but the deeper theory and implement need more works. I should consider more about how to recognize various sounds and analyze them through computer system. I think I can have some inspiration through the World Cafe next week which I can further build on my concept.

People's behavior and mental condition are influenced by their family environment, their experiences, and their sense of themselves. The maintenance of relations under the same roof is a subject that everyone needs to keep learning and to try.

LOME is an interactive home-installation. It can" listen" to how people behave and say under the roof, and the flower's condition reacts.

LOME can CAPTURE the following SOUND / VOICE at home :

~ Laughing

~ Love expression

~ Polite expression

~ Encouraging words

~ Praise

These features will raise the scale and the vibe at home will be judges as POSITIVE. It represents that the family environment is WARM and HARMONIOUS. The scale will keep within the safe area (green), and the flower will stay in FULL BLOOM.

Also, when these SOUND / VOICE be captured by LOME at home :

~ Crying

~ Loud noise

~ Shouting

~ Bad words

The vibe at home will be judged as NEGATIVE. It represents that the family environment is DARKand DYSFUNCTIONAL. The scale will drop to the dangerous area (red), and the flower will turn to black and die.

By this point, the dome represents home/family, and the flower represent the life and relationship between members. LOME works as a supervisor and reminder for people to be aware of the influence of their expression. It helps people to learn how to restrain themselves and form a healthy and warm home environment.

Reflection of class:

The first two contact sessions in week1 help me to form a basic structure of the connection between HCI and physical computing, also an initial image of which domain I want to work on.

For the first contact, I review this course by sharing the opinions raised by others. The main point I noticed is that Lorna pointed out the point for this course is not only about how big is the final exhibition/product but mainly values the process. It is valuable advice that I think I can recall my memory and experience in Design Thinking and Social mobile. Besides the final exhibition, the design process is the most important method to measure will the product obey the HCI principle and also make sense.

The second contact run around HCI challenges. The topic I working on is Well-being, Health & Eudaimonia. After the group discussion, I find that eudaimonia is like long-time happiness with more phycological affection. It reminds me to think about the mental health aspect when coming up with the design.

Poker's game is a good way to brainstorm and come up with ideas. It is difficult but also fun to play with which encourages me to think outside the box.

The work is done:

For the first week, it is a self-ideation process and reflects on which one or more HCI challenges can be considered in the design. I formed a basic structure of my design which I want to use voice as an interaction medium. Also, the concept of my design would be related to human beings’ mental comfort and relationship harmony.

How it relate:

My design concept reflects on my HCI challenge which is about metal happiness. Also, one of the examples in class which is kind of like a mood-triggered system minds me that although the mood is hard to be captured, what people say can reflect the mood. This is why I want to use voice as my interactive medium as it can be further transferred to the information about the atmosphere or temper.

Work to do:

After deciding on the main concept and interactive medium, the next step is to add details to my design. I need to think of what will be the format of the carrier of my interactive medium and how to make it “every day”, “make sense” and “fun to play with”. Also, I need to target my user group by detailing my concept and aim.

Hey there. My name is Jenny and I'm from Interactive Design.

So far, all I know about this course is that we need to develop a physically intelligent/smart/electronic object which can be fun to interact with and closely connect with the human being.

I am confident about the life cycle of the design process which includes idea generation, research, prototype testing, usability test... BUT I have nearly ZERO knowledge about programming related to hardware. And that's what I need to learn and discover.