Skip to Content

Test

Version 28.1 by Mathieu Jung-Muller on 2022/03/06 11:22

Problem statement and research questions

This project uses a Social Cognitive Engineering (SCE) approach to guide the design and research process. The SCE method provides a systematic approach to our study of robots for PwDs. The main goal of our application is to improve the well-being of the person with dementia (PwD) and of those living with them.

For the prototype that we have designed now, these are some research questions that we want to address.

  1. Are PwDs able to use our prototype smoothly? 
  2. Can medication/meal/activities reminders help PwDs to improve their lives?
  3. Can activity breakdown help people better manage their lives? The effects of using such technology in a natural setting.
  4. Does the robot have a positive effect on people’s emotional state?

suggestions:
1. Are the different stakeholders able to use our prototype smoothly?

  1. Does the prototype allow the PwD a greater autonomy in their day-to-day life?
  2. Does the prototype improve the emotional state of the PwD and their relatives?

Method

We target some PwDs that are similar to the personas we described. The PwDs will be observed during the evaluation as well as be reviewed after the evaluation to obtain the data and feedback.

Participants

For this study, we intend to find 20 people. The participants will all be early-stage dementia patients and we will make sure to include at least 5 males and 5 females to avoid gender imbalance. The group will include at least 5 people that are 50 or younger and we expect most of the other participants to be between 75 and 90. The participants will all join the same 3-hour session and will be given 5 to 10 minutes to perform their tasks.

Experimental design

Since our evaluating process is relatively short, we use within-subject, which means each participant goes through all conditions. In this way, our experiments tend to have more statistical power and less variability. 

Tasks

All participants will go through our designed testing process, which includes medication/meal/activity reminder, listening to the music, activity breakdown. 

  1. Medication/meal/activity reminder: The robot will remind the patient of daily activities, through which we can see the effectiveness according to their reactions.
    (1. Listen to music: PwDs will test some music functionalities in this part. PwDs can choose music by themselves by interacting with the robot or they can listen to random music which is picked according to their preferences.) for later?
  2. Activity breakdown: In this part, the robot will break down some complex tasks into a list of simple tasks and PwDs can follow the steps to accomplish complex tasks.

Measures

We are planning to measure our prototype both behaviorally and emotionally. Behavior data is the actions that PwDs take when interacting with the robot, which is the quantitative data. Emotional data is the change of expression and emotional state, which is the qualitative data.

Procedure

At the start of the evaluation, all participants will be gathered in the same room, together with the researchers and relatives. We will explain the whole evaluation process, provide simple instructions, explain that participation in the evaluations is voluntary, and participants are free to stop the evaluation at any time. Then, we will explain how the gathered data will be analyzed and help us to improve our prototype. All data will be kept private. We will also emphasize that if there is anything the participants dislike, they should let us know. The purpose of the study is to find out what they think of the prototype, and their honesty is greatly appreciated. Then, we will ask them to sign the consent form.
We will simulate a home setting, which is the most common scenario for PwDs. All PwDs are going to complete the evaluation separately.

  1. Our robot will give corresponding prompts. For example, “It’s time to have lunch”, “Medicine time!”, “Today is your birthday. Happy Birthday!”. 
  2. We will record the reaction and behavior of participants after they heard the prompts.
    (1. Our robot will ask: Do you want some music? Participants can choose the music they like through the user interface of our robot. Or they can play random music.) for later ?
  3. Observe and write down participants’ reactions and the change of emotional state.
  4. To test the activity breakdown function, a HPC who is familiar with the patient will have previously chosen a complex task the patient has difficulty doing, such as cooking a meal or doing some exercise. Pepper will say the different steps and accompany the PwD throughout the activity.
  5. Record the reactions of participants and evaluate how the is activity accomplished.

Material

  1. Instructions. To ensure the same situation for all participants, we will give out instructions before the start of the evaluation. All researchers and participants will go through the same evaluation process in general, but the activity will be different based on the particular situation of each PwD.
  2. Consent form. To protect the privacy of participants and ensure the evaluation process goes smoothly, we will ask participants to sign a consent form, indicating they are willing to take part in the evaluation and the data gathered from the experiment will be analyzed by researchers.
    (1. Questionnaire. To get the best performance of our robot, participants need to take a questionnaire about their music preferences. We will play music according to each participant’s preference.) for later?
  3. Pepper robot. Our robot is programmed using Choregraphe. The robot will have the same behavior and prompts for every participant.
  4. Video cameras. Two video cameras will be used to record all sessions from two opposing angles.

Results

After the experiments and the evaluation, all the data will be collected and analyzed. There can be many different situations in the evaluation process.

  1. In the calendar reminder session, almost all participants have the right behavior after the robot’s prompts. For example, they would take the medicine after hearing “Medicine time!”, they would eat food after the prompt “It’s time to have lunch!”. However, there are still two participants who cannot perform the right behavior. One of them cannot hear the prompts clearly, because he has hearing problems. The other one shows a confused expression after hearing the instruction, which indicates he has some problems with understanding the prompts.
  2. For the music session, 15 of the participants can complete the music questionnaire quite well, which helps us to understand their music preferences and will have a positive influence on later evaluation. However, three of them can finish the questionnaire to some extent, but they cannot remember the name of some artists or the name of songs they like. Two of them are not able to complete the questionnaire.
    When they are listening to the music, 18 of the participants show positive reactions, such as singing along, smiling, swinging to the music. The man with the hearing problems has little reaction. One of the participants who fail to complete the questionnaire also does not too much reaction. Maybe because he does not like the music or never heard the music before.
  3. In the Activity Breakdown session, 13 of the participants finish the tasks successfully. They can follow the robot’s instructions without any problems. Five of them follow the first few steps but fail in one of the tasks or do not have enough time to follow the robot. Two of them cannot perform the task from the beginning.

Discussion

Conclusions