prompt
stringlengths 1.67k
3.17k
|
|---|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 54, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 5, "sad": 5, "angry": 5, "neutral": 75, "surprised": 5, "fear": 5}, "eyes_state_percent": {"open": 86, "closed": 14}, "looking_state_percent": {"looking_at_screen": 62, "looking_away": 38}, "total_sampled_frames": 135, "most_frequent_emotion": "neutral"}, "transcript": "It's so boring. I don't want to do this homework at all. I'd rather play.", "previous_progress": {"focus": {"current": 45, "goal": 100}, "self_control": {"current": 80, "goal": 100}, "social": {"current": 62, "goal": 100}, "confidence": {"current": 70, "goal": 100}}, "previous_chat": "Supporter: How's your homework going?\nChild: Ugh, I just can't focus."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I know, homework can be so boring. I'd rather play too! Maybe we can make it a bit more fun or think of a reward for when you're done. I bet you can finish it if you try, and I'll cheer you on!", "updated_progress": {"focus": {"current": 50, "goal": 100}, "self_control": {"current": 80, "goal": 100}, "social": {"current": 62, "goal": 100}, "confidence": {"current": 70, "goal": 100}}, "feedback": "How about this: do homework for 10 minutes, then take a 5-minute play break, and repeat. We can make it a little game so it’s less boring."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 42, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 5, "sad": 50, "angry": 5, "neutral": 30, "surprised": 5, "fear": 5}, "eyes_state_percent": {"open": 71, "closed": 29}, "looking_state_percent": {"looking_at_screen": 64, "looking_away": 36}, "total_sampled_frames": 128, "most_frequent_emotion": "sad"}, "transcript": "I tried to do my reading homework, but I kept getting distracted and didn't finish. I really can't focus.", "previous_progress": {"focus": {"current": 28, "goal": 100}, "self_control": {"current": 77, "goal": 100}, "social": {"current": 53, "goal": 100}, "confidence": {"current": 61, "goal": 100}}, "previous_chat": "Supporter: Did you finish your homework?\nChild: Not yet. I got distracted again."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I get distracted easily too sometimes. It can be so hard to stay focused. Maybe I can read my book at the same time as you do your homework, so it feels like we're doing it together! We can check on each other and try to stay on track.", "updated_progress": {"focus": {"current": 33, "goal": 100}, "self_control": {"current": 77, "goal": 100}, "social": {"current": 53, "goal": 100}, "confidence": {"current": 61, "goal": 100}}, "feedback": "Maybe try to remove distractions around you. For example, turn off the TV and put your toys in another room until you're done. That might help you focus better."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 65, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 5, "sad": 15, "angry": 0, "neutral": 15, "surprised": 5, "fear": 60}, "eyes_state_percent": {"open": 81, "closed": 19}, "looking_state_percent": {"looking_at_screen": 58, "looking_away": 42}, "total_sampled_frames": 148, "most_frequent_emotion": "fear"}, "transcript": "I forgot to work on my science project and it's due tomorrow! I'm freaking out and I don't know what to do.", "previous_progress": {"focus": {"current": 37, "goal": 100}, "self_control": {"current": 79, "goal": 100}, "social": {"current": 58, "goal": 100}, "confidence": {"current": 74, "goal": 100}}, "previous_chat": "Supporter: Is your science project ready?\nChild: Oh no, I totally forgot about it!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh no, that is stressful. It happened to me once too. Maybe you can still do some parts of it tonight. I can stay up with you and help if you want! And if it's too much, maybe you could tell your teacher you forgot and ask for an extra day. It's going to be okay.", "updated_progress": {"focus": {"current": 40, "goal": 100}, "self_control": {"current": 79, "goal": 100}, "social": {"current": 58, "goal": 100}, "confidence": {"current": 74, "goal": 100}}, "feedback": "Try writing down the most important things you need to do for the project and do those first. If you can't finish everything, you could ask the teacher nicely for a bit more time."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 48, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 76, "closed": 24}, "looking_state_percent": {"looking_at_screen": 53, "looking_away": 47}, "total_sampled_frames": 104, "most_frequent_emotion": "sad"}, "transcript": "During the math test today, I kept getting distracted and now I'm pretty sure I failed it. I couldn't focus at all.", "previous_progress": {"focus": {"current": 22, "goal": 100}, "self_control": {"current": 81, "goal": 100}, "social": {"current": 66, "goal": 100}, "confidence": {"current": 55, "goal": 100}}, "previous_chat": "Supporter: How did your math test go?\nChild: Not good... I messed up."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm sorry it was hard. Tests can be really stressful. Sometimes I get distracted too. One bad test doesn't mean you failed everything. Next time, maybe try to take a deep breath if you start to lose focus. You've got this, and I'll help you practice if you want.", "updated_progress": {"focus": {"current": 17, "goal": 100}, "self_control": {"current": 81, "goal": 100}, "social": {"current": 66, "goal": 100}, "confidence": {"current": 55, "goal": 100}}, "feedback": "In the next test, try closing your eyes for a moment and taking a deep breath when you feel distracted. Then look back at your paper and focus on one question at a time."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 50, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 90, "surprised": 5, "fear": 5}, "eyes_state_percent": {"open": 83, "closed": 17}, "looking_state_percent": {"looking_at_screen": 71, "looking_away": 29}, "total_sampled_frames": 117, "most_frequent_emotion": "neutral"}, "transcript": "I got in trouble at school today because I was daydreaming in class. The teacher yelled at me for not paying attention.", "previous_progress": {"focus": {"current": 30, "goal": 100}, "self_control": {"current": 88, "goal": 100}, "social": {"current": 71, "goal": 100}, "confidence": {"current": 63, "goal": 100}}, "previous_chat": "Supporter: You look upset. What happened at school?\nChild: I got in trouble..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh man, I'm sorry your teacher yelled at you. It feels awful to get in trouble, especially if you didn't mean to do anything bad. I know you try to pay attention. Maybe you can sit closer to the teacher or ask for help staying focused. I'm sure you can do it.", "updated_progress": {"focus": {"current": 30, "goal": 100}, "self_control": {"current": 88, "goal": 100}, "social": {"current": 71, "goal": 100}, "confidence": {"current": 63, "goal": 100}}, "feedback": "Perhaps you could tell your teacher that you daydream sometimes without meaning to. Maybe they can let you sit at the front or give you a gentle reminder when you drift off. That might help."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 59, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 60, "angry": 0, "neutral": 30, "surprised": 5, "fear": 5}, "eyes_state_percent": {"open": 79, "closed": 21}, "looking_state_percent": {"looking_at_screen": 55, "looking_away": 45}, "total_sampled_frames": 110, "most_frequent_emotion": "sad"}, "transcript": "We had a group project today and I kept zoning out. My group members got mad at me for not helping enough.", "previous_progress": {"focus": {"current": 24, "goal": 100}, "self_control": {"current": 85, "goal": 100}, "social": {"current": 68, "goal": 100}, "confidence": {"current": 57, "goal": 100}}, "previous_chat": "Supporter: How was working with your group today?\nChild: Not great..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh no, that stinks. Group projects can be hard when it's tough to focus. I'm sure you didn't mean to slack off. Maybe you can explain to them that you sometimes get distracted. Next time you could ask for a specific part to do so you can focus on that. I bet you can show them you're trying.", "updated_progress": {"focus": {"current": 27, "goal": 100}, "self_control": {"current": 85, "goal": 100}, "social": {"current": 68, "goal": 100}, "confidence": {"current": 57, "goal": 100}}, "feedback": "You could tell your group you're sorry and that you'll do your best on the next part. Maybe ask them to give you a clear job in the project. That way you know exactly what to focus on."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 53, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 40, "angry": 0, "neutral": 50, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 74, "closed": 26}, "looking_state_percent": {"looking_at_screen": 59, "looking_away": 41}, "total_sampled_frames": 124, "most_frequent_emotion": "sad"}, "transcript": "I left my math textbook at school again and now I can't do my homework. Mom is so angry with me. I feel so stupid for forgetting.", "previous_progress": {"focus": {"current": 34, "goal": 100}, "self_control": {"current": 72, "goal": 100}, "social": {"current": 60, "goal": 100}, "confidence": {"current": 48, "goal": 100}}, "previous_chat": "Supporter: Hey, what's wrong?\nChild: I messed up. I forgot my book at school."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh no, forgetting stuff is so frustrating. I'm sorry your mom is angry. It doesn't mean you're stupid at all. We all forget things sometimes. Maybe we can find the homework problems online or borrow a friend's book. And we can think of a way to help you remember next time.", "updated_progress": {"focus": {"current": 36, "goal": 100}, "self_control": {"current": 72, "goal": 100}, "social": {"current": 60, "goal": 100}, "confidence": {"current": 48, "goal": 100}}, "feedback": "Maybe you could put a checklist in your backpack or set an alarm on your phone to remind you to pack everything before you leave school. That might help you not forget your book."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 55, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 50, "neutral": 30, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 82, "closed": 18}, "looking_state_percent": {"looking_at_screen": 63, "looking_away": 37}, "total_sampled_frames": 150, "most_frequent_emotion": "angry"}, "transcript": "I got yelled at in class for shouting out an answer without raising my hand. I was just excited but now I feel bad.", "previous_progress": {"focus": {"current": 59, "goal": 100}, "self_control": {"current": 25, "goal": 100}, "social": {"current": 80, "goal": 100}, "confidence": {"current": 67, "goal": 100}}, "previous_chat": "Supporter: Why do you look upset?\nChild: I got in trouble at school."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I sometimes blurt out answers too when I get excited. It's really hard to wait your turn, I know. Don't feel too bad. Maybe next time just take a breath and raise your hand first. The teacher will see you're trying, and I know you can do it!", "updated_progress": {"focus": {"current": 59, "goal": 100}, "self_control": {"current": 27, "goal": 100}, "social": {"current": 80, "goal": 100}, "confidence": {"current": 67, "goal": 100}}, "feedback": "When you know an answer, try to count to 3 in your head before you speak. Raise your hand while you count. It gives you a moment to remember to wait your turn."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 58, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 30, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 73, "closed": 27}, "looking_state_percent": {"looking_at_screen": 66, "looking_away": 34}, "total_sampled_frames": 158, "most_frequent_emotion": "sad"}, "transcript": "I kept getting out of my seat in class today because I felt super restless. My teacher got angry and made me stand in the back of the room.", "previous_progress": {"focus": {"current": 73, "goal": 100}, "self_control": {"current": 20, "goal": 100}, "social": {"current": 78, "goal": 100}, "confidence": {"current": 65, "goal": 100}}, "previous_chat": "Supporter: Hey, how was class today?\nChild: Not so great..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm sorry that happened. Sometimes it's really hard to sit still, I get it. I know you weren't trying to misbehave. Maybe you could ask the teacher if you can have a quick stretch break when you feel jumpy. Or you could try using a stress ball or fidget quietly at your desk. You're not bad, you just have lots of energy.", "updated_progress": {"focus": {"current": 73, "goal": 100}, "self_control": {"current": 23, "goal": 100}, "social": {"current": 78, "goal": 100}, "confidence": {"current": 65, "goal": 100}}, "feedback": "You might talk to your teacher and say that you get really restless. Maybe they will let you walk at the back for a minute or give you a special squishy ball to use so you can stay in your seat longer."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 49, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 78, "closed": 22}, "looking_state_percent": {"looking_at_screen": 52, "looking_away": 48}, "total_sampled_frames": 161, "most_frequent_emotion": "sad"}, "transcript": "I yelled at my friend and called him a mean name when I was angry. Now he's really upset and won't talk to me. I feel awful.", "previous_progress": {"focus": {"current": 57, "goal": 100}, "self_control": {"current": 26, "goal": 100}, "social": {"current": 40, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "previous_chat": "Supporter: You seem down. Did something happen with your friend?\nChild: I really messed up."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh no, I'm sorry. I know you didn't really mean what you said, you were just angry in the moment. Sometimes we say things we regret. The best thing to do is apologize. Tell him you're really sorry and that you didn't mean it. I'm sure he'll understand if he knows you feel bad about it.", "updated_progress": {"focus": {"current": 57, "goal": 100}, "self_control": {"current": 26, "goal": 100}, "social": {"current": 45, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "feedback": "Try writing a short note or text saying you're sorry and that you didn't mean what you said. Let him know that you value him as a friend. He might need a little time, but I bet he'll forgive you."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 60, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 55, "neutral": 15, "surprised": 5, "fear": 25}, "eyes_state_percent": {"open": 88, "closed": 12}, "looking_state_percent": {"looking_at_screen": 69, "looking_away": 31}, "total_sampled_frames": 173, "most_frequent_emotion": "angry"}, "transcript": "I got so angry playing my video game that I threw my controller and it broke. Now my parents are furious with me. I feel really bad.", "previous_progress": {"focus": {"current": 66, "goal": 100}, "self_control": {"current": 29, "goal": 100}, "social": {"current": 75, "goal": 100}, "confidence": {"current": 62, "goal": 100}}, "previous_chat": "Supporter: You look upset. What happened?\nChild: I messed up big time."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh man, I know that feeling. I've gotten so mad at a game before that I wanted to throw things too. I'm sorry your controller broke. It was an accident from being upset. I think if you apologize to your parents and show them you're trying to stay calm next time, they'll forgive you. We all lose our temper sometimes.", "updated_progress": {"focus": {"current": 66, "goal": 100}, "self_control": {"current": 29, "goal": 100}, "social": {"current": 75, "goal": 100}, "confidence": {"current": 62, "goal": 100}}, "feedback": "Next time you feel that angry, try to do something that won't hurt anything. Like, punch a pillow or squeeze a stress ball instead of throwing your controller. It can help you cool down without breaking stuff."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 62, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 30, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 70, "closed": 30}, "looking_state_percent": {"looking_at_screen": 47, "looking_away": 53}, "total_sampled_frames": 128, "most_frequent_emotion": "sad"}, "transcript": "My little brother grabbed my favorite toy, so I got really mad and I hit him. Then he started crying and my parents yelled at me. I know hitting was wrong, but I was just so angry.", "previous_progress": {"focus": {"current": 69, "goal": 100}, "self_control": {"current": 18, "goal": 100}, "social": {"current": 52, "goal": 100}, "confidence": {"current": 59, "goal": 100}}, "previous_chat": "Supporter: Why are you upset?\nChild: I got in trouble for fighting with my brother."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I get it, siblings can be SO annoying! I've felt that angry at my brother too. But hitting is not okay, and I know you feel bad about it. Maybe you should apologize to him when he's calm. Next time, if you feel that mad, try walking away or even yelling into a pillow instead of at him. I'm here for you.", "updated_progress": {"focus": {"current": 69, "goal": 100}, "self_control": {"current": 13, "goal": 100}, "social": {"current": 52, "goal": 100}, "confidence": {"current": 59, "goal": 100}}, "feedback": "When you get super angry, try to leave the room and cool down instead of hitting. You could take some deep breaths or talk to an adult about it. Then you can handle it without anyone getting hurt."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 54, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 60, "surprised": 0, "fear": 40}, "eyes_state_percent": {"open": 83, "closed": 17}, "looking_state_percent": {"looking_at_screen": 59, "looking_away": 41}, "total_sampled_frames": 121, "most_frequent_emotion": "neutral"}, "transcript": "I spent all of my allowance the same day I got it on a new video game. Now I don't have any money left for the book I really wanted, and my mom said I made a bad choice.", "previous_progress": {"focus": {"current": 64, "goal": 100}, "self_control": {"current": 22, "goal": 100}, "social": {"current": 67, "goal": 100}, "confidence": {"current": 72, "goal": 100}}, "previous_chat": "Supporter: Did you get that book you wanted?\nChild: No... I kind of messed up with my money."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Aw, that stinks. I actually did something similar before – I spent my money on candy and then had none left for a toy I really wanted. It feels pretty lousy. Maybe next time, you can save part of your allowance for later and spend a little bit now. It's hard to wait, but it might be worth it. Don't worry, you'll do better next time.", "updated_progress": {"focus": {"current": 64, "goal": 100}, "self_control": {"current": 24, "goal": 100}, "social": {"current": 67, "goal": 100}, "confidence": {"current": 72, "goal": 100}}, "feedback": "One idea: when you get your allowance, put half of it aside for savings and only spend the other half. That way you'll always have some saved up for the things you really want."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 57, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 80, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 91, "closed": 9}, "looking_state_percent": {"looking_at_screen": 72, "looking_away": 28}, "total_sampled_frames": 141, "most_frequent_emotion": "neutral"}, "transcript": "I was playing a board game with my friend, but I kept taking my turn too early and interrupting him. He got really annoyed at me for not waiting.", "previous_progress": {"focus": {"current": 70, "goal": 100}, "self_control": {"current": 31, "goal": 100}, "social": {"current": 76, "goal": 100}, "confidence": {"current": 68, "goal": 100}}, "previous_chat": "Supporter: How was game night with your friend?\nChild: I kind of messed it up..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oops, I know waiting for your turn can be really hard when you're excited. I've done that in games too and it bugged my friends. It's good that you noticed it. Maybe next time, you can try counting to ten or humming a little tune in your head while you wait so you don't jump in early. I'm sure your friend will forgive you if you say sorry.", "updated_progress": {"focus": {"current": 70, "goal": 100}, "self_control": {"current": 34, "goal": 100}, "social": {"current": 76, "goal": 100}, "confidence": {"current": 68, "goal": 100}}, "feedback": "Tell your friend you're sorry for jumping ahead. Next game, every time it's someone else's turn, try to quietly count to ten or take a deep breath. It can help you stay patient until it's your turn."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 52, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 80, "closed": 20}, "looking_state_percent": {"looking_at_screen": 54, "looking_away": 46}, "total_sampled_frames": 142, "most_frequent_emotion": "sad"}, "transcript": "Earlier today my mom was talking on the phone and I kept interrupting her because I wanted to ask her something. She got really angry at me for not waiting.", "previous_progress": {"focus": {"current": 60, "goal": 100}, "self_control": {"current": 44, "goal": 100}, "social": {"current": 79, "goal": 100}, "confidence": {"current": 71, "goal": 100}}, "previous_chat": "Supporter: Is everything okay? You seem bothered.\nChild: Mom got mad at me."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh, I know how that feels. Sometimes I get so excited I interrupt my parents too, and they get annoyed. It doesn't mean you're bad; you just had something important to say. Maybe next time you can gently tap her arm or say 'excuse me' and wait a little. That way she knows you need her without feeling like you're interrupting on purpose.", "updated_progress": {"focus": {"current": 60, "goal": 100}, "self_control": {"current": 46, "goal": 100}, "social": {"current": 79, "goal": 100}, "confidence": {"current": 71, "goal": 100}}, "feedback": "Try to remember to say 'excuse me' and wait until your mom looks at you before you start talking. It shows you're trying to be polite, and she'll be less likely to get mad."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 61, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 60, "neutral": 20, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 88, "closed": 12}, "looking_state_percent": {"looking_at_screen": 62, "looking_away": 38}, "total_sampled_frames": 137, "most_frequent_emotion": "angry"}, "transcript": "I yelled at my dad and said \"I don't care!\" when he told me to stop playing and do my homework. He got really mad and now I'm grounded. I feel bad for shouting at him.", "previous_progress": {"focus": {"current": 75, "goal": 100}, "self_control": {"current": 38, "goal": 100}, "social": {"current": 83, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "previous_chat": "Supporter: You seem upset. Everything okay at home?\nChild: I got into a big argument with my dad."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Uh oh, yelling at parents is tough. I know you didn't mean it like that, you were just frustrated. It happens to me too sometimes. It might help to say sorry to your dad later and tell him you were upset but didn't mean to shout. He might be angry now, but he'll understand if you show you regret it. Try to stay calm next time, even if you're annoyed.", "updated_progress": {"focus": {"current": 75, "goal": 100}, "self_control": {"current": 39, "goal": 100}, "social": {"current": 83, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "feedback": "When you feel like shouting at your dad, try to pause and take a breath. You could say, \"I'm upset right now\" instead of yelling. It might help him understand you without the shouting."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 40, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 40, "sad": 0, "angry": 0, "neutral": 60, "surprised": 0, "fear": 0}, "eyes_state_percent": {"open": 92, "closed": 8}, "looking_state_percent": {"looking_at_screen": 58, "looking_away": 42}, "total_sampled_frames": 140, "most_frequent_emotion": "neutral"}, "transcript": "Guess what? I actually finished all my homework by myself today, and it didn't even take that long! I managed to stay on task the whole time.", "previous_progress": {"focus": {"current": 40, "goal": 100}, "self_control": {"current": 73, "goal": 100}, "social": {"current": 82, "goal": 100}, "confidence": {"current": 77, "goal": 100}}, "previous_chat": "Supporter: Hey, how was homework today?\nChild: It went really well!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Wow, that's awesome! I'm so proud of you for finishing all your homework. See, you proved to yourself that you can do it when you focus. I bet it feels good to have it done, right? Maybe you can relax and do something fun now. Keep up the great work – this is a big win!", "updated_progress": {"focus": {"current": 50, "goal": 100}, "self_control": {"current": 73, "goal": 100}, "social": {"current": 82, "goal": 100}, "confidence": {"current": 77, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 47, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 50, "sad": 0, "angry": 0, "neutral": 40, "surprised": 10, "fear": 0}, "eyes_state_percent": {"open": 90, "closed": 10}, "looking_state_percent": {"looking_at_screen": 68, "looking_away": 32}, "total_sampled_frames": 119, "most_frequent_emotion": "happy"}, "transcript": "My teacher told me today that I did a great job paying attention in class! She actually noticed that I didn't get distracted like I usually do. I'm so happy!", "previous_progress": {"focus": {"current": 82, "goal": 100}, "self_control": {"current": 66, "goal": 100}, "social": {"current": 89, "goal": 100}, "confidence": {"current": 80, "goal": 100}}, "previous_chat": "Supporter: You seem really happy today!\nChild: My teacher praised me in class!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "That's fantastic news! I'm so happy for you. See, all your effort to pay attention is really showing. It must feel so good to have your teacher notice and praise you. You should be proud of yourself. This is proof you can do it, and I know you can keep it up!", "updated_progress": {"focus": {"current": 90, "goal": 100}, "self_control": {"current": 66, "goal": 100}, "social": {"current": 89, "goal": 100}, "confidence": {"current": 80, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 50, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 70, "surprised": 0, "fear": 30}, "eyes_state_percent": {"open": 85, "closed": 15}, "looking_state_percent": {"looking_at_screen": 72, "looking_away": 28}, "total_sampled_frames": 152, "most_frequent_emotion": "neutral"}, "transcript": "I got really angry at my sister today, but instead of yelling or hitting, I remembered what you told me. I went to my room and took some deep breaths. And it actually helped me calm down! I didn't end up yelling at all.", "previous_progress": {"focus": {"current": 58, "goal": 100}, "self_control": {"current": 30, "goal": 100}, "social": {"current": 74, "goal": 100}, "confidence": {"current": 79, "goal": 100}}, "previous_chat": "Supporter: How are you feeling today? You mentioned your sister was bugging you.\nChild: Yeah, she was, but you'll be proud of me..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh my gosh, that's amazing! I'm really proud of you for trying the deep breaths and walking away. It takes a lot of self-control to do that when you're super angry. And see, it worked! You stayed calm and no one got hurt or upset. I bet your sister was surprised too. This is a huge win. Keep using that strategy – you're getting really good at handling your anger!", "updated_progress": {"focus": {"current": 58, "goal": 100}, "self_control": {"current": 40, "goal": 100}, "social": {"current": 74, "goal": 100}, "confidence": {"current": 79, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 51, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 50, "sad": 0, "angry": 0, "neutral": 40, "surprised": 10, "fear": 0}, "eyes_state_percent": {"open": 94, "closed": 6}, "looking_state_percent": {"looking_at_screen": 75, "looking_away": 25}, "total_sampled_frames": 159, "most_frequent_emotion": "happy"}, "transcript": "I actually went up to a new girl at recess today and asked if I could play with her. And she said yes! We played soccer together and it was really fun. I think I made a new friend!", "previous_progress": {"focus": {"current": 82, "goal": 100}, "self_control": {"current": 69, "goal": 100}, "social": {"current": 30, "goal": 100}, "confidence": {"current": 88, "goal": 100}}, "previous_chat": "Supporter: Any news at school? How was recess?\nChild: I have something awesome to tell you!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "That's so awesome! I'm really proud of you for being brave and asking to join in. See, most of the time people do say yes if you ask nicely. It sounds like you had a great time. Making a new friend is a big deal, especially when you were feeling lonely before. This makes me so happy to hear! I knew you could do it.", "updated_progress": {"focus": {"current": 82, "goal": 100}, "self_control": {"current": 69, "goal": 100}, "social": {"current": 40, "goal": 100}, "confidence": {"current": 88, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 46, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 70, "surprised": 0, "fear": 30}, "eyes_state_percent": {"open": 87, "closed": 13}, "looking_state_percent": {"looking_at_screen": 65, "looking_away": 35}, "total_sampled_frames": 147, "most_frequent_emotion": "neutral"}, "transcript": "I apologized to my friend for what I said the other day. He actually forgave me and we played together again at lunch! I'm so relieved we're okay now.", "previous_progress": {"focus": {"current": 79, "goal": 100}, "self_control": {"current": 52, "goal": 100}, "social": {"current": 35, "goal": 100}, "confidence": {"current": 84, "goal": 100}}, "previous_chat": "Supporter: How did it go with your friend? Did you talk to him?\nChild: Yeah, I did and it went well!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm really glad to hear that! It takes courage to apologize, and you did it. See, I knew he would understand that you were sorry. Now your friendship is back on track. You must feel so relieved. This is a big deal – you fixed a problem by being honest and brave. Great job!", "updated_progress": {"focus": {"current": 79, "goal": 100}, "self_control": {"current": 52, "goal": 100}, "social": {"current": 43, "goal": 100}, "confidence": {"current": 84, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 48, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 80, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 89, "closed": 11}, "looking_state_percent": {"looking_at_screen": 61, "looking_away": 39}, "total_sampled_frames": 154, "most_frequent_emotion": "neutral"}, "transcript": "I actually cleaned my entire room today without my mom even asking me! When she came in, she was so surprised and happy.", "previous_progress": {"focus": {"current": 58, "goal": 100}, "self_control": {"current": 70, "goal": 100}, "social": {"current": 81, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "previous_chat": "Supporter: Hey, you're looking pretty proud. What's up?\nChild: I did something that made my mom really happy today."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Wow, that's fantastic! Cleaning your room all by yourself is a big deal. I bet your mom was super proud of you. And you should be proud of yourself too. Doing chores without being told shows you're really growing up and being responsible. Way to go! This is going to make things better at home, for sure.", "updated_progress": {"focus": {"current": 58, "goal": 100}, "self_control": {"current": 75, "goal": 100}, "social": {"current": 81, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 50, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 70, "surprised": 0, "fear": 30}, "eyes_state_percent": {"open": 86, "closed": 14}, "looking_state_percent": {"looking_at_screen": 55, "looking_away": 45}, "total_sampled_frames": 120, "most_frequent_emotion": "neutral"}, "transcript": "I actually managed to finish reading an entire book this week! I usually get bored and stop, but this time I read a bit every day and I did it.", "previous_progress": {"focus": {"current": 65, "goal": 100}, "self_control": {"current": 74, "goal": 100}, "social": {"current": 79, "goal": 100}, "confidence": {"current": 60, "goal": 100}}, "previous_chat": "Supporter: You look pleased. Anything cool happen?\nChild: I finished a book for the first time in a while!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "No way, that's awesome! You stuck with a book all the way through. I know that can be hard when reading usually gets boring for you. You should feel really proud of yourself for reading a bit every day and not giving up. Finishing a whole book is a big deal! Maybe you found a book that you really liked. Either way, you've shown you can do it. Keep it up – maybe you'll find another book that hooks you next!", "updated_progress": {"focus": {"current": 71, "goal": 100}, "self_control": {"current": 74, "goal": 100}, "social": {"current": 79, "goal": 100}, "confidence": {"current": 66, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 48, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 70, "surprised": 0, "fear": 30}, "eyes_state_percent": {"open": 85, "closed": 15}, "looking_state_percent": {"looking_at_screen": 57, "looking_away": 43}, "total_sampled_frames": 126, "most_frequent_emotion": "neutral"}, "transcript": "A new student looked lost today, so I helped her find her classroom and then I sat with her at lunch. She was really happy that someone helped her. It made me feel good to be nice.", "previous_progress": {"focus": {"current": 74, "goal": 100}, "self_control": {"current": 82, "goal": 100}, "social": {"current": 42, "goal": 100}, "confidence": {"current": 80, "goal": 100}}, "previous_chat": "Supporter: Anything special happen at school today?\nChild: There was a new student, and I did something nice."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "That is so awesome of you! Being the new kid is super hard, and you probably made her day so much better by helping and sitting with her. You basically became her first friend! You should feel really good about that. Not everyone would step up to help like you did. It shows what a kind and caring person you are. I'm proud to be your friend!", "updated_progress": {"focus": {"current": 74, "goal": 100}, "self_control": {"current": 82, "goal": 100}, "social": {"current": 47, "goal": 100}, "confidence": {"current": 80, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 53, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 50, "sad": 0, "angry": 0, "neutral": 50, "surprised": 0, "fear": 0}, "eyes_state_percent": {"open": 92, "closed": 8}, "looking_state_percent": {"looking_at_screen": 60, "looking_away": 40}, "total_sampled_frames": 139, "most_frequent_emotion": "happy"}, "transcript": "I just got this new video game I've been wanting forever, and it's amazing! I've been playing it since I got home. The graphics and levels are so cool.", "previous_progress": {"focus": {"current": 81, "goal": 100}, "self_control": {"current": 76, "goal": 100}, "social": {"current": 85, "goal": 100}, "confidence": {"current": 78, "goal": 100}}, "previous_chat": "Supporter: You seem super excited! What's up?\nChild: I finally got the new game I told you about!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Whoa, that sounds awesome! I know you've been talking about that game for weeks. I'm so happy you got it! Tell me, what's the coolest thing about it so far? Maybe this weekend we can play it together if you want. It’s always fun to share something you love with a friend. I'm really glad you're enjoying it!", "updated_progress": {"focus": {"current": 81, "goal": 100}, "self_control": {"current": 76, "goal": 100}, "social": {"current": 87, "goal": 100}, "confidence": {"current": 78, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 49, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 80, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 84, "closed": 16}, "looking_state_percent": {"looking_at_screen": 51, "looking_away": 49}, "total_sampled_frames": 155, "most_frequent_emotion": "neutral"}, "transcript": "I'm so bored. I finished my snack and now I'm just sitting here with nothing to do. I keep fidgeting and wandering around the house because I can't find anything fun right now.", "previous_progress": {"focus": {"current": 62, "goal": 100}, "self_control": {"current": 53, "goal": 100}, "social": {"current": 88, "goal": 100}, "confidence": {"current": 80, "goal": 100}}, "previous_chat": "Supporter: What's up? You look restless.\nChild: There's nothing to do, I'm just bored."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Boredom is the worst, I feel you. But maybe we can come up with something fun. What about drawing a comic of your own? You have those cool markers, right? Or we could both invent a silly story together right now. Sometimes when I'm bored, I also try doing a quick exercise, like jumping jacks, to get my energy out. Let's think of something – I'm sure we can turn this boredom into an adventure!", "updated_progress": {"focus": {"current": 63, "goal": 100}, "self_control": {"current": 53, "goal": 100}, "social": {"current": 88, "goal": 100}, "confidence": {"current": 80, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 54, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 80, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 81, "closed": 19}, "looking_state_percent": {"looking_at_screen": 58, "looking_away": 42}, "total_sampled_frames": 142, "most_frequent_emotion": "neutral"}, "transcript": "Sometimes I daydream that I have a pet dragon. I'd fly everywhere and never have to take the bus to school! It's silly, I know, but it's fun to imagine.", "previous_progress": {"focus": {"current": 80, "goal": 100}, "self_control": {"current": 69, "goal": 100}, "social": {"current": 86, "goal": 100}, "confidence": {"current": 72, "goal": 100}}, "previous_chat": "Supporter: What are you thinking about? You have that daydream look.\nChild: Haha, it's kind of silly but..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Whoa, a pet dragon would be so cool! That's not silly at all – I daydream about awesome stuff like that too. Imagine all the places you could fly! And maybe it could light campfires for us with its breath, haha. Daydreaming can be really fun. Your dragon idea just made me smile. We should totally draw what your dragon would look like or make up a story about it!", "updated_progress": {"focus": {"current": 80, "goal": 100}, "self_control": {"current": 69, "goal": 100}, "social": {"current": 86, "goal": 100}, "confidence": {"current": 73, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 63, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 70, "surprised": 0, "fear": 30}, "eyes_state_percent": {"open": 88, "closed": 12}, "looking_state_percent": {"looking_at_screen": 65, "looking_away": 35}, "total_sampled_frames": 116, "most_frequent_emotion": "neutral"}, "transcript": "My friends got annoyed with me today because we were playing a game and I kept changing the rules and telling everyone what to do. I was just trying to make it more fun, but they all got mad and said I was being bossy.", "previous_progress": {"focus": {"current": 65, "goal": 100}, "self_control": {"current": 79, "goal": 100}, "social": {"current": 25, "goal": 100}, "confidence": {"current": 68, "goal": 100}}, "previous_chat": "Supporter: You look frustrated. Did something happen during your game?\nChild: Yeah, I think I upset my friends."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh, I see. You just had cool ideas for the game and wanted to share them, right? I know you were trying to make it fun. But sometimes when we keep changing the rules or take charge too much, other people can feel upset. They want to have a say too. Maybe next time you can ask everyone if they like the new rule before using it, or let someone else decide a rule. That way everyone feels included. I'm sure your friends will give you another chance – you were just excited.", "updated_progress": {"focus": {"current": 65, "goal": 100}, "self_control": {"current": 79, "goal": 100}, "social": {"current": 30, "goal": 100}, "confidence": {"current": 68, "goal": 100}}, "feedback": "Next time you play, try to play by the rules everyone agreed on. If you have a new idea, ask the group, \"Hey, want to try a new rule?\" first. That way nobody feels bossed around and you can all have fun together."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 49, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 79, "closed": 21}, "looking_state_percent": {"looking_at_screen": 52, "looking_away": 48}, "total_sampled_frames": 131, "most_frequent_emotion": "sad"}, "transcript": "I accidentally told another classmate about my best friend's secret. I didn't mean to, it just slipped out! Now my best friend is really mad at me and says I betrayed her.", "previous_progress": {"focus": {"current": 72, "goal": 100}, "self_control": {"current": 35, "goal": 100}, "social": {"current": 29, "goal": 100}, "confidence": {"current": 71, "goal": 100}}, "previous_chat": "Supporter: You look guilty. Did something happen with your best friend?\nChild: I messed up and broke her trust..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh no, that's a tough situation. I know you didn't want to spill her secret on purpose. It just slipped because we're sometimes not thinking. It hurts that she's mad, but I think if you explain to her that you feel really sorry and you know it was wrong, she might forgive you in time. Let her know you understand why she's upset. Maybe give her some space if she needs it, but remind her that she’s your best friend and you never wanted to hurt her. I bet eventually she'll see that you regret it.", "updated_progress": {"focus": {"current": 72, "goal": 100}, "self_control": {"current": 35, "goal": 100}, "social": {"current": 33, "goal": 100}, "confidence": {"current": 71, "goal": 100}}, "feedback": "Go tell your friend how sorry you are and that you understand why she's hurt. Promise her you'll be more careful and won’t share her secrets again. She might need a little time to cool off, but don't give up on fixing your friendship."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 52, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 80, "closed": 20}, "looking_state_percent": {"looking_at_screen": 54, "looking_away": 46}, "total_sampled_frames": 125, "most_frequent_emotion": "sad"}, "transcript": "One of my friends told me today that I talk too much and I never let him speak. It made me feel really bad. I didn't realize I was doing that.", "previous_progress": {"focus": {"current": 74, "goal": 100}, "self_control": {"current": 32, "goal": 100}, "social": {"current": 37, "goal": 100}, "confidence": {"current": 75, "goal": 100}}, "previous_chat": "Supporter: What's wrong? You seem upset.\nChild: My friend said I'm too talkative..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm sorry he said that. I know it hurts, but he probably didn't mean to be super hurtful — maybe he was just frustrated at the moment. Sometimes when we get excited, we do end up talking a lot without noticing. It's cool that you realize it now. Maybe next time you hang out, you can try to ask him questions and really listen when he talks. That will show him you care about what he has to say. I'm sure he’ll notice you're trying, and he'll feel better about it. You got this!", "updated_progress": {"focus": {"current": 74, "goal": 100}, "self_control": {"current": 32, "goal": 100}, "social": {"current": 40, "goal": 100}, "confidence": {"current": 75, "goal": 100}}, "feedback": "Next time you're with your friend, make a little mental note to pause and let him share too. You could ask something like, \"How was your day?\" and then really listen. Showing that you want to hear him out will make him feel heard and not think you talk *too* much."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 50, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 76, "closed": 24}, "looking_state_percent": {"looking_at_screen": 60, "looking_away": 40}, "total_sampled_frames": 140, "most_frequent_emotion": "sad"}, "transcript": "I kind of want to join the soccer team at school, but I'm scared. What if I'm not good enough and the other kids don’t like me? I'd feel so embarrassed if I messed up in front of them.", "previous_progress": {"focus": {"current": 71, "goal": 100}, "self_control": {"current": 67, "goal": 100}, "social": {"current": 81, "goal": 100}, "confidence": {"current": 42, "goal": 100}}, "previous_chat": "Supporter: You mentioned soccer tryouts. Are you going to go?\nChild: I'm not sure... I'm pretty nervous about it."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I totally get being nervous. Joining a new team can be scary when you don't know everyone. But you love soccer, right? If you don't try, you'll never know. The other kids might turn out to be really nice, and even if you're not the best at first, you'll get better with practice. Everybody has to start somewhere. If someone does tease you for a mistake, remember everyone makes them — even pro players miss goals! And hey, if you join, you'll already have one friend cheering you on (me!). You can do this, and I'll be proud of you for trying.", "updated_progress": {"focus": {"current": 71, "goal": 100}, "self_control": {"current": 67, "goal": 100}, "social": {"current": 81, "goal": 100}, "confidence": {"current": 46, "goal": 100}}, "feedback": "It might help to go to the tryout just to see what it's like. You don't have to be perfect. Focus on having fun and showing your effort. Usually, teammates end up becoming friends once you start practicing together."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 44, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 80, "closed": 20}, "looking_state_percent": {"looking_at_screen": 57, "looking_away": 43}, "total_sampled_frames": 132, "most_frequent_emotion": "sad"}, "transcript": "We had a group assignment and whenever I tried to share my ideas, nobody listened. They just talked over me. It made me feel invisible and I got pretty upset.", "previous_progress": {"focus": {"current": 66, "goal": 100}, "self_control": {"current": 72, "goal": 100}, "social": {"current": 50, "goal": 100}, "confidence": {"current": 54, "goal": 100}}, "previous_chat": "Supporter: How was that group project in science class?\nChild: It was terrible. They ignored me the whole time."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm sorry they made you feel that way. It's really upsetting to be ignored, especially when you have ideas you want to share. Sometimes in groups, louder people take over, and it's not fair. Maybe next time, you could say, 'I have an idea too, can I share?' in a clear voice. Or if they still won't listen, talk to the teacher privately about it. You deserve to be heard just like everyone else. Your ideas are good, and don't forget that!", "updated_progress": {"focus": {"current": 66, "goal": 100}, "self_control": {"current": 72, "goal": 100}, "social": {"current": 53, "goal": 100}, "confidence": {"current": 54, "goal": 100}}, "feedback": "If you're in a group where you can't get a word in, try gently raising your hand or saying the person's name and then, \"Can I say something?\" It might get their attention. And remember, if the group isn't treating you right, it's okay to let the teacher know — they can help make sure everyone gets a turn to speak."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 45, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 77, "closed": 23}, "looking_state_percent": {"looking_at_screen": 50, "looking_away": 50}, "total_sampled_frames": 109, "most_frequent_emotion": "sad"}, "transcript": "I was supposed to clean my room and take out the trash today, but I totally forgot. I started playing and lost track of time. My mom got really mad at me for not doing my chores.", "previous_progress": {"focus": {"current": 28, "goal": 100}, "self_control": {"current": 63, "goal": 100}, "social": {"current": 79, "goal": 100}, "confidence": {"current": 70, "goal": 100}}, "previous_chat": "Supporter: You seem upset. Something happen at home?\nChild: Mom yelled at me because I forgot to do my chores."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh man, I know how that feels. Chores can be so boring and it's easy to forget about them when you're having fun. I'm sorry your mom is angry. Maybe you can apologize and do them now to show you're trying. One thing that helps me is setting a reminder or a timer to do chores, or turning it into a little game (like racing against the clock to clean up). Your mom will probably calm down when she sees you making an effort. You got this!", "updated_progress": {"focus": {"current": 31, "goal": 100}, "self_control": {"current": 63, "goal": 100}, "social": {"current": 79, "goal": 100}, "confidence": {"current": 70, "goal": 100}}, "feedback": "Next time, you could try writing down your chores on a note or setting an alarm on your phone to remind you. Also, doing them as soon as you get home (before playing) might help so they're out of the way. That way you won't forget."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 50, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 78, "closed": 22}, "looking_state_percent": {"looking_at_screen": 56, "looking_away": 44}, "total_sampled_frames": 118, "most_frequent_emotion": "sad"}, "transcript": "I feel like my parents like my sister more than me. They always praise her because she's calm and well-behaved, and I'm always the 'problem child'. It hurts, and I wish they'd treat me the same way.", "previous_progress": {"focus": {"current": 74, "goal": 100}, "self_control": {"current": 65, "goal": 100}, "social": {"current": 80, "goal": 100}, "confidence": {"current": 30, "goal": 100}}, "previous_chat": "Supporter: You look really down. Is something bothering you at home?\nChild: I feel like my parents don't like me as much as my sister."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "That sounds really painful. I'm sorry you feel that way. Parents can seem like they favor the calmer kid sometimes, but it doesn't mean they love you any less. You and your sister are different, and that's okay. Maybe they don't realize how you feel. It might help to tell them, like, \"I feel sad when you always praise her and not me.\" I'm sure they have things they're proud of about you, too. You have so many awesome qualities – like how creative and funny you are. They might just need a reminder to say it. Remember, you're just as important as your sister, no matter what.", "updated_progress": {"focus": {"current": 74, "goal": 100}, "self_control": {"current": 65, "goal": 100}, "social": {"current": 80, "goal": 100}, "confidence": {"current": 35, "goal": 100}}, "feedback": "If you can, try talking to your mom or dad one-on-one. Let them know that you sometimes feel less loved when they compare you to your sister. It might be a hard conversation, but it could help them understand your feelings and give you more support."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 60, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 60, "neutral": 30, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 88, "closed": 12}, "looking_state_percent": {"looking_at_screen": 64, "looking_away": 36}, "total_sampled_frames": 133, "most_frequent_emotion": "angry"}, "transcript": "I wanted to go to my friend's house after school, but my parents said no. They think I'm not responsible enough to go alone. I'm 14, but they still treat me like a little kid! It's so frustrating.", "previous_progress": {"focus": {"current": 78, "goal": 100}, "self_control": {"current": 50, "goal": 100}, "social": {"current": 83, "goal": 100}, "confidence": {"current": 67, "goal": 100}}, "previous_chat": "Supporter: You look really annoyed. What's up?\nChild: My parents won't let me go out on my own."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Ugh, that sounds frustrating. I know you just want a little freedom. Being 14 and still getting told no all the time can feel unfair. Your parents probably worry about you, but I get that it feels like they don't trust you. Maybe you can talk to them calmly and ask what you can do to show you're responsible. Like, offer to check in by phone or promise to be back by a certain time. If they see you understand their concerns, they might loosen up a bit. It's worth a try, and eventually they'll have to realize you're growing up.", "updated_progress": {"focus": {"current": 78, "goal": 100}, "self_control": {"current": 52, "goal": 100}, "social": {"current": 83, "goal": 100}, "confidence": {"current": 67, "goal": 100}}, "feedback": "Try making a deal with your parents: for example, say you'll text them as soon as you get to your friend's house and when you're leaving. If you show them you can stick to their rules and check in, they might be more likely to say yes next time."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 55, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 60, "neutral": 30, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 87, "closed": 13}, "looking_state_percent": {"looking_at_screen": 62, "looking_away": 38}, "total_sampled_frames": 131, "most_frequent_emotion": "angry"}, "transcript": "My parents took away my phone for a week because they say I spend too much time on it. I'm so upset! I know I was on it a lot, but taking it for a whole week feels unfair.", "previous_progress": {"focus": {"current": 79, "goal": 100}, "self_control": {"current": 48, "goal": 100}, "social": {"current": 85, "goal": 100}, "confidence": {"current": 74, "goal": 100}}, "previous_chat": "Supporter: You seem really mad. Did something happen?\nChild: Yeah, my parents just grounded me from my phone."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Aw man, losing your phone for a week is rough. I'd be upset too. I guess your parents were worried because you were on it a lot. Maybe they want you to do other things too. It does feel unfair, but maybe you can use this time to show them you can handle it. Like, if you do your homework and chores without being asked and find other stuff to do, they might give it back sooner or at least trust you more next time. When you get it back, you could try setting a limit for yourself so they see you're balancing things. It's hard, but you can get through this week. I'll be here to keep you company!", "updated_progress": {"focus": {"current": 79, "goal": 100}, "self_control": {"current": 49, "goal": 100}, "social": {"current": 85, "goal": 100}, "confidence": {"current": 74, "goal": 100}}, "feedback": "If you can, try talking calmly to your parents. Let them know you understand you spent a lot of time on the phone and you’re willing to set some limits. Maybe suggest a fair daily time limit. Showing you're mature about it might make them ease up."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 59, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 75, "closed": 25}, "looking_state_percent": {"looking_at_screen": 56, "looking_away": 44}, "total_sampled_frames": 136, "most_frequent_emotion": "sad"}, "transcript": "Remember how I had trouble in group projects before? Well, today we had another group assignment and I tried really hard to listen to everyone and not take over. And guess what? Our group worked really well together and the project turned out great!", "previous_progress": {"focus": {"current": 69, "goal": 100}, "self_control": {"current": 68, "goal": 100}, "social": {"current": 47, "goal": 100}, "confidence": {"current": 77, "goal": 100}}, "previous_chat": "Supporter: How was your group project today?\nChild: It actually went really well!"}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "That's fantastic! I do remember you were worried about group work. It sounds like you made a big effort to be a good team member, and it really paid off. I'm sure your group noticed and appreciated that you listened and let everyone contribute. You should feel proud of yourself. Working well with others is a big deal, and you showed you can do it. Way to go!", "updated_progress": {"focus": {"current": 69, "goal": 100}, "self_control": {"current": 68, "goal": 100}, "social": {"current": 52, "goal": 100}, "confidence": {"current": 77, "goal": 100}}, "feedback": ""}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 50, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 0, "neutral": 80, "surprised": 0, "fear": 20}, "eyes_state_percent": {"open": 82, "closed": 18}, "looking_state_percent": {"looking_at_screen": 60, "looking_away": 40}, "total_sampled_frames": 149, "most_frequent_emotion": "neutral"}, "transcript": "Sometimes I just feel like I'm the 'problem kid' because I'm always the one who gets yelled at or reminded to sit still. I wish I could just be like everyone else and not have to try so hard to do simple things right. It's really frustrating.", "previous_progress": {"focus": {"current": 68, "goal": 100}, "self_control": {"current": 34, "goal": 100}, "social": {"current": 77, "goal": 100}, "confidence": {"current": 20, "goal": 100}}, "previous_chat": "Supporter: You seem really down on yourself today. What's going on?\nChild: I'm tired of being the one who's always in trouble or messing up. I just want to be normal."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I hear you. It can feel really unfair and exhausting when it seems like you're always the one struggling with stuff that others find easy. But remember, 'normal' is kind of an illusion – everyone has things they struggle with, some just hide it better or have different challenges. You have to deal with stuff like sitting still or staying focused, and yeah, it's harder for you than for some people. But that doesn't mean you're a 'bad' kid or truly that different – it just means you have to find your own ways to succeed. And you are finding them, little by little. Try not to label yourself as the 'problem kid'. You have so many good qualities (like your creativity, kindness, humor, etc.) that maybe you don't get credit for when everyone's focused on the behavior stuff. It's frustrating, but things can get better over time as you figure out what strategies work for you. You're already working hard, and that's something to be proud of. You are not alone, and you are absolutely not a failure. Normal is overrated – just keep being the best you. It will click into place eventually, I promise.", "updated_progress": {"focus": {"current": 68, "goal": 100}, "self_control": {"current": 34, "goal": 100}, "social": {"current": 77, "goal": 100}, "confidence": {"current": 25, "goal": 100}}, "feedback": "Whenever you feel this way, it might help to write down a couple of things you're really good at or things that went right that day, no matter how small. Focusing on your strengths can remind you that you're more than whatever mistakes happened. And if you can, talk to a trusted adult about how you feel – they might give you some perspective or help teachers understand you better. You aren't the only one facing these challenges, and it doesn't define you."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 52, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 0, "angry": 60, "neutral": 30, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 86, "closed": 14}, "looking_state_percent": {"looking_at_screen": 61, "looking_away": 39}, "total_sampled_frames": 135, "most_frequent_emotion": "angry"}, "transcript": "I got in trouble today for cutting in line at the cafeteria. I was just really hungry and didn't want to wait, but some kids got mad and the teacher made me go to the end.", "previous_progress": {"focus": {"current": 71, "goal": 100}, "self_control": {"current": 28, "goal": 100}, "social": {"current": 74, "goal": 100}, "confidence": {"current": 66, "goal": 100}}, "previous_chat": "Supporter: Hey, what happened at lunch? You looked upset after.\nChild: I kinda messed up in the lunch line..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Oh man, I know waiting in line when you're hungry is tough. You just acted on impulse because you wanted food quick. But I guess the other kids got annoyed because it wasn't fair. It’s good you understand it was wrong. Next time, even if it's hard, try to stick it out in line. Or maybe find a friend to talk to in line to make the time go faster. It's not a huge mistake – just something to be careful about. You got this!", "updated_progress": {"focus": {"current": 71, "goal": 100}, "self_control": {"current": 30, "goal": 100}, "social": {"current": 74, "goal": 100}, "confidence": {"current": 66, "goal": 100}}, "feedback": "When you feel like cutting in line, try distracting yourself by chatting with someone or counting numbers in your head. It can make the wait feel shorter so you won't be as tempted to jump ahead."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 53, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 80, "closed": 20}, "looking_state_percent": {"looking_at_screen": 59, "looking_away": 41}, "total_sampled_frames": 130, "most_frequent_emotion": "sad"}, "transcript": "I was at the grocery store with my mom and I got really hyper and started running down the aisle. I ended up bumping into a shelf and a bunch of cereal boxes fell down. My mom was mortified and pretty angry with me. I feel bad because I caused a scene.", "previous_progress": {"focus": {"current": 66, "goal": 100}, "self_control": {"current": 27, "goal": 100}, "social": {"current": 70, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "previous_chat": "Supporter: Hey, what happened?\nChild: I was goofing around and knocked stuff over at the store. Mom was so mad."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "Yikes, I can imagine how that went. Grocery stores can be super boring, and I know when you get bored you get a little wild to entertain yourself. Still, knocking over the cereal probably gave your mom a heart attack! I’m sure it wasn't on purpose, but it's good that you feel bad – means you know it was wrong. Maybe when you start to feel that way, you could try to do something else quietly – like help your mom by being in charge of the list or pretend each aisle is a mission (but a slow, careful mission!). That way you're focused on a task and not just energy bouncing everywhere. I'm sure the store workers have seen stuff like that before, so don't worry too much about them. Just apologize to your mom, and show her you can behave better next time. You're not childish – you just have a lot of energy. You're learning to rein it in. I bet if you make up with your mom, things will get better.", "updated_progress": {"focus": {"current": 66, "goal": 100}, "self_control": {"current": 29, "goal": 100}, "social": {"current": 70, "goal": 100}, "confidence": {"current": 69, "goal": 100}}, "feedback": "Try turning shopping into a game for yourself: like quietly count how many people wear red shirts or see if you can spot all the items on the grocery list. Keeping your mind busy might keep your body out of trouble. Also, maybe ask your mom if you can push the cart or find items – giving you a job could channel your energy in a good way."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 47, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 79, "closed": 21}, "looking_state_percent": {"looking_at_screen": 61, "looking_away": 39}, "total_sampled_frames": 138, "most_frequent_emotion": "sad"}, "transcript": "I feel really lonely at school. During recess, I usually just walk around by myself because I don't know how to join the other kids' games. I wish I had someone to play with.", "previous_progress": {"focus": {"current": 83, "goal": 100}, "self_control": {"current": 79, "goal": 100}, "social": {"current": 15, "goal": 100}, "confidence": {"current": 50, "goal": 100}}, "previous_chat": "Supporter: How's school going? Did you play with anyone at recess?\nChild: Not really... I was alone again."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm sorry you feel lonely. It can be hard to join in when everyone seems to already have their friends. I bet there are other kids who feel shy too, even if it doesn't look like it. Maybe next recess, you could try smiling and saying, 'Hey, can I play too?' or find one person who looks friendly and ask to join them. I know it's scary, but most kids will say yes if you ask nicely. You deserve to have friends to play with. If I were there, I'd totally play with you!", "updated_progress": {"focus": {"current": 83, "goal": 100}, "self_control": {"current": 79, "goal": 100}, "social": {"current": 20, "goal": 100}, "confidence": {"current": 50, "goal": 100}}, "feedback": "You might try to start with just one classmate who seems nice. Maybe compliment something, like, \"Cool ball\" or \"You're really good at drawing\" and see if a conversation starts. Sometimes breaking the ice with one person can lead to making a friend."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 53, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 73, "closed": 27}, "looking_state_percent": {"looking_at_screen": 61, "looking_away": 39}, "total_sampled_frames": 143, "most_frequent_emotion": "sad"}, "transcript": "Some kids at school were calling me \"weird\" and making fun of me for being so hyper in class. It really hurt my feelings and now I just feel awful.", "previous_progress": {"focus": {"current": 77, "goal": 100}, "self_control": {"current": 58, "goal": 100}, "social": {"current": 21, "goal": 100}, "confidence": {"current": 54, "goal": 100}}, "previous_chat": "Supporter: You seem upset after school. What happened?\nChild: A couple of kids were picking on me..."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "I'm sorry those kids were so mean. Being called 'weird' stinks. But you know what? A lot of the coolest people are a little weird. It just means you're unique. Still, it's not okay for them to tease you. Maybe you can try to ignore them or stay with friends who appreciate you. If it keeps happening, definitely tell a teacher or an adult you trust. You don't deserve to be picked on. You're awesome the way you are, and those bullies are just being mean.", "updated_progress": {"focus": {"current": 77, "goal": 100}, "self_control": {"current": 58, "goal": 100}, "social": {"current": 22, "goal": 100}, "confidence": {"current": 58, "goal": 100}}, "feedback": "If those kids call you names again, try not to show them it upsets you. Sometimes bullies stop if they don't get a reaction. And remember to hang out with people who like you for who you are. If it gets really bad, talking to a teacher or counselor can help stop it."}<|eot_id|>
|
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an AI system that converts ADHD child questionnaire responses into a structured, numeric behavioral profile for diagnostic support. Use inference where exact values are not directly provided. Return valid JSON only.
<|eot_id|><|start_header_id|>user<|end_header_id|>
{"loudness": 52, "eye_state_and_emotion": {"emotion_distribution_percent": {"happy": 0, "sad": 50, "angry": 0, "neutral": 40, "surprised": 0, "fear": 10}, "eyes_state_percent": {"open": 74, "closed": 26}, "looking_state_percent": {"looking_at_screen": 64, "looking_away": 36}, "total_sampled_frames": 129, "most_frequent_emotion": "sad"}, "transcript": "My friends got annoyed with me today because I kept changing the rules in a game.", "previous_progress": {"focus": {"current": 73, "goal": 100}, "self_control": {"current": 75, "goal": 100}, "social": {"current": 27, "goal": 100}, "confidence": {"current": 66, "goal": 100}}, "previous_chat": "Supporter: Did something happen during the game?\nChild: Yeah, I think I upset my friends by being bossy."}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{"response": "It sounds like you were excited and had lots of ideas for the game. I know you only wanted everyone to have more fun. But sometimes if you change the rules too much or try to take control, other people can feel left out or annoyed. Maybe next time, stick with the agreed rules for a while, or ask if everyone is okay with a new rule before using it. That way they feel included. I'm sure they still want to play with you – they just got frustrated in the moment. You were just enthusiastic, not trying to be mean. Apologize and let them know you didn't mean to boss them around. I bet it'll be okay, and you guys will have fun next time.", "updated_progress": {"focus": {"current": 73, "goal": 100}, "self_control": {"current": 75, "goal": 100}, "social": {"current": 32, "goal": 100}, "confidence": {"current": 66, "goal": 100}}, "feedback": "Try following the rules everyone agreed on for most of the game. If you have a cool new idea, ask the group if they want to try it instead of just doing it. That way everyone feels like part of the decision."}<|eot_id|>
|
README.md exists but content is empty.
- Downloads last month
- 1