Inspiration

Social media has made information constant, fast, and overwhelming. While it connects us, it also contributes to mental fatigue, reduced attention span, and cognitive overload; especially among students. I was inspired to build a tool that helps users understand how digital consumption affects their cognition while offering supportive, stigma-free mental health guidance.

What it does

MindMirror is a web-based mental health and cognitive awareness platform that helps users reflect on their attention, memory, and mental well-being.

It includes:

  • A Reading Comprehension Test that measures Words Per Minute (WPM) and understanding through follow-up questions
  • A Digit Span Test to assess short-term memory and attention span
  • A Mental Health Chatbot powered by a fine-tuned AI model for supportive conversations
  • Curated academic research articles from universities on social media fatigue, attention, and cognitive health

Together, these features help users become more aware of their cognitive state in a non-clinical, accessible way.

How we built it

I built MindMirror using:

  • Flask for backend development and routing
  • A fine-tuned mental health chatbot model from Hugging Face
    (Kush26/Mental_Health_ChatBot)
  • WPM (Words per minute) calculating, comprehension scoring, and digit span evaluation
  • Research-based content sourced from university publications*

Challenges we ran into

  • Designing cognitive tests that are simple, fair, and meaningful without being clinical
  • Finding a lightweight quantized mental healt chatbot
  • Ensuring the chatbot remains supportive and responsible rather than diagnostic
  • Balancing technical functionality with ethical mental health considerations
  • Integrating academic research in a way that remains readable and engaging

Accomplishments that we're proud of

  • Combining cognitive assessment and mental health AI in a single platform
  • Using a domain-specific fine-tuned model instead of a generic chatbot
  • Grounding the platform in real academic research
  • Delivering a functional, testable prototype within the hackathon timeframe

Built With

Share this project:

Updates