Validating Koalacademy, a neuro-guided language learning platform

  • Author(s) / Creator(s)
  • Koalacademy is a language learning tool predicated on the subsequent memory effect
    (SME), which differentiates the brain activity between the successful or unsuccessful

    encoding of a studied word to memory, relaying this information back to the user in real-
    time. We take advantage of the SME in confirming or denying the encoding of words during

    the process of studying them, allowing for selective repetition of poorly studied words, thus
    improving the success-rate of learning. The present study is focussed on validating the
    underlying framework of Koalacademy, a scalable Brain Computer Interface (BCI) platform
    that is able to present stimuli and stream brain data in a timely fashion comparable to
    other traditionally validated means of obtaining electroencephalography (EEG) data from
    BCI headsets. The present study utilizes a comparison oddball task. We have two
    conditions, including a control condition using a single board computer—which brain data
    is streamed to and is triggered via a light sensor at the onset of stimulus on the
    Koalacademy platform—, and an experimental condition consisting of brain data streaming
    and triggered through Koalacademy. The present study is the first of two, while the latter
    aims to validate whether a cloud trained machine learning model based on data collected
    through Koalacademy is able to successfully predict subsequent recall in real-time.

  • Date created
  • Subjects / Keywords
  • Type of Item
    Conference/Workshop Poster
  • DOI
  • License
    Attribution-NonCommercial 4.0 International