Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision

Abstract To evaluate the quality of artificial visual percepts generated by emerging methodologies, researchers often rely on labor‐intensive and tedious human psychophysical experiments. These experiments necessitate repeated iterations upon any major/minor modifications in the hardware/software co...

Full description

Saved in:
Bibliographic Details
Main Authors: Na Min An, Hyeonhee Roh, Sein Kim, Jae Hun Kim, Maesoon Im
Format: Article
Language:English
Published: Wiley 2025-04-01
Series:Advanced Science
Subjects:
Online Access:https://doi.org/10.1002/advs.202405789
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850149910172663808
author Na Min An
Hyeonhee Roh
Sein Kim
Jae Hun Kim
Maesoon Im
author_facet Na Min An
Hyeonhee Roh
Sein Kim
Jae Hun Kim
Maesoon Im
author_sort Na Min An
collection DOAJ
description Abstract To evaluate the quality of artificial visual percepts generated by emerging methodologies, researchers often rely on labor‐intensive and tedious human psychophysical experiments. These experiments necessitate repeated iterations upon any major/minor modifications in the hardware/software configurations. Here, the capacity of standard machine learning (ML) models is investigated to accurately replicate quaternary match‐to‐sample tasks using low‐resolution facial images represented by arrays of phosphenes as input stimuli. Initially, the performance of the ML models trained to approximate innate human facial recognition abilities across a dataset comprising 3600 phosphene images of human faces is analyzed. Subsequently, due to the time constraints and the potential for subject fatigue, the psychophysical test is limited to presenting only 720 low‐resolution phosphene images to 36 human subjects. Notably, the superior model adeptly mirrors the behavioral trend of human subjects, offering precise predictions for 8 out of 9 phosphene quality levels on the overlapping test queries. Subsequently, human recognition performances for untested phosphene images are predicted, streamlining the process and minimizing the need for additional psychophysical tests. The findings underscore the transformative potential of ML in reshaping the research paradigm of visual prosthetics, facilitating the expedited advancement of prostheses.
format Article
id doaj-art-67f02e69d59642e39fec05a80bde3903
institution OA Journals
issn 2198-3844
language English
publishDate 2025-04-01
publisher Wiley
record_format Article
series Advanced Science
spelling doaj-art-67f02e69d59642e39fec05a80bde39032025-08-20T02:26:45ZengWileyAdvanced Science2198-38442025-04-011215n/an/a10.1002/advs.202405789Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial VisionNa Min An0Hyeonhee Roh1Sein Kim2Jae Hun Kim3Maesoon Im4Brain Science Institute Korea Institute of Science and Technology (KIST) Seoul 02792 Republic of KoreaBrain Science Institute Korea Institute of Science and Technology (KIST) Seoul 02792 Republic of KoreaBrain Science Institute Korea Institute of Science and Technology (KIST) Seoul 02792 Republic of KoreaBrain Science Institute Korea Institute of Science and Technology (KIST) Seoul 02792 Republic of KoreaBrain Science Institute Korea Institute of Science and Technology (KIST) Seoul 02792 Republic of KoreaAbstract To evaluate the quality of artificial visual percepts generated by emerging methodologies, researchers often rely on labor‐intensive and tedious human psychophysical experiments. These experiments necessitate repeated iterations upon any major/minor modifications in the hardware/software configurations. Here, the capacity of standard machine learning (ML) models is investigated to accurately replicate quaternary match‐to‐sample tasks using low‐resolution facial images represented by arrays of phosphenes as input stimuli. Initially, the performance of the ML models trained to approximate innate human facial recognition abilities across a dataset comprising 3600 phosphene images of human faces is analyzed. Subsequently, due to the time constraints and the potential for subject fatigue, the psychophysical test is limited to presenting only 720 low‐resolution phosphene images to 36 human subjects. Notably, the superior model adeptly mirrors the behavioral trend of human subjects, offering precise predictions for 8 out of 9 phosphene quality levels on the overlapping test queries. Subsequently, human recognition performances for untested phosphene images are predicted, streamlining the process and minimizing the need for additional psychophysical tests. The findings underscore the transformative potential of ML in reshaping the research paradigm of visual prosthetics, facilitating the expedited advancement of prostheses.https://doi.org/10.1002/advs.202405789artificial visionhuman psychophysical testmachine learningprosthetic vision
spellingShingle Na Min An
Hyeonhee Roh
Sein Kim
Jae Hun Kim
Maesoon Im
Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision
Advanced Science
artificial vision
human psychophysical test
machine learning
prosthetic vision
title Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision
title_full Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision
title_fullStr Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision
title_full_unstemmed Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision
title_short Machine Learning Techniques for Simulating Human Psychophysical Testing of Low‐Resolution Phosphene Face Images in Artificial Vision
title_sort machine learning techniques for simulating human psychophysical testing of low resolution phosphene face images in artificial vision
topic artificial vision
human psychophysical test
machine learning
prosthetic vision
url https://doi.org/10.1002/advs.202405789
work_keys_str_mv AT naminan machinelearningtechniquesforsimulatinghumanpsychophysicaltestingoflowresolutionphosphenefaceimagesinartificialvision
AT hyeonheeroh machinelearningtechniquesforsimulatinghumanpsychophysicaltestingoflowresolutionphosphenefaceimagesinartificialvision
AT seinkim machinelearningtechniquesforsimulatinghumanpsychophysicaltestingoflowresolutionphosphenefaceimagesinartificialvision
AT jaehunkim machinelearningtechniquesforsimulatinghumanpsychophysicaltestingoflowresolutionphosphenefaceimagesinartificialvision
AT maesoonim machinelearningtechniquesforsimulatinghumanpsychophysicaltestingoflowresolutionphosphenefaceimagesinartificialvision