Efficiency and internal noise for detection of suprathreshold patterns measured using simple reaction time

Area of Science:

• Visual Perception
• Human Factors
• Psychophysics

Background:

• Human visual performance is constrained by internal noise and inefficient sampling.
• Previous research focused on threshold contrast detection.
• Suprathreshold signal detection is crucial for many real-world visual tasks.

Purpose of the Study:

• To quantify internal noise and sampling efficiency for suprathreshold visual signals.
• To investigate how these factors change with increasing signal contrast.

Main Methods:

• Utilized simple reaction time (RT) measurements.
• Presented observers with suprathreshold Gabor stimuli in dynamic Gaussian white noise.
• Compared human RT variances to an ideal observer model to derive internal noise and sampling efficiency.

Main Results:

• Internal noise was found to remain constant across tested signal contrasts.
• Sampling efficiency demonstrated a positive correlation with increasing signal contrast.
• Visuomotor internal noise and sampling efficiency were successfully quantified.

Conclusions:

• Internal noise is a stable factor limiting visual performance even with strong signals.
• The visual system becomes more efficient at sampling information as signal contrast rises.
• Understanding these parameters is key to explaining human performance in supra-threshold visual tasks.