The aim of this tutorial was to investigate the application of psychophysical methods to measure perceptual sensitivity.
We took the following approach:
- We began by introducing a visual stimulus and task of interest (motion), and suggested a simple method of evaluating sensitivity to motion.
- We then realised that this method was subject to the observer's bias to respond "yes" or "no", and discussed how signal detection theory provides a framework for separating sensitivity from bias.
- We used an example experiment to collect hit and false alarm rates in a global motion detection task, and saw how such information is used to estimate sensitivity (d').
- We then considered how the physical stimulus might be translated into an 'internal response' via a 'delay-and-compare' circuit.
- We then stepped away from motion to consider an applied perception example that examined differences in facial emotion sensitivity between those with and without a diagnosis of schizophrenia.
- We then briefly considered an alternative experimental design (two-alternative forced-choice) that is less susceptible to bias.
- Finally, we finished with a look at the application of perceptual decision making concepts in the sporting arena.
In doing so, we have covered the following learning outcomes, either in part or in all aspects, in this tutorial:
- Explain how motion sensitivity can be measured and describe the importance of considering response bias.
- Define 'motion' and depict and interpret space-time representations.
- Construct and explain the operation of a 'delay-and-compare' motion detection circuit.
The following questions will allow you to evaluate your understanding of the topics covered in this tutorial. Should you get any of them incorrect, consider reviewing the relevant section.
In the context of a random-dot kinematogram, what does the term 'coherence' refer to?
Is there a potential concern with using the percentage of correct responses as a measure of sensitivity in a study with a yes/no design?
The key components for calculating sensitivity and bias from a yes/no task are:
A study reports that participants in Group A had a higher d' than participants in Group B. What does that mean?
Which of the following is a role of the 'delay neuron' in the example motion detection circuit?
What is a primary advantage of using a 2AFC design rather than a yes/no design to measure perceptual sensitivity?