Model Fitting: Fitting Human Performance to Signal Detection Theory
From human performance to model parameters
On the previous page, you were able to explore SDT through the relationship between data and model, but what if we observe human performance and we want to fit the model to this actual human data? This moves us from model exploration to model fitting, which you can do below.
The table of outcomes summarizes performance. ROC space shows the relationship between behavior and theory. And the model diagram illustrates an explanation of that performance in terms of SDT. As you perform trials of the task, the table of outcomes, ROC space, and the SDT model will change to reflect your current aggregate performance up until that point in the task. By observing how these different representations present and interpret your data, you can gain a deeper appreciation for the relationship between performance, data and theory.
You can select how many to perform, the of the stimulus on each trial, and the proportion of dots that exhibit when the signal is present. When you are ready, you can the task. At any time, you can temporarily , or permanently the task.
Each trial will begin with a fixation, +, then a stimulus, and finally a question mark, ?. Decide whether you think the signal is Present or Absent, and during the stimulus or question mark, respond by clicking to indicate a ‘Present’ response or to indicate an ‘Absent’ response.
Based on the stimulus and your response (or lack there of), you will then receive feedback indicating whether this trial resulted in a Hit, Miss, False Alarm, Correct Rejection, or No Response.
The table of outcomes summarizes your Hits, Misses, False Alarms, and Correct Rejections, along with your Hit Rate, False Alarm Rate, and overall Accuracy.
In ROC space, your performance is plotted as Hit Rate versus False Alarm Rate. All of the points with the same Sensitivity (d′) are illustrated with an Iso-Sensitivity Curve. All of the points with the same Bias (c) are illustrated with an Iso-Bias Curve.
The visual representation of the SDT model shows your calculated Sensitivity as the distance, d′, between the distributions. And it shows your calculated Bias as the location, c, of the threshold. The threshold divides the Signal + Noise Distribution into regions of Hits and Misses and divides the Noise Distribution into regions of Correct Rejections and False Alarms.