Basic Usage

First install the toolbox. After successful installation of psignifit 4 you can fit psychometric functions as follows:

In MATLAB, first get your data for each psychometric function in the form of a nx3 matrix (stimulus level | nCorrect | ntotal). It should look something like this example dataset:

data =    [...
0.0010,   45.0000,   90.0000;...  
0.0015,   50.0000,   90.0000;...  
0.0020,   44.0000,   90.0000;...  
0.0025,   44.0000,   90.0000;...  
0.0030,   52.0000,   90.0000;...  
0.0035,   53.0000,   90.0000;...  
0.0040,   62.0000,   90.0000;...  
0.0045,   64.0000,   90.0000;...  
0.0050,   76.0000,   90.0000;...  
0.0060,   79.0000,   90.0000;...  
0.0070,   88.0000,   90.0000;...  
0.0080,   90.0000,   90.0000;...  
0.0100,   90.0000,   90.0000];  

Second, you need a struct which allows you to change options for the fitting process. (The name of the struct does not matter; you may name it however you like.)

options             = struct;   % initialize as an empty struct

You can set the different options with lines of the form [name].[field]:

options.sigmoidName = 'norm';   % choose a cumulative Gauss as the sigmoid  
options.expType     = '2AFC';   % choose 2-AFC as the experiment type  
                                % this sets the guessing rate to .5 (fixed) and  
                                % fits the rest of the parameters  

BTW: Apart from the nAFC (2AFC, 3AFC, ... ) we provide two other options for expType: First 'YesNo' which enables a free upper and lower Asymptote and, second, 'equalAsymptote', which assumes that the upper and the lower asymptote are equal. You find a more detailed description of the types here.

With those ingredients a simple call to psignifit will fit your function:

result = psignifit(data,options);

You can find a complete list of the possible options in this wiki.

result is a struct which contains all information obtained from fitting your data. Perhaps of primary interest are the fit and the confidence intervals:

result.Fit
result.conf_Intervals

This gives you the basic result of your fit. The five values reported are:

1. the threshold
2. the width (difference between the 95 and the 5 percent point of the unscaled sigmoid)
3. lambda, the upper asymptote/lapse rate
4. gamma, the lower asymptote/guess rate
5. eta, scaling the extra variance introduced (a value near zero indicates your data to be basically binomially distributed, whereas values near one indicate severely overdispersed data)

The variable conf_Intervals returns credible intervals for the values provided in options.confP. By default these are 95%, 90% and 68%. With default settings you should thus receive a 5x2x3 array, which contains 3 sets of confidence intervals (lower and upper end = 2 values) for each of the 5 parameters.

We provide some basic plotting of the psychometric function and the posterior with the plotting functions.

For example, running

plotPsych(result)

plots the psychometric function given the data and options as provided and set above: