FastPTC (copyright RJ Baker, The University of Manchester, 2006)
This program was initially written to demonstrate the principle of psychoacoustic tuning curves
(PTCs). Since it's inception it has been used by several people in experimental studies - the
programme has been used to obtain PTCs on children as young as 6 years old. The principle
behind a PTC is that the listener has to detect a tonal signal that is near threshold (e.g. 10 dB
SL). A masker (narrow-band noise) is then introduced and the level of the masker is adjusted
so that the tone is just audible above the masker (masked threshold). Maskers away from the
tone frequency are less effective and thus need to be more intense to mask the tone. As the
masker frequency is swept from below to above the tone frequency a classic "V" shaped PTC
can be produced.
Traditionally PTCs are measured with the masker presented at discrete frequencies and the
masked threshold obtained at each masker frequency. These are then plotted to produce a
PTC. This software uses a type of "Bekesy" tracking to continuously sweep the masker across
the frequency range. The general outline of the software is given below.
The software consists of a main interface (fig. 1) which allows the parameters to be adjusted.
However, the program is loads all the parameters from a .ini file which must have all the
appropriate bits in it. An example of an .ini file is given in the zip archive of the program and an
annotated version can be found here: PTC-software.ini
Once the .ini file has been loaded the parameters can be adjusted if necessary (but better to
change the .ini file). You can also step through the list of conditions using the arrows on the
interface.
The tone level can be adjusted using the slider (or set in the .ini file) - it's up to you to decide
on the level.
The calibration button allows the system to be calibrated. The general principle is that in the
calibration routine (fig. 2) the attenuators (either software or hardware) is set to zero (i.e. max.
output) and a sinusoid is played out that can be measured appropriately. The hardware is then
adjusted appropriately to give the correct output - it is important to remember, that when using
the software attenuation the dynamic range is limited by the 16-bit scaling within the software
and the limitations of the sound card (24-bit scaling is on my todo list).
Once Calibrated and the tone level is set - the test can begin. Press the "Run PTC" button and
the following screen will appear (fig. 3). Read the instructions on the screen and click on the
start button. The general principle is: press the "space bar" when you can hear the tone - let
go when you can't (if you have a programmable game-pad, or joystick you can set one of the
triggers to "space" to make the task easier and usually quieter). If you want to stop the test at
any time, just press any other key and follow the instructions.
The output of the program is available in two forms. Firstly, you can look at a plot of the PTC
immediatly after it has been obtained (this can also be printed, or copied as an image). An
example of this output from the sofware can be seen in fig 4 . However, once a new PTC is
started, this is overwritten. The second output is into a .dat file (see example) . Which stores
the raw data in from the PTC. When a new PTC is run the file is appended to. The file name
can be changed by typing in the name into the main screen's "Results File Name" box.