Examples

In this chapter we describe four examples: 1) a closed-set identification task, 2) an adaptive speech in noise identification task, 3) an adaptive just noticeable difference task (gap detection), and 4) same as 3, but modified to the attention span and interest of young children. Each example consists of 1) a general description, 2) the concept, and 3) the implementation in XML.

It is advised to run the experiment before reading the details. The experiment files are stored in the APEX 4 directory under in the APEX 4 folder, together with sound files and figures of the respective experiments. Note again: if the experiment file has the extension “.apx” it remains an XML file that can be edited with OxygenXML. The results file will automatically have the extension “.results”. (needs to be done).

Overview of examples provided with APEX

calibration

calibration/calibration.apx

Short: Simple calibration use
Description: The experiment shows how to calibrate using the calibration screen. You can calibrate the left and right channel seperately.
How: Use the calibration element with a calibration profile and an id for each channel (left and right)

calibration/calibration-automatic-bk2250.apx

Short: example of automatic calibration with B&K SLM 2250 plugin
Description: calibrate left and right ear at the start of the experiment
How: The <soundlevelmeter> block specifies how to use the sound level meter. If the sound level meter is not found, the regular calibration dialog is shown.

calibration/calibration-automatic-dummyslm.apx

Short: example of automatic calibration with dummy SLM plugin
Description: Allow to experiment with automatic calibration without having a sound level meter handy
How: The <soundlevelmeter> block specifies how to use the sound level meter.

calibration/calibration-automatic-localisation.apx

Short: Automatic calibration using interface to sound level meter
Description: 13-loudspeaker sound source localisation experiment, with automatic calibration through the interface to the B&K SLM2250 Sound Level Meter
How: Each channel of the sound card has a gain. These gains are specified in the <calibration> element. Here the <soundlevelmeter> block specifies how to use the sound level meter. If the sound level meter is not found, the regular calibration dialog is shown.

calibration/calibration-connections.apx

Short
Description
How

childmode

childmode/childmode-car-full.apx

Short: Shows use of child mode: intro and outro movies, child panel
Description: The experiment starts after a silent introductory movie. The child needs to find the animal in one of the three cars. One car has a different sound than the two other cars.
How: Flash elements are introduced that read movie files with extension .swf

childmode/childmode-extramovies.apx

Short: Demonstration of extra movies (snail movie)
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three snails. One snail has a different sound than the two other snails. Childfriendly feedback is provided.
How: Flash elements of snails

childmode/childmode-full.apx

Short: Shows use of child mode: intro and outro movies, child panel, progressbar, shortcuts.
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The progressbar and feedback are also childfriendly.
How: Flash elements are introduced that read movie files with extension.swf, child panel activated by <panel> in <childmode> (<screens>) + button shortcuts are introduced (button 1 = 1, button 2 = 2, button 3 = 3)

childmode/childmode-full-L34.apx

Short: Shows use of child mode: intro and outro movies, child panel, progressbar for L34
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The progressbar and feedback are also childfriendly.
How: Flash elements, L34 implementation in <devices>

childmode/childmode-justmovies.apx

Short: Shows use of child mode: without intro and outro movies, but with the normal panel
Description: The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The same progressbar is used as in adult experiments, feedback is childfriendly.
How: Flash elements, same panel as adult experiments (no childmode selected in <screens>)

childmode/childmode-movies+intro.apx

Short: Shows use of flash movies, but without the actual childmode (no childpanel)
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. The same progressbar is used as in adult experiments. Feedback is childfriendly.
How: flash elements, same panel as adult experiments (no <panel>reinforcement.swf</panel> in childmode)

childmode/childmode-nofeedback.apx

Short: Shows use of child mode without special feedback movies
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs.
How: Flash elements, no special feedback with flash elements provided

childmode/childmode-noprogress.apx

Short: Shows use of child mode: intro and outro movies, but without the actual childmode (no childpanel)
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. Feedback is childfriendly.
How: Flash elements, no panel or progressbar (<showpanel>false</showpanel> and <progressbar>false</progressbar>)

childmode/childmode-onlyscreen-normalpanel.apx

Short: Shows use of child mode: intro and outro movies, but with the normal panel
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The same progressbar is used as in adult experiments. Feedback is childfriendly.
How: Flash elements, same panel as adult experiments (no childmode selected in <screens>)

childmode/childmode-poll.apx

Short: Shows use of child mode: instead of waiting a specific time, 0 is specified as length for intro, outro and feedback, which makes apex wait for the movies to finish.
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The same progressbar is used as in adult experiments.
How: Flash elements, same panel as adult experiments (no childmode selected in <screens>), length of intro and outro is specified in <childmode>

childmode/childmode-shortcuts.apx

Short: Shows use of child mode with shortcut keys
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The progressbar and feedback are also childfriendly.
How: Flash elements, button shortcuts are introduced (leftarrow = 1, downarrow = 2, rightarrow = 3)

childmode/childmode-skipintrooutro.apx

Short: Shows use of child mode: intro and outro movies, but without the child panel; also has button shortcuts to skip the intro or outro movie
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. Feedback is also childfriendly.
How: The F7 shortcut can be used to skip the intro and outro movies when <allowskip>true</allowskip> in <general>. Flash elements, button shortcuts are introduced (leftarrow = 1, downarrow = 2, rightarrow = 3, ’s’ = skipping intro or outro movie),

childmode/childmode-skipintrooutro-disabled.apx

Short: Skip intro or outro movie by hitting ’s’ on the keyboard is disabled.
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. Feedback is childfriendly.
How: Flash elements, skipping is disabled by setting <allowskip>false</allowskip> in <general>

childmode/childmode-snake-full.apx

Short: Shows use of child mode: intro and outro movies, child panel, progressbar
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three snakes. One snake has a different sound than the two other snakes. The progressbar and feedback are also childfriendly.
How: Flash elements of snakes, child panel activated by <panel> in <childmode> (<screens>)

childmode/childmode-waitforstart.apx

Short: Shows use of child mode, with no shortcut keys and waiting for start before every trial
Description: The experiment starts after a silent introductory movie. The child needs to find the stimulus in one of the three eggs. One egg has a different sound than the two other eggs. The next trial is only presented after selecting Start from the Experiment menu or pressing F5.
How: Flash elements + <waitforstart> function in <general>

connections

connections/connections-all-mixed-monostereo.apx

Short: The use of ALL connections
Description: Monaural presentations of digit 1. Level is changed by clicking the buttons quieter and louder
How: Connections defines how the different datablocks and filters are routed to the output device

connections/connections-byname.apx

Short: Try to autoconnect a stereo wavfile to a mono wavdevice by name
Description: Digits are presented dichotically, you can give a verbal response
How: Connections are used to route different datablocks and filters to an output device

connections/connections-byregexp.apx

Short: Try to autoconnect a stereo wavfile to a mono wavdevice using mode=regexp
Description: Dichotic presentation of digits, you can give a verbal response
How: Connections are used to route different datablocks and filters to an output device

connections/connections-bywildcard.apx

Short: Try to autoconnect a stereo wavfile to a mono wavdevice using mode=wildcard
Description: Dichotic presentation of digits, you can give a verbal response
How: Connections are used to route different datablocks and filters to an output device

connections/connections-negativechannel.apx

Short: Test the use of channel -1
Description: If something is connected to channel -1 of the output device, it should be ignored, no output when button wrong is clicked
How: Connections are used to route filters to output devices

controller

controller/controller-plugincontroller.apx

Short: Plugincontroller
Description: Gain of plugin controller varies according to user input
How: the democontroller is used here to demonstrate how a plugin controller can be implemented.

controller/osccontroller.apx

Short
Description
How

controller/osccontroller_avatar.apx

Short
Description
How

datablocks

datablocks/datablocks-combination.apx

Short: Combination of datablocks using sequential and simultaneous
Description: Play datablocks similtaneously and sequentially
How: Using the <simultaneous> and <sequential> tags in <stimulus>, datablocks can be organised and reused.

datablocks/datablocks-loop.apx

Short: Use the same datablock multiple times in a stimulus
Description: The datablock wavfile “één” is played twice within one stimulus
How: set the number of replications of the datablock by setting the number of the loop, here it is <loop>2</loop>

datablocks/datablocks-nodevice.apx

Short: demonstration of omission of device in datablock
Description: When there is only one device in the experiment, it does not need to be explicitly mentioned in each datablock.
How: APEX will automatically assign the device to each datatblock

datablocks/datablocks-repeat.apx

Short: Use the same datablock multiple times in a stimulus
Description: The datablock wavfile “één” is played four times within one stimulus
How: sequential datablocks within stimulus1

datablocks/datablocks-silence.apx

Short: Combination of silence and signal for 1 stimulus presentation
Description: The stimulus in this example contains first silence (a delay) before you hear “één”
How: Use sequential datablocks that include silence and wavfile/signals, e.g. also possible to put silence datablocks after the wavfile

datablocks/datablocks-simultaneous.apx

Short: Play datablocks simultaneously
Description: Demonstrate the use of simultaneous datablocks in a stimulus
How: When datablocks are listed in the <simultaneous> element in <stimulu> they are played simultaneously

filters

filters/filters-amplifier-defaultparameters.apx

Short: Show amplifier default parameters
Description: the stimulus wd1.wav is played in the left ear, when you click the button ’play with default parameters’, the stimulus is played again with the same gain (=0) (10 trials)
How: fixed_parameters

filters/filters-amplifier-setgain.apx

Short: Adapting device parameters: gain of device varies according to user input
Description: the stimulus wd1.wav and noise are played simultaneously in the left ear, when you click the button ’higher’/’lower’, the stimulus and the noise are played again with a different gain for the noise (= gain + 2/- 2, see output box ’parameterlist’)
How: adaptiveProcedure: gain is adjusted with stepsize = 2

filters/filters-complexprocessing.apx

Short: More complex processing, can be used for testing skips - 2 noise stimuli (sinus & wivineruis.wav) with different gains (filters)
Description: When you click on ’wrong’ you hear in both ears two noise stimuli and in the left ear 2 times a sentence, the same happens when you click on the ’wrong’ button again
How: trainingProcedure, creating a sinus as noise-stimulus

filters/filters-dataloop-2simultaneous.apx

Short: Shows use of dataloop generator : uses the same datablock twice (for 2 different dataloops)
Description: When you click on the button ’correct/wrong’, in both ears a noise stimulus is played and in the left ear ’one-two’(=stimulus1)/’silence-two’(=stimulus2) is presented - Dataloop: playing noise and according to the users input (correct/wrong) stimulus1 or stimulus2 is presented
How: 2 dataloops: noise for trial1 and trial2

filters/filters-dataloop.apx

Short: Shows use of dataloop generator - wivineruis.wav should play !not! continuously over trials (see filters - noisegen - apex:dataloop)
Description: When you click on the button ’wrong/correct’, in both ears a noise stimulus is played and in the left ear ’silence-two(twee)’(=stimulus2)/’one-two(een-twee)’(=stimulus1) is presented
How: trainingProcedure - dataloop generator - continuous: false

filters/filters-dataloop-continuous.apx

Short: Shows use of dataloop generator - wivineruis.wav should play continuously over trials (see filters - noisegen - apex:dataloop)
Description: The noise is playing continuously over trials. When you click on the button ’dataloop - silence/dataloop - 1’, in both ears a noise stimulus is played and in the left ear ’silence-two(twee)’(=stimulus2)/’one-two(een-twee)’(=stimulus1) is presented
How: trainingProcedure - dataloop generator - continuous: true

filters/filters-dataloop-jump.apx

Short: Shows use of jump parameter of dataloop generator
Description
How

filters/filters-dataloop-randomjump.apx

Short
Description
How

filters/filters-noisegen-setgain.apx

Short

Basegain of noise: 0 - Aapting device parameters - gain of device varies according to user input

Description

Noise and a stimulus (one-een) are presented simultaneously in the left ear, when you click on ’higher/lower’, you hear again the noise and stimulus in the left ear but the gain of the noise is adjusted with +2/-2 (=stepsize, see output box)

How

adaptiveProcedure - stepsize of gain - adapt_parameter (Procedure)

continuous:true

filters/filters-plugin.apx

Short
Description: example already somewhere else as well!
How

filters/filters-plugin-matlabfilter.apx

Short: Example using the matlabfilter plugin.
Description: The plugin will use matlab to process data, one buffer at a time. Note that you should make sure your system can find the Matlab eng library.
How: <plugin>matlabfilter</plugin> in <filters>.

filters/filters-plugin-scamblespectrumfilter.apx

Short: Demonstrate scramblespectrum filter
Description: The scramblespectrum filter will randomize the spectrum to reduce monaural spectral cues in a localisation experiment
How: parameters to the scramblespectrumfilter are specified in <filter>

filters/filters-plugin-scamblespectrumfilter_calibration.apx

Short
Description
How

filters/filters-pulsegen.apx

Short: Shows synchronisation for L34 using a soundcard pulse
Description: When you click on send, a pulse is presented in your left ear
How: filters: generator makes a soundcard pulse => 2 channels of stereo file are mixed into output ch 0, the pulse goes to ch 1 (left ear)

filters/filters-sinegenerator.apx

Short: Test of setting frequency parameter of sine generator - frequency of sinus varies according to user input
Description: A sinus is playing in the left ear when you click on one of the 3 buttons. If you click on 100Hz/1000Hz/defaults, the frequency of the sinus changes to 100Hz/1000Hz/10 000Hz.
How: filter: 1 generator: sinus (10 000Hz) - 3 stimuli: 3x silence with adjusted parameters (frequency: 100Hz/1000Hz/no change) - continuous: false

filters/filters-sinegen-setfrequency.apx

Short: Regression test for adapting device parameters - gain of device varies according to user input
Description: A sinus is played in the left ear (sinus is played continuously over trials), when you click the button ’higher’/’lower’, the sinus is played again with a different frequency for the sinus (= frequency + 100Hz/- 100Hz, see output box ’parameterlist’)
How: adaptiveProcedure: frequency (adapt_parameter) is adjusted with stepsize = 100Hz - datablock/stimulus: silence

filters/filters-sinegen-setgain.apx

Short: Regression test for adapting device parameters - gain of device varies according to user input
Description: A sinus is played in the left ear. The sinus is played continuously over trial but when you click on ’higher/lower’ the sinus is played again but the gain of the sinus is adapted with +5/-5 dB.
How: adaptiveProcedure - datablock/stimulus: silence - filter: generator: sinus - continuous: true - adapt_parameter: gain - (!basegain has to be low enough otherwise there is clipping!)

filters/filters-singlepulse-defaultparameters.apx

Short: Shows synchronisation using a soundcard pulse
Description: When you click on ’play with default parameters’, a pulse is presented in your left ear simultaneously with a stimulus (wd1.wav - een/one)
How: datablock/stimulus + filter: generator makes a soundcard pulse (negative polarity)

filters/filters-singlepulse-polarity.apx

Short: Shows synchronisation for L34 using a soundcard pulse with a negative polarity (filter - pulsegen)
Description: When you click on send, a pulse is presented in your left ear and a stimulus (datablock: _Input44_2.wav) is presented in your right ear
How: filters: generator makes a soundcard pulse with a negative polarity

filters/filters-vocoder.apx

Short: Demonstration of vocoder plugin.
Description: Play a sentence either vocoded or unvocoded
How: A vocoder-plugin filter has been added in the connections.

filters/filters-wavamplifier.apx

Short: Regression test for training procedure - Output: number according to last input
Description: When you click on the button ’play’: in the left ear/channel: “een”, is presented and in the right ear/channel: “twee”, 20dB louder compared to the left ear, is presented.
How: filter: amplifier => basegain (-10) + gain left channel(-10)/right channel(+10)

filters/filters-wavfadeincos.apx

Short: Regression test for fader (begin/end of stimulus is not abrupt)
Description: When you click on the button ’play fade/play NO fade’, in both ears a sinus is presented with at the beginning of the stimulus a fade-in/NO fade-in (click)
How: filter: fader in begin (cosine) => length: 400ms, direction: in (fade-in: begin of stimulus)

filters/filters-wavfadeinlin.apx

Short: Regression test for fader (begin/end of stimulus is not abrupt)
Description: When you click on the button play fade/play NO fade’, in both ears a sinus is presented with at the beginning of the stimulus a fade-in/NO fade-in (click)
How: filter: fader(linear) => length: 400ms, direction: in (fade-in: begin of stimulus)

filters/filters-wavfadeoutcos.apx

Short: Regression test for fader (begin/end of stimulus is not abrupt)
Description: When you click on the button ’play fade/play NO fade’, in both ears a sinus is presented with at the end of the stimulus a fade-out/NO fade-in (click)
How: filter: fader(cosine) => length: 400ms, direction: out (fade-out: end of stimulus)

filters/filters-wavfadeoutlin.apx

Short: Regression test for fader (begin/end of stimulus is not abrupt)
Description: When you click on the button ’play fade/play NO fade’, in both ears a sinus is presented with at the end of the stimulus a fade-out/NO fade-in (click)
How: filter: fader(linear) => length: 400ms, direction: out (fade-out: end of stimulus)

general

general/general-autosave.apx

Short: Automatically save results to apr-file after experiment
Description: Closed-set identification task with automatic saving of results
How: General - autosave

general/general-exitafter.apx

Short: Exit experiment when finished
Description: Closed-set identification task, exit after finish
How: General - exitafter=true

general/general-prefix-absolutepath.apx

Short: Get the datablock prefix from the absolute path
Description: Closed-set identification task
How: datablocks uri_prefix full path

general/general-prefix-frommainconfig.apx

Short: Get the datablock prefix from the main configfile
Description: Closed-set identification task
How: source=apexconfig

general/general-prefix-invalid.apx

Short: Get the datablock prefix from the main configfile
Description: Invalid prefix id specified: wavfile not found wd1.wav
How: source=apexconfig

general/general-scriptparameters.apx

Short: Demonstrate the use of general/scriptparameters
Description: Digits 1-7 are presented monaurally, you can click on a button to go to the next one
How: Scriptparameters - name=path, scriptparameters.js needed

general/general-showresults.apx

Short: Show results after experiment
Description: When the experiment is finished and the results are saved, you can choose to see the results
How: Results - showafterexperiment=true

general/general-xmlerror.apx

Short: Error in XML file
Description
How

interactive

interactive/invalid-entry.apx

Short: Change small parameter right before the start of the experiment
Description: GUI is shown with the element to be changed
How: Interactive

interactive/setgain.apx

Short: GUI to change some settings right before the experiment
Description: GUI appears to set/change some settings
How: Interactive

interactive/subject.apx

Short: GUI to change some settings right before the experiment
Description: GUI appears to set/change some settings
How: The name of the subject, in <results>/<subject> is set from the interactive dialog that appears before the experiment is loaded.

l34

l34/addinvalidfilter.apx

Short: Try to add wav filter to L34Device
Description: Since it is not possible to add a wav filter to an L34Device, an error occurs
How: see filters and device

l34/bilateral.apx

Short: Test of synchronized bilateral CI setup
Description: Test whether the stimuli are given from the first, the second or both L34 devices
How: Including two L34 devices, stimuli can be presented stimultaneously

l34/bilateral-singlepulse.apx

Short: Test of synchronized bilateral CI setup with pulse stimuli
Description: Test whether the stimuli are given from the first, the second or both L34 devices
How: Including two L34 devices, stimuli can be presented stimultaneously

l34/invaliddatablock.apx

Short: Try to parse invalid datablock, should return error
Description: datablock does not exist in folder stimuli, running the experiment gives an error.
How: uri_prefix is used to find stimuli, invalid.aseq does not exist

l34/L34DummySync.apx

Short: Shows synchronisation for L34 using a soundcard pulse
Description: 2 channels of stereo file are mixed into output ch 0, the pulse goes to ch 1
How: use dummyDeviceType and wavDeviceType and check connections

l34/L34Sync.apx

Short: Shows synchronisation for L34 using a soundcard pulse
Description: 2 channels of stereo file are mixed into output ch 0, the pulse goes to ch 1
How: use dummyDeviceType and wavDeviceType and check connections

l34/loudness_balancing.apx

Short: Loudness balancing between two stimulation CI electrodes
Description: Two signals are presented consecutively in the same CI, at two different stimulation electrodes. One electrode is the reference electrode and has a fixed current level, the comparison electrode has changing levels. The participant has to judge which signal is louder. An adaptive procedure is used to determine the balanced level.
How: Use of adaptiveProcedure, plugindatablocks, balancing.js needed

l34/mapping1.apx

Short: Use default map with current units between 1 and 255
Description
How: See defaultmap

l34/mapping2.apx

Short: Use real map to check mapping
Description
How: See defaultmap values

l34/r126wizard.apx

Short: R126 is the clinical fitting software
Description
How: Use <defaultmap> <from_r126/>

l34/rfgenxs-bilateral.apx

Short: Test of synchronized bilateral CI setup with an RFGenXS
Description: Test whether the stimuli are given on the first, the second or both channels
How: With an RFGenXS, stimuli can be presented stimultaneously

l34/setvolume.apx

Short: Change volume of CI stimuli (send simple XML file to L34 device)
Description: Change the volume of the CI stimuli to 10,50,70 or 100 current units
How: Add variable parameter id=l34volume, see device: volume 100 is default

l34/simpleaseq.apx

Short
Description
How

l34/simpleaseq-dummy.apx

Short
Description
How

l34/simpleaseq-examples.apx

Short: power-up test?
Description
How

l34/simpleaseq-sp12.apx

Short
Description
How

l34/triggertest.apx

Short
Description
How

l34/wordsaseq.apx

Short
Description
How

manual

manual/closedsetword.apx

Short: Closed-set identification of words in noise with figures
Description: A word is presented in noise and the subject responds by clicking on one of the 4 figures on the screen, repeated 3 times. Initial SNR is set via interactive GUI.
How: full experiment

manual/gapdetection.apx

Short: Gap detection
Description
How

manual/gapdetectionchild.apx

Short: gap detection for children
Description
How

manual/opensetidentification.apx

Short: Open set identification task
Description: A word is presented in noise and the subject responds by typing the word, repeated 3 times. Initial SNR is set via interactive GUI.
How: Full experiment

manual/sentenceinnoise.apx

Short: Sentences in noise task
Description: A sentence is presented in noise, the subject responds verbally, the test leader can indicatie whether the response was correct or incorrect, repeated 4 times. Initial SNR is set via interactive GUI.
How: Full experiment

manual/trainingprocedure.apx

Short: Training for constant procedure
Description: A stimulus is presented after the user had indicated the trial by pressing a button on the screen
How: apex:trainingProcedure

parameters

parameters/parameters-connection-filter.apx

Short: Change the channel of a connection + stimulus has gain
Description: Click on the right (or left button) to hear the sound in your right ear (or left ear).
How: Training procedure, Fixed (stimulus) and variable parameters (channel) + possibility in code to change the gain of the stimuli with filter (amplifier)

parameters/parameters-connection-soundcard.apx

Short: Change the channel of a connection using a parameter
Description: Click on the right (or left button) to hear the sound in your right ear (or left ear)
How: training procedure, Fixed (stimulus) and variable parameters (channel)

parameters/parameters-device-allchannels.apx

Short: Change parameter (gain) of device channel
Description: gain of device varies in BOTH channels according to user input (1 button = louder, 1 button = quieter)
How: adaptive procedure, fixed (stimulus) and variable (gain) parameters, stepsize of 2 dB

parameters/parameters-device-singlechannel.apx

Short: Change parameter (gain) of device channel
Description: Gain of device varies in LEFT channel according to user input (1 button = louder, 1 button = quieter)
How: Adaptive procedure, fixed (stimulus) and variable (gain) parameters, stepsize of 2 dB

parameters/parameters-filter.apx

Short: Change parameter (gain) of device channel
Description: gain of device varies in BOTH channel according to user input (1 button = louder, 1 button = quieter) + gain is visible on the screen
How: adaptive procedure, fixed (stimulus) and variable (gain) parameters, stepsize of 2 dB, Parameterlists are introduced that show Left or Right gain on the screen

parameters/parameters-filter-channel.apx

Short: Change parameter (gain) of device channel
Description: gain of device varies in RIGHT channel according to user input (1 button = louder, 1 button = quieter) + gain is visible on screen
How: adaptive procedure, fixed (stimulus) and variable (gain) parameters, stepsize of 2 dB, Parameterlists are introduced that show Left or Right gain on the screen

parameters/parameters-filter-channel-probe.apx

Short: Change parameter (gain) of device channel
Description: gain of device varies in LEFT channel according to user input (1 button = louder, 1 button = quieter) + gain is visible on screen
How: adaptive procedure, fixed (stimulus) and variable (gain) parameters, stepsize of 10 dB, Parameterlists are introduced that show Left or Right gain on the screen. Additionally there is a probe filter that saves the output to disk.

parameters/parameters-restore.apx

Short: Demonstrate restoring parameter values
Description: When a parameter is set from a stimulus, and subsequently a stimulus is played in which the parameter is not set, it should be restored to its default value
How: Parameter gain is set in stimulus1, and is not set in stimulus2. Therefore a gain of 0dB should be applied for stimulus2

parameters/parameters-spinbox.apx

Short: Adjust noise manually while presenting speech and noise
Description: gain of noise varies in right channel according to user input (1 button = arrow up, 1 button = arrow down) while the stimulus stays the same.
How: constant procedure, Fixed (stimulus) and variable (noise gain), stepsize of 1 dB, noise is generated by filter

parameters/parameters-wavfadeinout.apx

Short: Set parameters of fader
Description: The fader filter allows to apply an up and down ramp to a channel
How: There are two faders in <filters, faderin and faderout. Their parameters are set from the stimuli.

procedure

procedure/adaptive-1up-1down.apx

Short: 1up-1down Adaptive procedure: Frequency of ups-downs is the same - Experiment stops after 6 reversals
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in both ears. When you click again, the gain of the stimulus decreases/increases according to the button (correct/false)
How: adaptiveProcedure - stop_after_type: reversals - nUp/nDown - adapt_parameter (gain: gain of stimulus) -

procedure/adaptive-1up-2down.apx

Short: 1up-2down Adaptive procedure: Frequency of ups-downs is NOT the same - Experiment stops after 6 reversals
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in both ears. When you click again, the gain of the stimulus StaysTheSame/Decreases/Increases according to the button (correct/false) and the number of ups and downs
How: adaptiveProcedure - stop_after_type: reversals - nUp/nDown - adapt_parameter (gain: gain of stimulus)

procedure/adaptive-2up-1down.apx

Short: 2up-1down Adaptive procedure: Frequency of ups-downs is NOT the same - Experiment stops after 6 reversals
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in both ears. When you click again, the gain of the stimulus StaysTheSame/Decreases/Increases according to the button (correct/false) and the number of ups and downs
How: adaptiveProcedure - stop_after_type: reversals - nUp/nDown - adapt_parameter (gain: gain of stimulus)

procedure/adaptive_stopafter_presentations.apx

Short: Stop after a specified number of presentations (3) - should in this case stop after 6 trials (presentation: every trial is presented once => 2 trials => 2x3 = 6)
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in both ears. When you click again, the gain of the stimulus StaysTheSame/Decreases/Increases according to the button (correct/false) and the number of ups and downs
How: adaptiveProcedure - stop_after_type: presentations - adapt_parameter (gain: gain of stimulus)

procedure/adaptive_stopafter_presentations_invalid.apx

Short: Error message because #presentations is not equal to #stop_after
Description: Point of interest if you want to stop the experiment after some presentations! => #presentations = #stop_after
How: procedure: presentations - stop_after

procedure/adaptive_stopafter_reversals.apx

Short: Stop after a specified number of reversals (6) - changes between correct-false
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in both ears. When you click again, the gain of the stimulus Decreases/Increases according to the button (correct/false) and the stepsizes changes after 3 reversals
How: adaptiveProcedure - stop_after_type: reversals - adapt_parameter (gain: gain of stimulus) - stepsize: change_after: reversals

procedure/adaptive_stopafter_trials.apx

Short: Stop after a specified number of trials (10)
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in the left ear. When you click again, the gain of the stimulus Decreases/Increases according to the button (correct/false)
How: adaptiveProcedure - stop_after_type: trials - adapt_parameter (gain: gain of stimulus)

procedure/adjustment-pluginprocedure.apx

Short: Adjustment of stimuli with a pluginprocedure
Description: There are 6 buttons:
How: adaptiveProcedure - stop_after_type: trials - adapt_parameter (gain: gain of stimulus)

procedure/adp1.apx

Short: Regression test for ADP - Experiment stops after 4 reversals
Description: adaptive procedure, one of the following sequences are played: noise een noise, een noise noise, noise noise een => answer: one of the three sequences. Stepsize changes after 2 trials => stepsize determines the parameter ’snr’
How: adaptiveProcedure - adapt_parameter (snr: snr (order, not in dB) - intervals - stepsize determines the change in snr (1-2-3) not the value in dB

procedure/adp2.apx

Short: Regression test for ADP - Experiment stops after 10 reversals
Description: adaptive procedure, one of the following sequences are played: noise een noise, een noise noise, noise noise een => answer: one of the three sequences. Stepsize changes after 2 trials => stepsize determines the parameter ’snr’
How: adaptiveProcedure - adapt_parameter (snr: snr (order, not in dB) - intervals - stepsize determines the change in snr (1-2-3) not the value in dB

procedure/afc-choices.apx

Short: Regression test - 4 Intervals - 4 Choices, select interval 2-3
Description: Four intervals are possible= ’noise noise noise een’, ’noise noise een noise’, ’noise een noise noise’ and ’een noise noise noise’. However only possibility number “2” and “3” are selected
How: constantProcedure - intervals- count/possibilities: 4, select(ion): 2 and 3

procedure/afc-uniquestandard.apx

Short: Regression test for Intervals, Standard(reference signal), Uniquestandard(true/false = If uniquestandard is true and multiple standards are defined per trial, Apex will try to present another standard in each interval of the trial)
Description: In each trial, the 3 standards should be used (uniquestandard = true)
How: constantProcedure - intervals- uniquestandard: true

procedure/aid1.apx

Short: Regression test - Different stimuli can occur during one trial (according to the difficulty of the experiment (’snr’) (- no connections or gain!)
Description: adaptive procedure, one of the following stimuli are played: ’een’, ’twee’, ’drie’, ’vier’ or ’vijf’ => If you click correct/false the experiment becomes more difficult/easy corresponding with the ’snr’ parameter - the larger the values of snr, the easier the experiment
How: adaptiveProcedure - adapt_parameter (snr: snr (order, not in dB) - different stimuli (trial) - stepsize determines the change in snr (1-2-3-4-5) - changes afther 5 trials

procedure/aid-selectrandom.apx

Short: Regression test - Different stimuli can occur during one trial (according to the difficulty of the experiment (’snr’)
Description: adaptive procedure, one of the following stimuli are played: ’een’, ’twee’, ’drie’, ’vier’ or ’vijf’, each with a different gain => If you click correct/false the experiment becomes more difficult/easy corresponding with the ’snr’ parameter
How: adaptiveProcedure - adapt_parameter (snr: snr (order, not in dB) - different stimuli (trial) - stepsize determines the change in snr (1-2-3-4-5), gain(filters)

procedure/choices-randomstimulus-randomstandard.apx

Short
Description
How

procedure/continuous-waitbeforefirst.apx

Short: Shows how to start continuous filters etc before the first trial is presented
Description: The noise is playing continuously over trials (dataloop) - the noise starts 5s before the first trial (presenting: wd1.wav: ’een’) begins
How: procedure: time_before_first_trial: in seconds - filters: dataloop - continuous: true

procedure/cst-multistimulus.apx

Short: multiple stimuli per trial - one should be chosen randomly - 2 presentations (2trials => 2x2 = 4)
Description: The experiment has 2 trials (trial1: button 1-2-3 - trial2: button a-b-c) => In trial1/trial2=> correct answer is always: button1,buttonb.
How: constantProcedure - 2 trials with !each! 3 different stimuli but with the !same! buttons

procedure/extrasimple.apx

Short: Click on the corresponding button - Stop after 2 presentations (2trials => 2x2 = 4)
Description: The experiment has 2 trials with different stimuli (trial1: house - trial2: mouse) => In trial1/trial2 => correct answer is: house/mouse (stimulus is heard in the left ear)
How: constantProcedure - 2 trials with different stimuli corresponding with the buttons

procedure/fixedparameternotfound.apx

Short: Errormessage - fixed parameter with id=“test” not defined
Description: example of an error message about an undefined fixed parameter of a stimulus
How: stimuli - fixed parameters

procedure/heartrainprocedure.apx

Short: Regression test for heartrainprocedure.js
Description: example of an error message about an undefined fixed parameter of a stimulus
How: stimuli - fixed parameters

procedure/heartrainprocedure_short.apx

Short
Description
How

procedure/idn1.apx

Short: Matching of stimuli and buttons - Different trials (+ answers) - 1 presentations (6 trials)
Description: auditive stimulus 1 2 3 4 5 6 - input: buttons 1 2 3 4 5 6 => match the stimulus with the button
How: 6 trials with each trial a different correct answer - 6 stimuli and 6 buttons - order: sequential

procedure/idn1-mono.apx

Short: Matching of stimulus and button - 5 presentations (1 trial)
Description: auditive stimulus in right ear ’1’ - input: buttons 1 2 3 4 5 6 => match the stimulus with the button (stimulus1 => button1)
How: 1 trial - 6 stimuli and 6 buttons - 1 correct answer

procedure/idn1-skip.apx

Short: skip = 2: Number of trials that will be presented before the actual presentations start.
Description: skip=2 and presentations=2 => first 2 trials and then 2*6trials = 12 trials. If the order is sequential, the skipped trials will be the first skip trials from the trial list, repeated if necessary.
How: 2 presentations - 6 trials - skip = 2 (6 stimuli - 6 buttons)

procedure/idn1-waitforstart.apx

Short: Demonstrate use of wait for start
Description: The next trial is only presented after selecting Start from the Experiment menu or pressing F5.
How: <waitforstart> function in <general>

procedure/idn2.apx

Short: Matching of stimuli and buttons - Different trials (+ answers) - 1 presentations (6 trials)
Description: auditive stimulus 1 2 3 4 5 6 - input: buttons 1 2 3 4 5 6 => match the stimulus with the button
How: 6 trials with each trial a different correct answer - 6 stimuli and 6 buttons - order: random

procedure/idn2-skip.apx

Short: skip = 3: Number of trials that will be presented before the actual presentations start.
Description: skip=3 and presentations=2 => first 3 trials and then 2*6trials = 12 trials. If the order is random, the skipped trials will be picked from the trial list without replacement, repeating this procedure if necessary.
How: 2 presentations - 6 trials - skip = 3 (6 stimuli - 6 buttons)

procedure/input-during-stimulus.apx

Short: Input during stimulus is allowed
Description: skip=3 and presentations=2 => first 3 trials and then 2*6trials = 12 trials. If the order is random, the skipped trials will be picked from the trial list without replacement, repeating this procedure if necessary.
How: Input during stimulus: true - trials with buttons (input) - stimuli (output)

procedure/kaernbach.apx

Short: 1up-1down Kaernbach procedure: Frequency of ups-downs is the same - Up stepsize is 1, down stepsize is 7 - Experiment stops after 6 reversals
Description: When you click on the button ’correct or wrong’, you hear a stimulus ’wd1.wav (een/one)’ in both ears. When you click again, the gain of the stimulus decreases/increases according to the button (correct/false)
How: adaptiveProcedure - stop_after_type: reversals - nUp/nDown - adapt_parameter (gain: gain of stimulus) -

procedure/multiparameters-fixed.apx

Short: Adaptation of multiple parameters (order(fixed) & gain(adapt))
Description: Two buttons: louder - quieter => louder: increasing of order and gain with stepsize=2, quieter: decreasing of order and gain with stepsize=2 (see output box)
How: Adapt_parameter => order & gain - stepsize for gain(filter) & order(stimuli) = 2

procedure/multiparameters-invalid.apx

Short: Error message - parameter you want to be adapted is a fixed parameter (2nd adapt_parameter: order: is a fixed parameter)
Description: error message: only first adaptive parameter can be a fixed parameter
How: Adapt_parameter => gain(adapt) & order(fixed)

procedure/multiparameters-variable.apx

Short: Adaptation of multiple parameters: GainL & GainR
Description: 2 buttons: louder - quieter - when you click on louder/quieter the gain in right and left channel is increased/decreased with 2 dB (stepsize) - see output box
How: Adapt_parameter => gainL(adapt) & gainR(adapt)

procedure/multiprocedure-aid.apx

Short: Multiprocedure - 2 adaptiveprocedures with each their own parameters, procedure and trials
Description: 2 procedures => procedure1/procedure2: 1trial - button_correct/button_wrong - When you click correct, the experiment in the next trial of the same procedure becomes more difficult (you hear a higher number => larger value of snr = stimulus (twee, drie, vier, vijf: stepsize:1)
How: procedure => multiProcedure - larger_is_easier => snr(not in dB): larger value: easier (so if you click on the wrong_button => the experiment becomes easier)

procedure/multiprocedure.apx

Short: Multiprocedure - 2 constantprocedures with each their own parameters, procedure and trials
Description: 2 procedures => procedure1/procedure2: 3 trials - button1/button4 - button2/button5 - button3/button6 but different stimuli in each trial (Match the stimulus with the button)
How: procedure => multiProcedure

procedure/multiprocedure-choices.apx

Short: Multiprocedure - 2 constantprocedures with each their own parameters, procedure and trials - 2 different choices of intervals
Description: 2 procedures => procedure1/procedure2: 1trial - 3intervals - Match the stimulus (place of stimulus within noise) with the correct button
How: procedure => multiProcedure - intervals/choices - 3 choices - standardstimulus:noise

procedure/multiprocedure-choices-corrector.apx

Short: Multiprocedure - 2 constantprocedures with each their own parameters, procedure and trials, #presentations
Description: 2 procedures => procedure1: 3presentations/1trial => 3x - 3intervals: Match the stimulus (place of stimulus within noise) with the correct button
How: procedure => multiProcedure - intervals/choices - standardstimulus:noise

procedure/multiprocedure-constant-train.apx

Short: Multiprocedure - 2procedures (1constant - 1training) with each their own parameters, procedure and trials
Description: 2 procedures => constantprocedure1: 4presentations/1trial => 4x - 3intervals: Match the stimulus (place of stimulus within noise) with the correct button
How: procedure => multiProcedure - intervals/choices - standardstimulus:noise - constantProcedure - TrainingProcedure

procedure/multiprocedure-idn.apx

Short: Multiprocedure - 2constantprocedures with each their own parameters, procedure and trials
Description: 2 procedures => constantprocedure1: order: onebyone - 4presentations/1trial => 4x - 3intervals: Match the stimulus (place of stimulus within noise) with the correct button
How: procedure => multiProcedure - intervals/choices - standardstimulus:noise - constantProcedure

procedure/multiprocedure-mixed.apx

Short: Multiprocedure - 2constantprocedures with the SAME procedure, parameters and trials
Description: 2 procedures => constantprocedure1: 1presentations/3trials => 3x - 3buttons: 1-2-3 Match the stimulus (place of stimulus within noise) with the correct button
How: procedure => multiProcedure - constantProcedure

procedure/multiprocedure-onebyone.apx

Short: Multiprocedure-order! - 3constantprocedures with the SAME procedure, parameters and trials (different stimuli)
Description: 2 procedures => constantprocedure1: 2presentations/1trial => 2x - 2buttons: name of procedure & number: click on number (=1) if you have heard the stimulus (corresponding with the number)
How: procedure => multiProcedure: order: ONEBYONE - constantProcedure

procedure/multiprocedure-random.apx

Short: Multiprocedure-order! - 3constantprocedures with the SAME procedure, parameters and trials (different stimuli)
Description: 2 procedures => constantprocedure1: 2presentations/1trial => 2x - 2buttons: name of procedure & number: click on number (=1) if you have heard the stimulus (corresponding with the number)
How: procedure => multiProcedure: order: RANDOM - constantProcedure

procedure/multiprocedure-train-train.apx

Short: Multiprocedure-order - 2trainingProcedures with DIFFERENT procedure, parameters and trials (different stimuli)
Description: 2 procedures => trainingProcedure1: 3trials => 3x???? - 3buttons: click on a button and hear the corresponding stimulus
How: procedure => multiProcedure: order: ONEBYONE- trainingProcedure

procedure/multiscreen-idn1.apx

Short: Multiscreen - 1 procedure with for the first trials: screen 1 and the last trials: screen 2 => Same outcome as sequential Multiprocedure
Description: 6 trials - 1 presentation: 3 buttons(screen1): 1-2-3 and 3 buttons(screen2): 4-5-6 => Match the stimulus with the corresponding button
How: procedure => 1procedure but several screens: multiscreen- constantProcedure

procedure/multistandard-unique.apx

Short: Multiple standards - Unique standard
Description: 2 trials: trial1 = match one of the button with the noise-stimulus (standards/reference-signals = numbers)
How: 2 trials - constantProcedure

procedure/noanswer.apx

Short: Show warning if no answer is defined for a stimulus
Description: Error message => Cannot show feedback because: no screen was found for (button turns red)
How: procedure - trial1 - no answer (see comment)

procedure/nostandardfound.apx

Short: Show warning if no answer is defined for a stimulus
Description: Error message => Cannot show feedback because: no screen was found for (button turns red)
How: procedure - trial1 - no answer (see comment)

procedure/open-set-constant.apx

Short: Open set experiment - Constant stimuli
Description: 6 trials (1 presentation) - You hear a stimulus (1-2-3-4-5-6) in both ears - Typ the answer/stimulus (een, twee, drie, …) in the textbox below ’answer’.
How: trial - answer>ANSWER THAT THE SUBJECT HAS TO TYPE< - corrector xsi:type=“apex:isequal”

procedure/pause_between_stimuli.apx

Short: a pause of 500ms should be introduced between the stimulus and the standards
Description: 1 trials (10 presentations => 10x) - You hear a stimulus (1-2-3-4-5-6) in both ears - Typ the answer/stimulus (een, twee, drie, …) in the textbox below ’answer’.
How: pause between stimuli (in seconds)

procedure/pluginprocedure.apx

Short: Pluginprocedure => script: testprocedure
Description: When you click on a button with a certain number, the next time you have to click on the number you just have heard - parameter: stepsize & startvalue
How: pluginProcedure with a certain script

procedure/pluginprocedure-matrixtest.apx

Short
Description
How

procedure/randomchannel.apx

Short: Randomgenerator test
Description: the stimulus “een” & “twee” are played simultaneously but randomly in the left or right channel according the value returnd by random1
How: trainingProcedure - randomgenerator - connections - stimuli:datablocks: simultaneous

procedure/repeatuntillcorrect-endaftertrials.apx

Short: The first selected trial (of the first presentation) shoud be repeatedly presented (stop after 5 presentations, several trials)
Description: 5 presentations x 5 trials = 25x - When you click correct => the next trial begins / When you click wrong, the first trial of the first presentation is repeated
How: repeat_first_until_correct: true

procedure/repeatuntillcorrect-multitrial.apx

Short: The first selected trial (of the first presentation) shoud be repeatedly presented (stop after 5 reversals, multiple trials)
Description: 5 trials, stop after 5 reversals - When you click correct => the next trial begins / When you click wrong, the first trial of the first presentation is repeated
How: adaptiveprocedure - REPEAT-FIRST-UNTIL-CORRECT: TRUE - adapt_parameter

procedure/repeatuntillcorrect-singletrial.apx

Short: The first selected trial shoud be repeatedly presented: DOESN’T WORK! => single trial + adapt_parameter
Description: 1 trial, stop after 5 reversals - When you click correct => the next trial begins / When you click wrong, the first trial begins
How: adaptiveprocedure - repeat_first_until_correct: true - ADAPT_PARAMETER

procedure/selectrandom.apx

Short: Test of random stimulus output when more than one stimulus defined in the experiment
Description: Random selection of stimuli in a trial if there are more than one (=> sequential presentation of stimuli => more trials) - When you click on 1, ’een’, ’twee’, ’drie’ is presented randomly in both ears
How: constantprocedure - more than 1 stimulus in 1 trial => random selection of the stimulus

procedure/simple.apx

Short: Randomgenerator test
Description: the stimulus “house”/“mouse” are played in the left channel => click the right button (house/mouse)
How: constantProcedure - stimuli - datablocks

procedure/soundquality.apx

Short: Interval - select element ok, number 2
Description
How

procedure/time-between-trials.apx

Short: Demonstrates the use of feedback to enforce a pauze between trials
Description: If you click on a button, you hear a stimulus (1-2-3-4-5-6) in both ears => click on the button corresponding with the stimulus
How: feedback length (in ms)

procedure/train1.apx

Short: Trainingprocedure: demonstrate the presentation of a stimulus with a number according to last input
Description: when you click on 1/2/3/4/5/6 => you hear een/twee/drie/vier/vijf/zes
How: trainingProcedure with 6 trials and buttons

procedure/train2.apx

Short: Show the presentation of a stimulus with a random number out of the possible choices according to last input
Description: 3 buttons: 1-2 3-4 5-6 => when you click on 5-6, you hear either 5 or 6
How: trainingProcedure - multiple trials => button5 is the answer of the trial corresponding with stimulus5 & stimulus6

procedure/trial-nostimulus.apx

Short: Make a trial whithout a stimulus
Description: when you click on stimulus => apex shuts down
How: no stimuli - datablock

procedure/uniform-double.apx

Short: Random generator test - gain of stimulus is randomly chosen
Description: You hear ’een’ in your left ear - the gain of this stimulus changes randomly, each time you click the button ’1’
How: random generator - type: DOUBLE (fractional number, e.g. 0.1) - parameter => see filter

procedure/uniform-int.apx

Short: Random generator test - gain of stimulus is randomly chosen
Description: You hear ’een’ in your left ear - the gain of this stimulus changes randomly, each time you click the button ’1’
How: random generator - type: WHOLE(fractional number, e.g. -5) - parameter => see filter

randomgenerator

randomgenerator/multi-interval.apx

Short: Randomgenerator: random gain of each datablock of a multi-interval stimulus
Description: The stimulus “één één één” is heard each time by pressing on the “1” button, and the gain of each “één” has a random gain
How: The id of the gain is set to be the parameter of the randomgenerator, three intervals are defined in procedure

randomgenerator/randomchannel.apx

Short: Randomgenerator: random channel for stimulus
Description: The stimulus “één” is played on randomly the left or right channel
How: The value returnd by random1 decides whether it is channel 0 or channel 1, see <randomgenerators>

randomgenerator/uniform-double.apx

Short: Randomgenerator: random gain of stimulus presented in 1 channel, possibility to use non-integer values (double type)
Description: The stimulus “één” is heard in the right channel each time by pressing on the “1” button, and the gain of “één” has a random gain
How: See <randomgenerators>, the id of the gain is set to be the parameter of the randomgenerator, minimum value contains a decimal value

randomgenerator/uniform-int.apx

Short: Randomgenerator: random gain of stimulus presented in 1 channel, only possible to use integer values
Description: The stimulus “één” is heard in the left channel each time by pressing on the “1” button, and the gain of “één” has a random gain
How: See <randomgenerators>, the id of the gain is set to be the parameter of the randomgenerator, minimum and maximum values are integers

randomgenerator/uniform-int-invalid.apx

Short: Randomgenerator should have integer values, similar to uniform-int.apx
Description: Warning (no error) appears in messages window because random generator limits are not integer
How: Randomgenerator max value is 2.5

randomgenerator/uniform-smallint.apx

Short: Randomgenerator: random gain of stimulus presented in 1 channel, only possible to use integer values, only small range is used
Description: The stimulus “één” is heard in the left channel each time by pressing on the “1” button, and the gain of “één” has a random gain (small range in this case)
How: See <randomgenerators>, the id of the gain is set to be the parameter of the randomgenerator, minimum and maximum values are integers

results

results/results-confusionmatrix.apx

Short: Demonstrate the use of a confusionmatrix
Description: Results are converted into a confusionmatrix
How: Default results file for a confusionmatrix

results/results-procedureparameter-after.apx

Short: Demonstrate the use of result parameters
Description: Results are shown after the experiment is done.
How: Results: show results after experiment

results/results-procedureparameter-during.apx

Short: Demonstrate the use of real-time results
Description: Real-time results are shown during the experiment in a separate window
How: Results: show results during experiment

results/results-resultparameters.apx

Short: Demonstrate the use of resultparameters
Description: Resultparameters specified in <results>/<resultparameters> will be passed on to the resultviewer
How: These parameters will be made available in a hash params.

results/results-resultparams-localization-polarplot.apx

Short
Description
How

results/results-subject.apx

Short: Demonstrate how you can add the subjects name to the results file
Description: When the subject types his/her name in the interactive, this will automatically appear in the results file and be appended to the results filename
How: subject function in ’results’, ’interactive’ entry is added

screen

screen/button-shortcuts.apx

Short: shows how to use shortcuts to answer
Description: The screen shows different buttons: you can either click on them, or use a predefined shortcut button on the keyboard
How: shortcuts are implemented for the buttons

screen/currentfeedback.apx

Short: highlighting of the played stimulus
Description: the currently playing stimulus/button is highlighted
How: <showcurrent>true

screen/feedback-answer.apx

Short: feedback with picture and highlighting of the CLICKED element
Description: When clicking a button, a feedback picture (thumb up or down) is shown in the right panel and the CLICKED button is highlighted.
How: <feedback_on> clicked

screen/feedback-both.apx

Short: highlighted stimulus + feedback with picture and highlighting
Description: The currently playing stimulus/button is highlighted. When clicking a button, a feedback picture (thumb up or down) is shown in the right panel and the correct button is highlighted.
How: <showcurrent> true, <feedback> true

screen/feedback-ledfeedback.apx

Short
Description
How

screen/feedback-ledfeedback-localization.apx

Short
Description
How

screen/feedback-ledfeedback-showcurrent.apx

Short
Description
How

screen/feedback-multistimboth.apx

Short: highlighted stimulus + feedback with picture and highlighting for multiple stimuli
Description: The currently playing stimulus/button is highlighted. When clicking a button, a feedback picture (thumb up or down) is shown in the right panel and the correct button is highlighted.
How: <showcurrent> true, <feedback> true

screen/feedback-multistimnormal.apx

Short: feedback with picture and highlighting for multiple stimuli
Description: When clicking a button, a feedback picture (thumb up or down) is shown in the right panel and the correct button is highlighted.
How: <feedback> true

screen/feedback-multistimshowcurrent.apx

Short: highlighted stimulus for multiple stimuli
Description: the currently playing stimulus/button is highlighted
How: <showcurrent> true

screen/feedback-nohighlight.apx

Short: feedback without highlighting
Description: When clicking a button, a feedback picture (thumb up or down) is shown in the right panel and NO elements are highlighted.
How: <feedback_on>: none

screen/feedback-normal.apx

Short: feedback with picture and highlighting
Description: When clicking a button, a feedback picture (thumb up or down) is shown in the right panel and the correct button is highlighted.
How: <feedback> true

screen/feedback-onlywait.apx

Short: no visual feedback
Description: No feedback is given after clicking the correct/wrong button (no highlighting/picture). Subject has to wait between trials.
How: <feedback length= …> false

screen/feedback-plugin.apx

Short: Feedback using a plugin
Description: Own feedback can be tested if dummyfeedbackplugin is changed
How: dummyfeedbackplugin

screen/feedback-showcurrent.apx

Short: highlighted stimulus
Description: the currently playing stimulus/button is highlighted
How: <showcurrent> true

screen/fgcolor.apx

Short: shows how to change the font color of a button/label
Description: A label and button with a different font color are shown on the screen
How: <fgcolor>

screen/flash-disabled.apx

Short: Shows full use of flash movies, but without the actual childmode. Shows how to disable a button.
Description: the middle egg should not be clickable (disabled)
How: <disabled>true

screen/flash-flashfeedback.apx

Short: Shows full use of flash movies, but without the actual childmode
Description: Corresponding flash movies are shown during presentation of the stimuli and after clicking the correct/wrong button.
How: introduce flash elements: <flash>, <uri>, <feedback>

screen/flash-normalfeedback.apx

Short: Pictures instead of buttons
Description: Three eggs instead of buttons are shown.
How: flashelements for screen

screen/fullscreen.apx

Short: shows how to conduct an experiment in full screen modus
Description: A typical experiment is shown, but the Windows titlebar/taskbar/… is hidden
How: <fullscreen>true<fullscreen>

screen/general-itiscreen-annoyingtext.apx

Short: shows how to use an intertrial screen
Description: A screen is shown between two trials.
How: <screens> <general> <intertrialscreen>iti

screen/general-itiscreen-picture.apx

Short: shows how to use an intertrial picture
Description: A picture is shown between two trials.
How: <screens> <general> <intertrialscreen>iti

screen/layout-arc.apx

Short: example of different arc layouts for buttons
Description: Buttons are placed in several semi-circles on the screen
How: <arcLayout> upper, lower, left, right, full

screen/layout-arc-invalidwidth.apx

Short: Shows problems with arc layout for buttons
Description: Error when loading the experiment: something went wrong with the placing of the buttons.
How: <arcLayout> The number of x-values can’t exceed the width of the arc. In this case, the width has to be changed to 9 (instead of 2)

screen/layout-arcs.apx

Short: example of different arc layouts for buttons
Description: Buttons are placed in several semi-circles on the screen
How: <arcLayout> upper, lower, left, right, full

screen/layout-arc-single.apx

Short: example of simple archwise organisation of buttons
Description: Buttons are placed archwise/in a semi-circle on the screen.
How: <arcLayout> upper, lower, left or right

screen/layout-grid.apx

Short: example of gridlayout
Description: buttons are placed in a grid
How: <gridLayout>

screen/layout-grid-nested.apx

Short: example of nested gridlayout
Description: buttons are placed in nested grids
How: <gridLayout> in <gridLayout>

screen/layout-grid-rowcol.apx

Short: shows the use of the row and col attributes instead of x and y
Description: first screen you see is made with col & row, second screen with x & y
How: <gridLayout> replacing “x” & “y” by “row” & “col”. col = column in the grid, is the same as x. row = row in the grid, is the same as y

screen/layout-grid-rowcol-invalid.apx

Short: shows wrong usage of the row and col attributes
Description: first screen you see is made with col & row, second screen with x & y
How: <gridLayout> col should not be used together with x! col is the same as x, refers to column in the grid

screen/layout-grid-stretchfactors.apx

Short: shows how to change the size of the buttons
Description: Buttons showed on the screen have different sizes/stretchfactors.
How: Stretch factor for the columns/rows: a list of integers separated by comma’s. The width of the columns/rows will be proportional to the numbers.

screen/layout-grid-stretchfactors-invalid.apx

Short: Shows problems with changing the size of the buttons
Description: Error when loading the experiment: something went wrong when changing the size of the buttons.
How: The number of the column stretch factors has to be equal to its width.

screen/layout-hlayout.apx

Short: shows horizontal layout
Description: buttons are placed next to each other, in a horizontal layout
How: <hLayout>

screen/layout-mixed.apx

Short: example of mixed layouts
Description: buttons are placed in semi-circles and grids on the screen
How: <arcLayout>, <gridLayout>

screen/nomenu.apx

Short: shows how to hide the menu bar
Description: A typical experiment is shown, but the menu bar on top of the screen is hidden.
How: <showmenu>false<showmenu>

screen/panel-feedbackpictures.apx

Short: shows how to implement a feedback picture, other than the traditional thumb
Description: A green square is shown as feedback when clicking the correct button, a red square is shown when you give a wrong answer
How: <feedback_picture_positive> & <feedback_picture_negative>

screen/panel-feedbackpictures-filenotfound.apx

Short: shows error message for feedback picture
Description: Experiment doesn’t load, because the feedback pictures are not found.
How: Wrong naming of feedback pictures / pictures don’t exist

screen/panel-nopanel.apx

Short: shows how to hide the panel
Description: A typical experiment is shown, but the panel on the right side is hidden. Start the experiment by pressing F5.
How: <showpanel>false<showpanel>

screen/panel-showrepeatbutton.apx

Short: shows the use of a repeatbutton
Description: A repeatbutton is shown on the screen, just above the progressbar. When clicking the button, the last stimulus is repeated.
How: <repeatbutton>true<repeatbutton>

screen/panel-showstatuspicture.apx

Short: shows the use of a statuspicture
Description: A statuspicture (dot) is shown, just above the progress bar. The dot changes color: red when presenting the stimulus, green when waiting for an answer
How: <statuspicture>true<statuspicture>

screen/panel-showstopandrepeatbutton.apx

Short: shows how to use a stopbutton & repeatbutton
Description: A red stopbutton and a repeatbutton are displayed on the screen, just above the progressbar. When clicking the stopbutton, all output is immediately stopped & apex is shut down. When clicking the repeatbutton, the last stimulus is repeated.
How: <stopbutton> & <repeatbutton>

screen/panel-showstopbutton.apx

Short: shows how to use the stopbutton
Description: A red stopbutton is displayed on the screen, just above the progressbar. When clicking this button, all output is immediately stopped & apex is shut down
How: <stopbutton>true>stopbutton>

screen/prefix-apexconfig.apx

Short: shows how to use an ’apexconfig’ as a prefix for a filename
Description: idem
How: <uri_prefix source=“apexconfig”>pictures</uri_prefix>

screen/prefix-inline.apx

Short: shows how to use an ’inline’ as a prefix for a filename
Description: idem
How: <uri_prefix>../../pictures</uri_prefix>

screen/screenelements-all.apx

Short: example of a multitude of screenelements
Description
How

screen/screenelements-allbutflash.apx

Short: assembly of different kinds of screenelements
Description: Different labels and buttons are shown on the screen. First row containg a button and answerlabel, second row containing a label and a parameterlist, last row a textedit and picture
How: <button>, <answerlabel>, <label>, <parameterlist>, <textEdit>, <picture>

screen/screenelements-answerlabel.apx

Short: shows the use of an answerlabel
Description: An answerlabel is shown on the screen, containing the correct answer for the current trial.
How: add <answerlabel> to <gridLayout>

screen/screenelements-button-disabled.apx

Short: shows how to disable a button
Description: Two buttons are shown on the screen: nothing happens when clicking the disabled one
How: <disabled> true

screen/screenelements-checkbox.apx

Short: shows how to use a checkbox
Description: Checkboxes are shown on the screen: if a checkbox is clicked, the subject is assumed to have responded to the trial
How: <checkBox>

screen/screenelements-html.apx

Short: how to use a html page
Description: a html page is shown on the screen
How: <html> <page>

screen/screenelements-label.apx

Short: shows how to add text to a label
Description: screen shows a large label filled with text
How: <label> <text>

screen/screenelements-label-bgcolor.apx

Short: shows how to change background and font color of a label/button
Description: screen shows different labels/buttons with a different background color or font color
How: <bgcolor> changes the background color, <fgcolor> changes the font color

screen/screenelements-label-html.apx

Short: shows how to use a bold font by using html code
Description: screen shows the text written in html code
How: text

screen/screenelements-label-resize.apx

Short: shows how to resize the text on a label
Description: screen shows different labels, with and without resizing of the text
How: <text> split text into different lines with ’enter’ & ’

screen/screenelements-matrix.apx

Short: example of a matrix layout for buttons
Description: Screen shows buttons in a 2x2 matrix. After listening to the stimulus, you have to select one button in each column before clicking OK.
How: <matrix>

screen/screenelements-parameterlabel.apx

Short: shows how to implement a parameterlabel on the screen
Description: A parameterlabel is shown on the screen, i.e. a label containg the value of a parameter.
How: screen: <parameterlabel>, stimuli: define parameter

screen/screenelements-parameterlist.apx

Short: shows how to implement a parameterlist on the screen
Description: Parameters of the played stimuli are shown on the screen. This example shows two different parameterlists (two different trials).
How: <parameterlist>

screen/screenelements-picture-changepicture.apx

Short: example of a picture-button
Description: Click the picture after listening to the stimulus
How: <picture>

screen/screenelements-picture-notfound.apx

Short: shows error message for picture
Description: experiment doesn’t load, error message: picture not found
How: wrong <uri> (’abc’-picture doesn’t exist, can’t be found in the pictures-folder)

screen/screenelements-pictures.apx

Short: shows how to implement different pictures in the screen
Description: 4 pictures are shown
How: <picture>

screen/screenelements-pictures-disabled.apx

Short: shows how to disable a picture-button
Description: 4 pictures are shown on the screen, one is disabled (number 4)
How: <picturelabel> instead of <picture> + picture 4 is not included in the button group

screen/screenelements-pictures-feedback.apx

Short
Description
How

screen/screenelements-pictures-invalid.apx

Short: shows error message for picture
Description: experiment doesn’t load, error message: picture invalid
How: wrong file extension for picture (xxx in stead of .png)

screen/screenelements-pictures-resize2.apx

Short: example of screen layout with pictures
Description: Screen shows 4 pictures, after hearing a stimulus you should click one
How: <picture>

screen/screenelements-pictures-resize.apx

Short: example of screen layout with pictures
Description: Screen shows 4 pictures, after hearing a stimulus you should click one
How: <picture>

screen/screenelements-samename.apx

Short: Check whether everything is OK when two screenelements in different screens have the same name
Description: /
How: /

screen/screenelements-slider.apx

Short: shows how to implement a vertical slider on the screen
Description: Slider is shown on the screen. Sliding the bar doens’t change anything to the presented stimuli.
How: <slider>

screen/screenelements-slider-horizontal.apx

Short: shows how to implement a horizontal slider on the screen
Description: Slider is shown on the screen. Sliding the bar doens’t change anything to the presented stimuli.
How: <slider>

screen/screenelements-spinbox.apx

Short: example of a spinbox
Description: a spinbox is an input field that only accepts numbers and has 2 buttons to respectively increase or decrease its value
How: add <spinBox> to Layout, define: startvalue, minimum & maximum

screen/screenelements-spinbox-parameter.apx

Short: example of a spinbox with adjustable parameter
Description: a spinbox is an input field that only accepts numbers and has 2 buttons to respectively increase or decrease its value
How: add <spinBox> to Layout, define: startvalue, minimum & maximum, parameter to be adjusted

screen/screenplugin-emptyresult.apx

Short: shows example of error message
Description: Demonstrate a screen generated by javascript, show error message because of invalid xml
How: wrong naming of source <pluginscreens>

screen/stylesheet-elements.apx

Short: example of a different screenstyle
Description: screen has an other background color, buttons changed color, highlighting changed, …
How: <style_apex>

screen/textinput-font.apx

Short: example of font input
Description: screen shows two ’input fields’ with different fonts
How: <textEdit>

screen/textinput-inputmasks.apx

Short: example of various restricted inputs
Description: Screen shows four ’entry fields’, but input is restricted (for example, only numbers allowed)
How: <textEdit>, <inputmask> (Defined in the Qt documentation!)

screen/textinput-setfocus.apx

Short: example of restricted input
Description: Screen shows two ’entry fields’, but input is restricted (for example, only numbers allowed)
How: <textEdit>, <inputmask> numbers = only numeric input will be allowed

screenplugin

screenplugin/screenplugin-emptyresult.apx

Short
Description
How

soundcard

soundcard/soundcard-asio-multiplecard.apx

Short: Demonstrate the usage of multiple soundcards
Description: two soundcards are defined in <devices>
How: an extra soundcard is defined in <devices>, each with their own parameters

soundcard/soundcard-cardname-inline.apx

Short: Specify the name of the sound card to be used
Description: Specify the sound card to be used.
How: The name of the sound card to be used can be specified in devices/device/card. A list of devices in your system can be obtained by running apex from the commandline with parameter –soundcards

soundcard/soundcard-clipping.apx

Short: Demonstrate what happens when the output clips
Description: The output clips of the first two trials of this experiment
How: In the results file, the following will be shown to indicate clipping occured: <device id=“soundcard”> <clipped channel=“0”/> </device>

soundcard/soundcard-defaultsettings.apx

Short: Demonstrate default wavdevice settings
Description: Shows default settings in ’devices’
How: Devices: type of soundcard, number of channels (2), gain (0), samplerate (44100)

soundcard/soundcard-driver-asio.apx

Short: Demonstrate the usage of different drivers and their parameters
Description: In this example asio is used as a driver (basicwav-coreaduio/jack/portaudio show examples of different drivers)
How: Devices: <driver>asio</driver>

soundcard/soundcard-driver-coreaudio.apx

Short: Demonstrate the usage of different drivers and their parameters
Description: In this example coreaudio is used as a driver (basicwav-asio/jack/portaudio show examples of different drivers)
How: Devices: <driver>coreaudio</driver>

soundcard/soundcard-driver-jack.apx

Short: Demonstrate the usage of different drivers and their parameters
Description: In this example jack is used as a driver (basicwav-coreaduio/asio/portaudio show examples of different drivers)
How: Devices: <driver>jack</driver>

soundcard/soundcard-driver-portaudio.apx

Short: Demonstrate the usage of different drivers and their parameters
Description: In this example portaudio is used as a driver (basicwav-coreaduio/jack/asio show examples of different drivers)
How: Devices: <driver>asio</driver>

soundcard/soundcard-gain-invalidchannel.apx

Short: Demonstrate the misapplication of adjusting gain to the wrong channel
Description: Gain is applied to the wrong channel (channel = 10, when there are only 2 channels)
How: Devices: <gain channel=“10”>0</gain> -> it should be channel=“0” of channel=“1”, because there are only 2 channels defined.

soundcard/soundcard-mono-2channel.apx

Short: Demonstrate what happens when you use wavdevice with one channel not connected
Description: When you only use one of two defined channels you will not get an error, only a message: “not every inputchannel of soundcard is connected”
How: Devices: 2 channels defined, only 1 used in <connections>

soundcard/soundcard-portaudio-6channels.apx

Short: Demonstrate the usage of multiple channels
Description: Add more channels to your experiment
How: set more channels (e.g. <channels>6</channels, for 6 channels> and make connections with them so that the right stimulus is send to the right channel

soundcard/soundcard-portaudio-bigbuffersize.apx

Short
Description
How: <buffersize>8192</buffersize> is used

soundcard/soundcard-portaudio-cardnotfound.apx

Short: Demonstrate the misapplication of <card>
Description: <card> specifies the name of the sound card. If you use the wrong cardname, you will get an error when you try to open your experiment in apex.
How: wrong card name used in <card> -> ’Soundmax’ should be default or Soundmax should be defined in apexconfig.xml

soundcard/soundcard-portaudio-interactivecard.apx

Short: Demonstrate the misapplication of <card>
Description: <card> specifies the name of the sound card to be used. If you use the wrong cardname, you will get an error when you try to open your experiment in apex.
How: wrong card name used in <card> -> ’invalid’ should be default or an other name defined in apexconfig.xml

soundcard/soundcard-wav-padzero.apx

Short: Demonstrate the use of padzero
Description: Add the given number of empty (filled with zero’s) buffers to the output on the end of a stream. This avoids dropping of the last frames on some soundcards.
How: Devices: Add padzero function (e.g. <padzero>2</padzero>)

xmlplugin

xmlplugin/autostimulus.apx

Short: Demonstrate datablocks and stimuli generated by javascript
Description: The datablocks and stimuli are automatically generated in this experiment using javascript
How: using plugindatablocks and pluginstimuli, setting the pluginparameters and referring to the wanted javascript in <general> (here autostimulus.js)

xmlplugin/autotrials.apx

Short: Demonstrate trials, datablocks and stimuli generated by javascript
Description: The trials, datablocks and stimuli are automatically generated in this experiment using javascript
How: using plugintrials, plugindatablocks and pluginstimuli, setting the pluginparameters and referring to the wanted javascript in <general> (here autostimulus.js)

xmlplugin/datablockplugin.apx

Short: Demonstrate a datablock generated by javascript
Description: The datablock is automatically generated in this experiment using javascript
How: using plugindatablocks and referring to the wanted javascript in <datablocks> (here datablockgenerator.js)

xmlplugin/datablockplugin_l34.apx

Short: Demonstrate generating XML code for L34 directly from plugindatablocks and pluginstimuli
Description: The datablocks are automatically generated in this experiment using javascript
How: using plugindatablocks, pluginstimuli, setting the pluginparameters and referring to the wanted javascript in <datablocks> and <stimuli> (here datablockgenerator_l34.js)

xmlplugin/screenplugin.apx

Short: Demonstrate a screen generated by javascript
Description: The screen is automatically generated in this experiment using javascript
How: using pluginscreens, setting the pluginparameters and referring to the wanted javascript in <screens> (here screengenerator.js)

xmlplugin/screenplugin-inlibrary.apx

Short: Demonstrate the misapplication of a screen generated by javascript
Description: The screen is not generated in this experiment because the script source is invalid. You will get an error message when you try to open the experiment.
How: script source refers to <general> instead of the script itself. It should be: <script source=“screengenerator-library”/> in <screens> OR <scriptlibrary>screengenerator-library.js</scriptlibrary> in <general>

xmlplugin/screenplugin-invalidscript.apx

Short: Demonstrate the misapplication of a screen generated by javascript
Description: The screen is not generated in this experiment because the defined javascript is empty.
How: invalidxml.js is an empty/invalid javascript so no screen can be generated.

xmlplugin/screenplugin-invalidxml.apx

Short: Demonstrate the misapplication of a screen generated by javascript
Description: The screen is not generated in this experiment because the defined javascript is empty. You will get an error when you try to open the experiment.
How: the javascript contains an error so no screen can be generated.

xmlplugin/screenplugin-multiscreen.apx

Short: Demonstrate a screen generated by javascript
Description: The screen is automatically generated in this experiment using javascript
How: using pluginscreens, setting the pluginparameters and referring to the wanted javascript in <screens> (here screengenerator-multiscreen.js)

xmlplugin/stimulusplugin.apx

Short: Demonstrate stimuli generated by javascript
Description: The stimuli are automatically generated in this experiment using javascript
How: using pluginstimuli, setting the pluginparameters and referring to the wanted javascript in <stimuli> (here stimulusgenerator.js)

Example 1: Closed-set identification of words in noise with figures

General description of the experiment

See . This is an example of a closed-set identification test for children. A word (wave file) is presented and the child responds by clicking on one of four figures on the screen (figure [fig:closedset]). Subsequently, a new set of figures is shown, and again, a word corresponding to one of these figures is routed to the sound card. This is repeated three times. The three words are embedded in noise at a certain signal-to-noise ratio. In this example the level of the noise is fixed and the level of the word varies. At the beginning of the experiment APEX 4 queries for the SNR (signal to noise ratio, in dB). After having entered this value four figures will appear. Press Start to start the experiment and to hear the first stimulus (speech in noise). After the experiment has finished the results are written to a results file and the percentage correct is determined.

Conceptual

The experiment as described in the previous paragraph should first be translated to concepts understood by APEX 4. The main concepts in this example are datablock, stimulus, screen and procedure. For each of the 3 words to be presented to the subject, a wave file is available on disk. For each wave file, a datablock is defined, and for each resulting datablock a stimulus is defined. Everything that is defined is assigned an ID, to be able to refer to it later on. Therefore, now we have 3 stimuli with IDs , and .

We should also define the things to be shown on the screen during the experiment. This is done by defining a screen for each word. Each screen contains 4 pictures, of which one corresponds to the word. Each screen again gets an ID, in this case we name the screen by the pictures it contains. Therefore we now have 3 screens with ID , and .

To indicate the order in which the words should be presented, and together with which screen, a procedure is defined. In the procedure a number of trials is defined. Recall that a trial is the combination of a screen, a stimulus and a correct answer. Therefore a trial is a way to link each of our stimuli with a screen.

Now only the output logic remains to be defined. The idea is to continuously present a noise signal and to set the level of the words such that a certain SNR is obtained. To achieve this, we define 2 filters, the first one is a generator (i.e., a filter without input channels) that will generate the noise signal. The other one is an amplifier, which will amplify or attenuate the words to obtain the desired SNR.

In the following sections each of the elements of the file necessary to implement the latter concepts will be described in detail. The first one is .

Detailed description of various elements

      <procedure xsi:type="apex:constantProcedure">
        <parameters>
          <presentations>1</presentations>
          <order>sequential</order>
        </parameters>
        <trials>
          <trial id="trial_star">
            <answer>picturestar</answer>
            <screen id="screenstar_horse_vase_moon"/>
            <stimulus id="stimulus_star"/>
          </trial>
          <trial id="trial_fly">
            <answer>picturefly</answer>
            <screen id="screenknee_fly_mouse_star"/>
            <stimulus id="stimulus_fly"/>
          </trial>
          <trial id="trial_mouse">
            <answer>picturemouse</answer>
            <screen id="screenmouse_fly_star_moon"/>
            <stimulus id="stimulus_mouse"/>
          </trial>
        </trials>
      </procedure>

The attribute indicates that a constant stimuli procedure is used. This means that the procedure will select the next trial from the trial list and that it completes after every trial has been presented a certain number of times.

defines the behavior of the procedure
- every trial is presented once
- the three trials are presented sequentially, in the order that is specified in the experiment file
contains different elements to specify a trial. After selecting a trial the will show the specified screen and send the stimulus to the device. Each trial is given an ID (arbitrary name), eg , such that it can be referred to later on or viewed in the results file.
the correct answer, to be used by APEX 4 to determine whether the subject’s response is correct. Here, the subject gets the opportunity to click on one of four pictures. The result from the screen (the subject’s response) will be the ID of the element of the screen that was clicked. Therefore, in this example, we specify the ID of the picture corresponding to the stimulus that is presented.

A trial must be defined for all the words of the experiment.

        <corrector xsi:type="apex:isequal"/>

The corrector checks whether the response is correct or not. The attribute compares whether the two input values are exactly the same. In this example compares the answer specified under trial and the ID corresponding to the picture that has been clicked.

    <screens>
        <uri_prefix>closedset</uri_prefix>
        <reinforcement>
          <progressbar>true</progressbar>
          <feedback length="600">false</feedback>
        </reinforcement>
        <screen id="screenstar_horse_vase_moon">
          <gridLayout height="2" width="2">
            <picture id="picturestar" row="1" col="1">
              <path>star.jpg</path>
            </picture>
            <picture id="picturehorse" row="2" col="1">
              <path>horse.jpg</path>
            </picture>
            <picture id="picturevase" row="1" col="2">
              <path>vase.jpg</path>
            </picture>
            <picture id="picturemoon" row="2" col="2">
              <path>moon.jpg</path>
            </picture>
                </gridLayout>
          <buttongroup id="buttongroup1">
            <button id="picturestar"/>
            <button id="picturehorse"/>
            <button id="picturevase"/>
            <button id="picturemoon"/>
          </buttongroup>
          <default_answer_element>buttongroup1</default_answer_element>
        </screen>
        <screen id="screenknee_fly_mouse_star">
          <gridLayout height="2" width="2">
            <picture id="pictureknee" row="1" col="1">
              <path>knee.jpg</path>
            </picture>
            <picture id="picturefly" row="2" col="1">
              <path>fly.jpg</path>
            </picture>
            <picture id="picturemouse" row="1" col="2">
              <path>mouse.jpg</path>
            </picture>
            <picture id="picturestar" row="2" col="2">
              <path>star.jpg</path>
            </picture>
          </gridLayout>
          <buttongroup id="buttongroup2">
            <button id="pictureknee"/>
            <button id="picturefly"/>
            <button id="picturemouse"/>
            <button id="picturestar"/>
          </buttongroup>
          <default_answer_element>buttongroup2</default_answer_element>
        </screen>
        <screen id="screenmouse_fly_star_moon">
          <gridLayout height="2" width="2">
            <picture id="picturemouse" row="1" col="1">
              <path>mouse.jpg</path>
            </picture>
            <picture id="picturefly" row="2" col="1">
              <path>fly.jpg</path>
            </picture>
            <picture id="picturestar" row="1" col="2">
              <path>star.jpg</path>
            </picture>
            <picture id="picturemoon" row="2" col="2">
              <path>moon.jpg</path>
            </picture>
          </gridLayout>
          <buttongroup id="buttongroup3">
            <button id="picturemouse"/>
            <button id="picturefly"/>
            <button id="picturestar"/>
            <button id="picturemoon"/>
          </buttongroup>
          <default_answer_element>buttongroup3</default_answer_element>
        </screen>
      </screens>

contains several elements that can be referred to elsewhere in the experiment file (e.g., in above).

a relative path is specified here (relative with respect to the experiment file). Since APEX 4 knows the location of the experiment file, only the folder containing the wave files and pictures must be specified. It is also possible to give the absolute path, starting at the root. There are three ways to specify a prefix: by directly specifying an absolute path, by directly specifying a path relative to the experiment file or by referring to a prefix stored in . Please refer to Using prefixes for more information.
includes
- As the value is a progressbar will be displayed in the right hand bottom corner of the screen that indicates the percentage of trials that have been completed.
- duration of time after response (in msec) that APEX 4 waits before presenting the next trial. During this interval, feedback can be displayed. In this case, no feedback (thumb up, thumb down) is given as the value is .

For each word to be presented, a screen is defined. Each screen has an ID by which it can be referred to elsewhere in the experiment file (e.g. in to associate it with a stimulus).
- specifies how the four figures will be arranged on the screen. A GridLayout creates a regular grid on the screen with the specified number of rows and columns. In this example there are 2 rows and 2 columns. Each figure is defined by means of a element. Such a definition can be seen as associating a graphics file with an ID and specifying at what position of the layout it should be shown. In this case, jpeg files are used, but other formats are also possible (e.g., png, bmp, gif).
  - since the prefix as specified under is preprended to each it suffices to give the name of the .jpeg file in . The path is relative to the experiment file.
- defines a group of screen elements, namely those (four figures) that are displayed on the screen. The ID is defined before.
- As many elements can be defined in a screen, APEX 4 has no way to know which element contains the subject’s response. If, for example, a text box is shown and 2 buttons, it is unclear which is to be used to determine whether the answer is correct or not. Therefore, in the element is designated that contains the subject’s response. In the case of screen elements that are clicked in order to respond, the example is further complicated by the fact that we cannot specify just one of the elements (buttons, pictures), but the response rather comes from a group of elements. This is when a can be used to group together different screen elements.

     <datablocks>
        <uri_prefix>closedset/</uri_prefix>
        <datablock id="datablock_star">
          <device>wavdevice</device>
          <uri>star.wav</uri>
        </datablock>
        <datablock id="datablock_fly">
          <device>wavdevice</device>
          <uri>fly.wav</uri>
        </datablock>
        <datablock id="datablock_mouse">
          <device>wavdevice</device>
          <uri>mouse.wav</uri>
        </datablock>
        <datablock id="noisedata">
          <device>wavdevice</device>
          <uri>noise.wav</uri>
        </datablock>
        <datablock id="silence">
          <device>wavdevice</device>
          <uri>silence:500</uri>
        </datablock>
      </datablocks>

A datablock must be made for all wave files used in the experiment, including the noise (that will be used by the noise generator). In this example 5 datablocks are defined, 3 for the word files, 1 for the noise and 1 for silence. contains 5 elements and a prefix.

a relative path is specified here. It is also possible to give the absolute path, starting at the root (see Using prefixes).
For each wave file a datablock is made, with an ID. In datablock the special syntax is demonstrated for creating a datablock containing only silence (i.e., all samples are zeros). This is done to put silence before and after the word, to prevent the speech and noise from starting at the same time. The length of the silence datablock is specified in ms after the prefix . It is added before the signal (in element ), not before the noise.
Each datablock is associated to a (by means of the ID of the device).

In the next sections, the output logic will be defined. Figure [fig:ex1-output] gives an overview of the building blocks that are used in this example. On the left hand side the datablocks are shown. In the middle the noise generator and the amplifier and on the right hand side the sound card. As the words are to be amplified or attenuated according to the desired SNR, the corresponding datablocks are routed through the amplifier.

The connection window

In the next element the output device is specified.

     <devices>
        <device id="wavdevice" xsi:type="apex:wavDeviceType">
          <driver>portaudio</driver>
          <card>default</card>
          <channels>1</channels>
          <samplerate>44100</samplerate>
        </device>
      </devices>

All devices defined in the experiment file are grouped in . In this example there is only 1 element. Its ID is set to . As an ID is unique for an entire experiment file, it can be used later on to refer to this device.

The attribute tells APEX 4 that a sound card is used. The experiment only starts if all devices can be opened.
specifies the software driver to be used for sound output. If unsure, set it to .
specifies the name of the sound card to be used. The system default card can be used by specifying as a card name. Other card names can be defined in .
specifies the number of channels of the sound card to be used. The number of channels is restricted to the selected sound card, with a maximum of 2 for portaudio.
the sampling rate of the sound card; Not all sampling rates are supported by all devices, and some drivers automatically convert the sampling rate. Check your sound card documentation. The sample rate of the sound card should correspond to the sampling rates of all datablocks used with it. If not, an error message will be shown.

Filters must be defined whenever the signal (or noise) is manipulated. In this example the level of the noise remains constant and the signal is amplified or attenuated using an amplifier filter (loop of ).

    <filters>
        <filter xsi:type="apex:dataloop" id="noisegen"> (*@\label{xml:filter1}@*)
          <device>wavdevice</device>
          <channels>1</channels>
          <continuous>false</continuous>
          <datablock>noisedata</datablock>
          <basegain>-5</basegain>
          <gain id="noisegain">0</gain>
          <randomjump>true</randomjump>
        </filter>
        <filter xsi:type="apex:amplifier" id="amplifier">  (*@\label{xml:filter2}@*)
          <device>wavdevice</device>
          <channels>1</channels>
          <basegain>-5</basegain>
          <gain id="gain">0</gain>
        </filter>
      </filters>

contains individual elements, which specify a filter, or as a special case a generator (i.e., a filter without input channels).

on line [xml:filter1] the attribute tells APEX 4 that a dataloop generator has to be created. This is a generator that takes a datablock and loops it infinitely. The datablock to be looped is specified by its ID . The dataloop generator itself is assigned the ID .
on line [xml:filter2] the attribute tells APEX 4 that an amplifier has to be created. The gain of this amplifier will be varied to change the amplitude of the words and thus the SNR. It is assigned ID . The gain of the amplifier is made a variable parameter by assigning it ID
- The device with which the filter is associated
- The number of channels of the filter. The available number of channels is dependent on the type of filter. An amplifier can have any number of channels.
- If is set to , the filter or generator will keep on running in between two trials (i.e., when the subject is responding). In this example it stops when the stimulus stops playing ().
- The datablock with ID noisedata, specified under will be looped.
- the total gain of the amplifier is basegain+gain. Basegain cannot be a parameter, gain can be a parameter. The total gain of the complete output system should not be larger than 0 to avoid clipping of the signal. This is why basegain = -5.
- extent to which the signal is amplified.
- If is set to , when the dataloop is started, it will jump to a random sample in the datablock. Thereafter it is looped.

    <stimuli>
        <stimulus id="stimulus_star">
          <datablocks>
            <sequential>
              <datablock id="silence"/>
              <datablock id="datablock_star"/>
              <datablock id="silence"/>
            </sequential>
          </datablocks>
        </stimulus>
        <stimulus id="stimulus_fly">
          <datablocks>
            <sequential>
              <datablock id="silence"/>
              <datablock id="datablock_fly"/>
              <datablock id="silence"/>
            </sequential>
          </datablocks>
        </stimulus>
        <stimulus id="stimulus_mouse">
          <datablocks>
            <sequential>
              <datablock id="silence"/>
              <datablock id="datablock_mouse"/>
              <datablock id="silence"/>
            </sequential>
          </datablocks>
        </stimulus>
      </stimuli>

contains different , each with an ID.

can be combined in order (as opposed to .

This is repeated for all the stimuli in the experiment.

    <connections>
        <connection>
          <from>
            <id>_ALL_</id>
            <channel>0</channel>
          </from>
          <to>
            <id>amplifier</id>
            <channel>0</channel>
          </to>
        </connection>
        <connection>                (*@\label{xml:cha}@*)
          <from>
            <id>amplifier</id>
            <channel>0</channel>
          </from>
          <to>
            <id>wavdevice</id>
            <channel>0</channel>
          </to>
        </connection>               (*@\label{xml:chb}@*)
        <connection>                (*@\label{xml:chc}@*)
          <from>
            <id>noisegen</id>
            <channel>0</channel>
          </from>
          <to>
            <id>wavdevice</id>
            <channel>0</channel>
          </to>
        </connection>               (*@\label{xml:chd}@*)
      </connections>

defines how the different datablocks and filters are routed to the output device. The ID stands for all the datablocks. In this example they are routed to the first channel of the filter with ID amplifier (defined under ). In the amplifier the signal is amplified or attenuated and sent to the wavdevice on lines [xml:cha] to [xml:chb]. At the same time the noise, generated by a generator with ID noisegen, is sent to the same channel of the wavdevice. The channels are numbered from 0 onwards. The level of the noise remains constant and does not pass through an amplifier (lines [xml:chc] to [xml:chd]).

A visual representation of connections (see Figure [fig:ex1-output]) can be obtained by choosing “Show stimulus connections” under “Experiment”in the main APEX 4 menu (top left menu bar).

     <results>
      <page>apex:resultsviewer.html</page>
      <resultparameters/>
      <showduringexperiment>false</showduringexperiment>
      <showafterexperiment>true</showafterexperiment>
     </results>

Even if is not specified in the experiment file APEX 4 will deliver a results file in XML.

URL of the html page to show in the results window. The page should have the appropriate java script methods embedded. More example pages can be found in the APEX 4 resultsviewer folder.
Parameters to be passed to the results page. Each parameter will be set in hash params.
If true, an extra window will be created which will show the results of the current experiment while the experiment is being executed. Javascript embedded in the page will be executed upon each new trial.
If true, when the experiment is finished, a dialog box will appear querying whether results should be shown. If the answer is affirmative, a new window will be opened and the results will be shown after javascript processing.

    <interactive>
       <entry type="int" description="SNR in dB"
        expression="/apex:apex/filters/filter[@id='amplifier']/gain"
        default="0"/>
     </interactive>

If a small part of an experiment file has to be changed right before the start of an experiment (e.g. a start value, a gain value, the subject’s name), APEX 4 can show the experimenter a small containing the elements to be changed. This is accomplished by defining the element in the experiment file.

In this example we will modify the gain of the filter with ID to a value that is entered by the experimenter at the start of the experiment.

It is only possible to change the value of an existing element of the experiment file, elements cannot be added. For each element to be changed, an should be defined under . has four attributes that should be defined:

specifies the type of input element that will be shown. If it is , a spinbox¹ will be shown. If it is a plain text box will be shown. In this case a spinbox will be shown as a gain is always numeric.
defines the text to be shown in the next to the input element, such that the experimenter knows exactly what to fill in.
defines the element of the experiment file to be changed. It is specified by a so-called XPath expression ². For a description of XPath, we refer to the according standard or a good XML book.
specifies the default value to be shown in the input element.

Example 2: Identification of speech in noise using an adaptive method

General description of the experiment

See . This is an example of an adaptive speech in noise test. It determines the , the 50 percent correct point, using the 1-down, 1-up method described by @PM79. In this adaptive procedure the first sentence is repeated with increasing level until it is identified correctly. Subsequently, the SRT is determined by increasing or decreasing the level in steps of 2 dB, according to the last response. Other decision procedures (eg 2-down 1-up) can also be implemented using APEX 4. In this example the 5 sentences are scored on the basis of their keywords. The keywords are indicated in bold on the screen. The experimenter/clinician is seated in front of the screen and decides whether the subject has repeated the keywords (of the sentence) correctly, after which the correct or incorrect button is clicked(figure [fig:sentencenoise]). No feedback is provided. The starting level is given in signal-to noise ratio (the level of the noise remains the same, that of the signal varies). In this example speech and noise are routed to the same channel, i.e. one speaker or one earpiece of the headphone. The results are written to a results file.

Example of adaptive sentence in noise experiment

Conceptual

The experiment as described in the previous paragraph should first be translated to concepts understood by APEX 4. The main concepts in this example are datablock, stimulus, screen, procedure, a variable parameter (to change the gain) and fixed Parameter to show a sentence on the screen. For each sentence a datablock is defined, and for each resulting datablock a stimulus is defined. As always, everything that is defined, is assigned an ID, to be able to refer to it later on. In this example there are 5 stimuli with IDs , , , and . The procedure defines a number of trials. Recall that a trial is the combination of a screen (always the same in this example), a stimulus and a response.

As we are dealing with an adaptive procedure a fixed or variable parameter is adapted. In this example a variable parameter is adapted to change the gain of the sentence, and a fixed parameter is used to show a sentence on the screen. The screen also shows the signal-to-noise ratio (SNR) under test and the response alternatives “correct” and “incorrect”.

Now only the output logic remains to be defined. The idea is to continuously present a noise signal and to vary the level of the sentences adaptively. To achieve this, we define 2 filters, the first one is a generator (i.e., a filter without input channels) that will generate the noise signal. The other one is an amplifier, which will amplify or attenuate the sentences to obtain the desired SNR. Both are connected to one channel of the wavdevice.

In the following sections each of the elements of the XML file necessary to implement the latter concepts will be described in detail.

Detailed description of various elements

    <procedure xsi:type="apex:adaptiveProcedure">
      <parameters>
        <presentations>1</presentations>
        <order>sequential</order>
        <corrector xsi:type="apex:isequal"/>
        <nUp>1</nUp>
        <nDown>1</nDown>
        <adapt_parameter>gain</adapt_parameter>
        <start_value>0</start_value>
        <stop_after_type>presentations</stop_after_type>
        <stop_after>1</stop_after>
        <larger_is_easier>true</larger_is_easier>
        <repeat_first_until_correct>true</repeat_first_until_correct>
        <stepsizes><stepsize begin="0" size="2"/></stepsizes>
      </parameters>

      <trials>
        <trial id="trial_sentence1">
             <answer>button_correct</answer>
             <screen id="screen"/>
             <stimulus id="stimulus_sentence1"/>
        </trial>

        <trial id="trial_sentence2">
             <answer>button_correct</answer>
             <screen id="screen"/>
             <stimulus id="stimulus_sentence2"/>
        </trial>

        <trial id="trial_sentence3">
            <answer>button_correct</answer>
            <screen id="screen"/>
            <stimulus id="stimulus_sentence3"/>
        </trial>

        <trial id="trial_sentence4">
            <answer>button_correct</answer>
            <screen id="screen"/>
            <stimulus id="stimulus_sentence4"/>
        </trial>

        <trial id="trial_sentence5">
            <answer>button_correct</answer>
            <screen id="screen"/>
            <stimulus id="stimulus_sentence5"/>
        </trial>
      </trials>
    </procedure>

The attribute of refers to a procedure in which a parameter is changed according to the response of the subject. In this example the gain of the amplifier is adapted. The procedure selects the next trial from the trial list and it completes after every trial has been presented a certain number of times.

defines the behavior of the procedure
- every trial is presented once
- the trials are presented sequentially, in the order that is specified in the experiment file (if would be specified, they would be presented in random order).
- the corrector checks whether the response is correct or not. The attribute compares whether the two input values are exactly the same. In this example compares the answer specified under trial () and the ID corresponding to the picture that has been clicked (either or ).
- the level is increased after n (here 1) incorrect response(s); cf
- the level is decreased after n (here 1) correct response(s); cf
- to be adapted
- the experiment starts with gain=0 (input of user). The value here will be replaced by the entry value. Please refer to section [sec:Interactive] for more information.
- the experiment stops after a specified number of presentations of each trial is completed (it is also possible to stop after .
- the experiment stops after 1 presentation of each trial
- If , then larger values of the parameter are easier than smaller values. It is used to determine and .
- the first trial is repeated with increasing gain until it is identified correctly.
- from the beginning of the experiment (begin=0) the stepsize is 2 dB.
contains different elements to specify a trial. Once a trial is selected, will show the specified screen and send the stimulus to the device. Each trial is given an ID (arbitrary name), eg , such that it can be referred to later on or viewed in the results file. A trial must be defined for all the sentences of the experiment.
the correct answer, to be used by APEX 4 to determine whether the subject’s response is correct. In this example the experimenter will click on “correct” or “incorrect”. The result from the screen will be the ID of the element of the screen that was clicked ( or ).

    <screens>
            <reinforcement>
                <progressbar>true</progressbar>
                <feedback length="600">false</feedback>
            </reinforcement>

            <screen id="screen">
           <gridLayout height="2" width="1" id="main_layout" rowstretch="1,2">
                   <gridLayout width="4" height="4" columnstretch="1,4,2,2"
                   rowstretch="1,1,2,2" id="parameter_layout" row="1" col="1">

        <label id="snrlabel" row="1" col="1">
        <text>snr</text>
        </label>

        <parameterlabel id="snr" row="2" col="1">
          <parameter>gain</parameter>
        </parameterlabel>

        <label id="sentence" row="1" col="2">
        <text>sentence</text>
        </label>

        <parameterlabel id="sentence" row="2" col="2">
        <fontsize>12</fontsize>
        <parameter>sentence</parameter>
        </parameterlabel>

        </gridLayout>

        <gridLayout height="1" width="2" id="answer_layout" x="1" y="2">
        <button x="1" y="1" id="button_correct">
        <text>Correct</text>
         </button>
        <button x="2" y="1" id="button_wrong">
        <text>Incorrect</text>
        </button>

        </gridLayout>
        </gridLayout>
        <buttongroup id="buttongroup">
        <button id="button_correct"/>
        <button id="button_wrong"/>
        </buttongroup>
        <default_answer_element>buttongroup</default_answer_element>
        </screen>
        </screens>

contains element that is referred to in .

includes elements on progress and feedback
- If the value is a progress bar will be displayed in the right hand bottom corner of the screen that indicates the percentage of trials that have been completed.
- duration of the time after response (in msec) that APEX 4 waits before presenting the next trial. During this interval feedback can be displayed. In this example, no feedback (thumb up, thumb down) is given as the value is .
the screen has an ID by which it can be referred to in each trial. In this example the screen displays four blocks in the top left corner to indicate the SNR and the sentence. In addition, the labels “Correct” and “Incorrect” are displayed on the buttons at the bottom of the screen (cfr screenshot, figure…).
- places elements in an irregular grid. The screen is divided into sections according to the values. In this example there are two nested layouts. The main layout divides the screen in two main rows (height = 2, width = 1). The parameter layout divides the first of these rows in four rows and four columns, of which only the two first ones are filled. The answer layout divides the second main row in two columns, each with one button. The stretch factor for the columns is a list of integers, separated by comma’s. There should be as many integers as columns. The width of the columns is proportional to the numbers. With , and implies that the second row will be twice as wide as the first one. If , and and the second column will be four times as wide as the first and two times as wide as the 3rd and 4th. With the 3rd and 4th rows will be twice as wide as the 1st and 2nd.
- the labels on the left are fixed and display the “SNR” and “Sentence”. The button “Correct” and “Incorrect” are indicated at the bottom of the screen
- the blocks on the right display the values of the fixed parameters. Gain is the (variable) SNR level, and sentence is a fixed parameter, defined in
- can be specified in points
- defines a group of screen elements (those that are displayed on the screen). As many elements can be defined in a screen, APEX 4 has no way to know which element contains the subject’s response. Therefore, in the element is designated that contains the subject’s response. In the case of screen elements that are clicked in order to respond, the example is further complicated by the fact that we cannot specify just one of the elements (one button, one picture), but that the response rather comes from a group of elements. This is when a can be used to group together different screen elements.

    <datablocks>
      <uri_prefix>sentences/</uri_prefix>
      <datablock id="datablock_sentence1">
        <device>wavdevice</device>
        <uri>sent1.wav</uri>
      </datablock>
      <datablock id="datablock_sentence2">
        <device>wavdevice</device>
        <uri>sent2.wav</uri>
      </datablock>
      <datablock id="datablock_sentence3">
        <device>wavdevice</device>
        <uri>sent3.wav</uri>
      </datablock>
      <datablock id="datablock_sentence4">
        <device>wavdevice</device>
        <uri>sent4.wav</uri>
      </datablock>
      <datablock id="datablock_sentence5">
        <device>wavdevice</device>
        <uri>sent5.wav</uri>
      </datablock>
      <datablock id="noisedata">
        <device>wavdevice</device>
        <uri>noise.wav</uri>
      </datablock>
      <datablock id="silence">
        <device>wavdevice</device>
        <uri>silence:500</uri>
      </datablock>
    </datablocks>

contains a list of elements and a prefix.

a relative path is specified here (relative with respect to the experiment file). Since APEX 4 knows the location of the experiment file, only the folder containing the wave files and pictures must be specified. It is also possible to give the absolute path, starting at the root. There are 3 ways to specify a prefix: by directly specifying an absolute path, by directly specifying a path relative to the experiment file or by referring to a prefix stored in . Please refer to Using prefixes for more information.
for each wave file a datablock is made, with an ID.

In datablock the special syntax is demonstrated for creating a datablock containing only silence (i.e., zeros). This is done to put silence before and after the sentence, to prevent the speech and noise from starting at the same time. The length of the silence datablock is specified in ms after the prefix . It is added before the signal (in element ), not before the noise.
- the datablock is associated with the sound card with ID , as defined in the section.
- the URI is appended to the prefix defined above

    <devices>
        <device id="wavdevice" xsi:type="apex:wavDeviceType">
        <driver>portaudio</driver>
        <card>default</card>
        <channels>1</channels>
        <samplerate>44100</samplerate>
        </device>
    </devices>

All devices defined in the experiment file are grouped in .The ID is set to . As an ID is unique for an entire experiment file, it can be used later on to refer to this device (eg in datablocks).

the attribute tells APEX 4 that a sound card is used. The experiment only starts if all devices can be opened.
specifies the software driver to be used for sound output. If unsure, set it to .
specifies the name of the sound card to be used. The system default card can be used by specifying as a card name. Other card names can be defined in .
the sampling frequency of the sound card. Not all sampling rates are supported by all devices, and some drivers automatically convert the sampling rate. Check your sound card documentation. The sample rate of the sound card should correspond to the sampling rates of all used with it.

    <filters>
    <filter xsi:type="apex:dataloop" id="noisegen">
        <device>wavdevice</device>
        <channels>1</channels>
        <continuous>false</continuous>
        <datablock>noisedata</datablock>
        <basegain>-5</basegain>
        <gain>0</gain>
        <randomjump>true</randomjump>
        </filter>
    <filter xsi:type="apex:amplifier" id="amplifier" >
        <device>wavdevice</device>
        <channels>1</channels>
        <basegain>-5</basegain>
        <gain id="gain">0</gain>
    </filter>
    </filters>

contains individual elements, which specify a filter, or as a special case a generator (i.e., a filter without input channels).

The attribute tells APEX 4 that a dataloop generator has to be created. This is a generator that takes a datablock and loops it infinitely. The datablock to be looped is specified by its ID .
on line [xml:filter2] the attribute tells APEX 4 that an amplifier has to be created. The gain of this amplifier will be varied to change the amplitude of the words and thus the SNR. It is assigned ID . The gain of the amplifier is made a variable parameter by assigning it ID
- the device with which the filter is associated
- The number of channels of the filter. The available number of channels depends on the type of filter. An amplifier can have any number of channels.
- If set to the noise is presented during the speech, but not during the subject’s response.
- the datablock with ID noisedata, specified under will be looped.
- : the total gain of the amplifier is basegain+gain. The total gain of the complete output system should not be larger than 0 to avoid clipping of the signal. This is why basegain = -5.
- the gain value that is changed. E.g. if the targetamplitude is 65 dB SPL (see Calibration) the signal and noise have equal amplitude if gain = 0. Gain is changed by other modules.
- If is set to , when the dataloop is started, it will jump to a random sample in the datablock. Thereafter it is looped.

    <stimuli>
      <fixed_parameters>
        <parameter id="sentence"/>
      </fixed_parameters>

      <stimulus id="stimulus_sentence1">
        <datablocks>
        <sequential>
            <datablock id="silence"/>
            <datablock id="datablock_sentence1"/>
            <datablock id="silence"/>
        </sequential>
        </datablocks>

      <fixedParameters>
        <parameter id="sentence">This is <b> sentence </b> <b>one</b></parameter>
      </fixedParameters>
      </stimulus>

      <stimulus id="stimulus_sentence2">
        <datablocks>
        <sequential>
            <datablock id="silence"/>
            <datablock id="datablock_sentence2"/>
            <datablock id="silence"/>
        </sequential>
        </datablocks>

      <fixedParameters>
        <parameter id="sentence">This is <b> sentence </b> <b>two</b></parameter>
        </fixedParameters> </stimulus>

      <stimulus id="stimulus_sentence3">
        <datablocks>
        <sequential>
            <datablock id="silence"/>
            <datablock id="datablock_sentence3"/>
            <datablock id="silence"/>
        </sequential>
        </datablocks>

    <fixedParameters>
    <parameter id="sentence">This is <b> sentence </b> <b>three</b></parameter>
    </fixedParameters>
    </stimulus>

    <stimulus id="stimulus_sentence4">
        <datablocks>
        <sequential>
            <datablock id="silence"/>
            <datablock id="datablock_sentence4"/>
            <datablock id="silence"/>
        </sequential>
        </datablocks>

    <fixedParameters>
    <parameter id="sentence">This is <b> sentence </b> <b>four</b></parameter>
    </fixedParameters>
    </stimulus>

    <stimulus id="stimulus_sentence5">
        <datablocks>
        <sequential>
            <datablock id="silence"/>
            <datablock id="datablock_sentence5"/>
            <datablock id="silence"/>
        </sequential>
        </datablocks>

       <fixedParameters> <parameter id="sentence">This is <b> sentence
       </b> <b>five</b> </parameter> </fixedParameters> </stimulus>
    </stimuli>

contains different , each with an ID.

is used here to be able to show the sentence under test on the Screen. Fixed parameters are discussed in section [sec:Parameters].
- the fixed parameter is defined here and should also be defined in each
Each stimulus gets an ID (referred to in )
- here one refers to the described before; it includes the sentence file.
  - The sequence of is indicated here.
- sets the fixed parameter for each .
- indicates which words (the keywords) should appear in bold on screen.

This is repeated for all the sentences.

    <connections>
        <connection>
            <from>
            <id>_ALL_</id>
            <channel>0</channel>
            </from>
            <to>
            <id>amplifier</id>
            <channel>0</channel>
            </to>
        </connection>
        <connection>                (*@\label{xml:ch1}@*)
            <from>
            <id>amplifier</id>
            <channel>0</channel>
            </from>
            <to>
            <id>wavdevice</id>
            <channel>0</channel>
            </to>
        </connection>               (*@\label{xml:ch2}@*)
        <connection>                (*@\label{xml:ch3}@*)
            <from>
            <id>noisegen</id>
            <channel>0</channel>
            </from>
            <to>
            <id>wavdevice</id>
            <channel>0</channel>
            </to>
            </connection>           (*@\label{xml:ch4}@*)
    </connections>

defines how the different datablocks and filters are routed to the output device. The ID stands for all the datablocks. In this example they are routed to the first channel of the filter with ID amplifier (defined under ). In the amplifier the signal is amplified or attenuated and sent to the wavdevice on lines [xml:ch1] to [xml:ch2]. At the same time the noise, generated by a generator with ID noisegen, is sent to the same channel of the wavdevice. The channels are numbered from 0 onwards. The level of the noise remains constant and does not pass through an amplifier (lines [xml:ch3] to [xml:ch4]). Although noisedata is a and connected to amplifier it is not specified in and does not pass through amplifier.

A visual representation of connections can be obtained by choosing “Show stimulus connections” under “Experiment”in the main APEX 4 menu (top left menu bar).

     <results>
      <page>apex:resultsviewer.html</page>
      <resultparameters>
       <parameter name="reversals for mean">4</parameter>
      </resultparameters>
      <showduringexperiment>false</showduringexperiment>
      <showafterexperiment>true</showafterexperiment>
     </results>

Even if is not specified in the experiment file APEX 4 will save a results file in XML.

URL of the html page to show in the results window. The page should have the appropriate java script methods embedded. More example pages can be found in the APEX 4 resultsviewer folder.
Parameters to be passed to the results page. Each parameter will be set in hash params. In this example, the result will be calculated as the mean of the variable parameter (gap) at the last 4 reversals.
If true, an extra window will be created which will show the results of the current experiment while the experiment is being executed. Javascript embedded in the page will be executed upon each new trial.
If true, when the experiment is finished, a dialog box will appear querying whether results should be shown. If the answer is affirmative, a new window will be opened and the results will be shown after javascript processing.

    <interactive> <entry type="int" description="SNR start value"
      expression="/apex:apex/procedure/parameters/start_value" default="4" />
    </interactive>

In this example we will modify the gain of the filter with ID to a value that is entered by the experimenter at the start of the experiment.

a value is entered, here corresponding to SNR in dB. It is also possible to enter a string, e.g. name of the subject (see Reference Manual).
the interactive window will show 4 as default value; more than one entry may be defined.

Example 3: Gap detection: determining the Just Noticeable Difference

General description of the experiment

See . This is an example of a gap detection task: two stimuli are presented to the listener in random order, a stationary noise and an interrupted noise. During the presentation the intervals are highlighted. The subject’s task is to indicate the interval with the interruption(figure [fig:gapdetection]). Feedback is provided (thumb up, thumb down) and the minimal detectable gap is determined by means of an adaptive procedure (here 2-down, 1-up). Stimuli (wave files) are generated off line. This example only includes 5 wave files. Usually many more are generated. The results are written to a text file.

Example of adaptive gap detection experiment

Conceptual

The experiment as described in the previous paragraph should first be translated to concepts understood by APEX 4. The main concepts used here are procedure, screens, datablocks and stimuli.

For each wave file (noise with or without a gap) a is defined, and for each datablock a is defined. In this example there are 5 wave files, with gap sizes ranging between 1 and 5 ms. Their IDs are , , , and . In an adaptive procedure either a fixed or variable parameter is defined. In this example a fixed parameter is used, i.e. the gap size in the stimulus. The wave files with different gap sizes are stored on disk, and are assigned a certain gap value. This gap value is used to determine the gap threshold. Also, a is defined that shows the two response intervals indicating “1” and “2”.

To indicate the order in which the stimuli should be presented a is defined. An adaptive procedure is defined containing 1 with five variable stimuli (with gap) and the standard stimulus without the gap. Recall that a is the combination of a (always the same in this example), a stimulus and a correct answer. The standard and stimuli occur randomly in one of both intervals.

Now only the output logic remains to be defined. The idea is to present the standard and a variable stimulus after each other, and the stimulus to be presented depends on the response of the subject. Filters are not used since the signals are not changed.

In the following sections each of the elements of the XML file necessary to implement the latter concepts will be described in detail.

Detailed description of various elements

    <procedure xsi:type="apex:adaptiveProcedure">
       <parameters>
         <presentations>1</presentations>
         <order>sequential</order>
         <intervals count="2">
    	<interval number="1" element="button1"/>
    	<interval number="2" element="button2"/>
    	</intervals>
         <pause_between_stimuli>1000</pause_between_stimuli>
         <nUp>1</nUp>
         <nDown>2</nDown>
         <adapt_parameter>gap</adapt_parameter>
         <start_value>5</start_value>
         <stop_after_type>reversals</stop_after_type>
         <stop_after>5</stop_after>
         <min_value>0</min_value>
         <max_value>5</max_value>
         <larger_is_easier>true</larger_is_easier>
         <stepsizes>
           <change_after>reversals</change_after>
           <stepsize  begin="0" size="2"/>
           <stepsize begin="2" size="1"/>
         </stepsizes>
      </parameters>

      <trials>
        <trial id="trial1" >
         <screen id="screen1" />
         <stimulus id="stimulus1" />
         <stimulus id="stimulus2"/>
         <stimulus id="stimulus3"/>
         <stimulus id="stimulus4"/>
         <stimulus id="stimulus5"/>
         <standard id="standard"/>
        </trial>
      </trials>
    </procedure>

The attribute refers to a procedure in which a parameter is changed according to the response of the subject. In this example 1 trial is defined with different stimuli from which the adaptive procedure chooses.

defines the behavior of the procedure (eg., nr of presentations, order of presentation, response strategy).
- every trial is presented once.
- applies to the order of the trials. Since there is only 1 , it does not matter here wether the order is specified as or .
- the corrector checks whether the response is correct or not. In this example the number of choices in was 2. This means that will present the target stimulus in either interval 1 or 2. informs about the correct interval. also receives the clicked button from the screen and looks up the corresponding number in the element above and compares it with the number it received from .
- number of alternatives to choose from (here: 2 intervals) and the IDs of the buttons (defined under they correspond to.
- a pause of 1000 ms will be introduced between successive wave files;
- number of items after incorrect trials to increase the gap
- number of items after correct trials to decrease the gap
- the gap size cannot be smaller than 0.
- the gap size cannot be larger than 5
- the parameter to change, here the “gap”, see also
- the gap at which to start, here 5, see under
- reversals (it can also stop after a number of presentations or trials)
- number of reversals after which the procedure is stopped
- the number of reversals that are used for the average threshold
- here (the smaller the gap, the more difficult the task)
- the stepsize
- the is changed after a specified number of reversals. In this example a stimulus is skipped until the second reversal (start at 5, then 3, etc). Thereafter no stimulus is skipped.
only 1 trial is defined
- the ID of the trial
- the ID of the screen
- the ID of the stimulus
- the ID of the standard

    <screens>
      <reinforcement>
        <progressbar>true</progressbar>
        <feedback length="300">true</feedback>
        <showcurrent>true</showcurrent>
      </reinforcement>

      <screen id="screen1" >
        <gridLayout height="1" width="2" id="main_layout">
          <button x="1" y="1" id="button1" >
           <text>1</text>
          </button>
          <button x="2" y="1" id="button2" >
           <text>2</text>
          </button>
        </gridLayout>
      <buttongroup id="buttongroup1">
         <button id="button1"/>
         <button id="button2"/>
      </buttongroup>
        <default_answer_element>buttongroup1</default_answer_element>
     </screen>
    </screens>

contains several screen elements that can be referred to elsewhere in the experiment file (e.g. in above).
includes elements on progress and feedback
- if a progress bar will be displayed in the right hand bottom corner of the screen. The progress bar can indicate the percentage of trials that have been done or it shows when a reversal occurs in an adaptive procedure. In the latter case the progress bar will increase at every reversal while the number of trials varies.
- duration of the time after response (in msec) that APEX 4 waits before presenting the next trial. During this interval feedback can be displayed.
- if the interval is highlighted while a signal is presented.
each screen has an ID by which it can be referred to elsewhere in the experiment. In this example the screen displays two intervals.
- places elements in an irregular grid. The screen is divided into sections according to the values. In this example there is an equal number of rows (x) and columns (y).
  
  Gridlayout defines a group of screen elements (those that are displayed on the screen).
  - for each interval a button is specified; this button (interval) displays a number on the screen
  - the left interval denotes “1”, the right one denotes “2”.
- defines a group of screen elements (those that are displayed on the screen). As many elements can be defined in a screen, APEX 4 has no way to know which element contains the subject’s response. Therefore, in the element is designated that contains the subject’s response. In the case of screen elements that are clicked in order to respond, the example is further complicated by the fact that we cannot specify just one of the elements (one button, one picture), but that the response rather comes from a group of elements. This is when a can be used to group together different screen elements.

    <datablocks>
      <uri_prefix>Gapfiles</uri_prefix>
        <datablock id="g5ms" >
          <device>wavdevice</device>
          <uri>g5.wav</uri>
        </datablock>

        <datablock id="g4ms" >
          <device>wavdevice</device>
          <uri>g4.wav</uri>
        </datablock>

        <datablock id="g3ms"  >
          <device>wavdevice</device>
          <uri>g3.wav</uri>
        </datablock>

        <datablock id="g2ms" >
          <device>wavdevice</device>
          <uri>g2.wav</uri>
        </datablock>

        <datablock id="g1ms"  >
          <device>wavdevice</device>
          <uri>g1.wav</uri>
        </datablock>

        <datablock id="datablockref">
          <device>wavdevice</device>
          <uri>ref500.wav</uri>
        </datablock>
      </datablocks>

contains a list of elements and a prefix.

: a relative path is specified here. Since APEX 4 knows the location of the experiment file, only the folder containing the wave files and pictures must be specified.
for each wave file a unique datablock is defined by an ID. The standard signal (uninterrupted noise) must also be specified here.
each datablock includes the audio device to which the signal is routed
since the path is defined in it is not necessary to specify the entire path again

    <devices>
     <device id="wavdevice" xsi:type="apex:wavDeviceType">
     <driver>portaudio</driver>
     <card>default</card>
     <channels>2</channels>
     <gain>0</gain>
     <samplerate>44100</samplerate>
     </device>
    </devices>

all devices defined in the experiment file are grouped in the . Only 1 is used in this example. The ID is set to soundcard. As an ID is unique for an entire experiment file, it can be used later on to refer to this device.

the attribute tells APEX 4 that a sound card is used. The experiment only starts if all devices can be opened.
specifies the software driver to be used for sound output. If unsure, set it to .
specifies the name of the sound card to be used. The system default card can be used by specifying as a card name. Other card names can be defined in ApexConfig.
2 channels are specified here, because the signal is stereo.
the sampling frequency of the wave files. Not all sampling rates are supported by all devices, and some drivers automatically convert the sampling rate. Check your sound card documentation. The sample rate of the sound card should correspond to the sampling rates of all datablocks used with it.

    <stimuli>
       <fixed_parameters>
         <parameter id="gap"/>
       </fixed_parameters>

       <stimulus id="stimulus1" >
         <description>noisewithgap1</description>
         <datablocks>
           <datablock id="g5ms" />
         </datablocks>
         <fixedParameters>
           <parameter id="gap">5</parameter>
         </fixedParameters>
       </stimulus>

       <stimulus id="stimulus2" >
         <description>noisewithgap2</description>
         <datablocks>
           <datablock id="g4ms" />
         </datablocks>
           <fixedParameters>
            <parameter id="gap">4</parameter>
         </fixedParameters>
       </stimulus>

       <stimulus id="stimulus3" >
         <description>noisewithgap3</description>
         <datablocks>
           <datablock id="g3ms" />
         </datablocks>
          <fixedParameters>
           <parameter id="gap">3</parameter>
          </fixedParameters>
       </stimulus>

       <stimulus id="stimulus4" >
         <description>noisewithgap4</description>
         <datablocks>
           <datablock id="g2ms" />
         </datablocks>
           <fixedParameters>
            <parameter id="gap">2</parameter>
           </fixedParameters>
       </stimulus>

       <stimulus id="stimulus5" >
         <description>noisewithgap5</description>
         <datablocks>
           <datablock id="g1ms" />
         </datablocks>
           <fixedParameters>
             <parameter id="gap">1</parameter>
           </fixedParameters>
       </stimulus>

       <stimulus id="standard">
         <datablocks>
          <datablock id="datablockref"/>
         </datablocks>
           <fixedParameters>
            <parameter id="gap">0</parameter>
         </fixedParameters>
       </stimulus>
    </stimuli>

defines the auditory events, e.g. noise with gap and noise without gap.

- the gap is a fixed parameter that is identified by an ID
this element includes a description of the (variable) stimulus
- - the ID refers to the wave file corresponding to the datablock
- - includes information on the size of the variable gap

    <connections>
       <connection>
         <from>
           <id>_ALL_</id>
           <channel>0</channel>
         </from>
         <to>
           <id>wavdevice</id>
           <channel>1</channel>
         </to>
       </connection>

     </connections>

defines how the different datablocks and filters are routed to the output device. The ID stands for all the datablocks. In this example they are routed to 1 channel of the wavdevice.

A visual representation of connections can be obtained by choosing “Show stimulus connections” under “Experiment”in the main APEX 4 menu (top left menu bar).

    <results>
      <page>apex:resultsviewer.html</page>
      <resultparameters>
       <parameter name="reversals for mean">3</parameter>
      </resultparameters>
      <showduringexperiment>false</showduringexperiment>
      <showafterexperiment>true</showafterexperiment>
     </results>

Even if is not specified in the experiment file APEX 4 will save a results file in XML. The results file will display the correct answers, the reversals, the entire sequence of responses, and the average threshold based on the number of reversals and the magnitude of the corresponding gap parameter specified in the experiment file.

URL of the html page to show in the results window. The page should have the appropriate java script methods embedded. More example pages can be found in the APEX 4 resultsviewer folder.
Parameters to be passed to the results page. Each parameter will be set in hash params.
If , an extra window will be created which will show the results of the current experiment while the experiment is being executed. Javascript embedded in the page will be executed upon each new trial.
If , when the experiment is finished, a dialog box will appear querying whether results should be shown. If the answer is affirmative, a new window will be opened and the results will be shown after javascript processing.

     <general>
      <exitafter>true</exitafter>
     </general>

defines some general parameters.

if APEX 4 will close after the experiment has ended and results have been shown.

Example 4: Gap detection in child mode

General description of the experiment

See . This is an example of a gap detection task adapted to the interest of children: two stimuli are presented to the listener in random order, a stationary noise and an interrupted noise. It is the same experiment as example 3, but pictures (movies) of cars replace buttons on the screen, and a smiley panel is shown (figure [fig:gapchild]). During the presentation of stimuli the cars are animated. The child’s task is to indicate the stimulus with the interruption. Feedback is provided by smileys and the minimal detectable gap is determined by means of an adaptive procedure (here 2-down, 1-up). Only those elements that are specific to the child mode are described in this example.

Example of gap detection in child mode

Conceptual

The main concepts illustrated in this example are childmode, and reinforcement. Child mode involves a different panel, and intro/outro movies.

Detailed description of various elements

    <screens>
        <uri_prefix>car/</uri_prefix>
        <general>
          <showpanel>true</showpanel>
        </general>
        <reinforcement>
          <progressbar>true</progressbar>
          <feedback length="300">true</feedback>
          <showcurrent>true</showcurrent>
        </reinforcement>
        <childmode>
          <panel>reinforcement.swf</panel>
        </childmode>
        <screen id="screen1">
          <gridLayout height="1" width="2" id="main_layout">
            <flash row="1" col="1" id="button1">
              <path>stillcar.swf</path>
              <feedback>
                <highlight>rijden.swf</highlight>
                <positive>goodcar.swf</positive>
                <negative>badcar.swf</negative>
              </feedback>
            </flash>

            <flash row="1" col="2" id="button2">
              <path>stillcar.swf</path>
              <feedback>
                <highlight>rijden.swf</highlight>
                <positive>goodcar.swf</positive>
                <negative>badcar.swf</negative>
              </feedback>
            </flash>
          </gridLayout>

          <buttongroup id="buttongroup1">
            <button id="button1"/>
            <button id="button2"/>
          </buttongroup>
          <default_answer_element>buttongroup1</default_answer_element>
        </screen>
      </screens>

a relative path is specified here (relative with respect to the experiment file). This only applies to the movies in . There are 3 ways to specify a prefix: by directly specifying an absolute path, by directly specifying a path relative to the experiment file or by referring to a prefix stored in . Please refer to Using prefixes for more information.
- if a panel will be shown with smileys (see figure presented above)
- If a progress bar will be displayed on the right hand side with smileys. The progress bar shows when a reversal occurs in an adaptive procedure (while the number of trials varies).
- duration of the time after response (in msec) that APEX 4 waits before presenting the next trial. During this interval feedback can be displayed
- the interval is highlighted while a signal is presented. In this childfriendly example, the car corresponding to the present interval is moving.
replaces the “standard” panel, sets some defaults, enables intro/outro movies and changes the background colour
- the name of the movie file, per frame (smileys and panel are a collection of frames)

Each screen has an ID by which it can be referred to elsewhere in the experiment. In this example the screen displays two movies, each containing a car.

places elements in an irregular grid. The screen is divided into sections according to the values. In this example there are two columns and one row.
- replaces of the standard version
  - the name of the flash movie file on disk, for the case without animation (still car).
  - - the name of the movie file indicating the present interval is active (here the moving car)
    - the name of the movie file following a correct response (here a monkey waving)
    - the name of the movie file following an incorrect response (the trunk of an elephant)
defines a group of screen elements (those that are displayed on the screen). As many elements can be defined in a screen, APEX 4 has no way to know which element contains the subject’s response. Therefore, in the element is designated that contains the subject’s response. In the case of screen elements that are clicked in order to respond, the example is further complicated by the fact that we cannot specify just one of the elements (buttons, pictures), but that the response rather comes from a group of elements. This is when a can be used to group together different screen elements.

A spinbox is an input field that only accepts numbers and has 2 buttons to respectively increase or decrease its value ↩
see http://www.w3.org/TR/xpath ↩